Polish Journal of Microbiology · Polish Journal of Microbiology with the following points: P04 Œ 5, P05 Œ 5, P06 Œ 6, T09 Œ 6, T12 Œ 6 POLISH SOCIETY OF MICROBIOLOGISTS 00-725

P O L S K I E T O W A R Z Y S T W O M I K R O B I O L O G Ó WP O L I S H S O C I E T Y O F M I C R O B I O L O G I S T S

Polish Journal of Microbiology

formerly

Acta Microbiologica Polonica

2005

POLSKIE TOWARZYSTWO MIKROBIOLOGÓW

EDITORS

K.I. Wolska (Editor in Chief)

J. Dziadek, A. Kraczkiewicz-Dowjat, A. Skorupska, H. DahmE.K. Jagusztyn-Krynicka (Scientific Secretary)

EDITORIAL BOARD

President: Zdzis³aw Markiewicz (Warsaw, Poland)

Ryszard Chróst (Warsaw, Poland), Waleria Hryniewicz (Warsaw, Poland), Miros³aw Kañtoch (Warsaw, Poland),Donovan Kelly (Warwick, UK), Tadeusz Lachowicz (Wroc³aw, Poland), Wanda Ma³ek (Lublin, Poland),Andrzej Piekarowicz (Warsaw, Poland), Anna Podhajska (Gdañsk, Poland), Gerhard Pulverer (Cologne, Germany),Geoffrey Schild (Potters, Bar, UK), Torkel Wadström (Lund, Sweden), Jadwiga Wild (Madison, USA),

Miros³awa W³odarczyk (Warsaw, Poland)

EDITORIAL OFFICE

Miecznikowa 1, 02-096 Warsaw, Polandtel. 48 (22) 55 41 302, Tuesday and Thursday from 10 A.M. � till 2 P.M. only

fax 48 (22) 55 41 402

e-mail izabelaw@ biol.uw.edu.pl

Archives of Acta Microbiologica Polonica, from 2004 Polish Journal of Microbiology online

www.microbiology.pl\pjm\ at PTM Journals online www.microbiology.pl

Visit the home page to browse contents, gallery, links page and instructions to authorsin HTML and PDF formats

Editorial correspondence should be addressed to Editors of Polish Journal of Microbiology02-096 Warsaw, Miecznikowa 1, Poland

Correspondence regarding subscription and spedition of Polish Journal of Microbiology should be addressed toNational Institute of Public Health, Division of Clinical Microbiology and Prevention of Infections

00-725 Warsaw, Che³mska 30/34tel. 48 (22) 841 33 67, fax 48 (22) 841 29 49, e-mail: cls.edu.pl

QUARTERLY OF POLISH SOCIETY OF MICROBIOLOGISTS,PUBLISHED WITH THE FINANCIAL SUPORT

OF THE STATE COMMITTEE OF SCIENTIFIC RESEARCH

The individual sections of the State Committee for Scientific Research have creditedPolish Journal of Microbiology with the following points: P04 � 5, P05 � 5, P06 � 6, T09 � 6, T12 � 6

POLISH SOCIETY OF MICROBIOLOGISTS00-725 Warsaw, Che³mska 30/34

Front cover: Long chain of Aspergillus sp. spores at the ends of the phialides(courtesy of Jaros³aw Wi�niewski, M.Sc. and Magdalena Sobolewska Ph.D)

Typesetting and print: Publishing House Letter QualityWarsaw, Brylowska 35/38, tel. 631 45 18, 607 217 879

Circulation: 500 + 15

CONTENTS

ORIGINAL PAPERS

Genetic diversity among Lactococcus sp. and Leuconostoc sp. strains using PCR-RFLP of insertion sequences ISS1-type,IS904, IS982WALCZAK P., KONOPACKA M., OTLEWSKA A. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183

Lysostaphin as a potential therapeutic agent for staphylococcal biofilm eradicationWALENCKA E., SADOWSKA B., RÓ¯ALSKA S., HRYNIEWICZ W., RÓ¯ALSKA B. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 191

gyrA mutations in ciprofloxacin-resistant clinical isolates of Pseudomonas aeruginosa in a Silesian hospital � in PolandWYDMUCH Z., SKOWRONEK � CIO£EK O., CHOLEWA K., MAZUREK U., PACHA J., KÊPA M., IDZIK B.,WOJTYCZKA R.D. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 201

The prognostic and diagnostic markers of invasive candidiasis in patients during chemotherapyMAGRY� A., KOZIO£ � MONTEWKA M., STAROS£AWSKA E., GABCZYÑSKA B. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 207

Seroepidemiological studies of Chlamydia pneumoniae infections in 1�36 months old children with respiratory trackinfections and other diseases in PolandPODSIAD£Y E., FR¥CKA B., SZMIGIELSKA A., TYLEWSKA-WIERZBANOWSKA S. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 215

Occurence of serum class immunoglobulins interacting with specific antigens of Helicobacter pylori in patientswith unstable coronary artery disease and in symptomless individualsRECHCIÑSKI T., GRÊBOWSKA A., KURPESA M., RUDNICKA W., KRZEMIÑSKA � PAKU£A M., CHMIELA M. . . . . . . 221

Enzyme production and biotypes of vaginal Candida albicansCYBULSKI Z., KRZEMIÑSKA � JA�KOWIAK E., MAJEWSKI P., CHYLAK J., PAWLIK M. . . . . . . . . . . . . . . . . . . . . . . . . . . 227

Purification and characterization of two extracellular lipases from Pseudomonas aeruginosa Ps-xSAEED H.M., ZAGHLOUL T.I., KHALIL A.I., ABDELBAETH M.T. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 233

Extraction of milk-clotting enzyme produced by solid state fermentation of Aspergillus oryzaeSHATA H.M.A. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 241

Cloning and preliminary characterization of a GATC-specific $2-class DNA:m6A methyltransferase encodedby transposon Tn 1549 from Enterococcus spp.RADLIÑSKA M., PIEKAROWICZ A., GALIMAND M., BUJNICKI J.M. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 249

Co-occurence of urogenital mycoplasmas and group B streptococci with chlamydial cervicitisFRIEDEK D., EKIEL A., ROMANIK M., CHELMICKI Z., WIECHU£A B., WILK I., JÓ¯WIAK J., MARTIROSIAN G. . . . . 253

INSTRUCTIONS TO AUTHORS 257

Polish Journal of Microbiologyformerly Acta Microbiologica Polonica

2005, Vol. 54, No 3

Polish Journal of Microbiology2005, Vol. 54, No 3, 183�189

Genetic Diversity among Lactococcus sp. and Leuconostoc sp.Strains Using PCR-RFLP

of Insertion Sequences ISS1-type, IS904, and IS982

PIOTR WALCZAK*, MIROS£AWA KONOPACKA and ANNA OTLEWSKA

Institute of Fermentation Technology and Microbiology, Technical University of £ód�, £ód�, Poland

Received 3 February 2005, received in revised form 23 June 2005, accepted 24 June 2005

A b s t r a c t

PCR-RFLP analysis of commonly occurring insertion sequences ISS1-type, IS904 and IS982 in Lactococcus sp. andLeuconostoc sp. was used for the genetic differentiation of 17 strains of lactic acid bacteria. ISS1-type and IS982 werefound in all analysed strains while IS904 was present exclusively in strains belonging to Lactococcus sp. Amplificationof ISS1-type IS sequences resulted in formation of about 820 bp long amplicons, except of strains Lactococcus lactisssp. lactis E and Leuconostoc lactis R where extra DNA bands about 370 bp long were observed. Similarly for strains ofLeuconostoc lactis M and N, additional DNA bands about 280 bp long were present. TaqI digestion of ISS1-typeamplicons revealed that all analysed sequences belonged to the restriction type (ii) or (iii) for which major restrictionproducts were 543 and 147 bp long. Amplification of IS904 from all strains of Lactococcus sp. generated ampliconsabout 1260 bp long. In three strains of Leuconostoc sp. M, N and R, shorter amplicons about 880 bp were observedwhereas strains O and P did not contained IS904. Amplification of IS982 resulted in formation of amplicons about1000 bp long and no extra bands were observed for all tested strains. TaqI digestion of amplification products showedthat for strains C, I and F, G, H, belonging to Lactococcus sp. smaller DNA bands were visible suggesting that theycontain two different types of IS982.

K e y w o r d s: PCR-RFLP, insertion sequences, Lactococcus sp., Leuconostoc sp.

Introduction

Published data for genomes of Lactic Acid Bacteria (LAB) shows that insertion sequences are verycommon movable genetic elements present in the chromosome of bacteria as well as in plasmids. Commonsituation in LAB is that single strain carries from 3 to 8 different IS sequences, and each one can be presenteven in several copies (Klaenhammer et al., 2003). They contribute to genetic diversity of Lactic AcidBacteria due to the conservative and replicative transpositions within the chromosome, causing negativeor positive changes of bacterial phenotype. Bongers et al. (2003) observed that transposition of IS981,recovered activity of lactate production in a ldhB deficient strain of Lactococcus lactis by activation oftranscription of that gene. Transposition of ISS1 into chromosome of Lactococcus lactis resulted in muta-tion manifested with increased sensitivity of mutants to UV light (Duwat et al., 1997). Presence of at leasttwo identical IS sequences in bacterial chromosome can lead to large inversions of chromosomal DNAcaused by homologous recombination between them (Daveran-Mingot et al., 1998). Activation or silencingof genes by spontaneous transposition of insertion sequences within bacterial chromosome may serve asa mechanism of strain adaptation to environmental changes (Duval-Valentin et al., 2001). The presence of ISsequences can also contribute to the genetic stability of particular strain since it may generate spontaneousmutations in growing bacterial population. Stability of biochemical profiles of LAB, especially lactosefermentation and protease activity, is very important in dairy industry where bacterial starters are used forthe preparation of dairy products. IS mediated transposition leading to the induction of prophage excission

* Corresponding author. Mailing address: Institute of Fermentation Technology and Microbiology, Technical University of£ód�, Wólczañska 171/173, 90-530 £ód�, Poland. E-mail address: [email protected]

184 Walczak P. et al. 3

can be even more dangerous since it can stop fermentation process. ISS1-type insertion sequences belongingto IS6 family such as IS946, ISS1, iso-ISS1, IS1216, are usually associated with plasmids replicatingvia theta mechanism and therefore they are widely distributed among different bacterial genera, due to theplasmid exchange caused by conjugal transfer or plasmid uptake by competent cells (Romero andKlaenhammer, 1990; Schaefer et al., 1991). Insertion sequences of ISS1-type can be found in large plasmidsof Bacillus thuringiensis (pBtoxis) (Berry et al., 2002), Listeria inocua (pLI100) (Glaser et al., 2001), aswell as in Lactococcus lactis (pMRC01, pK214, pIL105) (Dougherty et al., 1998, Teuber et al., 1999, Anbaet al., 1995). Plasmids containing ISS1-type insertion sequences can transpose to bacterial chromosome andform cointegrates. The resolution of cointegrate leads to the excision of plasmid containing ISS1 and thesecond copy of IS is left in the chromosome. This feature was used for the construction of vectorsfor insertional mutagenesis of pGh9:ISS1 type (Maguin et al., 1996). The second very common insertionsequence usually present in the chromosome or plasmid DNA of LAB is IS904 belonging to the IS3 family.Its iso- forms are IS1076 (L, R) (Huang et al., 1990), IS1069 (Rauch et al., 1994) iso-IS904 (Rauch, 1990).In the chromosome locations of Lactococcus lactis IL1403, this sequence accompanies another one IS1077(Bolotin et al., 2001). IS904 can also be found in lactococcal plasmids such as pK214, pNZ4000 (Teuberet al., 1999; van Kranenburg and de Vos 1998; van Kranenburg et al., 1999; 2000) as well as in nisinsucrose conjugative lactococcal transposons Tn5276 and Tn5301 (Rauch and De Vos, 1992; Rauch et al.,1994; Horn et al., 1991). Transposons Tn5276 and Tn5301 belonging to the large conjugative transposons ofLAB, which contains IS904 and genes responsible for nisin A or Z biosynthesis, sucrose metabolism, biosyn-thesis of N5-(L-1-carboxyethyl)-L-ornithine (gene ceo) as well as genes coding excision/integration system(Xis/Int) of the transposon (Dodd et al., 1982; 1990). Chromosome of the Lactococcus lactis IL1403 con-tains only one copy of IS982 which transposase is inactive due to the mutation prematurely terminatingtranscription of this gene (Bolotin et al., 2001). IS982 sequence belonging to the separate family oftenresides on plasmid DNA such as pNZ4000 and pCI658 and is associated with eps gene cluster responsiblefor production of extracellular polysaccharides (van Kranenburg et al., 1997; Forde and Fitzgerald, 2003).However pCI658 contains iso-IS982 which is about 50 bp shorter than the normal sequence and most likelyits transposase is also inactive. Lactococcal citrate plasmid such as pKR223 from strain of Lactococcuslactis ssp.lactis var diacetylactis KR2 also contained IS982 sequence closely linked with restriction modifi-cation gene cluster (Twomey et al., 1998; 2000). Also citrate utilization genes found in other strain ofLactococcus lactis ssp. lactis var. diacetylactis CRL264 were located closely to the IS982 sequence (Lopezde Felipe et al., 1995). Knowledge about presence of particular insertion sequences and their stability in thechromosome of LAB is therefore of vital interest from the point of view of industrial application of certainstrains. Distribution of IS sequences in strains of LAB may also contribute to better characterization of themand elimination of duplicates from strain collections.

Experimental

Materials and Methods

Bacterial strains used in this study. Bacterial strains used in this study were originated from the collection of the Institute ofFermentation Technology and Microbiology, Technical University of Lodz or were isolated from kefir grains of polish origin (Table I).

DNA preparation. Plasmid and chromosomal DNA was isolated according to the method described by Anderson and McKay(1983).

PCR amplification of IS sequences. Primer sequences for amplification of ISS1-type, IS904 and IS982 were derived fromdata records published in NCBI Database (Table II). Underlined parts of oligonucleotides were identical to the beginning fragmentsof inverted repeats of corresponding IS sequences. In case where IRL and IRR sequences of IS were identical, only one primer wassufficient for the PCR procedure. Primer fragments located upstream of IR, marked in boldface, containing restriction sites forBamHI or EcoRI, were introduced additionally to facilitate possible cloning of PCR products. Amplification of ISS1 type ISsequences was performed in the following manner. About 20 ng of DNA template, 40 pmol of primer ISS1FR, 12.5 :l Red-TaqReadyMix DNA polymerase (Sigma-Aldrich) were mixed together and supplemented with PCR grade water to a total volume of25 :l. The amplification procedure consisted of one cycle of 2 min at 94°C, followed by 34 cycles for 1 min at 94°C, 1 min at 45°Cand 3 min at 72°C with final extension cycle for 2 min at 72°C was performed using Uno II thermocycler, Biometra, with tube lidheating block set for 105°C. No overlay oil was added to the tubes. The reaction mix for amplification of IS904 was the same exceptof primers replaced by IS904F and IS904R in the concentration of 20 pmol each. Similarly for amplification of IS982, 40 pmolof primer IS982FR was used. The amplification procedure for IS904 and IS982 consisted of one cycle of 3 min at 94oC, followed by29 cycles for 1 min at 94°C, 1 min at 40°C and 3min at 70°C with final extension cycle for 3 min at 70°C.

Agarose gel analysis. PCR products of IS amplification were analysed on 1% [w/v] agarose gel in TBE buffer. Amplified PCRproducts of IS sequences in amount of 10 :l were digested by TaqI (MBI Fermentas) for 1 h at 65°C according to the product

185Lactococcus sp., Leuconostoc sp. differentiation by PCR-RFLP of IS3

instruction. Restriction fragments were analysed on 2% [w/v] agarose gel in TBE buffer. Gells placed on UV transiluminator werephotographed with digital camera through yellow filter and obtained pictures were electronically inverted for better visualisationof separated DNA bands.

Results

Insertion sequences ISS1, iso-ISS1 and IS946 are very common in Lactic Acid Bacteria as well as inother gram positive organisms including Listeria, Staphylococcus, and Bacillus sp. Computer analysis ofpublished sequences of ISS1 type, revealed their genetic diversity in respect to size (807�809 bp), nucleotidesequence of inverted repeats IR and transposase gene tnpA as well as presence of restriction sites for dif-ferent enzymes. Insertion sequences ISS1 and IS946 contain 1, 3 or 4 restriction sites for TaqI and thisenzyme generates fragments with the following length: (i) � 661 and 147 bp; (ii) � 543, 147, 72 and 48 bp;(iii) � 543, 147, 48, 42 and 30 bp. Restriction enzyme EcoRV may cut some of ISS1 type molecules in onesite generating fragments of 377 and 431 bp whereas other types are not cut. Therefore, RFLP analysis ofISS1 type insertion sequences digested with one of these enzymes may be a simple method of their differen-tiation. Figure 1A shows results of agarose gel electrophoresis of PCR amplified ISS1-type insertionsequences from all tested strains of Lactococcus sp. and Leuconostoc sp. In one strain of Lactococcus lactisssp. lactis (E) and in three strains of Leuconostoc sp. (M, N, R), normal size of ISS1-type amplicons as wellas different shortened amplicons were observed. Size comparison of shortened amplicons revealed thatstrain Lactococcus lactis ssp. lactis (E) and Leuconostoc lactis (R) possessed extra amplicon of the samesize (about 370 bp), whereas strains of Leuconostoc lactis (M) and (N) contained shorter fragments of the

ISS1-type ISS1FR 5�-GCGGATCCGGTTCTGTTGCAAAGTTT-3� X62737, AF116286, AF036485

IS904 IS904F 5�-GCGGATCCTGGAAAGGTTATAATAAA-3� M27276, X52273,

IS904R 5�-CGAATTCTGGAAAGTCAACGAAAAA-3� X78469, X92946

IS982 IS982FR 5�-GCGGATCCAWACCCGAATTGCTAGTT-3� L34754, AF036485

Table IIPrimers used for PCR amplification of IS sequences

IS sequence Name Primer Sequence source

A Lactococcus lactis ssp. lactis var. diacetylactis Industrial strain / lac+, cit+

B Lactococcus lactis ssp. lactis var. diacetylactis Industrial strain / lac+, cit+

C Lactococcus lactis ssp. lactis Industrial strain / lac+, cit�

D Lactococcus lactis ssp. lactis Industrial strain / lac+, cit�

E Lactococcus lactis ssp. lactis Industrial strain / lac+, cit�

F Lactococcus lactis ssp. lactis Derivative of ATCC 11454

G Lactococcus lactis ssp. lactis Kefir isolate / lac+, cit�

H Lactococcus lactis ssp. lactis Kefir isolate / lac+, cit�

I Lactococcus lactis ssp. lactis Kefir isolate / lac+, cit�

J Lactococcus lactis ssp. lactis Kefir isolate lac+, cit�

K Lactococcus lactis ssp. lactis Kefir isolate / lac+

L Lactococcus lactis ssp.lactis Kefir isolate / lac+

M Leuconostoc lactis Industrial strain / lac+, cit+

N Leuconostoc lactis Industrial strain / lac+, cit+

O Leuconostoc mesenteroides Kefir isolate / lac+

P Leuconostoc mesenteroides Kefir isolate / lac+

R Leuconostoc lactis Kefir isolate / lac+, cit+

Table IBacterial strains and their properties

Strain symbol Strain name Description / genotype

Note. lac (+/�) � lactose fermentation, cit (+/�) citrate fermentation


same length being about 280 bp. Size of short amplicons were calculated form the separate agarose gelelectrophoresis experiment carried out in the presence of molecular size marker (BTL cat. No G004, pUC19digested with RsaI, HinfI and PvuII). Presence of such shorter than normal ISS1-type sequence in a particularstrain may be a unique marker differentiating it from other strains. However the nature and possible func-tion of shorten amplicons is not known. They can represent truncated forms of ISS1 type sequencestranspositionaly not functional due to the deletion of tnpA gene. This view strongly supports finding fromthe DNA sequence (Accession Number Z98171) published by Bourgoin (2002) for exopolysaccharide synthe-sis genes and insertion sequences of Streptococcus thermophilus. In the variable region of exopolysaccharidesynthesis genes, three ISS1-type sequences (ISS1SA, ISS1SB and ISS1SC) were found of which ISS1SB wastruncated form containing 62 first nucleotides and 270 last nucleotides of normal type ISS1 sequence. Thistruncated form being 332 bp long was named delta-ISS1SB. Figure 1B shows results of electrophoreticanalysis of TaqI digested amplification products. RFLP analysis revealed presence of two major DNA bandswith size of 550 and 150 bp which is very close to the expected values of 543 and 147 bp characteristic tothe (ii) or (iii) restriction pattern. Fragments smaller than 100 bp were not visible. All analyzed IS sequencesbelonged to the restriction type (ii) or (iii) according to the proposed classification, for which the largestfragment had 543 bp. DNA bands with size between 543 and 147 bp visible in lanes B, C, E, I, J, L, M,N and R were most likely originated from digestion of �upper� amplicons visible above ISS1 bands or fromshorten amplicons. It is also possible that those bands, are results of presence of two kinds of IS S1-typeIS sequences in the same strain. Such situation was demonstrated by Bourgoin et al. (1996) for strain ofStreptococcus thermophilus.

Insertion sequence IS904 formerly named as IS1069 or IS1076 is very common in the chromosome ofmany lactococcal strains (7 copies in Lactococcus lactis IL1403), (Bolotin et al., 2002). Figure 2A showsresults of agarose gel electrophoresis of PCR amplified IS904 insertion sequences from all tested strains.IS904 was present in all strains belonging to the genus Lactococcus and the size of obtained PCR ampliconswere the same, being about 1260 bp. In two strains of Leuconostoc sp. (O, P) IS904 was not present and inthe remaining ones (M, N, R), amplified sequence was shorter than that from strains of Lactococcus sp.Figure 2B shows results of electrophoretic analysis of TaqI digested amplification products. RFLP analysisof TaqI digestion products from strains of Lactococcus sp. showed their size of about 736 and 477 bp.Computer modeled restriction analysis of 5 published IS904 sequences or its iso- forms digested with TaqI,

Fig. 1. A � Agarose gel electrophoresis of amplification products with primer ISS1FR specific for ISS1 type IS sequences.B � Agarose gel electrophoresis of TaqI digested amplification products of ISS1-type IS sequences.

A, B, C, D, E, F, G, H, I, J, K, L � strains of Lactococcus sp., M, N, O, P, R � strains of Leuconostoc sp.


revealed presence of two recognition sites for that enzyme located at positions 75 and 797 of 1245 bp longentire IS sequence. Taking into account extra nucleotides present in primers IS904F and IS904R, TaqIrestriction fragments of PCR amplified sequences should have the following length 722, 456 and 82 bp.However the smallest fragment can be hardly visible when digestion products are analysed on 2% agarose gel.TaqI digestion of PCR products from strains of Leuconostoc sp. (M, N and R) were shorter than for originalIS904 sequence (about 550 and 330 bp). This results indicate that analysed strains of Leuconostoc sp. pos-sessed most probably a different type of an insertion sequence with front parts of inverted repeats having thesame oligonucleotide sequence. The length of the new sequence was estimated to be about 300�400 bpshorter than IS904. Question whether it is a new sequence or only deletion form of IS904 remains stillopened. The lack of IS904 in strains of Leuconostoc sp. or the presence of its shorter form, may be diagnosticfeature used for differentiation of Leuconostoc sp. and Lactococcus sp. associated with milk products.

Insertion sequences IS982 has been found in both already sequenced chromosomes of Lactococcus lactisstrains IL1403 and MG 1363 (Klaenhammer et al., 2002). Also Lactococcus lactis ssp. cremoris SK11contains this sequence (Yu et al., 1995). Nothing is known about presence of IS982 sequence in strains of

Fig. 2. A � Agarose gel electrophoresis of amplification products with primer IS904F and IS904R specific for IS904 sequences.B � Agarose gel electrophoresis of TaqI digested amplification products of IS904 sequences.


Fig. 3. A � Agarose gel electrophoresis of amplification products with primer IS982FR specific for IS982 sequences.B � Agarose gel electrophoresis of TaqI digested amplification products of IS982 IS sequences.



Leuconostoc sp. The typical length of this sequence varies from about 950 to 1011bp. Published sequencescontain a single restriction site for EcoRV and are not digested with TaqI enzyme. Figure 3A shows results ofagarose gel electrophoresis of PCR amplified IS982 insertion sequences from all tested strains of Lactococcus sp.and Leuconostoc sp. This sequence was present in all strains and obtained amplicons were uniform withrespect to their size. No extra bands were observed. Figure 3B shows results of electrophoretic analysis ofTaqI digested amplification products. RFLP analysis showed in some strains of Lactococcus sp. (C, I) and(F, G, H,) that obtained amplicons were partially digested with TaqI enzyme, what may suggest the presenceof two kinds of that sequence in one strain. Strains (C) and (I) had similar restriction pattern and strains (F),(G), (H) were characterized by the other type of restriction pattern. So far IS982 sequences published in thedata bases do not contain TaqI recognition sites and therefore detected paralogs are still waiting for therecovery and sequencing.

Discussion

Detection of insertion sequences ISS1-type, IS904, and IS982 in the chromosome and plasmid DNA ofLAB combined with RFLP analysis of TaqI digests of amplified sequences can be a valuable tool for analy-sis of genetic differences among closely related strains. Structural diversity of ISS1-type insertion sequencesmay be used for grouping strains according to their RFLP profile. The presence of certain type of ISS1sequences as well as their degenerated shortened forms in particular strain is often associated with thespecific plasmid which may function in strains of Lactococcus sp. (E) as well as in Leuconostoc sp. (R).Absence of IS904 or its presence in shorter form in strains of Leuconostoc sp. may be a method of straindifferentiation and allows to distinguish them from strains of Lactococcus sp. RFLP analysis of TaqI diges-tion of IS982 amplification product revealed the existence in certain strains of Lactococcus lactis, the newtype of such element which DNA is digested with this enzyme whereas, already known sequences, are not.It seems possible that such TaqI digested IS982 sequences are associated with the ability of strains to syn-thesize lantibiotic bacteriocin nisin. This view supports observation that strain of Lactococcus lactis ssp.lactis (H) having the same restriction pattern for TaqI digested IS982 as known nisin producer L. lactisATCC 11454 (F) was also able to produce bacteriocin (data not published). However, the nature and struc-ture of bacteriocin produced by strain Lactococcus lactis ssp. lactis (H) remains unknown. Also strain (G)represented the same TaqI restriction pattern of IS982 as (F) and (H) and therefore most probably it belongsto the same group. It seems possible that TaqI digested IS982 is therefore characteristic of nisin producingstrains having nisin-sucrose transposons such as Tn5276 and Tn5301 carrying nisin biosynthesis operons.Similarity between strains (F), (G) and (H) and fact that last two of them were isolated from the kefir grainssuggests that kefir microflora may produce nisin contributing to the probiotic properties of kefir. Howeverthese suggestions has to be confirmed experimentally.

Literature

A n b a J., E. B i d n e n k o, A. H i l l i e r, D. Ehr l ich and M.C. C h o p i n. 1995. Characterization of the lactococcal abiD1gene coding for phage abortive infection. J. Bacteriol. 177: 3818�3823.

A n d e r s o n D.G. and L.L. M c K a y. 1983. Simple and rapid method for isolating large plasmid DNA from Lactic Streptococci.Appl. Environ. Microbiol. 46: 549�552.

B e r r y C., S. O� N e i l, E. B e n - D o v, A.F. J o n e s, L. M u r p h y, M.A. Q u a i l, M.T.G. H o l d e n, D. H a r r i s,A. Z a r i t s k y and J. P a r k h i l l. 2002. Complete Sequence and Organization of pBtoxis, the Toxin-Coding Plasmid ofBacillus thuringiensis subsp. Israelensis. Appl. Environ. Microbiol. 68: 5082�5095.

B o l o t i n A., P. W i n c k e r, S. M a u g e r, O. J a i l l o n, K. M a l a r m e, J. W e i s s e n b a c h, S.D. E h r l i c h andA. S o r o k i n. 2001. The complete genome sequence of the lactic acid bacterium Lactococcus lactis ssp. lactis IL1403.Genome Res. 11: 731�753.

B o n g e r s R.S., M.H.N. H o e f n a g e l, M.J.C. S t a r r e n b u r g, M.A.J. S i e m e r i n k, J.G.A. A r e n d s, J. H u g e n h o l t zand M. K l e e r e b e z e m. 2003. IS981-Mediated Adaptive Evolution Recovers Lactate Production by ldhB TranscriptionActivation in a Lactate Dehydrogenase-Deficient Strain of Lactococcus lactis. J. Bacteriol. 185: 4499�4507.

B o u r g o i n F., G. G u e d o n, M. P e b a y, Y. R o u s s e l, C. P a n i s and B. D e c a r i s. 1996. Characterization of a mosaicISS1 element and evidence for the recent horizontal transfer of two different types of ISS1 between Streptococcusthermophilus and Lactococcus lactis. Gene 178: 15�23.

B o u r g o i n F. 2002. Streptococcus thermophilus variable locus including exopolysaccharide synthesis genes and insertionsequences. NCBI Accession Z98171.


D a v e r a n - M i n g o t M.-L., N. C a m p o, P. R i t z e n t h a l e r and P. L e B o u r g e o i s. 1998. A natural Large Chromo-somal Inversion in Lactococcus lactis Is Mediated by Homologous Recombination between Two Insertion Sequences.J. Bacteriol. 180: 4834�4842.

D o d d H.M., N. H o r n and M.J. G a s s o n. 1990. Analysis of the genetic determinant for production of the peptide antibioticnisin. J. Gen. Microbiol. 136: 555�566.

D o d d H.M., N. H o r n and M.J. G a s s o n. 1982. A lactococcal expression system for engineered nisins. Appl. Environ.Microbiol. 58: 3683�3693.

D u v a l - Va l e n t i n G., Ch. N o r m a n d, V. K h e m i c i, B. M a r t y and M. C h a n d l e r. 2001. Transient promoter forma-tion: a new feedback mechanism for regulation of IS911 transposition. The EMBO J. 20: 5802�5811.

D o u g h e r t y B.A., C. H i l l, J.F. W e i d m a n, D.R. R i c h a r d s o n, J.C. Ve n t e r and R.P. R o s s. 1998. Sequence andanalysis of the 60 kb conjugative, bacteriocin-producing plasmid pMRC01 from Lactococcus lactis DPC3147. Mol.Microbiol. 29: 1029�1038.

D u w a t P., A. C o c h u, S.D. E h r l i c h and A. G r u s. 1997. Characterization of Lactococcus lactis UV-Sensitive MutantsObtained by ISS1 Transposition. J. Bacteriol. 179: 4473�4479.

F o r d e A. and G.F. F i t z g e r a l d. 2003. Molecular organization of exopolysaccharide (EPS) encoding genes on the lactococcalbacteriophage adsorption blocking plasmid pCI658. Plasmid 49: 130�142.

G l a s e r P., L. F r a n g e u l, C. B u c h r i e s e r, A. A m e n d, F. B a q u e r o, P. B e r c h e, H. B l o e c k e r, P. B r a n d t,T. C h a k r a b o r t y, A. C h a r b i t, F. C h e t o u a n i, E. C o u v e, A. d e D a r u v a r, P. D e h o u x, E. D o m a n n,G. D o m i n g u e z - B e r n a l, E. D u c h a u d, L. D u r a n d, O. D u s s u r g e t, K.-D. E n t i a n, H. F s i h i, F. G a r c i a -D e l P o r t i l l o, P. G a r r i d o, L. G a u t i e r, W. G o e b e l, N. G o m e z - L o p e z, T. H a i n, J. H a u f, D. J a c k s o n,L.-M. J o n e s, U. K a r s t, J. K r e f t, M. K u h n, F. K u n s t, G. K u r a p k a t, E. M a d u e n o, A. M a i t o u r n a m,J. M a t a V i c e n t e, E. N g, G. N o r d s i e k, S. N o v e l l a, B. d e P a b l o s, J.-C. P e r e z - D i a z, B. R e m m e l,M. R o s e, C. R u s n i o k, T. S c h l u e t e r, N. S i m o e s, A. T i e r r e z, J.-A. Va z q u e z - B o l a n d, H. Vo s s,J. W e h l a n d and P. C o s s a r t. 2001. From the pathogenic to the innocuous: comparison of the Listeria monocytogenesand the Listeria innocua genomes. NCBI Accession NC_003383.

H o r n N., S. S w i n d e l l, H. D o d d, M. G a s s o n. 1991. Nisin biosynthesis genes are encoded by a novel conjugativetransposon. Mol. Gen. Genet. 228: 129�35.

H u a n g D.C., M. N o v e l, G. N o v e l. 1990. A transposon like element on the lactose plasmid of Lactococcus lactis subsp.lactis Z270. FEMS Microbiol. Lett. 61: 101�106.

K l a e n h a m m e r T., E. A l t e r m a n, F. A r i g o n i, A. B o l o t i n, F. B r e i d t, J. B r o a d b e n t, R. C a n o, S. C h a i l l o u,J. D e u t s c h e r, M. G a s s o n, M. v a n d e G u c h t e, J. G u z z o, A. H a r t k e, T. H a w k i n s, P. H o l s, R. H u t k i n s,M. K l e e r e b e z e m, J. K o k, O. K u i p e r s, M. L u b b e r s, E. M a g u i n, L. M c K a y, D. M i l l s, A. N a u t a,R. O v e r b e e k, H. P e l, D. P r i d m o r e, M. S a i e r, W. d e Vo s, B. W e i m e r, M. Z a g o r e c and R. S i e z e n.2002. Discovering lactic acid bacteria by genomics. Antonie van Leeuwenhoek 82: 29�58.

L o p e z d e F e l i p e F., C. M a g n i, D. d e M e n d o z a and P. L o p e z. 1995. Citrate utilization gene cluster of theLactococcus lactis biovar diacetylactis: organization and regulation of expression. Mol. Gen. Genet. 246: 590�599.

M a g u i n E., H. P r e v o s t, S.D. E h r l i c h and A. G r u s s. 1996. Efficient insertional mutagenesis in lactococci and othergram-positive bacteria. J. Bacteriol. 178: 931�935.

R a u c h P.J.G. 1990 Lactococcus lactis insertion element iso-IS904 coding for a transposase. NCBI Accession: X52273.R a u c h P.J., W.M. D e Vo s. 1992. Characterization of the novel nisin-sucrose conjugative transposon Tn5276 and its insertion

in Lactococcus lactis. J. Bacteriol. 174: 1280�1287.R a u c h P.J., M.M. B e e r t h u y z e n and W.M. d e Vo s. 1994. Distribution and evolution of nisin-sucrose elements in

Lactococcus lactis. Appl. Environ. Microbiol. 60: 1798�1804.R o m e r o D.A. and T.R. K l a e n h a m m e r. 1990. Characterization of insertion sequence IS946, an Iso-ISS1 element, isolated

from the conjugative lactococcal plasmid pTR2030. J. Bacteriol. 172: 4151�4160.S c h a e f e r A., A. J a h n s, A. G e i s and M. T e u b e r. 1991. Distribution of the IS elements ISS1 and IS904 in lactococci.

FEMS Microbiol. Lett. 80: 311�318.T e u b e r M., L. M e i l e, F. S c h w a r z. 1999. Acquired antibiotic resistance in lactic acid bacteria from food. Antonie van

Leeuvenhoek 76: 115�137.T w o m e y D.P., L.L. M c K a y and D.J. O� S u l l i v a n. 1998. Molecular characterization of the Lactococcus lactis LlaKR2I

restriction-modification system and effect of an IS982 element positioned between the restriction and modification genes.J. Bacteriol. 180: 5844�5854.

T w o m e y D.P., D e P.J. U r r a z a, L.L. M c K a y and D.J. O� S u l l i v a n. 2000. Characterization of AbiR, a novel multi-component abortive infection mechanism encoded by plasmid pKR223 of Lactococcus lactis subsp. lactis KR2. Appl.Environ. Microbiol. 66: 2647�2651.

v a n K r a n e n b u r g R. and W.M. d e Vo s. 1998. Characterization of multiple regions involved in replication and mobiliza-tion of plasmid pNZ4000 coding for exopolysaccharide production in Lactococcus lactis. J. Bacteriol. 180: 5285�5290.

v a n K r a n e n b u r g R., J.D. M a r u g g, I.I. v a n S w a m, N.J. W i l l e m and W.M. d e Vo s. 1997. Molecular character-ization of the plasmid-encoded eps gene cluster essential for exopolysaccharide biosynthesis in Lactococcus lactis. Mol.Microbiol. 24: 387�397.

v a n K r a n e n b u r g R., I.I. v a n S w a m, J.D. M a r u g g, M. K l e e r e b e z e m and W.M. d e Vo s. 1999. Exopolysaccharidebiosynthesis in Lactococcus lactis NIZO B40: functional analysis of the glycosyltransferase genes involved in synthesis ofthe polysaccharide backbone. J. Bacteriol. 181: 338-340.

v a n K r a n e n b u r g R., M. K l e e r e b e z e m and W.M. d e Vo s. 2000. Nucleotide sequence analysis of the lactococcalEPS plasmid pNZ4000. Plasmid 43: 130�136.

Yu W., I. M i e r a u, A. M a r s, E. J o h n s o n, G. D u n n y and L.L. M c K a y. 1995. Novel insertion sequence-like elementIS982 in lactococci. Plasmid 33: 218�225.



Lysostaphin as a Potential Therapeutic Agentfor Staphylococcal Biofilm Eradication

EL¯BIETA WALENCKA1, BEATA SADOWSKA1, SYLWIA RÓ¯ALSKA2,WALERIA HRYNIEWICZ3 and BARBARA RÓ¯ALSKA1*

1 Department of Immunology and Infectious Biology,2 Department of Industrial Microbiology and Biotechnology, Institute of Microbiology

and Immunology, University of £ód�, Banacha 12/16, 90-237 £ód�, Poland,3 National Institute of Public Health, Warsaw, Poland

Received 24 February 2005, received in revised form 8 July 2005, accepted 11 July 2005

A b s t r a c t

The aim was to study the activity of lysostaphin in monotherapy or in combination with oxacillin, towards biofilmsbuilt by clinical and reference S. aureus and S. epidermidis strains in the wells of microplate, in the chambersof a LabTekII chamber slide or on the polyethylene catheter. MICs of oxacillin and lysostaphin for planktonic bacteriawere determined according to the standards of NCCLS. BIC (Biofilm Inhibitory Concentration) was estimated bythe MTT assay. The integrity of biofilm treated with antimicrobials was also examined: by staining with FITC and laserscanning fluorescence confocal microscopy and visually by TTC reduction assay. Despite the fact that susceptibilityof planktonic cultures of 25 staphylococcal strains to lysostaphin action was various, we have demonstrated the effec-tiveness of lysostaphin in the treatment of biofilm, built not only on the flat surface of the microplates but also oncatheter�s surface. The synergistic effect of subBIC lysostaphin+oxacillin was observed for MSSA and MRSA biofilmsbut not for 1474/01 hVISA strain. Also BICOXA for S. epidermidis RP12 and A4c strains, but not for 6756/99 MRSEbiofilms was reduced when lysostaphin was simultaneously used.

K e y w o r d s: staphylococcal biofilms, antibiotics, lysostaphin

Introduction

Bacterial biofilms are described as polymer-dipped communities of cells which accumulate, in a pre-cisely controlled manner, on the abiotic or biotic surfaces (Lewis, 2001; Fux et al., 2003; Boles et al., 2004).Some naturally existing biofilms have a protective role for the host tissue homeostasis (biofilms on urogenitalepithelium, intestine epithelium, dental plaques), by preventing their colonization by exogenous pathogens(Prakash et al., 2003; Boles et al., 2004; Fux et al., 2003). However, it should be stressed that bacterial andfungal biofilms are also responsible for a number of diseases, such as native valve endocarditis, cysticfibrosis-associated pneumonia, middle ear infections, bone infections, bacterial prostatitis, periodontitis(Hall-Stoodley et al., 2004; Götz, 2002; Fux et al., 2003). Biofilms are also involved in the pathogenesis ofvarious infections related to implanted medical devices (urinary and vascular catheters, prosthetic heartvalves, prosthetic hip/knee, contact lenses etc.). Most of these infections have a chronic nature and, becauseof the intrinsic resistance of the biofilm cells to antibiotics and host defense sytems, such diseases are verydifficult to treat effectively. Many hypotheses were considered to explain the high biofilm resistance toantimicrobial agents: restricted penetration, decrease in bacterial metabolism and growth rate, increasein antibiotic-degrading enzymes accumulation and enhancement of exchanging rate of genes encoding forresistance (Lewis, 2001; Hall-Stoodley et al., 2004; Fux et al., 2003). Currently, the most accepted viewis that all the hypotheses are true, but none of them explains the resistance of various biofilms to killing by

* Coressponding author: prof. Barbara Ró¿alska, Department of Immunology and Infectious Biology, Institute of Microbiologyand Immunology, University of £ód�, Banacha 12/16, 90-237 £ód�, Poland, tel. 048-42-6354472, e-mail address: [email protected]

192 Walencka E. et al. 3

different groups of antimicrobial agents. Nevertheless, additional and now dominating is molecular biologybased explanation of antibiotic tolerance rather than resistance, induced within biofilm structure i.e. expres-sion of stress response genes, phase variation, persister-state and biofilm-specific phenotype development(Hall-Stoodley et al., 2004).

Many research groups investigated, alternative to antibiotics, potential strategies in preventing biofilmformation or its eradication. One possible approach is enzymatic removal of bacterial biofilms, but due tothe heterogeneity of the extracellular matrix, in most cases a mixture of enzymes may be necessary fora sufficient degradation of biofilm structure. Nevertheless, this strategy is useful in the eradication ofso called environmental biofilms, e.g. industrial pipelines (water, oil) or food processing equipment.Enzymatic biofilm disruption is a very attractive idea for the prevention or elimination of pathogenic biofilmscausing various medical problems. Enzymatic activity can be directed to biofilm matrix, allowing betterpenetration of subsequent antimicrobials used or to be directed to pathogen�s cell wall componentsand causing their lysis (Johansen et al., 1997; Kaplan et al., 2004). Endopeptidases secreted by differentbacterial species are examples of enzymes which can be used for this purpose. Two of them are now beingthoroughly investigated, the LasA protease produced by Pseudomonas aeruginosa and lysostaphin secretedby Staphylococcus simulans, both specifically active towards staphylococcal strains (Barequet et al., 2004;Wu et al., 2003). Lysostaphin, a 27-kDa endopeptidase which degrades the pentaglycine bridges in peptido-glycan bone of the cell wall, was shown as a potent antistaphylococcal agent, however, it acts much moreeffectively against S. aureus than S. epidermidis strains due to differences in cell wall composition of thesespecies. S. aureus and coagulase-negative staphylococci (CNS), mainly S. epidermidis, are known as theleading species in chronic polymer-associated infections of biofilm nature and resistant to antibiotictreatment. Therefore, we ask the question whether lysostaphin may be considered as an effective biofilmeradicating agent, since its strong activity against planktonic staphylococci is well documented (Climo et al.,1998, Kiri et al., 2002, von Eiff et al., 2003) and what advantages or limitations the use of lysostaphin aloneor with antibiotics, creates.

Experimental


Bacteria. The group of 25 staphylococcal strains chosen for investigation consisted of: clinical S. aureus (n = 10) andS. epidermidis (n = 9) isolates, S. aureus ATCC25923 and S. aureus ATCC29213 (the reference MSSA strains), S. aureus 1474/01(clinical hVISA, NIPH), S. epidermidis ATCC12228 (the reference MSSE strain), S. epidermidis RP12 (slime producing clinicalisolate, from the collection of A. Ljungh, Dept. Medical Microbiology, University of Lund, Sweden), S. epidermidis 6756/99(clinical MRSE, NIPH). Most clinical strains were isolated from medical device-associated infections (Table I). The organismswere stored in TSB with 15% glycerol at �70°C, and in each experiment the cultures were established from the original stock.

Antimicrobial agents and susceptibility testing against planktonic bacteria. The antibiotic oxacillin (disks 1 :g and tablets0.1 mg) was purchased from Mast Diagnostics (United Kingdom). Recombinant lysostaphin (from S. simulans, No. L 0761) wasobtained from Sigma, (St. Louis, USA). The susceptibility of staphylococcal strains to antimicrobial agents was determined by thestandard NCCLS disk diffusion and microdilution methods (National Commitee for Clinical Laboratory Standards, M7-A5, 2000).The concentration range of oxacillin used in the study was 0.125�128.0 :g mL�1 in CAMHB + 2% NaCl, whilst lysostaphin con-centration range tested was 0.0625�64.0 :g mL�1 in CAMH + 0.1% BSA (to prevent its nonspecific adherence to plastic surface).To specify the MICs, turbidometric (OD600) bacterial studies were carried out using the multifunction counter Victor2 (Wallac,Finland). MIC was estimated as the lowest concentration of antimicrobial agent which gave OD equal to the medium negativecontrol (below 0.05).

Biofilm formation. S. aureus or S. epidermidis from the stock cultures were grown for 24 h at 37°C on the agar plate. Next,a single colony of each strain was grown in 5 mL of TSB (Difco) supplemented or not with 0.25% D-(+)-glucose (TSBGlc).The overnight cultures were diluted 1:40 in TSBGlc. A final volume of 200 :L was added to each well of a 96-well tissue cultureplate (Nunclon� Surface, Nunc) or of 500 :L to each chamber of Lab Tek chamber slide II (Nalge-Nunc International, Napervile,Ill., USA). In order to allow bacteria to form biofilms the plates/chambers were incubated for 24 h at 37°C.

Stains for biofilm visualization. MTT (3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide, Sigma), TTC (2,3,5-tri-phenyltetrazolium chloride, Sigma), FITC (Fluorescein isothiocyanate Isomer I, Sigma) were used.

Biofilm MTT-staining. MTT assay was performed according to the method described by Kairo et al., (1999) with minormodifications. Briefly, after 24 h lasting biofilm formation the wells of microplate were emptied and filled with 150 :L of PBS perwell, then 50 :L of MTT solution (0,3% in PBS) was added and plates were incubated for 2 h at 37°C. At the end of incubationperiod, MTT was replaced with 150 :L of DMSO and 25 :L of glycine buffer (0.1 M, pH 10.2). For complete dissolving of formedpurple formazan crystals the plates were incubated for 15 min. at room temperature, with gentle agitation. The optical density of thewells containing biofilms was determined using a spectrophotometer (550 nm, Victor2 multifunction counter, Wallac, Finland).

Biofilm FITC-staining. Biofilms formed on the slides surface of Lab Tek chamber slide II device were stained with FITCsolution (0,1% in PBS) for 20 min. at room temperature. After staining chambers were emptied and very gently washed, once with

193Staphylococcal biofilm eradication3

PBS supplemented with 4% BSA and twice with PBS. After the washing, the chamber slides partitions were removed and the slideswere covered with cover glasses. The effects of biofilm FITC-staining were observed using the laser scanning confocal microscope(LSCM, Zeiss).

Biofilm bacterial cultures for antimicrobial test. Biofilms were prepared in 96-well microplates or in chamber slides deviceas described above. After 24 h of growth, the medium from the wells/chambers was removed by aspiration. The biofilms weretreated either with various concentrations of oxacillin or lysostaphin, or with a combination of oxacillin and subMIC concentrationof lysostaphin. Antimicrobial agents were diluted in CAMHB supplemented with 2% NaCl and 0,1% BSA. The incubation timewas 24 h at 37°C, then the medium containing the antimicrobial agent was gently aspirated and the viability of the biofilm remainingon the surfaces of the wells or chamber slides was stained with MTT or FITC as described above.

Confocal microscopy study. Examination of FITC-stained biofilms treated with antimicrobial agents was performed usinga LSM5 (Pascal) Laser Scanning Confocal Microscope (LSCM), equipped with Axiovert 2 (Zeiss) microscope with objectivePlan-Apochromat 100x (1.4 oil). Images were recorded at a 488 nm (argon laser) excitation and emission at 530 nm (long pass filterset). Digital image analysis of LSCM optical thin sections was performed with Pascal Zeiss software.

�Catheter study�. Intravascular catheter�s (Venflon, PTFE, Becton Dickinson, USA) segments (1 cm length) were preparedusing sterile instruments, placed into the eppendorf tubes (in triplicate) containing 1 mL of bacterial suspension (overnight cultureof S. aureus A3- MRSA, diluted 1:40 in TSBGlc), and incubated for 72 h at 37°C. Then catheter�s segments were rinsed with PBSand transferred to the new tubes containing: i) medium, ii) oxacillin, iii) lysostaphin, or iiii) oxacillin and subMIC lysostaphin for3 or 24 h incubation at 37°C. Next, biomaterial segments, gently rinsed with PBS were moved to the fresh TSBGlc medium withTTC (one drop of 1% TTC in PBS) and incubated for 24 h at 37°C. The presence of red stained bacterial biofilm (reduction by livebacteria of colourless TTC to the red insoluble formazan crystals) was estimated as described previously (Sadowska et al., 1998).

Data analysis and presentation. Each assay was performed in duplicate or triplicate on at least two occasions. Data wereexpressed as the mean from 2�3 evaluations obtained from each experiment.

Results

Most S. aureus and S. epidermidis strains used in this study were clinical isolates, mainly from patientssuffering from medical-devices-associated infections, as presented in Table I. These strains, after theirinitial identification, were obtained from hospital microbiological laboratories. In our hands, strains werereidentified using selected typing methods (hemolysis, clumping factor and tube coagulase tests) and 32IDStaph (BioMerieux) system.

Using the standard antimicrobial susceptibility tests, according to the 2000� guidelines of NCCLS, theMICs of oxacillin and lysostaphin were determined. The obtained MICs values of each strain cell suspen-sion tested on two occasions were similar (differences not exceeding one dilution), therefore single valuesare presented (Table II). For the suspension cultures of S. aureus and S. epidermidis strains, oxacillin MICsranged from 0.250 to >128 :g mL�1. However, most S. aureus clinical strains (7/10) were highly resistant,with MICOXA >128 :g mL�1, whereas most S. epidermidis strains (10/12) had MICs range from 1.0 to

S. aureus

ATCC25923 reference MSSA, ATCC*

ATCC29213 reference MSSA, ATCC

1474/01 clinical hVISA, NIPH**

A3 drain

A7 hip prosthesis associated abscess

B1 orthopedic wound

C1 tracheostomic tube

D5 tracheostomic tube

D8 venous catheter�s tip

D13 drain

E1 tracheostomic tube



Table IList of staphylococcal strains used in these studies

Strain Origin / characteristics

* � American Type Culture Collection, ** � National Institute of Public Health, Poland

S. epidermidis

ATCC12228 reference MSSE, ATCC

RP12 slime positive, University of Lund, Sweden

6756/99 clinical MRSE, NIPH

A4a hip prosthesis

A4b hip prosthesis

A4c hip prosthesis

C10 blood

C11 blood

C12 blood

C13 blood

C15 blood

C27 blood

Strain Origin / characteristics


16 :g mL�1. These 25 staphylococcal strains underwent tests of susceptibility to lysostaphin. WithinS. aureus group, lysostaphin MICs ranged from 0.0625 to 0.500 :g mL�1, with median MIC = 2.7 :g mL�1

for more than 92% of strains. One exception was S. aureus E4 strain whose MICLYS was 32 :g mL�1.The median MIC for 12 S. epidermidis strains was much higher � 7.0 :g mL�1, and for 91% of strainsthe range started from 0.5 and ended at 16 :g mL�1. Generally, for most staphylococcal strains there was nocorrelation between MICOXA and MICLYS values. However, S. aureus E4 which presented the highestMICLYS = 32.0 :g mL�1 was also highly resistant to oxacillin (MICOXA>128 :g mL�1. A similar result wasdemonstrated for 6756/99 MRSE strain (MICLYS = 32.0 :g mL�1, MICOXA>16.0 :g mL�1). On the otherhand, 1474/01 hVISA strain, whose MICOXA exceeded 128 :g mL�1 had the lowest MICLYS among all25 tested strains, which was 0.0625 :g mL�1.

The preparation of S. aureus and S. epidermidis biofilms, on hydrophobic polystyrene (96-well micro-plate) or negatively charged glass (Lab Tek chamber slide II) surfaces, produced high bacterial yield. AfterMTT staining, optical density (OD550) readings ranged from 1.8 to 3.1 (Fig. 1), therefore the isolates werecategorized as strong biofilm producers. The degree of biofilm formation was commonly higher when bac-teria were initially and finally grown in TSBGlc than in TSB without glucose, however, for some strains(mainly among S. aureus group) the opposite effect was observed (data not shown). Cells growing asbiofilms (in chamber slide devices) stained with FITC, were characterized without the disturbance of theirstructure using laser fluorescence scanning microscope. LSCM images showed multilayered clumps of bac-teria, surrounded by less dense material, attached to the surface (Fig. 3-A, B). S aureus strains producedbiofilms which were, on average, 2.5 times thinner than those formed by S. epidermidis. Biofilm of S. aureusATCC29213, 24 h after set-up, reached 2.66 ± 0.56 :m of thickness, whereas that of S. epidermidis RP12was 6.35 ± 0.86 :m. When cultures incubation time was prolonged, further growth of biofilms wasobserved and at the end (72 h) biofilm of S. aureus ATCC29213 and S. epidermidis RP12 reached thethickness of 4.09 ± 0.78 :m and 8.02 ± 0.61 :m, respectively.

In order to test the hypothesis that lysostaphin and beta-lactam antibiotics act as synergistic agents towardstaphylococci, the effect of the oxacillin or lysostaphin (alone or in combination) on cell viability in biofilmswas examined. BICs (Biofilm Inhibitory Concentration) of antimicrobials were determined by MTT reductionassay measuring the active metabolism of bacteria that survived the antimicrobials action. Biofilms of all13 S. aureus, as well as 12 S. epidermidis strains were highly resistant to oxacillin (BICs>128�256 :g mL�1).The absorbances of the oxacillin treated S. aureus biofilms were the same after 24 h as at the time ofantibiotic application (OD550 2.9�3.1). Similarly, the absorbances of S. epidermidis biofilms showed a mini-mal range of changes. One exception was biofilm of S. epidermidis RP12 strain treated with oxacillin at the

S. aureus

ATCC 25923 (MSSA) 16 0.250 0.125

ATCC 29213 (MSSA) 14 0.250 0.250

1474/01 (hVISA) 0 >128.0 0.0625

A3 0 >128.0 0.250

A7 0 >128.0 0.500

B1 14 2.0 0.250

C1 16 1.0 0.250

D5 0 >128.0 0.250

D8 15 1.0 0.250

D13 13 2.0 0.250

E1 0 >128.0 0.250

E4 0 >128.0 32.0

E7 6 >128.0 0.500

Table IISusceptibility to oxacillin and lysostaphin of S. aureus and S. epidermidis strains. MIC for planktonic culture determined

by disk-diffusion and broth microdilution methods

Strain

Oxacillin Lysostaphin

Inhibition(∅mm)

MICOX

:g mL �1

MICLYS

:g mL �1

S. epidermidis

ATCC 12228 22 0.125 4.0

RP12 16 1.0 8.0

6756/99 (MRSE) 6 16.0 32.0

A4a 17 2.0 1.0

A4b 17 2.0 2.0

A4c 17 1.0 1.0

C10 0 16.0 8.0

C11 10 2.0 2.0

C12 14 2.0 0.500

C13 14 2.0 16.0

C15 0 128.0 2.0

C27 0 128.0 8.0

Strain Inhibition(∅mm)

MICOX

:g mL �1

MICLYS

:g mL �1

Oxacillin Lysostaphin


concentration of 128 :g mL�1, where the absorbance dropped after 24 h from 2.2 to 0.49. For the evaluationof biofilm susceptibility to lysostaphin, two concentration ranges of the enzyme were used, chosen on thebasis of different MICs values established for S. aureus and S. epidermidis strains growing in suspension.For generally more susceptible planktonic cultures of most S. aureus (MICsLYS 0.063�0.5 :g mL�1), lyso-staphin used against biofilms was at the concentrations of 2�64 :g mL�1. For more resistant planktonicS. epidermidis (MICsLYS 0.5�16 :g mL�1), lysostaphin concentration range used for biofilm eradicationwas 8�256 :g mL�1. S. aureus biofilms demonstrated various susceptibility to the lysostaphin; BICLYS rangefor 8/13 strains was from 4 to 32 :g mL�1, set when the absorbance dropped near the baseline establishedfor negative control well (OD550 = 0.05�0.09). On the other hand, for 5/13 S. aureus strains BICLYSexceeded the maximal concentration used � >64 :g mL�1. S. epidermidis biofilms were significantly moreresistant to lysostaphin, BICLYS determined for 10/12 strains was >256 :g mL�1, for one strain (RP12) itwas 128 :g mL�1 (OD550 dropped from 2.237 to 0.092) and for the other one (A4c) � 16 :g mL�1 (OD550dropped from 2.762 to 0.096) (Fig. 1).

Further experiments concerned the possible synergistic effect of the antibiotic and subinhibitory concen-tration of enzyme, towards biofilm cultures. For this purpose, 3 S. aureus and 3 S. epidermidis strains,

3.5

3

2.5

2

1.5

1

0.5

0

ATC

C 2

5923

ATC

C 2

9213

1474

/01

A3

A7

B1

C1

D5

D8

D13 E

1

E4

E7

OD

550

A

BIC

biofilm treated with LYS biofilm non treated

3.5

3

2.5

2

1.5

1

0.5

0

ATC

C 1

2228

RP

12

6756

/99

A4a

A4b A4c

C10

C11

C12

C13

C15

C27

OD

550

BIC

B

biofilm treated with LYS biofilm non treated

Fig. 1. Biofilm formation of S. aureus (A) and S. epidermidis (B) strains and their lysostaphin susceptibility,measured by the MTT reduction assay in the wells of 96-well microplate


whose biofilms were characterized as differently susceptible to lysostaphin, and, of course, highly resistantto oxacillin, were included in these studies. The applied �biofilm MTT viability� test, revealed non-signifi-cant differences between replicate wells, both within and between test dates, therefore single values arepresented in Table III. When lysostaphin was used in subBIC (established earlier), together with differentoxacillin concentrations, MTT reduction assay revealed that effective inhibition in biofilm growth could beachieved with a much lower antibiotic concentration. This was demonstrated for biofilms of S. aureusATCC29213 (MSSA) and S. aureus A3 (MRSA), BICOXA dropped from >128 to 4.0 and 32.0 :g mL�1,respectively. The synergistic action of oxacillin and lysostaphin was also demonstrated for S. epidermidisRP12 and clinical S. epidermidis A4c biofilms. Unfortunately, such a good effect was not observed for1474/01 hVISA and 6756/99 MRSE strains included in this part of the study (Table III, Fig. 2).

The activity of lysostaphin towards biofilm cultures was confirmed by laser scanning confocal micro-scopy. Microscopic examination of S. aureus and S epidermidis biofilms, prepared in chamber slides, whichwere treated with lysostaphin (16 :g mL�1) for 24 h, demonstrated a disruption of S. aureus biofilm andloosening of S. epidermidis biofilm structure (Fig. 3-A1, B1). Oxacillin alone had no effect on biofilmintegrity, even when used at the highest concentration, whereas oxacillin used together with subBIC oflysostaphin caused partial biofilm disruption (data not shown).

The incubation of polyethylene catheter�s segments, for 72 h at 37°C, with overnight culture of clinicalMRSA strain (S. aureus A3) resulted in biofilm formation. Their presence on extra- and intraluminal surfacesof control segments was demonstrated as red stained bacterial deposit, which was a result of the reduction ofcolourless TTC to the red insoluble formazan crystals by live bacteria (Fig. 4a). When the colonized catheter�ssegment was immersed in the medium with oxacillin (4 :g mL�1, according to NCCLS borderline concentra-tion describing MRSA) and incubated at 37°C for 24 h, no visible effect of antibiotic action was seen (Fig. 4b).Lysostaphin alone used at a concentration of 8 :g mL�1 (BICLYS established earlier for S. aureus A3 strain)caused complete biofilm eradication as soon as after 3 h incubation (Fig. 4c). In order to demonstrate synergis-tic effect of antibiotic and lysostaphin, catheter�s samples colonized by S. aureus A3 were incubated at 37°Cfor 24 h with oxacillin (4 :g mL�1) together with lysostaphin in subBIC concentration (4 :g mL�1). Also inthis case total biofilm eradication was achieved (lack of red stained bacterial deposit).

S. aureus

ATCC 29213 (MSSA) >128 8 4 (+4)

A3 (MRSA) >128 8 32 (+4)

1474/01 (hVISA) >128 4 >128 (+2)

Table IIIThe influence of oxacillin or/and lysostaphin on biofilm viability, evaluated by MTT reduction assay in 96-well

microplate

StrainBIC

OX

(:g mL�1)BIC

LYS

(:g mL�1)BIC

OX (+LYS)

(:g mL�1)

S. epidermidis

RP12 >128 64 32 (+64)

A4c >128 16 4 (+8)

6756/99 (MRSE) >128 >128 >128 (+64)

BIC OX (+LYS)

(:g mL�1)BIC

LYS

(:g mL�1)BIC

OX

(:g mL�1)Strain

Fig. 2. Effect of lysostaphin (LYS), oxacillin (OXA) or both (LYS+OXA) on (A) S. aureus ATCC29213� MSSA, (B) S. aureus A3 � MRSA, (C) S. aureus 1474/01 � hVISA biofilms, measured by the MTT

reduction assay in the wells of 96-well microplate.1 (A-C) � biofilm non-treated (positive control); 2 (A-C) � biofilm treated with OXA 128�2 :g mL�1; 3 � biofilmtreated with LYS: (A, B) � 32�4 :g mL�1, (C) � 16�2 :g mL�1; 4, 5 � biofilm treated with OXA 128�2 :g mL�1

+ LYS: (A, B) � 4 :g mL�1, (C) � 2 :g mL�1; 6 � medium + MTT (negative control)


Discussion

It is now well accepted that bacteria form groups and respond as groups and that individual bacteria inbiofilm community rapidly diversify, which increases the capability of the group and provides a form of�biological insurance� (Boles et al., 2004). Changeability of the bacteria within biofilm structure causestheir extremely high resistance to antimicrobial agents and host immune system (Prakash et al., 2003; Fuxet al., 2004; Hall-Stoodley et al., 2004). Biofilm infections are the major medical problems, with S. aureus

Fig. 3. Laser scanning confocal microscopy (LSCM) images of (A) � S. aureus A3 and(B) S. epidermidis A4c biofilms, formed in chamber slides and stained with FITC.A1, B1 � biofilms of S. aureus A3 and S. epidermidis A4c, respectively, treated for 24 h withLYS 16 :g mL�1. The square panel are a plain view and the side panels are vertical cross sections,

respectively

Fig. 4. Eradication activity of oxacillin or lysostaphin alone and in combination, against S. aureus A3 biofilm formed on catheter�sextra- and intraluminal surfaces, measured by the TTC reduction assay.

a � biofilm non-treated (positive control); b � biofilm treated for 24 h with OXA (4 :g mL�1); c � biofilm treated for 3 h with LYS (8 :g mL�1);d � biofilm treated for 24 h with OXA (4 :g mL�1) + LYS (4 :g mL�1)


and coagulase-negative staphylococci, mainly S. epidermidis, as the leading species responsible for chronicpolymer-associated infections (Götz, 2002; Vuong and Otto, 2002; Costa et al., 2004; Lindsay and Holden,2004). This was the main reason why in the present paper we wanted to test whether lysostaphin, alone or incombination with oxacillin, may be considered as an effective staphylococcal biofilm eradicating agent,which was indicated for planktonic cells in the reports of many authors (Climo et al., 1998, 2001; Kiri et al.,2002; von Eiff et al., 2003).

Twenty five staphylococcal strains, 13 of S. aureus and 12 of S. epidermidis were included in this study(Table I and Table II). Within S. aureus group, lysostaphin MICs ranged from 0.0625 to 0.500 :g mL�1 andfor S. epidermidis strains the range started from 0.5 and ended at 16 :g mL�1. Such lysostaphin MICsranges, different for S. aureus and S. epidermidis strains were also demonstrated by Climo et al., (1998,2001), Kiri et al., (2002), Wu et al., (2003). It is known that lysostaphin is capable of cleaving the cross-linking pentaglycine bridges in peptidoglycan of cell wall and that differences in the S. aureus andS. epidermidis susceptibility are mediated by increased incorporation to the muropeptide of serine and alanine,instead of glycine (Climo et al., 1998, 2001; Kiri et al., 2002; von Eiff et al., 2003).

For all 25 staphylococcal strains their biofilm formation ability was evaluated using the MTT assay andlaser scanning confocal microscopic (LSCM) observations. Optical density (OD550) readings after MTTstaining ranged from 2.1 to 3.1 (Fig. 1), therefore the isolates were categorized as strong biofilm producers,however, as revealed by LSCM, S. epidermidis formed more biomass than S. aureus (72 h after set-up8.02 ± 0.61 :m and 4.09 ± 0.78 :m for S aureus ATCC29213 and S. epidermidis RP12, respectively). Inmany reports it is stressed that the main factors involved in S. epidermidis and S. aureus biofilm formationare not the same and they are not known to the same extent (Mack et al., 2004). The same methods (MTT,LSCM) were used for the evaluation of oxacillin or/and lysostaphin action on biofilm viabillity and integ-rity. A single standard method for the biofilm susceptibility testing is still lacking, so it is very difficult tocompare the already published results obtained for biofilms assayed under different conditions. Since wewanted to know how many bacteria survive the incubation of biofilms with antimicrobial agent, our choicewas to use MTT (3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide) measuring active metabo-lism of cells. The reproducibility of our observations proved that this method was accurate and proper. Itwas demonstrated that biofilms of both species were highly resistant to oxacillin (BICs>128�256 :g mL�1),which means that irrespective of different level of susceptibility (low or high MICOXA of planktonic cells),none of the biofilms was killed by oxacillin used at such a concentration (Fig. 1).

It should be pointed out that the first authors who demonstrated lysostaphin activity against staphylococcalbiofilms were Wu et al., (2003). They also proposed the speculation on the possible explanation for themechanisms of biofilm eradication by lysostaphin, such as rapid lysis of adherent cells, which may besufficient to destabilize biofilm matrix and allow their detachment. Also, in our study lysostaphin was shownto be effective in biofilm eradication, however with different concentrations used. According to what wasdescribed earlier in many reports, the synergistic effect of lysostaphin with oxacillin combination worksfor planktonic cells (Kiri et al., 2002), and the same effect could be expected against biofilm, which wasdemonstrated in our study. We have shown that oxacillin at a concentrations of 4.0 and 32.0 :g mL�1, whenapplied for 24 h with subBIC of lysostaphin, was effective in the killing of biofilm formed by S. aureusATCC29213 and A3 strains, respectively. A similar result (significant decrease in BICOXA values) wasdemonstrated for biofilms of two (RP12, A4c) S. epidermidis strains (Table III, Fig. 2). However, usinglysostaphin in subBIC had also unexpected limitations, since applying it together with oxacillin did notresult in the reduction of oxacillin concentration which could be effective in the killing of 1474/01 hVISAand 6756/99 MRSE biofilms. We suggest that unpredictable synergistic effect of lysostaphin low doses,combined with antibotics, could be the main limitation for such a therapeutic strategy.

Similarly to Wu et al., (2003) we have examined biofilm microscopically. In our study laser scanningconfocal microscopy of FITC-stained biofilm was used, which allowed us to observe not only differences inS. aureus and S. epidermidis biofilm thickness and architecture but also antibiotic and lysostaphin influence,more precisely than SEM used by these authors. Our results are in some discrepancy with theirs sincelysostaphin used at a concentration of 16 :g mL�1 did not clear the biofilm from the surface, howeverit managed to disrupt it. Wu et al., (2003) demonstrated such an effect using a higher lysostaphin concen-tration. Nevertheless, we do hope that biofilm structure partially disrupted by the action of lytic enzymeprobably could be enough for a more effective antibiotic activity and/or immune mechanisms of the host.The observed synergistic effect of lysostaphin and oxacillin, shown by MTT-reduction assay, was con-firmed by the study on �catheter model�. It was another method successfully introduced to our study, basedon the reduction by live bacteria of tetrazolium salt � TTC to unsoluble red formazan crystals. The presence


of red-stained bacterial biofilm on extra- and intraluminal catheter�s surfaces and their disappearing afterincubation with antimicrobials were easy to follow (Fig. 4 a-c).

The findings of this in vitro study suggest that lysostaphin might be considered for treatment of implantor catheter-associated infections, caused by staphylococci, mainly S. aureus. However, it should be stressedthat still we are far from being able to use lysostaphin in clinics, although many studies revealed that lyso-staphin possesses a potent anti-staphylococcal activity, for example in in vivo models of rabbit aortic valveendocarditis and nasal colonization in a cotton rat model or in vitro against bacteria isolated from anteriornares and blood (Climo et al., 1998; Patron et al., 1999; Kokai-Kun et al., 2003; von Eiff et al., 2003). Thepromising results published by Wu et al., (2003) and the results of our study on the lysostaphin activity(alone or in combination with antibiotics) against the staphylococcal biofilms, should be considered withsome caution due to: the unpredictable lysostaphin susceptibility of a given strain, which has to be estab-lished experimentally, the possibility of generation of lysostaphin-resistant strains when too low concentra-tions are used, or development of anti-lysostaphin antibodies when prolonged and repeated therapy withhigh lysostaphin doses is introduced (Boyle-Vavra et al., 2001; Climo et al., 2001; Dajcs et al., 2002; Kiriet al., 2002). Nevertheless, enzymatic detachment of medical biofilms seems to be a new way to increase orreplace the ineffective in many cases antibiotic therapy (Barequet et al., 2004; Kaplan et al., 2004).

Literature

B a r e q u e t I.S., G.J. B e n S i m o n, M. S a f r i n, D.E. O h m a n and E. K e s s l e r. 2004. Pseudomonas aeruginosa LasAprotease in treatment of experimental staphylococcal keratitis. Antimicrob. Agents Chemother. 48: 1881�1687.

B o l e s B.R., M. T h o e n d e l and P.K. S i n g h. 2004. Self-generated diversity produces �insurance effects� in biofilm commu-nities. PNAS. 101: 16630�16635.

B o y l e - Va v r a S., R.B. C a r e y and R.S. D a u m. 2001. Development of vancomycin and lysostaphin resistance in a methi-cillin-resistant Staphylococcus aureus isolate. J. Antimicrob. Chemother. 48: 617�625.

C l i m o M.W., K. E h l e r t and G.L. A r c h e r. 2001. Mechanism and suppression of lysostaphin resistance in oxacillin-resis-tant Staphylococcus aureus. Antimicrob. Agents Chemother. 45: 1431�1437.

C l i m o M.W., R.L. P a t r o n, B.P. G o l d s t e i n and G.L. A r c h e r. 1998. Lysostaphin treatment of experimental methicillin-resistant Staphylococcus aureus aortic valve endocarditis. Antimicrob. Agents Chemother. 42: 1355�1360.

C o s t a S.F., M.H. M i c e l i and E.J. A n a i s s i e. 2004. Mucosa or skin as source of coagulase-negative staphylococcalbacteraemia? Lancet Inf. Dis. 4: 278�286.

D a j c s J.J., B.A. T h i b o d e a u x, D.O. G i r g i s, M.D. S h a f f e r, S.M. D e l v i s c o and R.J. O� C a l l a g h a n. 2002.Immunity to lysostaphin and its therapeutic value for ocular MRSA infections in the rabbit. Invest. Ophthalmol. Visual Sci.43: 3712�3716.

F u x C.A., P. S t o o d l e y, L. H a l l - S t o o d l e y and J.W. C o s t e r t o n. 2003. Bacterial biofilms: a diagnostic and therapeuticchallenge. Expert Rev. Anti-infect. Ther. 1: 667�683.

G ö t z F. 2002. Staphylococcus and biofilms. Mol. Microbiol. 43: 1367�1378.H a l l - S t o o d l e y L., J.W. C o s t e r t o n and P. S t o o d l e y. 2004. Bacterial biofilms: from the natural environment to infec-

tious diseases. Nature Rev. 2: 95�106.J o h a n s e n C., P. F a l h o l t and L. G r a m. 1997. Enzymatic removal and disinfection of bacterial biofilms. Appl. Environ.

Microbiol. 63: 3724�3728.K a i r o S.K., J. B e d w e l l, P.C. T y l e r, A. C a r t e r and M.J. C o r b e l. 1999. Development of a tetrazolium salt assay for

rapid determination of viability of BCG vaccines. Vaccine. 17: 2423�2428.K a p l a n J.B., C. R a g u n a t h, K. Ve l l i y a g o u n d e r, D.H. F i n e and N. R a m a s u b b u. 2004. Enzymatic detachment

of Staphylococcus epidermidis biofilms. Antimicrob. Agents Chemother. 48: 2633�2636.K i r i N., G. A r c h e r and M.W. C l i m o. 2002. Combinations of lysostaphin with $-lactams are synergistic against oxacillin-

resistant Staphylococcus epidermidis. Antimicrob. Agents Chemother. 46: 2017�2020.K o k a i - K u n J.F., S.M. W a l s h, T. C h a n t u r i y a and J.J. M o n d. 2003. Lysostaphin cream erdicates Staphylococcus

aureus nasal colonization in a cotton rat model. Antimicrob. Agents Chemother. 47: 1589�1597.L e w i s K. 2001. Riddle of biofilm resistance. Antimicrob. Agents Chemother. 45: 999�1007.L i n d s a y J.A. and M.T.G. H o l d e n. 2004. Staphylococcus aureus: superbug, supergenome? Trends Microbiol. 12: 378�385.M a c k D., P. B e c k e r, I. C h a t t e r j e e, S. D o b i n s k y, J.K.-M. K n o b l o c h, G. P e t e r s, H. R o h d e and

M. H e r r m a n n. 2004. Mechanisms of biofilm formation in Staphylococcus epidermidis and Staphylococcus aureus: func-tional molecules, regulatory circuits, and adaptive responses. Int. J. Med. Microbiol. 294: 203�212.

National Commitee for Clinical Laboratory Standards. 2000. Methods for dilution antimicrobial susceptibility tests for bacteria thatgrow aerobically. Approved standard fifth edition. M7-A5: 1�54.

P a t r o n R.L., M.W. C l i m o, B.P. G o l d s t e i n and G.L. A r c h e r. 1999. Lysostaphin treatment of experimental aorticvalve endocarditis caused by a Staphylococcus aureus isolate with reduced susceptibility to vancomycin. Antimicrob. AgentsChemother. 43: 1754�1755.

P r a k a s h B., B.M. Ve e r e g o w d a and G. K r i s h n a p p a. 2003. Biofilms: a survival strategy of bacteria. Curr. Sci. 85:1299�1307.


S a d o w s k a B., M. W i ê c k o w s k a, W. R u d n i c k a and B. R ó ¿ a l s k a. 1998. Variation in slime/biofilm production bystaphylococcal strains. AAM J. 7: 203�209.

Vo n E i f f C., J.F. K o k a i - K u n, K. B e c k e r and G. P e t e r s. 2003. In vitro activity of recombinant lysostaphin againstStaphylococcus aureus isolates from anterior nares and blood. Antimicrob. Agents Chemother. 47: 3613�3615.

V u o n g C. and M. O t t o. 2002. Staphylococcus epidermidis infections. Microb. Infect. 4: 481�489.W u J.A., C. K u s u m a, J.J. M o n d and J.F. K o k a i - K u n. 2003. Lysostaphin disrupts Staphylococcus aureus and Staphylo-

coccus epidermidis biofilms on artificial surfaces. Antimicrob. Agents Chemother. 47: 3407.


gyrA Mutations in Ciprofloxacin-resistantClinical Isolates of Pseudomonas aeruginosa in a Silesian Hospital in Poland

ZENOBIA WYDMUCH1, OLGA SKOWRONEK-CIO£EK1, KRZYSZTOF CHOLEWA2,URSZULA MAZUREK2, JERZY PACHA1, MA£GORZATA KÊPA1, DANUTA IDZIK1

and ROBERT D. WOJTYCZKA1

1 Department of Microbiology and 2 Department of Molecular Biology and Medical Genetics,Medical University of Silesia, Sosnowiec, Poland

Received 24 March 2005, received in revised form 6 June 2005, accepted 7 June 2005

A b s t r a c t

Among 73 clinical isolates of Pseudomonas aeruginosa 48 strains were ciprofloxacine (CIP) susceptible and 25 CIPresistant (Minimal inhibitory concentration � MIC>32 :g/ml � 14 strains) or of intermediate susceptibility to CIP(MIC≥1,5 � 32 :g/ml � 11 isolates). Mutations in the quinolone-resistance-determining region (QRDR) of gyrA genewere searched in groups of CIP resistant and of intermediate susceptibility to CIP isolates. Two methods: restrictionfragment length polymorphism (RFLP) analysis and DNA sequencing analysis allowed to detect three different muta-tions. The nucleotide substitutions observed led to the following amino acid replacements: Thr-83 → Ile, Asp-87→Asn,Asp-87→Gly. One mutated strain among the group of mutants analyzed showed double mutation (Thr-83→ Ile,Asp-87→Gly) and additional silent mutation (Val-103 →Val); whilst the rest of the isolates showed differentsingle missense mutations. The most frequently detected mutation in the gyrA gene (16 out of 25 mutants) was theThr-83→ Ile substitution.

K e y w o r d s: Pseudomonas aeruginosa, ciprofloxacin, resistance, mutation, gyrA

Introduction

Pseudomonas aeruginosa possesses intrinsic mechanisms of resistance to a wide variety of antibiotic.Fluoroquinolones (FQs) are broad-spectrum antibiotics, which are known to be effective in the treatment ofa wide range of infections. Ciprofloxacin has emerged as one of the most effective quinolones againstP. aeruginosa (Mouneimñe et al., 1999). However, the extensive use of fluoroquinolones has resultedin an increasing incidence of FQs resistance (Yonezawa et al., 1995). Opportunistic infections causedby Pseudomonas aeruginosa, especially quinolone-resistant isolates, pose a serious medical problem. Theantibacterial effect of the FQs depends on the inhibition of bacterial topoisomerases: DNA gyrase andtopoisomerase IV (Bearden and Danziger, 2001; Dricla and Zhao, 1997). DNA gyrase and topoisomeraseIV are heterotetrameric proteins composed of two subunits designated A and B. The genes encoding the Aand B subunits are referred to as gyrA and gyrB (DNA gyrase) or parC and parE (DNA topoisomerase IV).Both subunits constitute the active form of the enzyme: A2B2 (Dricla and Zhao, 1997). DNA gyrase activityis strongly inhibited by quinolones. Alterations in DNA gyrase or topoisomerase IV caused by mutations inthe QRDR appear to play a major role in fluoroquinolone resistance in clinical isolates of P. aeruginosa(Yoshida et al., 1990). Recently, several species of bacteria have been studied in order to determine theinfluence of gyrA mutation on quinolone resistance (Weigel et al., 1998).

The authors have not found any Polish publications concerning mutations in gyrA gene in fluoroquinolone-resistant clinical isolates of P. aeruginosa.

The aim of our study was to find mutations responsible for ciprofloxacin-resistance in gyrA gene ofP. aeruginosa clinical strains, isolated in Silesia region.

202 Wydmuch Z. et al. 3

Experimental


Bacterial isolates. 73 clinical isolates of Pseudomonas aeruginosa were isolated at Microbiological Laboratory, Saint BarbaraHospital No 5 in Sosnowiec, Poland, in 2003 from various clinical materials, collected from infirmary and hospitalized patients,showing this species infections.

Antimicrobial susceptibility testing. Initial MIC profiles were screened using the disc diffusion method in accordance withthe recommendations of the National Committee for Clinical Laboratory Standards (NCCLS, 2000). The MICs of ciprofloxacinwere determined by the E-test (AB Biodisk, Solna, Sweden). The range of CIP concentration in susceptibility tests was between0.002 and 32 :g/ml. The results were evaluated after 24 hours of incubation at 35°C. The Pseudomonas aeruginosa strain ATCC27853 was included as a control.

PCR analysis. Primers were designed to amplify the DNA fragment including the putative quinolone resistance-determi-ning region (Yoshida et al., 1990). For the QRDR of gyrA (GenBank access number L29417), a pair of primers, PaGA 1(5�-TGACGGCCTGAAGCCGGTGCAC-3�) and PaGA 4 (5�-TATCGCATGGCTGCGGCGTTG-3�) (Takenouchi et al., 1999), wasused. These primers allowed amplification of the gyrA gene region, including codons 38 to 122, to encompass the region containingcodons 67 to 106. The amplification procedure comprised of initial denaturation at 94°C for 3 min followed by 38 cycles ofdenaturation (30 s at 94°C), annealing (30 s at 55°C), and polymerization (60 s at 72°C), and then a final extension cycle: 10 min at72°C. The reactions were conducted in a final volume of 50 :l with 2.5 U of Taq DNA polymerase (Fermentas).

PCR-RFLP analysis. PCR products were treated with Cfr 42 I enzyme (an isoschizomer of SacII), (MBI Fermentas) at 37°Cfor 2 hours, and the restriction fragments were separated in 3% low-melting-point agarose gel and visualized by means of ethidiumbromide staining (Takenouchi et al., 1999). The Cfr 42 I site (CCGC¯GG) is present in the wild-type gyrA gene between nucleotides512 and 517, whereas it is absent in the mutant gyrA gene.

Nucleotide sequencing. Prior to the sequencing reaction, the PCR products were purified by isopropanol precipitation with2M NaClO4 and glycogen in low-TE buffer. Amplicons were analyzed by direct sequencing using a BigDye Terminator CycleSequencing Ready Reaction Kit (PE Applied Biosystems). DNA sequencing analysis was performed with the same primers as thoseused for PCR. Sequence reactions products were purified with the use of SigmaSpin columns (Sigma-Aldrich) and air-dried. Sampleswere heated at 96°C for 4 min in 5 :l of loading buffer (25 mM EDTA, pH 8,0 with 50 mg/ml blue dextran and deionized formamidwith a ratio of 1:5) and chilled rapidly with ice before loading with 5% denaturing polyacrylamide gel in an Applied Biosystem377 DNA Sequencer according to the manufacture�s instructions (Kureishi et al., 1994; Yonezava et al., 1995; Nakano et al., 1997;Takenouchi et al., 1999). The nucleotide sequences obtained both for sense and antisense primers were compared with the GeneBankDatabase (http://www.ncbi.nlm.nih.gov/blast) in order to check their homology.

Results and Discussion

The first step of our study was to perform a drug susceptibility test. Using the E-test method, we evalu-ated the lowest ciprofloxacin concentration visibly preventing growth of P. aeruginosa, (MIC) (Fig. 1). Thestrains analyzed showed a various susceptibility to ciprofloxacin. Figure 1 presents ciprofloxacin MICvalues for analyzed strains of P. aeruginosa. 48 isolates (65.7%) were CIP susceptible (MIC ≤1 :g/ml),whilst 14 strains (19.2%) were CIP resistant with MIC >32 :g/ml, and 11 strains (15.1%) were of inter-

16

14

12

10

8

6

4

2

0

Num

eber

of s

trai

ns

0.064

MIC (:g/ml)0.094

0.1250.19

0.250.32

0.94 1 1.5 3 320.64

0.50.38

0.75 2

Fig. 1. Ciprofloxacin MIC profile for different strains of P. aeruginosa

203gyr A mutations in CIP-resistant P. aeruginosa3

mediate susceptibility to CIP with MIC ≥1.5�32 :g/ml. Only the resistant strains and the isolates of inter-mediate susceptibility were subjects of our further studies (25 strains). The PCR reaction confirmed thepresence of the gyrA gene in all of the strains. In order to detect mutations two methods were used for eachexamined analyzed strain: PCR-RFLP with Cfr 42 I (SacII) and DNA sequencing analysis. PCR-RFLPanalysis with Cfr 42 I was performed as a screening method. The restrictase SacII recognizes the CCGC↓GGsite, present only in a wild type gyrA gene between 512 and 517 nucleotides at codon 83, therefore thismethod allows mutation at the Cfr 42 I site to be detected, however it does not exclude the possibility ofmutation elsewhere. In our study, using the PCR-RFLP method, we found that 9 out of 25 isolates (36%)have Cfr 42 I restriction site and 16 strains (64%) did not have the site recognized by this enzyme (Table I).Of the 16 isolates without Cfr 42 I restriction site, 14 showed resistance to CIP; furthermore, the MICsvalues were above the maximum E-test concentration (>32 :g/ml) (Table II).

In order to perform a more accurate analysis of the entire fragment of the gyrA gene, a DNA sequencinganalysis was conducted for each examined strain of P. aeruginosa. Using this method we discovered gyrAmutation in 23 strains (92%). Gene alterations were located at codon 83 and 87. We found 15 strains witha single mutation Thr-83→ Ile (13 strains resistant to CIP, 2 of intermediate susceptibility to CIP); 7 strainswith a single mutation Asp-87→Asn (all of them of intermediate susceptibility to CIP), and 1 strain (#47)with a double mutation and additional silent mutation Thr-83→ Ile, Asp-87→Gl, Val-103→Val (resistantto CIP) (Table I, Fig. 2).

Cfr42 I site present Wild type � � � 2 (8)

Single mutation Asp-87 → Asn GAC→AAC 7 (28) �

Cfr42 I site not detected Single mutation Thr-83 → Ile ACC→ATC 15 (60) �

Double mutation Thr-83 → Ile ACC→ATCAsp-87 → Gly GAC→GGC 1 (4) �

Silent mutation Val-103 → Val GTA→GTC

Total number of strains with and without mutation in the gyrA gene 23 (92) 2 (8)

Total number of analyzed strains 25 (100)

Table ITypes of point mutations in gyrA and ciprofloxacin MIC range for the resistant or of intermediate susceptibility

P. aeruginosa strains

PCR-RFLP method

DNA sequencing method Number and (%)of strains

with mutation

Number and (%)of strains

without mutationType of mutation Amino acidreplacement

Nucleotidereplacement

ATCC 27853 0.25 Thr (ACC) Asp (GAC) Val (GTA)

2 3 Asn (AAC)

3 3 Asn (AAC)

4 2 Asn (AAC)

5 1.5

6 2 Asn (AAC)

8 2

10 2 Ile (ATC)

26 2 Asn (AAC)

27 > 32 Ile (ATC)

36 2 Asn (AAC)

40 > 32 Ile (ATC)

43 1,5 Asn (AAC)

Table IICorrelation between MIC values and presence of mutations in the gyrA gene of P. aeruginosa

Strainnumber

MIC:g/ml

Amino acid alteration in gyrA geneat codon

83 87 103

47 > 32 Ile (ATC) Gly (GGC) Val (GTC)

54 > 32 Ile (ATC)

67 > 32 Ile (ATC)

69 > 32 Ile (ATC)

70 > 32 Ile (ATC)

71 > 32 Ile (ATC)

76 > 32 Ile (ATC)

91 > 32 Ile (ATC)

93 > 32 Ile (ATC)

94 > 32 Ile (ATC)

95 > 32 Ile (ATC)

102 > 32 Ile (ATC)

119 2 Ile (ATC)

1038783

MIC:g/ml

Strainnumber

Amino acid alteration in gyrA geneat codon

204W

ydmuch Z

. et al.3

Fig. 2. Nucleotide and amino acid sequence matching between isolate #47 with double mutation and silent mutation of gyrA gene and wild type of gyrA gene(GB ACC L29417, Kureishi et al., 1994)

205gyr A mutations in CIP-resistant P. aeruginosa3

Akasaka and co-workers (2001) collected clinical isolates of P. aeruginosa, and sequenced bacterialDNA to search for type II topoisomerase mutations. They found mutations in gyrA gene in 119 out of 150isolates with reduced susceptibilities to levofloxacine (79.3%); the replacement(s) of amino acid(s) referredto: Thr-83→ Ile or Ala, Asp-87→Asn, Gly or Tyr. The principal replacement observed by the researcherswas Thr-83→ Ile (74.7%). In our study we found similar types of mutation in gyrA gene at codon 83 and 87.As a result of nucleotide substitutions we detected the following mutations: Thr-83→ Ile, Asp-87→Asn,Asp-87→Gly. In addition to point mutations in the gyrA gene followed by a single amino acid change in22 strains: 7 Asp-87→Asn and 15 Thr-83→ Ile, we found one strain with one double mutation and addi-tional silent mutation: Thr-83→ Ile, Asp-87→Gly, Val-103→Val. Moreover, the replacement Thr-83 →Ilewas the most frequent (65.2%).

The point mutation followed by Thr-83→ Ile (ACC?ATC) amino acid replacement was also reported tobe the most frequent by Mouneimné et al. (1999), Yonezawa and co-authors (1995), and Nakano�s group(1997). Contrary to our study, mutation at codon 87: Asp-87→Asn has been reported only sporadically byother authors (Yonezawa et al., 1995; Nakano et al., 1997).

The silent mutation Val-103→Val observed in the present study was one of 7 different types of silentmutation in the gyrA gene reported by Takenouchi and co-authors (1999). Takenouchi�s group found twodouble mutations, three of which were previously unknown: Ala-67→Ser, Asp-87→Gly, Ala-84→Pro andGln-106→Leu. One of them, Asp-87→Gly, was also described by us in strain #47 carrying a doublegyrA mutation, and showed resistance to CIP ≥32 :g/ml. The resistance of mutant Thr-83→ Ile to FQ, asmentioned in other articles, was also observed in our study (87.5%) (Mouneimné et al., 1999; Weigel et al.,1998; Akasaka et al., 2001; Nakano et al., 1997; Takenouchi et al., 1999).

According to Ball (1994), suitable drug dosing and medical procedure supervision can minimize or evenprevent bacterial resistance to quinolones. He emphasizes that bacteria causing intrahospital infectionsthrough steady contact with a wide range of chemotherapeutics easily develop resistance to drugs and, asa consequence, the drug resistance phenomenon occurs mostly among hospitalized patients. In our study,10 of the 14 isolates (71.4%) recognized as CIP resistant (MIC ≥32 :g/ml) were obtained from hospitalizedpatients, which confirms Ball�s (1994) observation.

Our results and the data of other authors show that the analyzed sequence of the gyrA gene playsa crucial role in counteracting ciprofloxacin and DNA gyrase, however, the participation of a mutationlocalized in different regions of P. aeruginosa genome cannot be excluded. A wide spectrum of genes hasrecently been analyzed with regard to FQ resistance, e.g. mutations in parC, mexR or nfxB genes have beenthe subject of intense discussion (Nakajima et al., 2002; Nakano et al., 1997). Further studies on the mecha-nism of P. aeruginosa resistance to FQs are necessary. Such studies may not only improve the efficiency ofFQ therapy and help to prevent the new bacterial resistance phenomenon, but also can help to synthesizenew drugs that enable initiation of the more effective therapy of P. aeruginosa infections.

Literature

A k a s a k a T., M. T a n a k a, A. Ya m a g u c h i and K. S a t o. 2001. Type II topoisomerase mutations in fluoroquinolone-resistant clinical strains Pseudomonas aeruginosa isolated in 1998 and 1999: role of target enzyme in mechanism offluoroquinolone resistance. Antimicrob. Agents Chemother. 45: 2263�2268.

B a l l P. 1994. Bacterial resistance to fluoroquinolones: lessons to be learned. Infection 22: 140�147.B e a r d e n D.T. and L.H. D a n z i g e r. 2001. Mechanism of action and of resistance to quinolones. Pharmacotherapy. 21:

224S�232S.D r i l i c a K. and X. Z h a o. 1997. DNA gyrase, topoisomerase IV, and the 4-quinolones. Microbiol. Mol. Biol. Rev. 61: 377�392.K u r e i s h i A., J.M. D i v e r, B. B e c k t h o l d, T. S c h o l l a a r d t, L.E. B r y a n. 1994. Cloning and nucleotide sequence of

Pseudomonas aeruginosa DNA gyrase gyrA gene from strain PAO1 and quinolone-resistant clinical isolates. Antimicrob.Agents Chemother. 38: 1944�1952.

M o u n e i m n é H., J. R o b e r t, V. J a r l i e r and E. C a m b a u. 1999. Type II topoisomerase mutations in ciprofloxacin-resistant strains of Pseudomonas aeruginosa. Antimicrob. Agents Chemother. 43: 62�66.

N a k a j i m a A., Y. S u g i m o t o o, H. Y o n e y a m a and T. N a k a e. 2002. High-level fluoroquinolone resistance inPseudomonas aeruginosa due to interplay of the MexAB-OprM efflux pump and the DNA gyrase mutation. Microbiol. Immun.46: 391�395.

N a k a n o M., T. D e g u c h i, T. K a w a m u r a, M. Ya s u d a, M. K i m u r a, Y. O k a n o and Y. K a w a d a. 1997. Muta-tions in the gyrA and parC genes in fluoroquinolone-resistant clinical isolates of Pseudomonas aeruginosa. Antimicrob.Agents Chemother. 41: 2289�2291.

National Committee for Clinical Laboratory Standards. 2000. Methods for dilution antimicrobial susceptibility tests for bacteriathat grow aerobically. 5th edn. Approved standard. NCCLS document M7-A5. Wayne, Pennsylvania.

206 Wydmuch Z. et al. 3

T a k e n o u c h i T., E. S a k a g a w a, M. S u g a w a r a. 1999. Detection of gyrA mutations among 335 Pseudomonas aeruginosastrains isolated in Japan and their susceptibilities to fluoroquinolones. Antimicrob. Agents Chemother. 43: 406�409.

W e i g e l L.M., Ch.D. S t e w a r d and F.C. T e n o v e r. 1998. gyrA mutations associated with fluoroquinolone resistance ineight species of Enterobacteriaceae. Antimicrob. Agents Chemother. 42: 2661�2667.

Yo n e z a w a M., M. T a k a h a t a, N. M a t s u b a r a, Y. W a t a n a b e and H. N a r i t a. 1995. DNA gyrase gyrA mutationsin quinolone-resistant clinical isolates of Pseudomonas aeruginosa. Antimicrob. Agents Chemother. 39: 1970�1972.

Yo s h i d a H., M. B o g a k i, M. N a k a m u r a and S. N a k a m u r a. 1990. Quinolone resistance-determining region in theDNA gyrase gyrA gene of Escherichia coli . Antimicrob. Agents Chemother. 34: 1271�1272.


The Prognostic and Diagnostic Markers of Invasive Candidiasisin Patients During Chemotherapy

AGNIESZKA MAGRY�1, MARIA KOZIO£-MONTEWKA1, EL¯BIETA STAROS£AWSKA2

and BEATA GABCZYÑSKA2

1 Medical University, Department of Clinical Microbiology, Lublin, Poland2 Oncology Hospital, Lublin, Poland

Received 31 March 2005, received in revised form 25 May 2005, accepted 27 May 2005

A b s t r a c t

The aim of the work was the early detection of Candida spp. in clinical samples of patients with carcinoma ovariorumundergoing chemotherapy by comparing three indicators of candidiasis: presence of mannan and yeast DNA in thebloodstream and colonization of mucosal membranes by Candida species as a prognostic marker of deep candidiasis.Thirty-one women with carcinoma ovariorum, during chemotherapy without symptoms of deep fungal infections, wereexamined twice over a six-day period. C. albicans was the dominant organism isolated from mucosal membranes. Twoserum samples were positive for mannan on the first day of examination. All these patients were previously colonizedby Candida spp. on mucous membranes. Four patients were positive on the last day of examination. Three of thesepatients were colonized by Candida spp. C. albicans infection was detected early in 4 out of 12 clinical samples bya combination of PCR and mannan-detecting methods. Colonization increases the risk of deep candidiasis. PCR andantigen detection are fast and reliable methods for early detection of Candida in bloodstream. For patients at risk,the clinical samples must be tested by at least two independent methods.

K e y w o r d s: cancer, candidiasis, chemotherapy

Introduction

Severe fungal infections, especially candidiasis, represent a serious medical problem in immunocompro-mised patients during anticancer chemotherapy. The mortality rates among patients of risk for Candidainfection remain high, ranging from 50�80%, despite the adequate treatment (Buchman et al., 1990; Flahautet al., 1998; Pagno et al., 1999; Richardson and Kokki, 1999).

Fungal infections appear mainly in people with dysfunctions of the immune system. In these patients theopportunistic and endogenous Candida strains that colonize superficial sites or non-sterile body fluids arethe major etiologic agents of life-threatening infections. The factors predisposing to the systemic fungalinfections are immunosuppresion, disturbance of residual bacterial flora by antibiotics, extensive surgicalprocedures, and AIDS. Systemic fungal infections are a great problem in patients receiving cancer therapy.In these patients profound neutropenia following chemotherapy is a serious risk factor for opportunisticfungal infections associated with high mortality risk (Holmes et al., 1994; Füsle, 1997; Rüchel, 1997; Pfaller,1995; Murray et al., 1999). Disseminated candidiasis disturbs established schemes for chemotherapy andprolonging pause between courses reduces chances of successful treatment by about 20�30%. Thus, inpatients of risk, the empiric antifungal therapy is usually recommended. On the other hand, this therapy maybe unnecessarily toxic and costly, and it may play a part in spreading resistance among Candida species(Buchman et al., 1990; Elie et al., 1998; Shin et al., 1997). High mortality rates in neutropenic patientstreated for cancer may be reduced only by earliest possible diagnosis of systemic candidiasis with appropriateand effective treatment. Current methods for detection of candidiasis in clinical settings are poor at best.

Corresponding author. Mailing address: Agnieszka Magry� Department of Clinical Microbiology, Medical University of Lublin,ul. Chod�ki 1, 20-123 Lublin, Poland. Phone: (0-81) 740-58-33. Electronic mail address: [email protected]

208 Magry� A. et al. 3

Too much time passes when employing microbiological cultures, while serological tests for antibody detec-tion in immunocompromised patients are useless in clinical settings (Buchman et al., 1990; Flahaut et al.,1998; Richardson and Kokki, 1991; Shin et al., 1999). The standard laboratory diagnostic approach tosystemic candidiasis is based on fungal growth in culture, and is often delayed due to the relatively slowgrowth of yeasts and inadequate sensitivity of the method in early recognition of infection. Thus, a lot ofeffort is put into developing rapid and more sensitive diagnostic methods of systemic fungal infections andidentifying the virulence determinants of isolated strains.

The invasive fungal infection occurs as a result of an increase in the biomass of the pathogenic fungi.The infection increases in aggressiveness as it spreads from an initial site of colonization, ultimatelypenetrating the host�s protective barriers and damaging tissues. In the processes within bloodstream, thereis an increase in fungal specific products that can potentially be utilised as markers of systemic infection.In this regard, techniques have been developed to detect yeast�s proteins, metabolites, DNA and poly-saccharides (Richardson and Kokki, 1999; Sendid et al., 1999). The tests for the detection of antigens insera-glucan, a major polysaccharide of the cell wall and the highly immunogenic mannan, of the Candidaare commercially available for few years. Early detection of systemic fungal infection by these methods, orby PCR assays, should improve the survival time in patients at risk. The crucial step in these patients is thetimely start of therapy with antimycotic drugs before presentation of clinical symptoms.

The aim of our work was the early detection of Candida spp. in clinical samples of patients with carci-noma ovariorum undergoing chemotherapy by comparing three indicators of candidiasis: 1) presence ofmannan in the bloodstream; 2) presence of yeast DNA in the bloodstream; and 3) colonization of mucosalmembranes by Candida species as a prognostic marker of deep candidiasis.

Experimental


Patients and clinical data. Thirty-one patients aged 34�54 years (mean 42,6 years) with confirmed carcinoma ovariorum wereincluded in our study. The patients were qualified for chemotherapy and treated with the aggressive anti-cancer drug Taxol, andcisplatin according to Polish standards for carcinoma ovariorum treatment. Presently, Taxol and cisplatin are considered the mosteffective treatment for carcinoma ovariorum, with haematological toxicity from level II to IV (WHO scale). The treatment,comprising six courses, involved standard doses of Taxol and cisplatin at intervals of 21 days. The study was performed after 3,4 course of chemotherapy.

Clinical samples and strains identification. In order to evaluate the colonization of mucosal membranes by Candida species,nasal, pharyngeal and vaginal swabs from 31 cancer patients were plated onto Sabouraud agar and CHROMagar Candida. Theswabs were collected twice over a six-day period. The plated swabs were incubated at 35°C for 48�72 h and examined daily forevidence of growth. When a sample was found to be positive for Candida, the strain was identified. Isolates from CHROMagarCandida were identified according to the manufacturer�s instructions. Isolates from Sabouraud agar were first identified micro-scopically by Gram staining to assess the cell morphology and then the germ tube test was performed. Isolates were also identif iedon the basis of biochemical features using the ID 32C (bioMérieux) test.

Detection of soluble antigen of Candida in serum. Sera for the serological tests comprised 62 samples from 31 cancer patientsand 21 samples from 21 healthy donors.

Two procedures were used to detect the soluble antigen of Candida in serum. The first method was the immunoenzymatic test(ELISA) for detection of mannan in serum (Platelia Candida; BioRad). Before starting the test all tested sera and standards wereheat-treated in the presence of EDTA to dissociate the immune complexes and to precipitate the serum proteins that may interferewith the test. The supernatants were tested after a final, clarifying centrifugation. Microplate wells were coated with monoclonalantibodies directed against $ 1�5 oligomannosides of Candida. 50 µl of conjugate (peroxidase-labelled anti-mannan MAb) weresimultaneously incubated with 50 µl of supernatant. After incubation, the wells were washed and the reaction was revealed byincubation in the dark with 200 µl chromogenic substrate. The enzymatic reaction was stopped by adding 1,5N H2SO4. The opticaldensity was determined at 450 nm. The limit of detection of the assay is 0,25 ng of mannan per ml of serum tested. All reactionswere performed in duplicate.

The second method was based on latex agglutination, using commercial test Pastorex Candida (BioRad). Briefly: 100 µl oftreatment reagent was added to 300 µl of serum and placed on heat at 100°C for 3 minutes. After centrifugation at 10 000 g for10 min., the supernatant was tested. 40 µl of supernatant fluid was deposited on the card, 10 µl of Candida latex was added and thecard was shaked for 10 min. (160 rev./min.). A positive reaction was indicated by the agglutination of the latex particles.

PCR detection of Candia albicans DNA in human whole blood samples. Isolation of genomic DNA: Clinical specimens(n = 62) examined for yeasts� DNA presence were obtained from 31 cancer patients and included the whole blood samples. Allsamples were collected twice in a period of one week. In order to remove the erythrocytes from the blood specimens, 100 µl of thewhole blood was added to 1 ml of sterile distilled water. Each sample was incubated on a shaker for 5 min. at RT. The lysate wascentrifuged at 3,000×g for 10 min. and the pellet was resuspended in 2 ml of sterile distilled water, vortexed and centrifuged againat 3,000×g for 10 min. The erythrocyte-free pellet was washed in 2 ml of 20 mM Tris-HCl (pH 8.3), centrifuged for 10 min. at3,000 × g, resuspended in 100 µl of RE lysis buffer (A&A Biotechnology) to which 10 µl of yeast lysis enzyme (ICN) was added.

209Diagnosis of invasive candidiasis in cancer patients3

The mixture was incubated for 1h at 37°C. After this step, 200 µl of universal lysis buffer LT (A&A Biotechnology) and 20 µl ofprotease K (1.7 mg/ml; Qiagen) were added and the mixture was incubated 20 min. at 37°C and then the sample was heated at 75°Cfor 5 min. The sample was centrifuged at 10,000 rpm for 3min. and supernatant was applied to QiAmp spin column (Qiagen),centrifuged at 10,000 rmp for 1 min. and washed twice with 500 µl of washing buffer A1 (A&A Biotechnology). DNA was elutedwith 100 µl of eluting buffer RE (10 mM TRIS, pH 8.5; A&A Biotechnology) preheated to 75°C. The sample was incubated at RTfor 5 min. and centrifuged at 10,000 rpm for 1 min. The purified DNA preparation was kept at �20°C until PCR.

Primers and PCR amplification: Two oligonucleotides derived from Candida albicans rRNA fragments were used as the outerprimers. The sequences of these oligonucleotides are the following:

camt1 5�-CACCCGATCCGCCTCCTACCGAAG-3�camt2 5�cGTCTGCCCGATCCGTACCTCCGT-3�.

These primers amplify a 1039-bp fragment in the rRNA genes of C. albicans. The amplification was performed in a 50 µl reactionmixture containing: 50 mM KCl, 10 mM Tris-HCl (pH 8.3), 1.5 mM MgCl2, 100 µl (each) deoxynucleotide triphosphates (dATP,dCTP, dGTP, dTTP; A&A Biotechnology) and 1.25 U of Delta2 DNA polymerase (A&A Biotechnology) and 10 pmol (each)primer. A 2.5 µl aliquot of the extracted DNA was added to the mixture. PCR was performed in a thermocycler (Peqlab,Biotechnologie GmbH) as follows. The reaction was initiated by incubation at 95°C for 5 min. Then, the first cycle included 30 sec.of denaturation at 94°C, 1 min. of annealing at 64°C and 1 min. of elongation at 72°C. This first step was followed by 30 cycles.The PCR product was then analysed on a 2% agarose gel stained with ethidium bromide. The positive samples were testedby nested-PCR.

For nested-PCR, 1 µl of the PCR product from the first amplification and 10 pmol of Candida albicans-specific inner primerswere mixed in a fresh reaction mixture. The sequences of these oligonucleotides are: camt3 5�-ATATATTAgTCTCCACCCGA-3�camt4 5�-atagAGTATAACCACCCGAT-3�. These generated PCR product of 317-bp of Candida albicans. The second amplifica-tion was performed in similar conditions with the exception of annealing temperature (47°C for 1 min.) and elongation temperature(72°C for 30 sec.). The amplified product was analysed on a 2% agarose gel stained with ethidium bromide.

Each reaction was carried out in duplicate. To avoid sample contamination, precautions suggested by Kwok and Higuchi (Kwokand Higuchi, 1989) were used. Cross-contamination by aerosols was reduced by physical separation of laboratory rooms used forreagent preparation, sample processing, and DNA amplification. Other precautions included UV irradiation for microcentrifugetubes, racks, surfaces of laboratory benches, and instruments. Such laboratory procedures as autoclaving of buffers and distilledwater, use of fresh lots of previously aliquoted reagents, combined use of positive-displacement pipetters and aerosol-resistantpipette tips, frequent changing of gloves, premixing reagents, addition of DNA as the last step, and testing negative controls,including omission of either the primer or the DNA template during PCR, were used. Appropriate negative controls were includedin each test run, including controls omitting either the primer or the DNA template during PCR assay. In all experiments thenegative controls always tested negative. Positive standard for PCR used 1ml of purified DNA for Candida albicans.

Results

Routine culture of the swabs from mucosal membranes on Sabouraud agar demonstrated, that 4 patients(13%) had Candida spp. on at least two sites (Pfaller, 1995; Yeo and Wong, 2002). Candida strainscolonized pharynx (4 cases) and vagina (3 cases) most frequently. Nose swabs were positive in 2 cases.Moreover, Candida spp. strains were present in 6 other patients at only one site.

Of all Candida species isolated from patients, 10 were identified as C. albicans. These strains wereisolated from 6 patients. 5 strains of C. glabrata were isolated from 4 patients. One strain of C. kruzei wasisolated from 1 patient (Table I).

3or 2 sites 4 C. albicans C. albicans C. albicans1 C. albicans � C. albicans3 C. glabrata � C. glabrata6 � C. albicans C. albicans

1 site 2 � C. glabrata �5 � C. albicans �7 � � C. glabrata8 C. albicans � �9 � C. albicans �

10 C. glabrata � �11 � C. kruzei �

none 12�31 � � �

Table IThe positive culture results on the first day of examination

Noof occupied

sitesPatients

Site of Candida isolation

vagina nose pharynx


Thirty-one patients were classified into 3 groups according to the number of sites occupied by Candidaspp. The first group (patients 4, 1, 3 and 6; 12.9%) comprised patients colonized by yeasts on two or threesites. The second group (patients 2, 5, 7, 8, 9, 10 and 11; 22.5%) corresponds to patients colonized at onlyone site by Candida spp. The third group (patients 12�31; 70.9%) comprised patients with no signs ofCandida present on mucosal membranes (Table I). On the sixth day of examination we did not observeany other cases of yeast�s colonization on mucosal membranes.

Each ELISA experiment for circulating mannan detection in sera was performed in duplicate. Therepeatability of the optical density (OD) values on a single microtiter plate corresponded to a coefficient ofvariation of <10%. The sensitivity of detection was 0.25 ng of mannan per ml. In this study we considered>0.25 ng/ml to be the cutoff level since 100% of the healthy controls had mannanemia values that wereequall or less than this value.

The results obtained for each of the 31 individual patients are summarized in Table II. Table II comparesthe methods used to identify markers of candidiasis: routine culture indicating mucous membranes coloniza-tion, circulating mannan detection and detection of C. albicans DNA as markers of disseminated candidiasis.All patients have been classified into 3 groups according to the number of sites occupied by Candida spp.The patients� classification was based on the mucosal membranes occupation by Candida spp., becausecolonization is regarded as one of the prognostic markers of deep candidiasis.

In patients no 1 and 3 from the first group, a high mannan concentration was detected in the serumsamples early (on the first day of examination). Mannnan concentration was above 1 ng/ml in both patients,while the latex agglutination method gave positive (no 1) and questionable (no 3) results. One of thesepatients (no 1) was colonized by C. albicans. For this patient PCR identification confirmed disseminatedcandidiasis caused by endogenous species. The phenotypic identification of the strain that colonized themucosal membranes (C. albicans) matched the PCR reaction.

Nucleic acid of C. albicans was also detected early by the PCR technique in the blood sample of patientno 4 six days before the appearance of mannan in the patient�s serum. It is worthy to note that in this caseC. albicans colonized three anatomical sites. In the case of patient no 6, who was colonized by C. albicanson two sites, no other markers of candidiasis were detected on the first or the second day of examination.

In the second group of patients it was not possible to detect antigenemia or C. albicans DNA in bloodsamples on the first day of examination. We observed one case of positive case of mannanemia six days later(patient no 2).

In patients from the third group, we did not observe signs of Candida present on mucosal membranes orother markers of candidiasis in serum or blood samples.

Disseminated C. albicans infection was detected early in 4 out of 12 (33%) clinical samples, by a combi-nation of PCR and mannan-detecting ELISA methods. Moreover, the PCR method allowed us to detect4 positive cases of deep C. albicans infections, whereas positive mannanemias were identified in 3 cases

3 or 2 sites 4 C. albicans < 0.25 � +/+ 1 � +/+1 C. albicans 1.5 + +/+ 1.5 + +/+3 C. glabrata >1 +/� � 1 � �6 C. albicans 0 � � 0 � �

1 site 2 C. glabrata < 0.25 � � 1 � �5 C. albicans < 0.25 � � < 0.25 � �7 C. glabrata 0 � � 0 � �8 C. albicans 0 � � 0 � �9 C. albicans 0 � � < 0.25 � �

10 C. glabrata 0 � � 0 � �11 C. kruzei 0 � � 0 � �

None 12�31 None 0 � � 0 � �

Table IIComparative identification of Candida spp. by routine phenotypic identification methods versus antigen

detection and PCR/nested PCR method

Noof occupied

sites

Patients�no

First examination Second examination

Routinephenotypic

identification

antigen conc.by ELISA

(ng/ml)

PCR/nestedPCR

latexagglutin.

antigen conc.by ELISA

(ng/ml)

latexagglutin.

PCR/nestedPCR


(Table III). All patients were at the higher risk of the candidiasis as the routine culture indicated fungalcolonization of the mucosal membranes. In four patients with C. glabrata present on mucosal membranes,an increase in mannan concentration was observed in 2 clinical samples. PCR method did not identify DNAof C. albicans in the blood of these subjects.

When microbiological and serological investigations were positive, antifungal treatment at standard dos-ages was performed.

Discussion

Infections due to Candida spp. remain the most frequent complications in cancer patients despite anincrease in systemic fungal infections over the last few years. Candidiasis is the most prominent infectionwith a high mortality rate in patients at risk (Pagno et al., 1999).

For many years Candida albicans was regarded as the most frequent etiologic agent involved in almostall cases of candidiasis, but recent reports indicate an increase in candidiasis sustained by non-albicansspecies: mainly Candida krusei, Candida tropicalis and Candida glabrata (Pagno et al., 1999; Shinet al., 1997; Yamamura et al., 1999; Abbas et al., 2000). Our study confirms this finding: Candidaalbicans was the predominant species isolated from mucosal membranes, but other non-albicans specieswere also present.

Along with the appearance of life-threatening Candida non-albicans strains, the strains resistant to anti-fungal drugs appeared. It may be a direct result of the widespread use of antifungal agents for the prophy-laxis and treatment of candidiasis (Pfaller, 1995; Yamamura et al., 1999; Morgenstern et al., 1999). In manyoncological hospitals over the world, antimycotic prophylaxis is routinely prescribed and administrated atthe start of chemotherapy, also in the absence of positive culture results.

In order to reduce the mortality rate among the group of patients at risk there is a need for rapid, sensitiveand specific tests for appropriate antifungal therapy before presentation of clinical symptoms. Nowadayspatient evaluation, mycological cultures, diagnostic imaging, and biopsies are the standard clinical andlaboratory approaches in diagnosis and therapeutic monitoring of systemic candidiasis. Unfortunately, thesemethods very often lack sensitivity in the early recognition of infection and are imprecise as markers ofcomplete eradication (Richardson and Kokki, 1999). In the context of these disadvantages, new approachesto early diagnosis of disseminated candidiasis are being developed.

In our work we compared traditional diagnostic methods with novel methods � serological and molecu-lar. The routine culture allowed us to detect mucosal membranes colonization in 4 patients undergoingchemotherapy. Although the importance of colonization is unclear, in most instances precedes fungemia andis regarded as an independent risk factor for systemic fungal infection (Safdar et al., 2001).

The probability of deep candidiasis was evaluated by serological methods detecting circulating mannanin patients� blood. We tested 62 serum samples drawn from cancer patients, with detection of mannan in6 cases. Subjects with mannanemia also showed signs of Candida sp. on mucosal membranes. Mannanantigenemia detection for the immunodiagnosis is now one of the most widely studied antigens in patientswith candidiasis. Many investigators (Richardson and Kokki, 1999; Sendid et al., 1999; Yeo and Wong,2002) suggest that positive mannan results may correlate with invasive candidiasis. Furthermore, studieshave also shown a correlation between detectable mannanemia and tissue invasive by Candida spp. in

C. albicans/6 12 1 (8%) 2 (16%) 2 (16%) 2 (16%)

C. glabrata/4 8 1 (12.5%) 2 (25%) 0 (0%) 0 (0%)

C. kruzei/1 2 0 (0%) 0 (0%) 0 (0%) 0 (0%)

0/ 20 40 0 (0%) 0 (0%) 0 (0%) 0 (0%)

Table IIIResults of mannan detection versus Candida albicans DNA detection in clinical samples

Candida spp./noof patients

Noof clinicalsamples*

No of clinical samples positive for the following:

Mannanemia of ≥ 1 ng/ml PCR identification

I examin. II examin.

* the clinical samples from each patient were collected twice

I examin. II examin.


patients� with candidemia (Yeo and Wong, 2002). In our study, we used two tests to detect mannan in serumsamples of examined persons.

The latex agglutination test has been widely used as the first commercially available antigen detectiontest. The latex agglutination test has shown good specificity, but poor sensitivity, due to the rapid clearanceof the antigen from patient�s sera and a low amount of circulating antigen (Richardson and Kokki,1999; Füsle, 1997; Sendid et al., 1999).

By using the ELISA method, the detection limit has been improved up to 0,25 ng of mannan per ml.This resulted in an increase in sensitivity, and allowed detection of mannanemia in 13% of patients at risk.However, these results are still disappointing because in 5% of patients with negative antigen detection,the symptoms of disseminated infection were clinically observed (data not shown). This illustrates oneof the major limitations of the fungemia detection tests: the transient character of mannan circulation.Several mechanisms have been proposed to account for this observation, including the quick degradation ofmannose oligomers by serum mannosidases, the binding of the mannose oligomers to soluble serumproteins (mannose binding protein C3) or membranous receptors of phagocytes (Sendid et al., 1999).As a consequence, sensitivity rises with the number of serum samples available from each patient. Theperformance of mannan detection in serum varies with the frequency of testing in a given patient. Regularantigen monitoring of high-risk patients is recommended as a means of increasing the sensitivity of the testand advancing the date of the first positive result (Yeo and Wong, 2002).

The doubling time for yeasts is long � an hour or more, compared to 20 min. for bacteria. Furthermore,most clinical specimens are contaminated by bacteria that compete for nutrients and thereby can add todoubling times. For this reason, specimens that are believed to contain yeasts are generally plated out ontoa selective medium that contains antibacterial agents. The specific identification of fungus requires 3 to5 days (Buchman et al., 1990). With the PCR-based amplification procedure, contamination by the bacterialand human cells is avoided and because of the high sensitivity of the method it was possible to detect a smallamount of yeast DNA in blood.

In our work, 62 clinical samples from cancer patients undergoing chemotherapy were tested for thepresence of C. albicans nucleic acid in blood. 33% (4 samples) were PCR positive and 25% (3 samples)were mannan-positive. All patients with disseminated Candida albicans fungemia were colonized by thesame species. The PCR method appears to be more sensitive than mannan detecting immunoenzymaticmethod. In our work the difference in the sensitivity of both tests was not significant, but other authors(Buchman et al., 1990; Richardson and Kokki, 1999, Shin et al., 1999; Yeo and Wong, 2002) indicate thehigher sensitivity of PCR based methods compared with other non-culture tests.

In the presented work, we did not detect other than Candida albicans species by PCR. However, the highspecificity of species-specific primers demonstrated by many investigators (Buchman et al., 1990; Holmeset al., 1994; Elie et al., 1998; Shin et al., 1997; Shin et al., 1999) confirm their value in early detection anddifferentiation of Candida species present in clinical materials.

The conclusions from our work are: (i) colonization increases the risk of disseminated candidiasis occur-rence and may be used as its prognostic marker; (ii) PCR and mannan detection are fast and reliable methodsfor early detection of Candida in bloodstream; (iii) in order to increase the rate of detection of systemicfungal infections, the clinical samples should be collected at least twice; (iv) for patients at risk, the clinicalsamples must be tested by at least two independent methods.

Literature

A b b a s J., G.P. B o d e y and H.A. H a n n a. 2000 Candida krusei fungemia: an escalating serious infection in immunocompro-mised patients. Arch. Intern. Med. 160: 2659�2664.

B u c h m a n T.G., M. Rossier, W. M e r z and P. C h a r a c h e. 1990. Detection of surgical pathogens by in vitro DNA amplifi-cation. Part I. Rapid identification of Candida albicans by in vitro amplification of a fungus-specific gene. Surgery 108:338�347.

E l i e C.H., T.J. L o t t, E. R e i s s and C.J. M o r r i s o n. 1998. Rapid identification of Candida species with species-specificDNA probes. J. Clin. Microbiol. 36: 3260�3265.

F l a h a u t M., D. S a n g l a d, M. M o u o d, J. B i l l e and M. R o s s i e r. 1998. Rapid detection of Candida albicans in clini-cal samples by DNA amplification of common regions from C. albicans-secreted aspartic proteinase genes. J. Clin. Microbiol.36: 395�401.

F ü s l e R. 1997. Diagnosis of fungal infections. Mycoses 40 (Suppl. 2): 13�15.H o l m e s A.R., R.D. C a n n o n, M.G. S h e p e r d and H.P. J e n k i n s o n. 1994. Detection of Candida albicans and other

yeasts in blood by PCR. J. Clin. Microbol. 32: 228�231.


K w o k S. and R. H i g u c h i. 1989. Avoiding false positives with PCR. Nature (London) 339: 237�238.M o r g e n s t e r n G.R., A.G. P r e n t i c e, H.G. P r e n t i c e, J.E. R o p n e r, S.A. S c h e y and D.W. W a r n o c k. 1999.

A randomized controlled trial of itraconazole versus fluconazole for the prevention of fungal infections in patients withhaematological malignances. Br. J. Haemathol. 105: 901�911.

M u r r a y P.R., E.J. B a r o n, M.A. P f a l l e r, F.C. T e n o v e r and R.H. Yo l k e n. 1999. Manual of clinical microbiology.7th Edition, ASM Press, Washington 1184�1241.

P a g a n o L., A. A n t i o n o r i and A. A m m a s s i a r i. 1999. Retrospective study of candidemia in patients with haematologicalmalignances. Clinical features, risk factors and outcome of 76 episodes. Eur. J. Haematol. 63: 77�85.

P f a l l e r M.A. 1995. Epidemiology of candidiasis. J. Hosp. Inf. 30 (Suppl): 329�338.R i c h a r d s o n M.D. and M.H. K o k k i. 1999. New perspectives in the diagnosis of systemic fungal infections. Ann Med. 31:

327�335.R ü c h e l R. 1997. Clinical presentation of invasive Candida mycoses. Mycoses 40 (Suppl. 2): 17�20.S a f d a r A., V. C h a t u r v e d i, E.W. C r o s s, S. P a r k, E.M. B e r n a r d, D. A r m s t r o n g and D.S. P e r l i n. 2001.

Prospective study of Candida species in patients at a comprehensive cancer center. Antimicrob. Agents Chemother. 45:2129�213.

S e n d i d B., M. T a b a u r e t, J.L. P o i r o t, D. M a t h i e n, J. F r u i t and D. P a u l a i n. 1999. New enzyme immunoassaysfor sensitive detection of circulating Candida albicans mannan and antimannan antibodies: useful combined test for diagno-sis of systematic candidiasis. J. Clin. Microbiol. 37: 1510�1517.

S h i n J.H., F.S. N o l t e and C.J. M o r r i s o n. 1997. Rapid identification of Candida species in blood cultures by a clinicallyuseful PCR method. J. Clin. Microbiol. 36: 1454�1459.

S h i n J.H., F.S. N o l t e, B.P. H o l l o w a y and C.J. M o r r i s o n. 1999. Rapid identification of up to three Candida species ina single reaction tube by a 5' exonuclease assay using fluorescent DNA probes. J. Clin. Microbiol. 37: 165�170.

Ya m a m u r a D.L.R., C. R o t s t e i n, L.E. N i c o l l e and S. I o a n n o u. 1999. Candidemia at selected Canadian sites: resultsfrom the Fungal Disease Registry, 1992�1994. Can. Med. Assoc. J. 160: 493�499.

Ye o S.F. and B. W o n g. 2002. Current status of nonculture methods for diagnosis of invasive fungal infections. Clin. Microbiol.Rev. 15: 465�484.



Seroepidemiological Studies of Chlamydia pneumoniae Infectionsin 1� 36 Months Old Children

with Respiratory Tract Infections and Other Diseases in Poland

EDYTA PODSIAD£Y1*, BEATA FR¥CKA2, AGNIESZKA SZMIGIELSKA2

and STANIS£AWA TYLEWSKA-WIERZBANOWSKA1

1 National Institute of Hygiene, Chocimska 24, 00-791 Warsaw, Poland,2 Department of Pediatrics and Pediatric Nephrology, Medical University of Warsaw,

Marsza³kowska 24, 00-576 Warsaw, Poland

Received 20 April 2005, received in revised form 23 June 2005, accepted 11 July 2005

A b s t r a c t

Presence of specific IgM, IgG and IgA antibodies against Chlamydia pneumoniae was evaluated in children aged 1 weekto 36 months to investigate the role of C. pneumoniae in respiratory infections and other diseases. Serum samples wereobtained from 150 hospitalized children, including 123 children presenting the clinical symptoms of various respiratorytract infections, two children with acute diarrhoea, two children with meningitis, 14 children with urinary tract infection,and 9 children with non-infectious diseases. Levels of specific C. pneumoniae IgM, IgG and IgA serum antibodies weremeasured by enzyme-linked immunoassay (ELISA). C. pneumoniae IgM antibodies were detected in 16 (13.0 %) of123 children with respiratory tract infections. Specific IgG antibodies were found in sera of 11 children under 12 monthsold. Among 27 children without symptoms of a respiratory tract disease, specific C. pneumoniae IgM were found in two(7.4%) children, including one child with meningitis and another child with urinary tract infection. Specific IgA antibodieswere not found in any tested child. All cases of C. pneumoniae infections were identified within two calendar years outof eight that were analyzed, i.e. in 1997 and 2000. The incidence of C. pneumoniae infections varied seasonally, withmost children infected in autumn. C. pneumoniae IgM antibodies were detected more often in girls (17.9%) then inboys (7.2%). C. pneumoniae infections occur among small children in central Poland. The results of this study indicatethat C. pneumoniae may play a role in the etiology of respiratory tract infections in infants and young children.

K e y w o r d s: children, infants, respiratory tract infections, Chlamydia pneumoniae, antibodies

Introduction

Chlamydia pneumoniae is a human pathogen with a wide distribution all over the world. This pathogenis responsible for upper and lower respiratory tract infections such as pharyngitis, sinusitis, bronchitis andpneumonia and probably myocarditis, erythema nodosum and reactive arthritis. The list of diseases associa-ted with C. pneumoniae infection is growing. It is postulated that the microorganism is involved in thedevelopment of chronic diseases such as asthma, sarcoidosis, atherosclerosis, Guillain-Barre syndrome,Reiter�s syndrome, and Alzheimer disease (Dowel et al., 2001; Grayston, 2000; Kuo et al., 1995).

It has been recognized worldwide that 50% of adults have significant levels of specific serum antibodiesagainst C. pneumoniae (Kuo et al., 1995). Considerable data are available on C. pneumoniae respiratorytract infections in adults, but relatively little is known on the role of this pathogen in infants and youngchildren. Thus the role of C. pneumoniae infections in pediatric patients is uncertain. Few reports onthe association of C. pneumoniae with respiratory tract infections in children have been published so far.These studies have been performed in Gambia, Sudan, Thailand, the Philippines, Sweden, Finland, Germany,Switzerland, USA, and Chile (Forgie et al., 1991; Heiskanen-Kosma et al., 1999; Herrmann et al., 1994;Likitnukul et al., 2003; Lund-Olsen et al., 1994; Saikku et al., 1988; Tagle et al., 2000), with divergent andsometimes equivocal results.

* Corresponding author: Edyta Podsiadly, National Institute of Hygiene, Chocimska 24, 00-791 Warsaw, Poland; tel. + 48 22542 12 50; e-mail:[email protected]

216 Podsiad³y E. et al. 3

The aim of our study was to investigate the presence of specific C. pneumoniae IgM, IgG and IgAantibodies in Polish children with respiratory tract and other diseases, aged 1�36 months.

Experimental


Patients and specimens. A randomly selected group of 150 children aged from 1 week to 36 months admitted to the Depart-ment of Pediatrics and Pediatric Nephrology, Medical University of Warsaw in 1995�2003, with clinical symptoms of respiratorytract infections or other diseases was studied. The study group included 23 newborns (age 1�4 weeks), 75 infants (age 5 weeks-12 months) and 52 children age 13�36 months. Sixty-seven girls and 83 boys were enrolled. Patients originated from the Mazowszedistrict � in central part of Poland. Diagnoses included recurrent respiratory tract infections in 37 patients, pneumonia in 28 patients,bronchitis in 28 patients, pharyngitis in 24 patients, laryngitis in two patients, otitis media in four patients, acute diarrhoea in twopatients, purulent meningitis in two patients, and urinary tract infection in 14 patients. Nine children were diagnosed witha non-infectious disease, including dietary indiscretion, urinary tract malformation.

Chest roentgenogram was performed in all children with respiratory infections. A diagnosis of pneumonia was based on radio-logical findings including hyperinflation, prominent bronchovascular makings or diffuse interstitial and patchy alveolar infiltrates.

A diagnosis of bronchitis was based on clinical findings including wheezing, cough, dyspnea or tachypnea and normal chestradiograph. Children with more then six incidents of upper or lower respiratory tract infections within a year were categorized ashaving recurrent respiratory tract infections.

Blood samples were taken from each patient prior to administration of antibiotic treatment.Laboratory tests included: leukocyte count and differential white blood cell count, CRP and serological tests.Serology. Anti-chlamydial IgM, IgG and IgA antibodies were determined by enzyme-linked immunoassay (ELISA), with

a major outer membrane protein as an antigen. For IgM antibodies, Chlamydia pneumoniae IgM ELISA (Vircell, Granada, Spain)test was used. This test has 91% sensitivity and 98% specificity (Numazaki et al., 1996; Gutierrez et al., 2002). For IgG and IgA,ELEGANCE Chlamydia pneumoniae IgG & IgA ELISA (Bioclone, Sydney, Australia) tests were used. The ELEGANCE Chlamydiapneumoniae ELISA test has 71.7% sensitivity and 95.8% specificity for IgG and 73.9% sensitivity and 92.9% specificity for IgA(Kishimoto, 1990; Kishimoto, 1996;. Ekman, 1993). Presence of IgM antibodies was regarded as an indication of acute infection.The presence of C. pneumoniae IgG antibodies in children aged up to 12 months was assumed to be maternal origin.

Results

Among 150 tested children, C. pneumoniae IgM and IgG antibodies were detected in sera of 18 (12%)and 12 (8%), respectively.

In children with symptoms of a respiratory tract infection, specific serum C. pneumoniae IgM antibodieswere detected in 16 (13.0%) of 123 patients. Specific IgG antibodies were found only in sera of 12 childrenyounger then 12 months. Specific IgG antibodies were not detected in children with specific C. pneumoniaeIgM. Specific IgA antibodies were not found in any of 150 tested children.

Acute C. pneumoniae infection was detected in patients with various respiratory tract diseases (Table I).Most commonly, it was found in children with bronchitis � 32.1% (9/28), including 77.7% (7/9) of childrenwith wheezing. C. pneumoniae infections were also detected in 10.8% (4/37) of children with recurrentrespiratory tract infections, 8.3% (2/24) of children with pharyngitis, and in one child (out of two) withlaryngitis. C. pneumoniae infection was detected neither in patients with otitis media (0/4) nor in patientswith pneumonia (0/26). Among 27 children with a non-respiratory illness, specific serum C. pneumoniaeIgM antibodies were found in two (7.4%) patients, including one child with meningitis and another one withurinary tract infection. Specific IgM and IgG antibodies were not detected in children with non-infectionsdiseases and in children with acute diarrhoea (Table I).

Serological evidence of C. pneumoniae infection was found in 6.3% of infants, 13.5% of children aged1�2 years and in 46.7% of those aged 2�3 years. Specific C. pneumoniae antibodies were found morecommonly in girls (12/67; 17.9%) then in boys (6/83; 7.2%).

Only three children with C. pneumoniae infection presented fever, including a 10-month-old infant withpharyngitis, a 16-month-old child with bronchitis complicated by seizures and a 9-month-old infant withpurulent streptococcal meningitis. In children with acute C. pneumoniae infection, average leukocyte countin peripheral blood was 12 100/mm3 with range 7 000�27 000 (normal values 5 000�17 500/mm3 ). Leuko-cytosis of 27 000/mm3 was found in a 22-month-old girl with acute bronchitis. Differential white blood cellcount showed predominant lymphocytes from 51 to 80% in 8 children in this group, CRP level was slightlyelevated and averaged 0.6 with range 0.01�0.8 mg/dL (normal level <0.01 mg/dL). Among the remainingchildren with symptoms of a respiratory tract infection and no specific C. pneumoniae IgM antibodies,23 children presented fever, leukocyte count ranged from 5 000 to 24 000 and averaged 11 600/mm3, again

217C. pneumoniae infections in children3

with differential white blood cell count showing predominant lymphocytes from 51 to 75% in 33 children,and CRP averaged 0.5 mg/dL.

Serological testing revealed that all cases of C. pneumoniae infections occurred within only two out ofeight analyzed years (Figure 1). In 1997, specific C. pneumoniae antibodies were detected in 4 of 16 testedchildren (25.0 %), and in 2000, 14 (25.0%) of 56 examined children were seropositive. In contrast, negativeresults were obtained in serum samples collected in other years (1995, 1996, 1998, 1999, 2001 and 2002) inall 78 tested children. In addition, it appears that the incidence of C. pneumoniae infections varies seasonally.Cases were detected mostly in November (1997 � 2/4, 2000 � 9/14), October (2000 � 2/14) and December

Recurrent RTI * 37 4 0 4 (10.8)

Bronchitis 28 9 0 9 (32.1)

Pneumonia 28 0 6 0

Pharingitis 24 2 5 2 (8.3)

Otitis media 4 0 1 0

Laryngitis 2 1 0 1

Meningitis 2 1 0 1

UTI ** 14 1 0 1 (7.1)

Acuta diarrhoea 2 0 0 0

Non-infectious diseases 9 0 0 0

Table IPrevalence of C. pneumoniae serum antibodies in 1�36 months old children

with various diseases

DiseaseNo

testedNo seropositives# No (%) C. pneumoniae

infections***IgM IgG$

* RTI � respiratory tract infections, ** UTI � urinary tract infections, *** The presence of spe-cific C. pneumoniae IgM antibodies was considerd an acute infection, # Specific C. pneumoniaeIgA antibodies were not detected in any tested serum, $ Specific C. pneumoniae IgG antibodieswere only detected in children under 12 months of age and were assumed to be maternal origin.

0

5

10

15

20

25

30

1995 1996 1997 1998 1999 2000 2001 2002

years

%

April October November December Years

Fig. 1. C. pneumoniae infections by years and months* Specific C. pneumoniae antibodies were not detected in any child in January, February,

March, May, June, July , August and September


(1997� 2/4, 2000 � 2/14). Among children tested in April, there was one child with C. pneumoniae infec-tion. Specific C. pneumoniae antibodies were not found in serum samples collected in January, February,March, May, June, July, August and September.

Discussion

Among children younger than 3 years old with respiratory tract diseases, acute C. pneumoniae infectionswere detected in 13.0% of them. The youngest patient with C. pneumoniae infection detected serologicallywas one-month-old. The frequency of C. pneumoniae etiology in respiratory tract infections in pediatricpopulations varies from 0% up to more than 18% (Likitnukul et al., 2003; Lund-Olsen et al., 1994; Saikkuet al., 1988; Tagle et al., 2000; Hammerschlag et al., 2003; Schmidt et al., 2002; Baer et al., 2003; Chirgwinet al., 1991; Block et al., 1995). These differences are probably related to the population studied and thepresence or absence of local outbreaks in the respective communities during the investigation period as wellas to applied specific diagnostic procedures (Hammerschlag et al., 2003).

It is well known that C. pneumoniae infections spread as long-lasting epidemics that occur at irregularintervals of several years� duration (Grayston, 2000; Kuo et al., 1995). There is a paucity of data on respira-tory tract infections caused by C. pneumoniae in children in Poland. Similarly, there are no data onC. pneumoniae epidemics in Poland. Our results may suggest that such epidemics occurred in 1997 and2000. Additionally, a seasonal trend for C. pneumoniae cases throughout a year was observed, with a peakof incidence in autumn. These results are in accordance with studies by T. Heiskanen-Kosma et al. whoobserved incidence peak of C. pneumoniae infections in October (Heiskanen-Kosma et al., 1998).

Specific C. pneumoniae antibodies have been detected most frequently among children aged 2�3 years.Serological evidence of C. pneumoniae infection was found in 46.7% children in this age group. Majorityof cases were associated with recurrent respiratory tract infections. C. pneumoniae antibodies were alsodetected in children without symptoms of respiratory track diseases: a 9-month-old girl with meningitis andin a 3-month-old boy with a urinary tract infection. Detection of specific C. pneumoniae antibodies, inparticular IgM class, in children without respiratory tract infections suggest the occurrence of asymptomaticchlamydial infections in early childhood. In Finland, it was shown that 4.0 to 6.0% of 2�4 years old healthychildren were seropositive. It confirms probability of the occurrence of asymptomatic C. pneumoniae infec-tions in this age group (Tuuminen et al., 2000).

The significance of asymptomatic C. pneumoniae infections and their sequelae requires further investi-gations. In adult populations, C. pneumoniae antibodies are present with significantly higher frequency inmen then in women. This is, however, at variance with our results in a group of children aged 1�36 months.C. pneumoniae antibodies were detected more often in girls then in boys.

These results are in accordance with a study by Lin et al. who found higher prevalence of C. pneumoniaeantibodies in females then in males among subjects aged 6 months to 20 years (Lin et al., 2004).

Most of the seroepidemiological studies on C. pneumoniae infections in children were performed withthe MIF method. According to the recent recommendations of Centers for Disease Control and Prevention(USA) and the Laboratory Centre for Disease Control (Canada) (Dowell et al., 2001) this technique isconsidered a reference method, although there are studies showing low sensitivity of this method in children(Kutlin et al., 1998).

Simultaneous application of MIF and culture in the examination of children shows that MIF is nota sensitive technique for detecting C. pneumoniae antibodies in children under 5 years old. Since positiveculture results were achieved in seronegative children. According to some studies only 20% to 30% ofchildren with culture-documented C. pneumoniae infection had antibodies detectable by MIF (Schmidt et al.,2002; Hyman et al., 1995; Emre et al., 1994). Kutlin et al. found that over 80% of culture-positive andMIF-negative children had antibodies to C. pneumoniae detected by immunoblotting (Kutlin et al., 1998).

The lack of standardized methods, including serology and PCR, makes it difficult to perform a specificmicrobiologic diagnosis. Although there is currently no validated diagnostic marker for C. pneumoniaeinfections, and the value of the EIA method in diagnosis C. pneumoniae infections is an open issue,detection of C. pneumoniae IgM antibodies in several children below 3 years of age may suggest thatC. pneumoniae infections occur in early childhood. Moreover, frequent detection of C. pneumoniae anti-bodies in children with respiratory tract infections suggests that the bacterium may be an etiological agentof respiratory diseases in infants and young children.

Acknowledgement. The studies were supported by a research grant KBN 4PO5D 038 17.

219C. pneumoniae infections in children3

Literature

B a e r G., G. E n g e l c k e, M. A b e l e - H o r n, U.B. S c h a a d and U. H e i n i n g e r. 2003. Role of Chlamydia pneumoniaeand Mycoplasma pneumoniae as causative agents of community-acquired pneumonia in hospitalized children and adolescents.Eur. J. Clin. Microbiol. Infect. Dis. 22: 742�745.

B l o c k S., J. H e d r i c k, M.R. H a m m e r s c h l a g, G.H. C a s s e l and J.C. C r a f t. 1995. Mycoplasma pneumoniae andChlamydia pneumoniae in pediatric community-acquired pneumonia: comparative efficacy and safety of claritromycin vs.erythromycin ethylsuccinate. Pediatr. Infec. Dis. J. 14: 471�477.

C h i r g w i n K., P.M. R o b l i n, M. G e l l i n g, M.R. H a m m e r s c h l a g and J. S c h a c h e r. 1991. Infection with Chlamy-dia pneumoniae in Brooklyn. J. Infect. Dis. 163: 757�761.

D o w e l l S.F., R.W. P e e l i n g, J. B o m a n, G.M. C a r l o n e, B.S. F i e l d s, J. G u w n e r, M.R. H a m m e r s c h l a g,L.A. J a c k s o n, C.C. K u o, M. M a a s s, T.O. M e s s m e r, D.F. T a l k i n g t o n, M.L. T o n d e l l a and S.R. Z a k i.2001. Standardizing Chlamydia pneumoniae assays: recommendations from the Centers for Disease Control and Prevention(USA) and the Laboratory Centre for Disease Control (Canada). Clin. Infect. Dis. 33: 492�503.

E k m a n M. 1993. Evaluation of serological methods in the diagnosis of Chlamydia pneumoniae during an epidemic in Finland.Eur. J. Clin. Microbiol. Infec. Dis. 12: 756�760

E m r e U., P.M. R o b l i n and M. G e l l i n g. 1994. The association of Chlamydia pneumoniae infection and reactive airwaydisease in children. Arch. Pediatr. Adolesc. Med. 148: 727�729.

F o r g i e I.M., K.P. O� N e i l l, N. L l o y d - E v a n s, M. L e i n o n e n, H. C a m p b e l l and H.C. W h i t t l e. 1991. Etiologyof acute lower respiratory tract infections in Gambian children: II. Acute lower respiratory tract infection in children agesone to nine years presenting at hospital. Pediatr. Infect. Dis. J. 10: 42�47.

G r a y s t o n J.T. 2000. Background and current knowledge of Chlamydia pneumoniae and atherosclerosis. J. Infect. Dis. 181(suppl 3): S402�411.

G u t i e r r e z J., J. M e n d o z a, F. F e r n a n d e z, J. L i n a r e s - P a l o m i n o, M.J. S o t o and M.C. M a r o t o. 2002. ELISAtest to detect Chlamydophila pneumoniae IgG. J. Basic Microbiol. 42: 13�18.

H a m m e r s c h l a g M.R. 2003. Pneumonia due to Chlamydia pneumoniae in children: epidemiology, diagnosis, and treatment.Pediatr. Pulmonol. 36: 384�390.

H e i s k a n e n - K o s m a T., M. K o r p p i, A. L a u r i l a, C. J o k i n e n, M. K l e e m o l a and P. S a i k k u. 1999. Chlamydiapneumoniae is an important cause of community-acquired pneumonia in school-aged children: serological results of a pro-spective, population-based study. Scan. J. Infect. Dis. 31: 255�259.

H e i s k a n e n - K o s m a T., M. K o r p p i, C. J o k i n e n, S. K u r k i, L. H e i s k a n e n, H. J u v o n e n, S. K a l l i n e n,M. S t e n, A. T a r k i a i n e n, P.R. R o n n b e r g, M. K l e e m o l a, H. M a k e l a and M. L e i n o n e n. 1998. Etiologyof childhood pneumonia: serologic results of a prospective, population-based study. Pediatr. Infec. Dis. J. 17: 986 �991.

H e r r m a n n B., M.A. S a l i h, B.E. Yo u s i f, O. A b d e l w a h a b and P.A. M a r d h. 1994. Chlamydial etiology of acutelower respiratory tract infections in children in Sudan. Acta Paediatr. 83: 169�172.

H y m a n C.L., P.M. R o b l i n, C.A. G a y d o s, T.C. Q u i n n, J. S c h a c h e r and M.R. H a m m e r s c h l a g. 1995. Preva-lence of asymptomatic nasopharyngeal carriage of Chlamydia pneumoniae in subjectively healthy adults: assessment bypolymerase chain reaction-enzyme immunoassay and culture. Clin. Infect. Dis. 20: 1174�1178.

K i s h i m o t o H. 1990. Research into the comparison between healthy adults and those suffering from C. pneumoniae (TWAR)acute respiratory infection through serum based epidemiological examination. J. Infect. Dis. 64: 986�992.

K i s h i m o t o H. 1996. The foundation and clinical serum based diagnosis of Chlamydia pneumoniae. Sphere Scien. Cure 12:2250�2256.

K u o C.C, L.A. J a c k s o n and L.A. C a m p b e l l. 1995. Chlamydia pneumoniae (TWAR). Clin. Microbiol. Rev. 8: 451�461.K u t l i n A., P.M. R o b l i n and M.R. H a m m e r s c h l a g. 1998. Antibody response to Chlamydia pneumoniae infection in

children with respiratory illness. J. Infect. Dis. 177: 720�724.L i k i t n u k u l S., P. N u n t h a p i s u d and N. P r a p p h a l. 2003. Prevalence of Chlamydia pneumoniae infection in Thai

children with community-acquired pneumonia. Pediatr. Infect. Dis. J. 22:749�750.L i n T.M., C.C. K u o, W.J. C h e n, F.J. L i n and H.L. E n g. 2004. Seroprevalence of Chlamydia pneumoniae infection in

Taiwan. J. Infect. 48:91�95.L u n d - O l s e n I., A. L u n d b a c k, J. G n a r p e and H. G n a r p e. 1994. Prevalence of specific antibodies to Chlamydia

pneumoniae in children with acute respiratory infections. Acta Paediatr. 83: 1143�1145. Numazaki K., T. Ikebe and S. Chiba. 1996. Detection of serum antibodies against Chlamydia pneumoniae by ELISA. FEMS

Immunol. Med. Microbiol. 14: 179�183.S a i k k u P., P. R u u t u, M. L e i n o n e n, J. P a n e l i u s, T.E. T u p a s i and J.T. G r a y s t o n. 1988. Acute lower-respira-

tory-tract infection associated with chlamydial TWAR antibody in Filipino children. J. Infect. Dis. 158: 1095�1097.S c h m i d t S.M., C.E. M u l l e r, B. M a h n e r and S.K. W i e r s b i t z k y. 2002. Prevalence, rate of persistence and respiratory

tract symptoms of Chlamydia pneumoniae infection in 1211 kindergarten and school age children. Pediatr. Infect. Dis. J.21: 758�762.

T a g l e M., R. K o g a n, P. R o j a s, L. R u b i l a r, R. V i d a l and E. P a y a. 2000. Diagnosis of Chlamydia pneumoniae incommunity-acquired pneumonia in children in Chile. Acta Paediatr. 89: 650�653.

T u u m i n e n T., S. Va r j o, H. I n g m a n, T. W e b e r, J. O k s i and M. V i l j a n e n. 2000. Prevalence of Chlamydiapneumoniae and Mycoplasma pneumoniae immunoglobulin G and A antibodies in a healthy Finnish population as analyzedby quantitative enzyme immunoassays. Clin. Diagn. Lab. Immunol. 7: 734�738.



Occurrence of Serum Class G Immunoglobulins Interactingwith Specific Antigens of Helicobacter pylori in Patients

with Unstable Coronary Artery Disease and in Symptomless Individuals

TOMASZ RECHCIÑSKI1, ANETA GRÊBOWSKA2, MA£GORZATA KURPESA1,WIES£AWA RUDNICKA2, MARIA KRZEMIÑSKA-PAKU£A1 and MAGDALENA CHMIELA2

1 Second Chair and Clinic of Cardiology, Medical University of £ód�, ul. Kniaziewicza 1/5, 91-347 £ód�2 Department of Infectious Biology, University of £ód�, ul. Banacha 12, 90-237 £ód�, Poland

Received 7 January 2005, received in revised form 25 February 2005, accepted 28 June 2005

A b s t r a c t

An impact of Helicobacter pylori on the process of atherogenesis may be related to the intensity of humoral responseagainst selected specific antigens of this bacteria. We performed serological studies in which the recognition of 7 selectedantigens was possible. The investigated group consisted of 56 patients hospitalized due to unstable angina pectoris. Thecontrol group consisted of 29 symptomless volunteers. The levels of class G serum immunoglobulins interacting withglycine extract (GE) of H. pylori antigens were assessed by ELISA test in both groups. The same sera were tested bythe Milenia blot H. pylori IgG system. In this assessment the presence of IgG antibodies interacting with antigens ofmolecular weight of 120, 87, 64, 35, 30, 26, and 20 kDa was estimated separately for every listed antigen. The resultsrevealed significant differences between investigated groups in the prevalence of anti-GE IgG (unstable angina � 100%vs. controls � 60%) and in the level of such antibodies expressed as total optical density units � OD450 (6.1 ± 3.0 vs.3.4 ± 3.0 respectively, p<0.05). However, anti-GE IgG detected in the sera of patients as well as controls reacted withsimilar frequency with selected H. pylori antigens: highly specific (120, 87, 64, 30 kDa) and specific (35, 26, and20 kDa). We conclude, that although H. pylori infection is so common and mainly associated with gastroduodenalsymptoms, it is also recognized by serological methods with high prevalence in patients with coronary artery disease,and less frequently in symptomless individuals. The humoral response against H. pylori in class G immunoglobulinsin patients with unstable angina is characterized by higher levels of anti-H. pylori IgG but not by the higher prevalenceof serum IgG interactions with the highly specific and specific H. pylori antigens. Such infection could be considered asa cofactor for atherogenesis by inducing strong humoral response against surface antigens of this bacteria.

K e y w o r d s: immunity, H. pylori, atherosclerosis, unstable angina

Introduction

Helicobacter pylori is one of the most common pathogens of humans causing a lifelong infection ofgastric mucosa in >50% of the global population. This infection has been proved to be an important patho-genic factor for some gastro-duodenal diseases like type B gastritis, ulcer disease, mucosa-associated lym-phoid tissue lymphoma and gastric adenocarcinoma (Rosenstock et al., 1997; Howden et al., 1996). Recentstudies point to a possible role of H. pylori in extradigesive diseases, like Sjörgrene syndrome, thyroiddiseases or Schönlein-Henoch purpura (Gasbarini et al., 1998). Also vascular diseases like Raynaud phe-nomenon or headaches are considered to be linked with this infection. Although human atherosclerosis isa multifactoral disease, some possible links between this chronic process, infections and inflammation havebeen emphasized (Watanabe et al.,1996). A relationship between H. pylori seropositivity and atherosclero-sis in general and coronary heart disease in particular has been reported in mid 1990s (Mendall et al., 1994;Patel et al., 1995); however the role of this infection in atherogenesis is still controversial � in some largestudies, there was no correlation shown between H.pylori, cardiovascular risk factors and ischaemic heart

Corresponding author: T. Rechciñski, MD, 2nd Chair and Clinic of Cardiology, Medical University of £ód�, Biegañski Hospital,Kniaziewicza 1/5, 91-347 £ód�, Poland, e-mail: [email protected]

222 Rechciñski T. et al. 3

disease (Murray et al., 1995; Ridker et al., 2001). Since then, the methods of serological detection ofH. pylori have become more sophisticated and accurate. This is a problem of great importance, as thisbacteria shares a number of antigens with other microorganisms, giving false-positive cross-reactions inserological studies (Johansen et al., 1995; Paziak-Domañska et al., 2000). In the recent studies it was foundthat patients with unstable angina demonstrated enhanced humoral response to a glycine acid extract ofH. pylori (Rechciñski et al., 2002). Glycine extract is a mixture of H. pylori surface antigens whose molecu-lar weight ranges from 14 to 120 kDa. It is to be emphasized that in previous studies, the composition ofcirculating IgGs in respect to different specific and highly specific H. pylori antigens in patients with anginapectoris was not estimated or compared with symptomless individuals. In our present study, we intended tofind out whether the enhancement of humoral response was generalized or rather limited to the immuno-globulins of a strictly limited idiotype (antigenic specificity). The aim of this study was also to find out howprevalent the patients with unstable angina are when infected by more virulent strain of H. pylori, charac-terized by the presence of cytotoxin-associated protein A (CagA), highly specific for this bacteria.

Experimental

Material and Methods

Fifty-six males (age 30�65 years) hospitalized due to chest pains in the Cardiology Department were included in the study afteran unstable angina pectoris according to Braunwald definition was diagnosed. Coronary angiography confirmed the atheromaticbackground of symptoms (Braunwald, 1989). In this group, the prevalence of previous myocardial infarction was 52%, arterialhypertension � 72%, diabetes mellitus � 18%, hyperlipidemia � 62%, nicotinism � 18%, rate of the revascularisation (percutaneoustransluminal coronary angioplasty or coronary artery by-pass graft) � 58%. The control group consisted of 29 symptomlessage-matched males. All controls had negative history of ischaemic heart disease or gastric symptoms. The comparison of preva-lence of risk factors of atherosclerosis in investigated group and controls is presented in Table I. Table II shows the pharmacologicaltreatment obtained by patients and controls prior to blood sampling.

ELISA. Blood from antecubital vein was obtained from all study participants for serological tests. The sera were stored at atemperature of �70°C. The level of anti-H.pylori IgG was estimated using ELISA method with GE of H. pylori CCUG 17874 as thecoating antigen and with rabbit anti-human IgG antibodies labeled with horseradish peroxidase HRP (Dako, Glostrup, Denmark).The serum samples for anti-GE IgG ELISA were diluted from 1:500 to 1:128000. The results were expressed as total optical density

values measured in 450 nm wave length. The details of this assay are described elsewhere (Rechciñski et al., 1997). The seradiluted 1:500 were also tested for optical density (OD450) of IgG reacting with the recombinant cytotoxin-associated proteinA CagA using ELISA (A. Covacci, IRS Siena, Italy).

Another method used in this study to assess the presence of IgG antibodies against four highly specific antigens and threespecific antigens of H. pylori was Milenia blot H. pylori IgG test (DPC Biermann GmbH, Bad Nauheim, Germany). The antigensinvestigated in this kit were: cytotoxin-associated protein A (CagA) of molecular mass 120 kDa, vacuolising cytotoxin A (VacA)� 87 kDa, urease subunit A (UreA) � 30 kDa, urease subunit B (UreB) � 64 kDa, as well as antigens of molecular weight of 35, 26,and 20 kDa. The combination of positive results for separate antigens made it possible to diagnose H. pylori infection, as recom-mended by the manufacturer.

Statistical methods. Chi2 test was used to assess the significance of differences in results observed between the studied groups.

Results

The IgG antibodies reacting with GE of H. pylori antigens were detected by ELISA in all patients withunstable angina and in 60% of symptomless individuals (p<0.05). The level of anti-GE IgG expressed astotal optical density for the wave length of 450 nm (OD total) ranged in the group of patients with unstable

Table IThe comparison of risk factors of atherosclerosis

in investigated group and controls

Table IIComparison of pharmacological treatment in patients

with unstable angina and controls

Diabetes mellitus 18% 3%

Arterial hypertension 72% 13%

Hypercholesterolaemia 62% 12%

Nicotinism 18% 33%

Risk factor Unstable anginapectoris

Controls

Aspirin 100% 6%

Statins 60% 0

Beta-blockers 90% 0

ACE* � inhibitors 74% 0

Compound Unstable anginapectoris

Controls

*ACE � angiotensin converting enzyme

223H. pylori IgG system and atherogenesis3

angina from 0.585 to 13.922 (mean 6.1 ± 3.0) and in the controls � from 0.422 to 8.308 (mean 3.4 ± 3.0)� p<0.05 (Table III). Using Milenia blot, 75% of cardiac patients, and 48% of controls were H. pyloripositive � p<0.05. The prevalence of anti-CagA IgG assessed by ELISA was observed in 66% of cardiacpatients and in 70% of controls � with the difference being nonsignifficant � p>0.05. The total opticaldensity in ELISA test for anti-CagA ranged in the group of patients with unstable angina from 0.09 to 2.489(mean 0.8 ± 0.70), in controls � from 0.067 to 2.485 (mean 0.6 ± 0.7) and did not vary significantly betweentwo groups (Table IV).

Despite the significant differences between both studied groups in the prevalence of H. pylori infectiondetected by ELISA or Milenia blot, no statistical difference was found in the occurrence of IgG antibodiesagainst selected highly specific antigens when they were analyzed separately (Table V). Antibodies inter-acting with CagA were detected by Milenia blot in 59% patients with unstable angina and in 48% controls(p>0.05), the prevalence of IgG interacting with VacA was 57% and 38%, respectively (p>0.05). Also IgGsinteracting with two subunits of urease were detected with a similar rate in both groups � 71% vs. 48% forUreA (p > 0.05) and 50% vs. 38% for UreB (p>0.05), respectively. Also, the rate of detection of IgG anti-bodies against specific 35 kDa protein did not vary significantly between the compared groups: 21% vs.14%, respectively (p>0.05). Antibodies of IgG class against protein of molecular weight of 26 kDa were

CagA 120kDa 59 � (80) 48 � (100)

VacA 87kDa 57 � (76) 38 � (79)

ure A 30kDa 71 � (95) 48 � (100)

ure B 64kDa 50 � (67) 38 � (79)

35kDa 21 � (29) 14 � (29)

26kDa 71* � (95) 45* � (93)

20kDa 73* � (98) 41* � (96)

Table VThe prevalence of serum IgG interactions with selected H. pylori antigens detected by Milenia blotH. pylori IgG system. Analysis performed for all subjects in compared groups and restricted only to

H. pylori-seropositive individuals in both groups

AntigensPercent of positive serum interactions with selected H. pylori antigens detected by Milenia blot

Angina pectoris all subjects � (seropositive ones) Controls all subjects � (seropositive ones)

* � significance p< 0.05

Angina pectoris 0.585 � 13.922 6.1 ± 3.0* 0.09 � 2.489 0.8 ± 0.7

Controls 0.422 � 8.308 3.4 ± 3.0* 0.06 �2.485 0.6 ± 0.7

Table IVThe levels of anti-H. pylori GE IgG and anti-CagA IgG expressed as optical density

* � significance p< 0.05

ELISA

Group total OD 450 1:500�1:128000 OD 450 1:500

IgG anti-GE range IgG anti-GE mean IgG anti-CagA range IgG anti-CagA mean

Angina pectoris 100* 66 75+ 59

Controls 60* 70 48+ 48

Table IIIThe prevalence of anti-H. pylori GE IgG and anti-CagA IgG assessed by ELISA or Milenia blot

Group

Percent of positive results

ELISA Milenia blot

GE 14-120kDa IgG CagA IgG 20-120kDa IgG CagA IgG

* and + � significance p< 0.05


detected in 71% of patients with unstable angina and in 45% of controls (p<0.05), and the antibodiesagainst 20 kDa protein were found in 73% vs. 41% (p<0.05), respectively. However, when the analysis ofthe occurrence of IgG antibodies specific for 26 and 20 kDa proteins was only limited to H. pylori-positiveindividuals in both groups (Milenia blot), the prevalence of such antibodies ranged from 86 to 98%, andthere were no statistical differences between these subgroups.

Discussion

The differentiation of various antigens interacting with the sera of the investigated humans made theserodiagnosis of H. pylori infection more reliable (Faulde et al., 1993; Nilsson et al., 1997), but it was notused in clinical practice due to the fact that this method seemed to be expensive and laborious. The results ofour studies concerning the similar prevalence of anti-CagA IgG in patients with coronary artery disease incomparison with controls, remain in accordance with previous reports (Koenig et al., 1999; Whincup et al.,2000). Such high prevalence of this type of antibodies even in the group of healthy individuals, may bejustified by the asymptomatic H. pylori infections. In our previous study, the frequency of asymptomaticinfections in healthy donors ranged from 50 to 60% as detected by 13C-urea breath test (Wi�niewska et al.,2002). Another possible explanation of this phenomenon is antigenic mimicry with host proteins detectablein both normal and atherosclerotic arteries tissues (Franceschi et al., 2002). In previous studies it was alreadysuggested that generalized elevated levels of IgG, IgA, IgE (but not IgM), help to predict first myocardialinfarction or sudden cardiac death in population of healthy men with abnormalities in the lipid profile(Kovanen et al., 1998), but the specificity of these immunoglobulins was not investigated. In our studies itwas confirmed that patients with the documented coronary atherosclerosis in comparison with symptomlessindividuals had higher levels of IgG reacting with H. pylori antigens. Previously, we showed that the highesttiters of anti-GE IgG were even more prevalent in the patients with atherosclerosis than in H. pylori-infected patients with gastroduodenal symptoms (Rechciñski et al., 2002; Chmiela et al., 2003). The tendencyof the patients with coronary artery disease to intensively produce IgG to various antigens was considered.However, we could not see any significant difference in the production of IgG to mycobacterial Hsp65among such patients, patients with active tuberculosis, dyspeptic patients and healthy controls (Chmielaet al., 2003). In the present study we did not find any significant difference in the prevalence of serum IgGinteraction with H. pylori highly specific and specific antigens between the group of coronary atherosclero-sis and controls. Obviously, we are not free to assume the absence of atheromatic plaques in coronaryvessels of symptomless individuals in the control group, since recently it has been proved that coronaryatherosclerosis is a process which begins already in asymptomatic teenagers (Tuzcu et al., 2001). Neverthe-less, according to the clinical interview, the plaques in coronary arteries of these individuals, if present,were stable. The role of immunoglobulins in the process of initiation or destabilization of atheromatouslesions may be elucidated by recent studies of Sims et al., 2001. These authors found in their immunohisto-logical studies, that gamma globulin leakage from the lumen into arterial wall is observed when arteriesexhibit a breakdown of sub-endothelial lamina, which results in subsequent lipids and inflammatory cellsentry. Class G immunoglobulins seem to have the best properties to participate in this process, as they havethe smallest sizes and the longest clearance in comparison with IgA and IgM. One cannot rule out the factthat also the size of the reacting antigen may be important for this phenomenon. Interestingly enough, it waspreviously found, that the smallest antigens of investigated bacteria are recognized by serum antibodiespredominantly in the organisms of infected children, in contrast with the infected adults, whose organismsused to recognize mainly the larger bacterial proteins (Chmiela et al., 1998). Although we did not observesuch a phenomenon in the patients with atherosclerosis or in symptomless subjects, it is possible thatgastrotoxic activity of aspirin used for treating the patients with coronary artery disease may lead to thefailure of mucosal barrier and an easier penetration of some bacterial antigens to the circulatory system andthe induction of stronger humoral response.

We conclude, that class G immunoglobulins interacting with H. pylori surface antigens, and circulatingin blood vessels of the patients with unstable angina pectoris, can be characterized by higher concentrationthan in symptomless individuals, and by a similar specificity. The strong humoral response against H. pyloriantigens may play a role in the maintenance of coronary heart disease.

Acknowledgements. We thank Dr. Antonello Covacci (IRIS, Siena, Italy) for providing us with flCagA protein and Mrs.Aleksandra Siwicka for her linguistic assistance. This study was supported by grant from KBN 3P05E04524

225H. pylori IgG system and atherogenesis3

Literature

B r a u n w a l d E. 1989. Unstable angina: a classification. Circulation 80: 410�414.C h m i e l a M., M. £ a w n i k, E. C z k w i a n i a n c, T. R e c h c i ñ s k i, I. P ³ a n e t a - M a ³ e c k a and W. R u d n i c k a. 1998.

Systemic humoral response to Helicobacter pylori in children and adults. Archivum Immunologiae et Therapiae Experimen-talis 4: 161�167.

C h m i e l a M., M. K o w a l e w i c z - K u l b a t, A. M i s z c z a k, M. W i � n i e w s k a, T. R e c h c i ñ s k i, K. K o ³ o d z i e j,J.D. K a s p r z a k, T. W a d s t r o m and W. R u d n i c k a. 2003. A link between Helicobacter pylori and/or Chlamydiaspp. infections and atherosclerosis. FEMS Immunol. Med. Microbiol. 3: 187�192.

F a u l d e M., J. C r e m e r and L. Z ö l l e r. 1993. Humoral immune response against Helicobacter pylori as determined byimmunoblot. Electrophoresis 14: 945�951

F r a n c e s c h i F., A. S e p u l v e d a, A. G a s b a r i n i, P. P o l a, N.G. S i l v e r i, G. G a s b a r i n i, D.Y. G r a h a m andR.M. G e n t a. 2002. Cross-reactivity of anti-CagA antibodies with vascular wall antigens; possible pathogenic link betweenHelicobacter pylori infection and atherosclerosis. Circulation 106: 430�434.

G a s b a r i n i A., F. F r a n c e s c h i, G. C a m m a r o t a, P. P o l a and G. G a s b a r i n i. 1998. Vascular and immunologicaldisorders associated with Helicobacter pylori infection. Ital. J. Gastroenterol. Hepatol. 30: 115�118.

H o w d e n C.W. 1996. Clinical Expressions of Helicobacter pylori Infection. Am. J. Med. 100: 275�345.J o h a n s e n H.K., A. N o r g a a r d, L.P. A n d e r s e n, P. J e n s e n, H. N i e l s e n and H. H o i b y. 1995. Cross-reactive

antigens shared by Pseudomonas aeruginosa, Helicobacter pylori, Campylobacter jejuni and Haemophilus influenze maycause false-positive titers of antibodies to Helicobacter pylori. Clin. Diagn. Lab. Immunol. 2: 149�155.

K o e n i g W., D. R o t h e n b a c h e r, A. H o f f m e i s t e r, M. M i l l e r, G. B o d e, G. A d l e r, V. H o m b a c h, W. M ä r z,M.B. P e p y s and H. B r e n n e r. 1999. Infection with Helicobacter pylori is not a major independent risk factor for stablecoronary disease. lack of a role of cytotoxin-associated protein A � positive strains and absence of a systemic inflammatoryresponse. Circulation 100: 2326�2331.

K o v a n e n P.T., T. M ä n t ä r i, T. P a l o s u o, V. M a n i n e n and K. A h o. 1998. Prediction of myocardial infarction indyslipidemic men by elevated levels of immunoglobulins classes A, E, and G, but not M.. Arch. Inter. Med. 158: 1434�1439.

M e n d a l l M., P. G o g g i n, M. M o l i n e a u x, Y. L e v y, T. T o o s y, D. S t r a c h a m and A.J. C a m m. 1994. Relation ofHelicobacter pylori infection and coronary heart disease. Br. Heart J. 71: 437�439.

M u r r a y L.J., K.B. B a r n f o r d, O. R e i l l y P.J.D., E.E. M c C r u m and A.E. E v a n s. 1995. Helicobacter pylori infec-tion: relation with coronary heart disease and cardiovascular risk factors. Br. Heart J. 74: 497�501.

N i l s s o n I., Å. L j u n g h, P. A l e j u n g and T. W a d s t r ö m. 1997. Immunoblot assay for serodiagnosis of Helicobacterpylori infections. J. Clin. Microbiol. 35: 427�432.

P a t e l P., M.A. M e n d a l l, D. C a r r i n g t o n, D. S t r a c h a n, E. L e a t h a m, N. M o u l i n e a u x, J. L e v y, C. B l a k e s t o n,C.A. S e y m o u r, A.J. C a m m and T.C. N o r t h f i e l d. 1995. Association of Helicobacter pylori and Chlamydiapneumoniae infections with coronary heart disease and cardiovascular risk factors. BMJ. 311: 711�714.

P a z i a k - D o m a ñ s k a B., M. C h m i e l a, A. J a r o s i ñ s k a, F.A. M a j e e d, E. C z k w i a n i a n c, I. P ³ a n e t a - M a ³ e c k aand W. R u d n i c k a. 2000. The importance of cross-reactions in the interpretation of ELISA tests in serodiagnosis ofH. pylori infections. Pediatria Wspó³czesna. Gastroenterologia, Hepatologia i ¯ywienie Dziecka 2: 1�4.

R e c h c i ñ s k i T., M. C h m i e l a, E. M a ³ e c k a - P a n a s, I. P ³ a n e t a - M a ³ e c k a and W. R u d n i c k a. 1997. Serologi-cal indicators of Helicobacter pylori infection in adult dyspeptic patients and healthy blood donors. Microbiol. Immunol. 40:387�393.

R e c h c i ñ s k i T., J.D. K a s p r z a k, M. C h m i e l a, M. K r z e m i ñ s k a - P a k u ³ a and W. R u d n i c k a. 2002. Patientswith unstable angina pectoris present increased humoral response against Helicobacter pylori in comparison with patientswith aggravated dyspepsia. Acta Microbiol. Pol. 51: 339� 344.

R i d k e r P.M., J. D a n e s h, L. Yo u n g m a n, R. C o l l i n s, M. S t a m p t e r, R. P e t o and C.H. H e n n e k e n s. 2001.A prospective study of Helicobacter pylori seropositivity and the risk for future myocardial infarction among socioeconomi-cally similar U.S. men. Ann. Intern. Med. 135: 184�188.

R o s e n s t o c k S., L. K a y, C. R o s e n s t o c k, L.P. A n d e r s e n, O. B o n n e v i e and T. J o r g e n s e n. 1997. Relationbetween Helicobacter pylori infection and gastrointestinal symptoms and syndromes. Gut 41: 169�176.

S i m s F.H., J.B. G a v i n, S. E d g a r and T. K o e l m e y e r. 2001. Diffusion of gamma globulin into arterial wall identifieslocalized entry of lipid and cells in atherosclerosis. Coron. Art. Dis. 12: 21�30.

T u z c u E.M., S.R. K a p a d i a, E. T u t a r, K.M. Z i a d a, R.E. H o b b s, P.M. M c C a r t h y, J.B. Yo u n g and S.E. N i s s e n.2001. High prevalence of coronary atherosclerosis in asymptomatic teenagers and young adults. Circulation 103: 2705�2710.

W a t a n a b e T., S. H a r a o k a and T. S h i m a o k a. 1996. Inflammatory and immunological nature of atherosclerosis. Int.J. Cardiol. 54: S25�S34.

W h i n c u p P., J. D a n e s h, M. W a l k e r, L. L e n n o n, A. T h o m s o n, P. A p p l e b y, Ch. H e w k e y and J. A t h e r t o n.2000. Prospective study of potentially virulent strains of Helicobacter pylori and coronary heart disease in middle-aged men.Circulation 101: 1647�1652.

W i � n i e w s k a M., H.O. N i l s s o n, L. B ¹ k - R o m a n i s z y n, T. R e c h c i ñ s k i, W. B i e l a ñ s k i, I. P ³ a n e t a -M a ³ e c k a, M. P ³ o n k a, S. K o n t u r e k, T. W a d s t r o m, W. R u d n i c k a and M. C h m i e l a. 2002. Detection ofspecific Helicobacter pylori DNA and antigens in stool samples in dyspeptic patients and healthy subjects. Microbiol.Immunol. 46: 657�665.



Enzyme Production and Biotypes of Vaginal Candida albicans

ZEFIRYN CYBULSKI1, EL¯BIETA KRZEMIÑSKA-JA�KOWIAK, PRZEMYS£AW MAJEWSKI2,JERZY CHYLAK3 and MAREK PAWLIK4

1 Wielkopolska Cancer Center, 2 Department of Clinical Pathomorphology,3 Department of Medical Microbiology, University of Medical Sciences, Poznañ,

4 Central Hospital, Lutycka str., Poznañ, Poland

Received 28 December 2004, received in revised form 4 March 2005, accepted 17 June 2005

A b s t r a c t

Candidial vulvovaginitis is one of the most common forms of vaginal infection. However, the origin of the infectingorganism is sometimes doubtful. Therefore, epidemiological investigation can help to recognize routes of infectionspreading. The aim of the present study was to determine the ability to produce esterases by clinical isolates ofC. albicans and to find the relationship between their serotypes. Also, it was intended to determine the ability of thesestrains to produce proteases and lipases as well as the ability of the strains to assimilate carbohydrates. 46 strainsof C. albicans isolates from the vagina of women suffering from vulvovaginitis were examined. Three main kinds ofesterases were distinquished by their spectra of hydrolytic activity toward "-naphthyl acetate, $-naphthyl propionateand indoxyl acetate. The strains were grouped into four categories: three categories in which esterase patterns wereobserved and one category in which esterase bands were not observed. On the basis of the 20 carbon sources assimilated,the C. albicans strains were categorized into 11 biotypes with the major biotype accounting for 21 (45.7%) strains. Theexamination of proteolytic activity using casein and albumin enabled to divide the strains into four groups. All of theexamined strains belonged to serotype A and all of them expressed lipolytic activity. Esterase electrophoretic patternsand biotypes based on proteolytic activities were compared with the ability to assimilate carbon from various sources.

K e y w o r d s: C. albicans, vulvovaginitis, biotypes, esterases

Introduction

The use of epidemiological markers to type C. albicans in communicable diseases is recognized as beingvery important for the following purposes: detection of sources and routes of infections; distinction betweenrecurrent and chronic infection; identification of endemic strains; selection of specific immunoprophylaxisand immunotherapy (al Rawi and Kavanagh, 1999; Kubota, 1998; Magliani et al., 2002; Mendling et al.,2000; Moraes et al., 2000). In the case of Candida vulvovaginitis, the origin of the infecting organism isunclear. It has been suggested that there are many women who carry C. albicans in the vagina withoutsubjective symptomatology, often with low Candida concentration. These observations are in agreementwith the view that C. albicans may be either human commensal or a pathogen (Kubota, 1998; Marai, 2001).C. albicans is a common opportunistic pathogen in HIV-infected patients. Diabetes mellitus increases therate of vaginal colonization and infection with C. albicans spp. (Goswami et al., 2000; Haberland-Carrodeguas et al., 2002; Taylor et al., 2000). However, C. albicans is a potent allergen in some situationsand it has been suggested that local hypersensitivity to these fungi is a factor which prolongs recurrentvaginal candidiasis (Moraes et al., 2000).

Various methods have been used to type Candida isolates, with the objective of developing epidemiologi-cal tools. The most widely used methods include biotyping, serological typing, comparison of the susceptibil-ity or resistance response to antifungal chemicals, and the ability of the strains to produce hydrolytic enzymes(Kantarcioglu 2002; Kurnatowska, 1998; Mendling et al., 2000; Mercure et al., 1996; Quindos et al., 1996).

Hydrolytic enzymes of C. albicans have been implicated as virulence factors, and particular emphasishas been given to the extracellular protease, esterase and lipase (De Bernardis et al., 1999; Kantarcioglu2002; Kurnatowska, 1998). The goals of the present study were: firstly, to determine the patterns of esterase

228 Cybulski Z. et al. 3

electrophoretic types (zymotypes) among clinical isolates of vaginal C. albicans, secondly, to find out whetherthe zymotypes could be used to define strains of C. albicans for epidemiological purposes, thirdly, to deter-mine the serotype of the strains. Moreover, the aim was to compare C. albicans strains with respect to theiresterase patterns, their capability of secreting protease and lipase and their ability to assimilate carbohydrates.

Experimental


46 strains of Candida albicans were isolated from vagina of women suffering from vulvovaginitis. The age of patients rangedfrom 17 to 67 years. Samples were taken from the wall or fluid (pool) of the vaginal canal with a sterile cotton swab and immediatelystreaked on Sabouraud glucose plates agar. Isolates were found to be C. albicans if they showed germ tube production in rabbitserum and if they were able to produce chlamydospores on Nickerson-Mankowski medium. Apart from that, strains of Candidaalbicans were typed using API 20C AUX (bioMerieux). Proteolytic activities of the strains examined were determined usinghuman albumin and bovine casein as substrates according to the procedure described by Staib (1965). Lipolytic activities weredetermined using the medium containing Tween 80 according to Werner (1966). API 20 C AUX test was performed according tothe manufacturer�s instruction. For each of the isolates the serotype was determined by slide agglutination test according to Hannulaet al. (2001). The determination of the esterase patterns was performed using the techniques described by Goullet and Picard (1991)and Branger et al. (1990) which were adapted in order to examine strains of C. albicans. The single colony of each examined strainof C. albicans growing on Sabouraud agar was inoculated into Sabouraud broth then used to determine the esterase pattern.

C. albicans was grown overnight at 37°C in 40 ml of Sabouraud broth in Erlenmayer flasks being vigorously shaken. Aftercentrifugation the yeasts were washed in 0.075 M Tris-0.06 M glycine buffer (pH 8,7), resuspended in 1.5 ml of the same buffer andthen disrupted 10 times by refrigeration and thawing. Debris was removed at 10.000×g for 15 min. at 4°C. The supernatantscontaining at least 6 mg of protein per 1 ml were stored at �20°C until used for electrophoresis. Each strain was cultured andextracted at least twice. Horizontal slab polyacrylamide-agarose gel electrophoresis and an estimation of electrophoretic mobilitywere performed as described by Goullet and Picard, 1991 and Uriel 1966. The detection of hydrolytic activity to a naphthyl acetate,b naphthyl propionate and indoxyl acetate was carried out by staining with the dye Fast Blue.

Results

All of the examined strains belonged to serotype A and all of them showed lipolytic activity. The exami-nation of esterase electrophoretic motility allowed to differentiate the examined strains into four groups ofwhich one group was constituted by strains not producing esterases detected in the present study. All of the

Fig. 1. Esterase patterns of C. albicansElectrophoretic mobility of Candida albicans esterases in hori-zontal slab polyacrylamide � agarose gel electrophoresis. Threeesterase patterns of C. albicans are presented. All patterns showesterases which hydrolysed a-naphthyl acetate, b-naphthylpropionate and indoxyl acetate. 27 of 46 examined C. albicansstrains exhibited esterase activity detectable in used method.

three substrates used: "-naphthyl acetate, $-naphthylpropionate and indoxyl acetate, were hydrolysed byfungal esterases. 15 of the C. albicans strains exhibitedan esterase pattern consisting of two distinctly stainedbands. 7 strains showed a simple esterase pattern con-sisting of a single band and the pattern of 5 strains con-sisted of three bands (zymotypes A, B, C, respectively).The mobility of these enzymes and esterase patterns ofthe examined strains are shown in Fig. 1. 19 strainsof C. albicans did not exhibit esterase activity detectedin this examination.

Basing on proteolytic activities, the strains were di-vided into four groups. 23 of 46 examined strains exhib-ited proteolytic activity to human albumin and bovinecasein, 10 strains were active with respect to only bovinecasein and 11 strains were active only with respectto human albumin. 2 strains did not exibit proteolyticactivity with respect to both substrates used (Table I).

On the basis of 20 carbon sources assimilated, theC. albicans isolates were categorized into 11 biotypeswith the major biotype accounting for 21 (45.7%) strainsof numerical profile 2 576 174. 13 of them hydrolyzedboth substrates: human albumin and bovine casein. 8 ofthe strains from these 21 strains belonged to pattern Aof esterase production. The relationship between esterase

229C. albicans biotypes3

I + + 23I � + 10III + � 11IV � � 2

Table IBiotypes according to proteolytic

activity of C. albicans

BiotypeProteolytic activity

H.A.a B.C.b

Numberof strains

H.A.a � human albumin, B.C.b � bovine casein(+) � presence of proteolysis(�) � absence of proteolysis

No of strains 3 4 13 1 8 2 2 9

Table IIThe comparison of 21 strains of C. albicans exhibiting the same biochemical

properties

Numerical profile(2576174) of

21 strains examined

Proteolytic activitywith respect to Lack of

proteolysis

Esterase pattern

HAa BCb HA&BC 0

HAa human albumin, BCb bovine caseinA, B, C � zymogram characterized by two, one and three esterase bands, respectively0 � lack of esterase pattern.

CBA

patterns and proteolytic activity of the strains exhibiting the most common biotype is shown in Table II.The remaining 25 strains belonging to ten different biotypes showed a variety of esterase paterns and theydiffered in their proteolytic activities to the substrates used, making any comparison impossible.

Discusion

Candida albicans is an important cause of a wide spectrum of diseases including vulvovaginitis (Farinaet al., 2000; Linhares et al., 2001; Marai, 2001; Mendling et al., 2000). The presence of Candida vulvovagini-tis cannot be definitively identified by clinical criteria (Linhares et al., 2001; Novikova et al., 2002). Amongmethods used in yeast infections diagnosis, the most important are cultures of these microorganisms(Linhares et al., 2001). The identification of C. albicans is based on a number of phenotypic characteristics,of which carbohydrate assimilation is of primary importance. For these purposes we used the commercialAPI 20 C AUX system, which is commonly used for diagnosis of Candida spp. (Ellabib et al., 2001; Mercureet al., 1996; Smith et al., 1999; Wadlin et al., 1999).

When microbial strain typing methods are compared, the most important characteristics are typeability,reproducibility and discriminatory power. The results of the present study confirm those of other authors,and namely that the API system shows good reproducibility, sensitivity and specificity (Wadlin et al., 1999).However, the discriminatory power of API system is not high, and, if epidemiological investigation hasto be performed, results of examination with the use of Api system should be supplemented with othermethods. For example electrophoretic typing of esterases and biotyping based on proteolytic activity can beused as additional methods for differentiating strains. Other authors have already indicated the value ofusing a combined typing system (Quindos et al., 1996).

We tentatively grouped the C. albicans strains into four categories, three with esterase patterns, and onewithout an esterase band. Because the C. albicans isolates show serological differences according to theirgeographical origin (Hannula et al., 2001; Mercure et al., 1996), serological examination of C. albicansseems to be very important from the epidemiological point of view. In our investigation, all examinedstrains belonged to the A serotype.

It must be pointed out that the application of esterase zymogram in bacterial taxonomy and epidemio-logy is often used (Branger et al., 2003; Chetoui et al., 1998; Gilot and Andre, 1995 and 1996; Giver et al.,1998). The examination of properties of C. albicans enzymes is widely applied (De Bernardis et al., 1990;Pichova et al., 2001; Tsuboi et al., 1996; Vazquez-Reyna et. al., 1999) and may be useful for epidemio-logical purposes. Tsuboi et al 1996 showed that the induction level of C. albicans extracellular esterasewas found to be correlated with fungal growth. Protease production is one of the most relevant factors ofCandida spp. virulence in mucosal diseases, including vaginitis (De Bernardis et al., 1990 and 1999;Kantarcioglu and Yucel, 2002; Pichova et al., 2001; Rodrigues et al., 1999; Smolenski et al., 1997).

C. albicans is significantly more proteolytic than the non-albicans yeasts (Wu and Samaranayake, 1999).Vaginal candidiasis may be observed as an acute disease which is characterized by a presence of inflamationsymptoms. The cause of development of vaginal candidiasis may be reinfection from a gastrointestinalreservoir, sexual transmission, or impaired host defence mechanisms and enhanced Candida virulence(Kubota, 1998). There is clear evidence that proteolytic C. albicans strains were more virulent than thenon proteolytic ones (De Bernardis et al., 1999). In our study, 95.7% of the investigated strains exhibitedproteolysis, when albumin, albumin and casein or casein were used as substrates (Table I). Similar results of


proteolytic activity are reported by Kantarcioglu et al., 2002, who showed that 95% of clinical C. albicansisolates exhibited proteolysis on media with bovine serum albumin. On the other hand, Odds et al. (1983)reported that only one-third of C. albicans strains isolated from the female genital tract produced proteinaseagainst human albumin. In the present study, the analysis of albumin and casein proteolysis is used inbiotyping of C. albicans isolates. We found that 23 (50%) of the strains exihibited lysis of both theseproteins and only two strains did not reveal proteolytic activity. It seems probable, then that the examinationof lipolytic activity are of minor importance in epidemiological investigation of vaginal C. albicans.

Literature

a l - R a w i N. and K. K a v a n a g h. 1999. Characterisation of yeasts implicated in vulvovaginal candidosis in Irish women. Br.J. Biomed. Sci. 56: 99�104.

B r a n g e r C., C. G a r d y e, J.O. G a l d b a r t, C. D e s c h a m p s and N. L a m b e r t. 2003.Genetic relationship betweenmethicillin-sensitive and methicillin-resistant Staphylococcus aureus strains from France and from international sources:delineation of genomic groups. J. Clin. Microbiol. 41: 2946�51.

B r a n g e r C., P. G o u l l e t, A. B o u t o n n i e r and J.M. F o u r n i e r. 1990. Correlation between esterase electrophoretictypes and capsular polysaccharide types 5 and 8 among methicillin-susceptible and methicillin-resistant strains of Staphylo-coccus aureus. J. Clin. Microbiol. 28: 150�1.

C h e t o u i H., E. D e l h a l l e, P. M e l i n, M.J. S t r u e l e n s, R. D e R y c k, P. O s t e r r i e t h and P. D e M o l. 1998.Typing of nosocomial strains of Serratia marcescens: Comparison of pulsed-field gel electrophoresis of macrorestrictionfragments with biotyping, esterase typing and ribotyping. Res. Microbiol. 149: 137�143.

D e B e r n a r d i s F., L. A g a t e n s i, I.K. R o s s, G.W. E m e r s o n, R. L o r e n z i n i, P.A. S u l l i v a n and A. C a s s o n e.1990. Evidence for a role for secreted aspartate proteinase of Candida albicans in vulvovaginal candidiasis. J. Infect. Dis.161: 1276�83.

D e B e r n a r d i s F., F. M o n d e l l o, G. S c a r a v e l l i, A. P a c h i, A. G i r o l a m o and L. A g a t e n s i. 1999. High aspar-tyl proteinase production and vaginitis in human immunodeficiency virus-infected women. J. Clin. Microbiol. 3: 1376�80.

E l l a b i b M.S. and I.A. E l J a r i n y. 2001. In vitro activity of 6 antifungal agents on candida species isolated as causativeagents from vaginal and other clinical specimens. Saudi Med. J. 22: 860�3.

F a r i n a C., V. M a l i g h e t t i, J.P. L o m b a r t, M. T o y b, Z. Yo u s s o u f, S. C a l i g a r i s, A. M a t t e e l l i, G. C h a t e l,M. A r o s i o and G. C a r o s i. 2000. Yeasts from vaginal exudates in the Comoros Islands. J. Mycol. Med. 10: 91�93.

G i l o t P. and P. A n d r e. 1995. Characterization of five esterase from Listeria monocytogenes and use of their electrophoreticpolymorphism for strain typing. Appl. Environmental Microbiol. 61: 1661�65.

G i l o t P. and P. A n d r e. 1996. Serotyping and esterase typing for analysis of Listeria monocytogenes populations recoveredfrom foodstuffs and from human patients with listeriosis in Belgium. J. Clin. Microbiol. 34: 1007�10.

G i v e r L., A. G e r s h e n s o n, P.O. F r e s k g a r d and F.H. A r n o l d. 1998. Directed evolution of a thermostable esterase.Proc. Natl. Acad. Sci. 95: 12809�13.

G o u l l e t P. and B. P i c a r d. 1991. Pseudomonas aeruginosa isolate typing by esterase electrophoresis. FEMS Microbiol. Lett.78: 195�200.

G o s w a m i R., V. D a d h w a l, S. T e j a s w i, K. D a t t a, A. P a u l, R.N. H a r i c h a r a n, U. B a n e r j e e and N.P.K o c h u p i l l a i. 2000. Species-specific prevalence of vaginal candidiasis among patients with diabetes mellitus and itsrelation to their glycaemic status. J. Infect. 41: 162�6.

H a b e r l a n d - C a r r o d e g u a s C., C.M. A l l e n, F.M. B e c k, W.J. B u e s c h i n g, S.L. K o l e t a r and P. S u n d s t r o m.2002. Prevalence of fluconazole-resistant strains of Candida albicans in otherwise healthy outpatients. J. Oral Pathol. Med.31: 99�105.

H a n n u l a J., B. D o g a n, J. S l o t, E. O k t e and S. A s i k a i n e n. 2001. Subgingival strains of Candida albicans in relationto geographical origin and occurrence of periodontal pathogenic bacteria. Oral Microbiol. Immunol. 16: 113�8.

K a n t a r c i o g l u A.S. and A. Yu c e l. 2002. Phospholipase and protease activities in clinical Candida isolates with reference tothe sources of strains. Mycoses 45: 160�5.

K u b o t a T. 1998. Chronic and recurrent vulvovaginal candidiasis. Nippon Ishinkin Gakkai Zasshi. 39: 213�8.K u r n a t o w s k a A.J. 1998. Activity of hydrolytic enzymes of Candida albicans strains isolated from patients with periodontal

and membrane mucosae of oral cavity diseases. Mycopathologia 141: 105�9.L i n h a r e s L.M., S.S. W i t k i n, S.D. M i r a n d a, A.M. F o n s e c a, J.A. P i n o t t i and W.J. L e d g e r. 2001. Differentia-

tion between women with vulvovaginal symptoms who are positive or negative for Candida species by culture. Infect. Dis.Obstet. Gynecol. 9: 221�5.

M a g l i a n i W., S. C o n t i, A. C a s s o n e, F. D e - B e r n a r d i s and L. P o l o n e l l i. 2002. New immunotherapeutic strate-gies to control vaginal candidiasis. Trends Mol. Med. 8: 121�6.

M a r a i W. 2001. Lower genital tract infections among pregnant women: a review. East Afr. Med. J. 78: 581�5.M e n d l i n g W., M. P i n t o D e A n d r a d e, J. G u t s c h m i d t, R. G a n t e n b e r g, W. P r e s b e r and G. S c h o n i a n.

2000. Strain specificity of yeasts isolated from different locations of women suffering from vaginal candidosis, and theirpartners. Mycoses 43: 387�92.

M e r c u r e S., S. S e n e c h a l, P. A u g e r, G. L e m a y and S. M o n t p l a i s i r. 1996. Candida albicans serotype analysis byflow cytometry. J. Clin. Microbiol. 34: 2106�12.

231C. albicans biotypes3

M o r a e s P.S., S. d e L i m a G o i a b a and E.A. T a k e t o m i. 2000. Candida albicans allergen immunotherapy in recurrentvaginal candidiasis. J. Investig. Allergol. Clin. Immunol. i10: 305�9.

N o v i k o v a N., E. Ya s s i e v i c h and P.A. M a r d h. 2002. Microscopy of stained smears of vaginal secretion in the diagnosisof recurrent vulvovaginal candidosis. Int. J. STD AIDS. 13: 318�22.

O d d s F.C., A.B. A b b o t t, R.L. S t i l l e r, H.J. S c h o l e r, A. P o l a k and D.A. S t e v e n s. 1983. Analysis of Candidaalbicans phenotypes from different geographical and anatomical sources. J. Clin. Microbiol. 18: 849�57.

P i c h o v a I., L. P a v l i c k o v a, J. D o s t a l, E. D o l e j s i, O. H r u s k o v a - H e i d i n g s f e l d o v a, J. W e b e r, T. R u m land M. S o u c e k. 2001. Secreted aspartic proteases of Candida albicans, Candida tropicalis, Candida parapsilosis andCandida lusitaniae. Inhibition with peptidomimetic inhibitors. Eur. J. Biochem. 268: 2669�77.

Q u i n d o s G., V. L i p p e r h e i d e, B. B a r t u r e n, R. A l o n s o, J. B i k a n d i, R. S a n - M i l l a n, M. T e l l a e t x e,L. R i b a c o b a and J. P o n t o n. 1996. A new method of antibiotyping yeasts for subspecies discrimination and distribu-tion in human clinical specimens. Eur. J. Epidemiol. 12: 55�62.

R o d r i g u e s A.G., P.A. M a r d h, C. P i n a - Va z, J. M a r t i n e z d e O l i v e i r a and A.F. d a F o n s e c a. 1999. Is thelack of concurrence of bacterial vaginosis and vaginal candidosis explained by the presence of bacterial amines? Am.J. Obstet. Gynecol. 181: 367�70.

S m i t h M.B., D. D u n k l e e, H. V u and G.L. W o o d s. 1999. Comparative performance of the RapID Yeast Plus System andthe API 20C AUX Clinical Yeast System. J. Clin. Microbiol. 37: 2697�8.

S m o l e n s k i G., P.A. S u l l i v a n, S.M. C u t f i e l d and J.F. C u t f i e l d. 1997. Analysis of secreted aspartic proteinasesfrom Candida albicans: purification and characterization of individual Sap1, Sap2 and Sap3 isoenzymes. Microbiology 143:349�56.

S t a i b F. 1965. Serum-proteins as nitrogen source for yeastlike fungi. Sabouraudia 4: 187�193.T a y l o r B.N., C. F i c h t e n b a u m, M. S a a v e d r a, J. S l a v i n s k y I I I, R. S w o b o d a, K. W o z n i a k, A. A r r i b a s,

W. P o w d e r l y and P.L. F i d e l J r. 2000. In vivo virulence of Candida albicans isolates causing mucosal infections inpeople infected with the human immunodeficiency virus. J. Infect. Dis. 182: 955�9.

T s u b o i R., H. K o m a t s u z a k i and H. O g a w a. 1996. Induction of an extracellular esterase from Candida albicans andsome of its properties. Infect. Immun. 64: 2936�40.

U r i e l J. 1966. Method of electrophoresis in acrylamide-agarosa gels (in French). 1966. Bull. Soc. Chim. Biol. (Paris). 48: 969�82.Va z q u e z - R e y n a A.B., P. P o n c e - N o y o l a, C. C a l v o - M e n d e z, E. L o p e z - R o m e r o and A. F l o r e s - C a r r e o n.

1999. Purification and biochemical characterization of two soluble alpha-mannosidases from Candida albicans. Glycobiology9: 533�7.

W a d l i n J.K., G. H a n k o, R. S t e w a r t, J. P a p e and I. N a c h a m k i n. 1999. Comparison of three commercial systems foridentification of yeasts commonly isolated in the clinical microbiology laboratory. J. Clin. Microbiol. 37: 1967�70.

W e r n e r H. 1966. Studies on the lipase activity in yeasts and yeast-like fungi (in German). Zbl. Bakt. I Orig. 200: 113�124.W u T. and L.P. S a m a r a n a y a k e. 1999. The expression of secreted aspartyl proteinases of Candida species in human whole

saliva. J. Med. Microbiol. 48: 711�20.



Purification and Characterization of Two Extracellular Lipasesfrom Pseudomonas aeruginosa Ps-x

HESHAM M. SAEED1, TAHA I. ZAGHLOUL1, AHMED I. KHALIL2

and MOHAMED T. ABDELBAETH1

1 Institute for Graduate Studies and Research, Department of Bioscienceand Technology University of Alexandria, Chatby 21526, Alexandria, Egypt,

2 Institute of Graduate Studies and Research, Department of Environmental Studies,University of Alexandria, Chatby 21526, Alexandria, Egypt

Received 2 March 2005, received in revised form 25 May 2005, accepted 27 May 2005

A b s t r a c t

Two different extracellular lipases were isolated and purified from Pseudomonas aeruginosa Ps-x to apparent homo-geneity using ammonium sulfate precipitation followed by ion exchange chromatography on Q- and S-Sepharose column.Both of the purified lipases are monomeric protein with molecular weight of 15.5 and 54.97 KDa respectively. Theoptimal activities of the enzymes were at 45 and 50°C and pHs 10.0 and 9.0. Calcium ions increase thermostability ofboth purified lipases I and II. The purified lipase I showed no metal ion dependence for its activity since EDTA up to10 mM has no effect on the enzyme activity. However purified lipase II showed slight inhibition by EDTA at the sameconcentration. Moreover, a serine protease inhibitor, PMSF showed an inhibitory effect on both purified enzymes.

K e y w o r d s: Lipases, Pseudomonas aeruginosa Ps-x, 16S rRNA

Introduction

Lipases (triacyl glycerol acylhydrolases, EC 3.1.1.3) constitute a diverse and ubiquitous family of enzymesthat in biological system initiate the catabolism of fats and oils by hydrolyzing the fatty acylester bonds ofacylglycerols (Carriere et al., 1994 and Jose et al., 2004). Most of lipases remain active in a variety of organicsolvents, where they can catalyze various transformations other than hydrolytic reaction by which they aredefined (Margolin and Klibanov, 1987). Lipases are widely distributed in nature and have been found inmany species of animals, plants, bacteria, yeast and fungi. Although their wide distribution, the enzymesfrom microorganisms are most interesting because of their potential application in various industries rangingfrom the use in laundry detergent to stereospecific biocatalysts (Maliszewska and Przemyslaw, 1992). Mostof the microbial lipases are secreted into the culture medium and they differ from one another in theirphysical and biochemical properties. Since each industrial application requires specific properties of lipases,there is still an interest in additional lipases that could be used in new applications (Jaeger et al., 1994; Lambitand Goswami, 2002 and Kyu et al., 2005). The synthesis and secretion of extracellular lipases by variousmicroorganisms appear to be controlled in a variety of ways which are only now beginning to be elucidatedand the limited data on this subject have been reviewed by Jaeger et al., 1994. One of the best studied case,P. aeruginosa, lipase only appears in the culture medium at the end of logarithmic growth when a numberof other hydrolytic exoenzymes are also released. However, lipases are produced during logarithmic growthin a minimal media when growth is dependent for carbon and energy upon hydrolysis of either a triglycerideor a detergent such as a Tween (Jaeger et al., 1994 and Stuer et al., 1986). This study described the produc-tion, purification and characterization of two extracellular lipases from P. aeruginosa Ps-x.

Abbreviations: SDS-PAGE; sodium dodecyl sulfate polyacrylamide gel electrophoresis, Q-Sepharose; Quaternary amino methylSepharose, EDTA; ethylene diamine tetraacetic acid.

1 Corresponding author: tel: 002034297005, fax: 002034285792, e-mail: [email protected]

234 Saeed H.M. et al. 3

Experimental


Bacterial isolate identification. The bacterial isolate used in this study was a gift from Dr. Yossry Gohar, Department ofMicrobiology, Faculty of Science, Alexandria University, Egypt. This isolate was tested for lipase production using nutrient agarmedium, pH 7.5 containing per liter: peptone, 10.0 g; NaCl, 5.0 g; CaCl2, 0.1 g; and agar, 20.0 g. The medium was sterile byautoclaving and let to cool to 45°C after that 1% of sterile Tween-20 or Tween-80 was added to the medium and mixed well thenpoured (20� 25 ml) into 100 mm Petri dishes. The identification of the bacterial isolate was based on cell morphology, colonymorphology, growth on nutrient broth and nutrient agar as well as several biochemical tests. The identification process was per-formed at the Fermentation and Biotechnology Center, El-Azhar University, Cairo. To confirm the biochemical tests results forisolate identification, 16S rRNA (rDNA) technique was carried out. DNA was isolated and purified according to Sambrook et al.,1989. Amplification of the 16S rDNA gene from the genome was carried out by polymerase chain reaction (PCR) using primersdesigned to amplify 16S rRNA gene. The forward primer was 5�-AGAGTTTGATCMTGGCTCAG-3� and the reverse primer was5�- TACGGYTACCTTGTTACGACTT-3�. The polymerase chain reaction analysis was performed with 100 ng of genomic DNA ina final volume of 50 :l, including a reaction buffer 1x, 30 pmole of each primer, and 2 units of Taq polymerase. Thermocyclingconsisted of an initial denaturation of 5 minutes at 94°C and of 30 cycles of 1 minute at 94°C (denaturation), 1 minute at 55°C(primers annealing), and 1.5 minutes at 72°C (extension). Polymerase chain reaction products were analyzed for purity check on1% agarose gel by electrophoresis, stained with ethidium bromide (0.5 :g/ml), and visualized using ultraviolet transillumination.

DNA sequencing. DNA was sequenced by the dideoxy chain termination method according to Sanger et al., 1977 using ABIPrism Ready Reaction Dye Terminator Sequencing Kit and analyzed on an ABI 377 automated sequencer. The nucleotide sequenceobtained about 326 base pairs were then analyzed using nucleotide Blast search data base and have been deposited in the GenBanksequence data base and have the accession number AF419219.

Lipase assay. Lipase activity was determined colorimetrically according to Kordel et al., 1991, where two solutions wereprepared for the assay. Solution 1 contained 90 mg of pNPP (p-nitrophenyl palmitate), dissolved in 30 ml propane-2-ol. Solution 2contained 2 g Triton X-100 and 0.5 g gum Arabic dissolved in 450 ml (Tris-HCl 50 mM) buffer at pH 8.0. The assay solution wasprepared by adding 1 ml of solution 1 to 9 ml of solution 2 drop wise to get an emulsion that remained stable for 2 hrs. The assaymixture contained 900 :l of the emulsion and 100 :l of the appropriately diluted enzyme solution. The liberated p-nitrophenol wasmeasured at 410 nm using Novospek, Pharmacia spectrophotometer. One unit of enzyme was defined as the amount of enzyme thatreleases 1 :mol of p-nitrophenol from the substrate.

Protein analysis. Total protein concentrations of cell free supernatant and purified samples were assayed by the method ofBradford (Bradford, 1976) using a calibration curve established with bovine serum albumin as a standard. Proteins in these prepara-tions were analyzed by SDS-PAGE on 10% gels according to the method of Laemmli (Laemmli, 1970). Gel analysis and molecularweight determination was done using Alpha Imager 1200 Tm gel documentation system.

Monitoring bacterial growth and extracellular lipase production. The growth and activity of the extracellular lipase wasmonitored throughout the growth of the P. aeruginosa Ps-x strain on modified Williams Basal medium II (Williams et al., 1990)containing per liter: NH4Cl, 5g; NaCl, 5g; K2HPO4, 3.0 g; KH2PO4, 4.0 g; MgCl2×6H2O, 1.0 g and yeast extract 1.0 g. Deionizedwater was added to approximately 1 L, then the pH was adjusted to 7.5 with 10 N NaOH and autoclaved. Cells were activated bygrowing them overnight at 37°C on nutrient agar plates. Several recently growing colonies were transferred to a 50 ml of produc-tion medium and incubated at 37°C overnight with agitation at 150 rpm. Then, 1 ml was taken to inoculate 100 ml of the productionmedium and allowed to grow at 37°C with shaking at 150 rpm. Growth was monitored by measuring the absorbance at 420 nm.At the indicated time, 1.0 ml of the growing cultures was taken and centrifuged in a microcentrifuge at 8,000 rpm for 2 minutes andthe supernatants were assayed for extracellular lipase using p-nitrophenylpalmitate as a substrate as described earlier.

Enzyme production and purification. The extracellular lipase produced by P. aeruginosa Ps-x was first purified by salting outprecipitation using ammonium sulfate at 80% saturation. P. aeruginosa cells were cultured aerobically in 500 ml productionmedium containing per liter: 5 g NaCl, 0.05 g CaCl2×2H2O, 10 g yeast extract and 10 ml Tween 20 for 18 hours at 37°C withagitation at 150 rpm. The culture was then centrifuged at 8,000 rpm for 20 minutes at 4°C using the small SS-34 small rotor. Solidammonium sulfate was then added slowly to the culture filtrate at 70% saturation with gentle stirring on ice bath. After that, themixture was allowed to stand at 4°C for overnight. The mixture was then centrifuged at 12,000 rpm for 30 minutes using the aboverotor. Pellet was dissolved in 5 ml 0.02 M Tris-HCl pH 8.0 and dialyzed overnight against 2 liters of the same buffer. The proteincontent and the lipase activity were determined as described earlier. The concentrated dialyzed cell free supernatant was thenapplied to a column (2.5×7 cm) containing Quaternary amino methyl Sepharose fast flow (Q-Sepharose), which previously wasequilibrated with 20 mM Tris-HCl, pH 8.5. The column was washed with 3 bed volumes of the same buffer at a flow rate of 60 ml/hourand the bound proteins were eluted with a linear gradient of NaCl (0�1.0 M) in the same buffer. Active fractions that contain lipaseenzyme were then polled and concentrated using ammonium sulphate as mentioned before. The concentrated and dialyzed enzymewas further purified using S-Sepharose fast flow column (2.5×5 cm) prequilibrated with 20 mM Tris-HCl, pH 8.5 at a flow rate of45 ml/hour. The column was washed with three bed volumes of the same buffer and the bound proteins were eluted using lineargradient of NaCl (0�0.5 M) in the same buffer. Lipase containing fractions were polled and concentrated as described before.

pH optima, temperature and thermostability studies. Extracellular lipase optimum pH was determined over a pH valuesrange from 6.0 to 10.0. Sodium phosphate buffer 0.1 M was used for pH 6.0 and 7.0, Tris-HCl 0.1 M for pH 8.0 and 9.0 andcarbonate buffer 0.1 M for pH 10.0 and 11.0. The temperature optimum of the purified P. aeruginosa lipases was determinedover a temperature range of 25�90°C in 50 mM Tris-HCl buffer pH 8.0. Thermostability of the purified enzymes was examined ata temperature range 40�70°C for different time intervals (10�60 minutes) in 50 mM Tris-HCl buffer pH 8.0 in absence and inthe presence of 5.0 mM CaCl2. The residual activity was determined by taking 25 ml of the enzyme solution after specified time(10�60 minutes) and the assay was carried out as described before.

Effect of some compounds on the activity of the purified enzymes. To examine the effect of EDTA (ethylenediaminetetraacetic acid), SDS (sodium dodecylsulphate), PMSF (phenylmethane sulfonyl fluoride) and DMSO (dimethyl sulfoxide) on

235Lipases from P. aeruginosa3

the activity of the purified lipases, different concentrations of these compounds were prepared and the enzymes solution werepre-incubated with these compounds for 30 minute on ice after which the residual activity was determined as described before usingp-nitrophenyl palmitate substrate.

Effect of some metal ions on the activity of the purified enzymes. For determining the effect of metal ions on lipase activity,the purified enzymes were pre-incubated with 1�10 mM of the following metal ions MgCl2, ZnCl2, CaCl2 and NaCl for 30 minuteon ice and then the residual activity was determined.

Results and Discussion

Identification of bacterial isolate. Based on cell morphology and colony morphology, growth on nutrientbroth and nutrient agar as well as several biochemical tests, the bacterial isolate was identified as Pseudo-monas aeruginosa. To confirm the biochemical tests results for bacterial isolate identification, 16S rRNAmethodology was carried out. The sequencing data obtained utilizing this strategy indicated that the isolateunder study was 98% P. aeruginosa and the nucleotide sequence was deposited in the GenBank database,and given the accession number AF419219.

Monitoring the bacterial growth and extracellular lipases enzymes production. The growth and extracel-lular lipase production level of P. aeruginosa Ps-x was monitored on modified William�s Basal medium IIsupplemented with 2% Tween-20. Figure 1 illustrates the growth behavior with incubation time and lipaseproduction. The level of extracellular lipase enzyme production of P. aeruginosa increased gradually at theend of log phase (6 hours after inoculation) and reached its maximum level (74.66 U/ml) after 48 hours ofinoculation. It was reported that the extracellular lipase production normally appears in the fermentationmedium when the bacterial cell growth reach to the end of the logarithmic. Moreover, the observation ofsimilar growth rates for Pseudomonas strains during the logarithmic growth phase is in accordance with theconcept that the production of these enzymes is advantageous to the microorganism only when nutrientsbecome limiting i.e., during the late log phase or early stationary phase (Lee and Rhee, 1993).

0 .01

0 .1

1

10

0 2 5 7 9 12 48

Time after inoculation (hrs )

Ab

420 n

m

0

10

20

30

40

50

60

70

80

Lip

ase

act

ivity

Ab .420 nm . U /ml

Fig. 1. Monitoring of growth and extracellular lipase production by P. auroginosa strain Ps-xon modified basal medium

Purification and characterization of P. aeruginosa extracellular lipases. P. aeruginosa lipasesenzymes were purified using (NH4)2SO4 selective precipitation (70%) followed by ion exchange chromato-graphy on Q and S-Sepharose Fast Flow columns. It was found that the unbound proteins were elutedfirst using 20 mM Tris-HCl, pH 8.5 buffer giving only one protein peak that overlapped with the lipolyticactivity. Upon using NaCl gradient (0.0�1.0 M) in the same buffer, one major and one minor protein peaksappeared as shown in Figure 2. Surprisingly, the minor protein peak showed a lipolytic activity as well.This result indicated that, at least P. aeruginosa Ps-x strain under study produced two extracellular lipasesof different biochemical characters. Thus, upon using Q-Sepharose Fast Flow anion exchanger, lipaseenzyme contained in the first peak (LipI) was purified to 6.36 fold with specific activity of 165.5 (Units/mgprotein/ml). While lipase enzyme contained in the second minor peak that eluted with NaCl gradient (LipII)


was found to be purified to almost 3.77 fold and with specific activity of 98.2 (Units/mg protein/ml) asindicated in Table 1. Interestingly, the eluted lipase I (LipI) was found to be almost pure enzyme as judgedby SDS-PAGE as shown in Figure 4 just by selective precipitation with ammonium sulphate followed byanion exchanger chromatography on Q-Sepharose column. Figure 3 show the fractionation pattern of lipaseII on S-Sepharose cation exchanger. It was found that; one distinctive protein peak was appeared in theelution buffer 20 mM Tris-HCl pH 7.4 that overlapped with the lipase activity. By applying NaCl gradient

0

0.02

0.04

0.06

0.08

0.1

1 11 21 31 41 51 61

Fraction Number

Ab

28

0 n

m

0

20

40

60

80

NaC

l

Act

ivity

Ab.280 nm. Units/ml NaCl gradient (M)

0.0

1.0

0.2

0.4

0.8

0.6

Fig. 3. Fractionation pattern of extracellular lipase II (LipII) from P. aeruginosa on S-Sepharose Fast Flow column

0

1

2

3

4

5

6

1 11 21 31 41 51 61 71 81

Fraction Number

Ab

280 n

m

0

10

20

30

40

50

60

70

80

Act

ivity

Ab.280 nm. U/ml NaCl gradient (M)

0.0

1.0

0.2

0.4

0.8

0.6

NaC

l

Fig. 2. Fractionation pattern of the extracellular lipase enzyme produced by P. aeruginosa upon using Q-SepharoseFast Flow anion exchange column

(NH4)

2SO

4 concentrated and dialyzed enzyme 18 496.5 7448 19 26 1 100

Q-Sepharose ColumnFirst Peak (Lipase I) 8 503.2 4025.6 3.04 165.5 6.36 54

Second Peak (Lipase II) 20 88.4 1768.6 0.9 98.2 3.77 23.7

S-Sepharose ColumnSecond Peak (Lipase II) 18 38.4 691.2 0.1 384 14.76 9

Table IPurification of extracellular lipases produced by P. aeruginosa Ps-x

Purification stepsVolume

(ml)

Lipaseactivity

Units/ml a

TotalUnits

Proteincontentmg/ml

Specificactivity

U/mg protein

Foldpurification

Recovery(%)

a One unit of enzyme is the amount of enzyme that liberate one micromole of D-nitrophenol per minute at 37°C


(0�1.0 M) in 20 mM Tris-HCl pH 7.4, two major protein peaks appeared that retained no lipolytic activity.Ion-exchange chromatography on S-Sepharose Fast Flow resulted in increase in both the specific activityand fold purification, 384 and 14.74 versus 98.2 and 3.77 after Q-Sepharose Fast Flow column (Table I).SDS-PAGE showed that both of the purified lipases were composed of a single type of subunit witha molecular weight of 15.5 (lipase I) and 54.97 KDa (lipase II) respectively as shown in Figure 4 and 5.

pH and temperature optima of the purified lipase. The pH range of purified P. aeruginosa lipase I andII was detectable over a wide range between 6�10 with an optimum pH value at 9.0 and 10 respectively.It has been reported that most of lipases produced by Pseudomonas sp. have pH optimum around pH 7.0�9.5(Lee and Rhee, 1993; Dong et al., 1999). It was found that, P. aeruginosa purified lipase I showed lipolyticactivity over a wide range of temperature from 40 to 70°C with an optimum temperature of 50°C which isin agreement with that of P. aeruginosa EF2 and significantly higher than that of Pseudomonas PACIRand psychrotrophic strain of P. fluorescens (Gibert et al., 1991; Yong and Rhee, 1993 and Lee et al., 1993).On the other hand, purified lipase II showed an optimum temperature of 45°C.

Temperature and pH stability of the purified enzymes. Temperature stability of the purifiedP. aeruginosa lipase I and lipase II was investigated for a period of one hour at temperature range from40�70°C in absence and in the presence of calcium ions. It was noticed that, the residual activity for bothlipase I and lipase II decreased as the exposure time and temperature increased in absence of calciumchloride. It was found that, after 60 minutes of exposure to 70°C, purified lipase I retained about 47.74%of its original activity. On the other hand purified lipase II retained about 36.7% of its original activityafter exposure to 70°C for 60 minutes. Thermostability in the presence of calcium ions indicated that, ata temperature of 40 and 50°C, CaCl2 not only stabilize lipase I against thermal inactivation but also enhancedthe lipolytic activity of the purified enzyme (146.19 and 116.4% respectively after exposure time of60 minutes). On the other hand, CaCl2 at the same concentration and under identical assay condition had noeffect on thermostability of purified lipase II at all studied temperatures (from 40�70°C). These resultsindicated that P. aeruginosa lipase I that have a molecular weight of 15.5 KDa, can be stabilized againstthermal inactivation by the addition calcium chloride ion while calcium chloride ion has no effect on thethermostability of the purified lipase II (molecular weight 54.97 KDa). The effect of metal ions on thermo-stability and activity of some enzyme has been investigated before and it has been reported that metal ionssuch as Ca2+, Mn2+ and Co2+ can stabilize and activate some enzymes such as xylose isomerases which bindtwo metal ions, one cation is directly involved in catalysis and the second is mainly structural (Whitlow et al.,1991 and Kasumi et al., 1982). The role of calcium ions in relation to thermostability was also investigated

Fig. 4. Sodium dodecyl sulfate polyacrylamide gel (10%)electrophoresis of P. aeruginosa cell free supernatant (Lane 2),ammonium sulfate concentrated and dialyzed sample (Lane 3),Q-Sepharose purified lipase I (Lane 4), and Q-Sepharoseunbound eluted proteins (Lane 5). Lane 1 represents molecular

weight markers protein

Fig. 5. Sodium dodecyl sulfate gel (10%) electrophoresis ofpurified lipase II from P. aeruginosa (lane 2). Lane 1 represents

molecular weight markers protein

KDa

97.4

66.2

43.0

30.0

20.1

14.2

1 2 3 4 5 1 2

97.4

66.2

43.0

30.0

20.1


and extensively studied in Thermoactinomyces vulgaris subtilisin-type serine-protease thermitase and it wasfound that, thermitase contains three Ca2+-binding sites; one of them is not present in its mesophilic homo-logues (Teplyakov et al., 1990). A thermophilic homologue of thermitase, the Bacillus AK1 protease, containsone more Ca2+ than thermitase does, and it is significantly more kinetically stable than thermitase in thepresence of Ca2+. It has been reported that, in certain P. aeruginosa lipase, the Ca2+ ion might be involved inthe correct positioning of the histidine residue of the catalytic triad, since three of the Ca2+ ion ligand arecontained in a loop together with the histidine residue (Mohamed et al., 2003 and Umesh et al., 2003).Thus, it was clear from these results that, both P. aeruginosa purified lipase I and II can be affected by theexposure to temperature in presence and in absence of CaCl2 metal to different degree. Difference in thermo-stability of both enzymes may be explained by the presence of metal binding site that can bind for examplecalcium ions and hence increase thermostability and enhance the lipolytic activity of purified lipase I butnot lipase II that lack this binding site.

One of the most important characteristic features of the enzymes involved in detergent industries is thepH stability of these enzymes. Stability of P. aeruginosa lipases against pH was examined at pH�s rangefrom 6�10 for 20 days (480 hours) at room temperature and it was found that, both enzymes are stable atpHs from 6�9 for 20 days. Interestingly, both enzymes showed some sort of activation after 36 hoursof storage at pHs from 6�8 and this activation phenomenon increased gradually with the incubation time atindicated pHs. On the other hand both purified lipase I and II showed pH instability at pH 10 and weregradually inactivated as the incubation time increases at pH 10 and the residual activities were found to be28.69 and 50.0 % for lipase I and II respectively after 20 days. One possible explanation for this observationis that different pH values affected the ionization properties of some amino acid residues in the enzyme,which may be resulted in the variation of the activity and stability of the enzymes (Dong et al., 1990).

Effect of some metal ions and some compounds on lipase activity. The effect of some metal ions onthe activities of P. aeruginosa purified lipases was also investigated. It was found that, NaCl, CaCl2 andMgCl2 enhance the lipolytic activity of purified lipase I and II and the residual activity was higher in thecase of CaCl2 than that of NaCl and MgCl2 (Table II). This observation is also exhibited by other Pseudomo-nas lipases (Fox and Stepaniak, 1983 and Yamamoto and Fujiwara, 1988). The observation that lipaseactivity was significantly enhanced in the presence of these metal ions probably reflects the ability of thesesalts to react with free fatty acids adhering to the oil droplets to diminish interfacial charge effects and/or toincrease droplet surface area. Interestingly, ZnCl2 at a concentration of 10 mM inhibits the lipolytic activityfor both lipase I and II and the inhibition was higher in the case of lipase I than lipase II as shown inTable II. This inhibitory effect of Zn+2 is also exhibited by the other Pseudomonas lipases (Yamamoto andFujiwara, 1988) and could probably be due to the direct interaction of metal with the catalytic site, althoughthe alteration of the properties of the interface must be considered and it may be attributed to their bindingof the thiol group of the enzyme or sulfhydryl groups which may be present in the active center of somemicrobial lipases (Schrag et al., 1991).

None 100 100

NaCl 111.04 101.01

CaCl2

122.04 270.32

MgCl2

97 108.89

ZnCl2

106.72 30.64

Table IIEffect of some compounds on the activity of P. aeruginosa purified lipase I and II

CompoundLipase residual activity* (%)

Lipase I Lipase II

* Lipase residual activity (%) reflects the percentage of enzyme activity at a given compound concentration when compared to the enzymaticactivity at zero concentration and under identical assay conditions.

EDTA 20.03 100

DMSO 173.5 173.5

SDS 94.54 101.21

PMSF 15 41.84

Lipase IILipase I

Lipase residual activity* (%)Compound

The effect of EDTA, DMSO, SDS and PMSF was also studied. Data of Table II indicated that the divalentmetal-chelating agent, EDTA at a concentration of 10 mM showed slight inhibitory effect on the activity oflipase I and no effect on the activity of lipase II. It was reported that lipase enzyme produced byP. aeruginosa EF2 is not affected by metal chelating agent, EDTA which indicate that this kind of lipolyticactivity is independent of metal ions (Gibert et al., 1991) in contrast to some other metal ions dependent


Pseudomonas lipases (Fox and Stepaniak, 1983 and Yamamoto and Fujiwara, 1988). Dimethyl sulfoxide(DMSO) enhanced the lipolytic activity of both purified lipase I and II and this observation was agreed withthe lipase of P. aeruginosa YS-7. It was found that sodium dodecyl sulphate (SDS) at a concentration of10 mM showed little inhibition of purified lipase II while the same concentration had no effect on theactivity of purified lipase I (94.541% and 101.219% respectively). Since most of the microbial true lipasescontain a catalytic triad Ser-Asp-His, inhibition study using PMSF was carried out. PMSF at a concentra-tion of 10 mM showed a drastic inhibitory effect on both lipases as shown in Table II and the residualactivity was found to be 15% and 41.84% for lipase I and II respectively. This inhibitory effect of PMSFalso exhibited for some Pseudomonas sp. (Schrag et al., 1991; van Oort et al., 1989 and Svendsen et al., 1995).

Conclusion. In the present work, two lipases from Pseudomonas aeruginosa Ps-x were isolated, purifiedand biochemically characterized. When compared with lipases from other Pseudomonas sp. (Stuer et al.,1986; Brady et al., 1990; Sugiura, 1977 and Mencher and Alford, 1967), these lipases differ in the followingproperties: molecular weight, wider stable pH and temperature ranges and purified lipase II being of highermolecular weight compared to other Pseudomonas sp. lipases. The satisfactory pH and thermal stability willmake these two lipases very attractive for future synthetic applications. Genomic library was constructedfrom purified P. aeruginosa DNA and screened for extracellular lipases. A number of potentials positiveslipases producing clones were obtained. Future research will focus on the sequencing, subcloning andoverexpression of these genes in E. coli. Moreover, studies of the substrate specificities, stereospecificitiesand some aspects of the kinetics of their secretion into the fermentation medium will be undertaken.

Literature

B r a d y L., A.M. B r z o z o w s k i, Z.S. D e r e w e n d a, E. D o d s o n, G. D o d s o n, S. T o l l e y, J.P. T u r k e n b e r g,L. C h r i s t i a n s e n, B. H u g e j e n s e n, L. N o r s k o v, L. T h i m and U. M e n g e. 1990. A serine protease triadforms the catalytic center of a triacylglycerol lipase. Nature 343: 767�770.

B r a d f o r d M.M. 1976. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the prin-ciple of protein-dye binding. Anal. Biochem. 72: 248�254.

C a r r i e r e F., Y. G a r g u r i, H. M o r e a u, S. R a n s a c, E. R o g a l a s k a and R. Ve r g e r. 1994. Gastric lipases: cellular,biochemical and kinetic aspects. pp. 181�205. In: P. Woolley, S.B. Petersen (eds). Lipases-their structure, biochemistry andapplication. Cambridge, UK.Cambridge University Press.

D o n g H., S. G a o, S. H a n and S. C a o. 1990. Purification and characterization of a Pseudomonas sp. Lipase and its proper-ties in non-aqueous media. Biotechnol. Appl. Biochem. 30: 251�256.

F o x P.F. and L. S t e p a n i a k. 1983. Isolation and some properties of extracellular heat stable lipases from Pseudomonasfluorescens strain AFT 36. J. of Diary Research. 50: 77�89.

G i b e r t E.J., A. C o r n i s h and C.W. J o n e s. 1991. Purification and properties of extracellular lipase from Pseudomonasaeruginosa EF2. J. General Microbiol. 137: 2223�2229.

J a e g e r K.E.H., S. R a n s a c, B.W. D i j k s t r, C. C o l s o n, M. v a n H e u v e l and O. M i s s e t. 1994. Bacterial lipases.FEMS Microbiol. Rev. 15: 29�63.

J o s e M.P., C. O r t i z, M. F u e n t e s, G. F e r n a n d e z, J.M. G u i s a n and R.F. L a f u e n t e. 2004. Use of immobilizedlipases for lipase purification via specific lipase-lipase interaction. J. Chrom. A. 1038: 267�273.

K o r d e l M., B. H o f m a n n, D. S c h o m b u r g and R.D. S c h m i d. 1991. Extracellular lipase of Pseudomonas sp. strainATCC 21808: purification, characterization, crystallization, and preliminary X-ray diffraction data. J. Bacteriol. 173: 4836�4841.

K a s u m i T., K. H a y a s h a i and N. T s u m u r a. 1982. Roles of magnesium and cobalt in the reaction of glucose isomerasefrom Streptomyces griseofuscus S-41. Agric. Biol. Chem. 9: 21�30.

K y u R.K., D.Y. K w o n, S.H. Yo o n, W.Y. K i m and K.H. K i m. 2005. Purification, refolding and characterization of recom-binant Pseudomonas fluorescens lipase. Protein Expression and Purification. 39: 124�129.

L a m b i t K. and P. G o s w a m i. 2002. Isolation of a Pseudomonas lipase produced in pure hydrocarbon substrate and its appli-cation in the synthesis of isoamyl acetate using membrane immobilized lipase. Enz. Microb. Tech. 31: 727�735.

L a e m m l i U.K. 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227: 680.L e e Y.P., G.H. C h u n g and J.S. R h e e. 1993. Purification and characterization of Pseudomonas fluorescens SIK W1 lipase

expressed in E. coli. Biochem. Biophys. Acta 1169: 156�164.L e e S.Y. and J.S. R h e e. 1993. Production and partial purification of lipase from Pseudomonas putida 3SK. Enz. Microb. Tech.

15: 617�623.M a r g o l i n A.L. and A.M. K l i b a n o v. 1987. Peptide synthesis catalyzed by lipases in anhydrous organic solvents. J. Am.

Chem. Soc. 109: 3802�4.M a l i s z e w s k a I. and M. P r z e m y s l a w. 1992. Production and some properties of lipase from Penicillium citrinum. Enz.

Micrb. Tech.. 14: 190�193.M o h a m e d K., P. Va n G e l d e r, W. B i t t e r and J. T o m m a s s e n. 2003. Role of the calcium ion and the disulfide bond in

the Burkholderia glumae lipase. J. Mol. Cat. B: Enzymatic 22: 329�338.M e n c h e r J.R. and J.A. A l f o r d. 1967. Purification and characterization of the lipase of Pseudomonas fragi. J. Gen. Microbiol.

48: 317�328.


S a m b r o o k J., E.F. F r i t s c h and T. M a n i a t i s. Molecular cloning. A laboratory Manual. Cold Spring Harbor Laboratory,NY. 1989.

S a n g e r F., S. N i c k l e n and A.R. C o u l s o n. 1977. DNA sequencing with chain terminating inhibitors. Proc. Natl. Acad.Sci. USA. 74: 5463�5467.

S t u e r W., K.E. J a e g e r and U.K. W i n k l e r. 1986. Purification of extracellular lipase from Pseudomonas aeruginosa.J. Bacteriol. 168: 1070�1074.

S c h r a g J.D., L. Yu n g e, W. S h a n g and C. M i r o s l a w. 1991. Ser-His-Glu triad forms the catalytic site of the lipase fromGeotrichum candidum. Nature 351: 761�764.

S v e n d s e n A., K. B o r c h, M. B a r f o e d, T.B. N i e l s e n, E. G o r m s e n and S.A. P a t k a r. 1995. Biochemical proper-ties of cloned lipases form Pseudomonas family. Biochem. Biophys. Acta 1259: 9�17.

S u g i u r a M., T. O i k a w a, K. H i r a n o and T. I n u k a i. 1977. Purification, crystallization, and properties of triacylglycerollipase from Pseudomonas fluorescens. Biochim. Biophys. Acta 488: 353�358.

T e p l y a k o v A.V., I.P. K u r a n o v a, E.H. H a r u t y u n y a n, B.K. Va i n s h t e i n, C. F r o m m e l, W.E. H o h n e andK.S. W i l s o n. 1990. Crystal structure of thermitase at 1.4 A resolution. J. Mol. Biol. 214: 261�279.

U m e s h K., A.U. J i n w a l, R. R o y, C. A b h i j i t, A. C h o w d h u r y, A.P. B h a d u r i c and P.K. R o y. 2003. Purificationand characterization of an alkaline lipase from a newly isolated Pseudomonas mendocina PK-12CS and chemoselectivehydrolysis of fatty acid esters. Bioorg. Med. Chem. 11: 1041�1046.

v a n O o r t M.G., A.M. D e v e e r, R. D i j k m a n, M.L. T j e e n k, H.M. Ve r h e i j, G.H. d e H a a s, E. W e n z i g andF. G ö t z. 1989. Purification and substrate specificity of Staphylococcus hyicus lipase. Biochemistry 28: 9278�9285.

W h i t l o w M., A.J. H o w a r d, B.C. F i n z e l, T.L. P o u l o s, E. W i n b o r n e and G.L. G i l l i l a n d. 1991. A metal-media-ted hydride shift mechanism for xylose isomerase based on the 1.6 A Streptomyces rubiginosus structures with xylitol andD-xylose. Proteins 9: 153�173.

W i l l i a m s C.M., C.S. R i c h t e r, J.M. M a c k e n z i e and J.C.H. S h i h. 1990. Isolation, identification, and characterizationof a feather-degrading bacterium. J. App. Environ. Microbiol. 56: 1509�1515.

Ya m a m o t o K. and N. F u j i w a r a. 1988. Purification and some properties of a castor-oil hydrolyzing lipase from Pseudomonassp. Agri. Biological. Chem. 52: 3015�3021.

Yo n g S.L. and J.S. R h e e. 1993. Production and partial purification of a lipase from Pseudomonas putida 3SK. Enz. Microb.Technol. 15: 617� 623.


Extraction of Milk-clotting Enzyme Produced by Solid State Fermentationof Aspergillus oryzae

HODA M.A. SHATA

Microbial Chemistry Department, National Research Center, El Behoos Street,Dokki, Cairo, Egypt

Received 9 March 2005, received in revised form 7 June 2005,accepted 10 June 2005

A b s t r a c t

Studies on the extraction of milk-clotting enzyme after solid-state fermentation (SSF) of wheat bran by a local strain ofAspergillus oryzae LS1 were done. The extraction of the enzyme was found to be depended on different parameters likenature of extractant, soaking time, temperature etc. From different inorganic and organic extractants, calcium chloride(0.05%) and glycerol (40%) were found to be the best solvents for leaching out milk-clotting enzyme. The optimumvolume of calcium chloride was 5 ml/g biomass. An extraction time of 180 min. at 30°C and 100 rpm was sufficient toextract out nearly 28% of the enzyme (2666.7 U/g biomass). Most of the enzyme (about 98.71%) was recovered in fourrepeated extractions.

K e y w o r d s: milk-clotting enzyme, extraction, solid-state fermentation, Aspergillus oryzae

Introduction

Solid-state fermentation is the one in which microorganisms secret the necessary enzyme for degrada-tion of the available substrate molecules in order to meet their nutritional requirement. In this system thefermented mass consists of non-utilized solid substrate containing microbial cells, spores and product forwhich a number of co-metabolites are formed during the course of fermentation.

Solid-state fermentation is the fermentation in the absence of free liquid and recovery of fermentationproducts requires its extraction from the solid fermented medium with a suitable solvents or solutions (Tungaet al., 1998). From the bulky solid mass, getting the product out of the system has many problems (Lonsaneand Krishnaiah, 1992). Attempt has been made by researchers to isolate the desired product from the fer-mented mass by various techniques due to its implication on process economics (Bjurstrom, 1985 and Caltonet al., 1986). With the initial moisture levels used for fermentation, squeezing of the solid medium itselfhardly yields any extract and even if there is any, the volume might not be sufficient for complete extraction.The medium must therefore, be soaked for some time in an adequate amount of a suitable extractant for totalrecovery of the product. A common extractant is distilled or deionized water (Silman, 1980; Wang et al.,1984; Yano et al., 1991 and Ghidyal et al., 1993). Bhumibhamon (1986) used distilled water for extractionof glucoamylase enzyme from solid-state on rice bran. Malathi and Chakraborty (1991) also used distilledwater for extraction of alkaline protease from solid wheat bran fermentation. Tunga et al., 1998 also reportedthat fermented mass was soaked with water for two hours at room temperature to extract proteolytic enzyme.Other extractants have also been used for extract other enzymes. Yang and Chiu (1987) used sodium chloridesolution for protease extraction. Shata (1999) extracted glucoamylase by sodium chloride solution. Rivera-Munoz et al. (1991) used 0.002 M succinate buffer for lipase and protease while Castilho et al. (2000)extracted pectinases from wheat bran by acetate buffer at pH 4.4.

This paper presents, studies on milk-clotting enzyme extraction from wheat bran, a cheap agro-residue,fermented by a local strain of Aspergillus oryzae LS1.The study includes the effect of some factors whichinfluence the efficiency of leaching out of the enzyme and its efficacy in the leaching technique.

242 Shata H.M.A. 3

Experimental


Microorganisms. Aspergillus oryzae LS1 a local strain obtained from the Microbial Resource Center at Cairo (MIRCEN), AinShams University was used through out this work.

Inoculum preparation. Fungal spores were prepared on Czapek�s Dox agar medium in Petri dish incubated for 7 days at 30°Cto assure good sporulation. The amount of inocula determined as colony-forming units (CFU) was prepared by scraping agar discsof 1.5 cm diameter aseptically from agar plate.

Fermentation. Wheat bran (10 gram) in 500 ml conical flask sterilized at 121°C for 20 min. were mixed with 15 ml of a sterilesalt solution containing (g/l): magnesium chloride 15×10�3, and fructose, 8 previously sterilized at 110°C for 10 min. Each flaskwas inoculated with 186×106 colony-forming units. Incubation was carried out at 30oC for three days.

Extraction. The extraction of the enzyme from the fermented biomass was carried out with distilled water, tap water anddifferent salt solutions of, potassium chloride, magnesium chloride, calcium chloride and sodium chloride. The investigation wasalso, carried out to see the effect of different organic solvents like glycerol, methanol, ethanol and acetone in concentration of 5%.

Unless otherwise stated, the extraction was conducted in 250 ml conical flask containing one gram of fermented biomass and5 ml of solvent solution. Each type of solvent was added separately to the fermented biomass and kept for 180 min. on a rotaryshaker at 100 rpm at 30°C. Then the extract was collected by filtration through Whatman No 1 filter paper and the clear solutioncontaining milk-clotting enzyme was assayed.

To optimize the extraction process, different experiments were carried out by varying the volume and solvent concentration inaddition to time, temperature and physical state (agitation or stationary) keeping all other conditions at optimum levels.

Milk-clotting enzyme assay. The assay of milk-clotting enzyme activity was carried out according to the standard proceduredescribed by Carlson et al. (1985) with some modifications using 12% (w/v) dried skim milk in 0.01 M of calcium chloride solutionas substrate. The reaction mixture contained 0.5 ml of enzyme and 1 ml of substrate and incubated at 70°C. The enzyme activitywas calculated according to Otani et al. (1991) as follows: Milk-clotting activity units = 2400/TXS/E where T is the time necessaryfor the crude fragment formation, S is the volume of milk, and E is the volume of enzyme.

Protein estimation. Protein content was estimated by the method of Ohanistti and Bar (1978).

Results

Influence of solvent type. Effect of different solvents on milk-clotting enzyme extraction from solid-state fermented wheat bran is presented in Figs. 1 and 2. Inorganic and organic solvents beside tap anddistilled water were used.

0.00 0.02 0.04 0.06 0.08 0.10500

1000

1500

2000

2500

MgCl2

NaCl

KCl

CaCl2

###

Milk

-clo

ttin

g a

ctivity (

U/g

bio

mass)

Fig. 1. Effect of different concentrations of magnesium chloride, sodium choride, potassium chorideand calcium choride on extraction of milk-clotting enzyme from SSF

Salt conc. g/100 ml

243Milk-clotting enzyme extraction from SSF by A.oryzae3

Among four inorganic salt solutions tested (sodium chloride, potassium chloride, calcium chloride andmagnesium chloride), calcium chloride in concentration of 0.05% (w/v) gave the best extraction of milk-clotting enzyme from the fermented solids. As the concentration of calcium chloride increased, the amountof extracted enzyme increased reaching its maximum value at 0.05% (w/v). Addition of different concentra-tions of potassium chloride, sodium chloride or magnesium chloride, gave no significant results comparedto calcium chloride solution.

Among four organic solvents (methanol, glycerol, ethanol, and acetone) investigated in concentrationof 5% (v/v), glycerol gave the maximum leaching of milk-clotting enzyme. To optimize the concentration ofglycerol in the solution used as solvent, experiment was carried out by varying the glycerol concentrationfrom 5% to 60% keeping other parameters constant. As the concentration of glycerol increased milk-clottingenzyme recovery increased (Fig. 3). The increasing trend was observed up to 40%, beyond which there wasa slight decrease. Therefore, calcium chloride of 0.05% (w/v) was used through the next experiments.

Optimization of solvent volume. Fig. 4 shows the effect of different volumes of 0.05% calcium chlo-ride solution used as extractant on recovery of milk-clotting enzyme from the fermented wheat bran in SSF.

Fig. 2. Effect of different organic solvents and water extraction of milk-clotting enzyme from SSF

Acetne Ethanole Methanol Glycerol Tap water Dist. water0

500

1000

1500

Solvents

Milk

-clo

tting

act

ivity

U/g

bio

mas

s

Fig. 3. Effect of different concentrations of glycerol on extraction of milk-cloting enzyme from SSF

0 10 20 30 40 50 60

1600

1700

1800

1900

2000

2100

2200

Milk

-clo

tting

act

ivity

U/g

bio

mas

s

Glycerol % (v/v)

244 Shata H.M.A. 3

The range investigated was 2.5�30 ml/g biomass with 3 h soaking time at 30°C and 100 rpm. The totalactivity of the extract increased up to 5 ml/g biomass, above which it remained more or less constant. How-ever, the amount of extracted protein increased with the increase of volume of calcium chloride up to 20 ml.

Effect of physical state. Fig. 5 shows the effect of physical state on extraction of milk-clotting enzyme.It was found that agitation was quite satisfactory for maximum recovery of the enzyme. So it was selected asthe best condition for enzyme recovery.

Effect of extraction time. Soaking of the fermented solid substrate with calcium chloride solution at theoptimum ratio 5 ml/g biomass was done at 30°C and at 100 rpm for different periods varying from 15 min to48 hours. As shown from Fig. 6 maximum amount of milk-clotting enzyme was extracted after 180 min.Longer extraction time did not result in significant gain of recovery.

Effect of temperature on extraction process. To evaluate the effect of temperature on leaching pro-cess, a series of experiments were carried out at 10, 30, 40, 50, 60 and 70°C, keeping the other experimentalconditions at optimum. The results presented in Table I show that 30°C was the optimum temperature forextraction of the enzyme. At higher temperature above 30°C the activity decreased.

Fig. 4. Effect of calcium chloride volume on extraction of milk-cloting enzyme from SSF

Calcium chloride volume (ml)

0 5 10 15 20 25

1000

1200

1400

1600

1800

2000

2200

2400

BM

ilk-c

lott

ing a

ctivity (

U/g

bio

mass)

2

4

6

CP

rote

in m

g/m

l

Milk

-clo

tting

act

ivity

U/g

bio

mas

s

Fig. 5. Effect of phisical state on extraction of milk-cloting enzyme from SSF

Milk

-clo

tting

act

ivity

(U/g

bio

mas

s)

0

500

1000

1500

2000

2500

3000

Aggitation Static


Repeated extraction. Repeated extractions were carried out for recovery of the most milk-clotting en-zyme from the fermented solid mass. From Table II, it was observed that most of milk-clotting enzyme wasrecovered during four repeated extractions. About 28% of the total activity was found in each of the firstthree extractions.

Fig. 6. Effect of soaking time on extraction of milk-cloting enzyme produced in SSF

0 50 100 150 200 250

1200

1400

1600

1800

2000

2200

2400

2600

2800

Milk

-clo

tting

act

ivity

(U/g

bio

mas

s)

Time (min.)

Recovery in the 1st washing 533 2667 533.3 28 28

Recovery in the 2nd washing 527 2637 1060.8 29 57

Recovery in the 3rd washing 527 2637 1588.3 28 86

Recovery in the 4th washing 264 1319 1852.0 13 99

Recovery in the 5th washing 5 27 1857.3 1 100

Table IIMilk-clotting activity recovery in five repeated extractions with fresh calcium chloride solution

Number of recovery stageMCA

U/ml U/g

CumulativeMCA U/ml

CumulativeMCA U/ml %

MCA U/ml%

MCA = Milk-clotting activity

10 16 80 3 97

20 26 133 5 95

30 505 2526 100 0

45 231 1154 46 54

50 53 267 11 89

60 20 100 4 69

70 0 0 0 0

Table IEffect of temperature on extraction of milk-clotting enzyme

Temperature°C

MCA

U/ml U/g

Lossof activity %

Remainingactivity %

MCA = Milk-clotting activity

246 Shata H.M.A. 3

Discussion

Solid-state fermentation is fermentation in the absence of free liquid, and recovery of the fermentationproduct requires its extraction from the solid fermented medium. Among the salt solutions tested, calciumchloride (0.05% w/v) gave the best extraction of milk-clotting enzyme from the fermented solids. However,it�s interesting to notice that the extraction of milk-clotting enzyme increased with all salt solutions,as compared with tap or distilled water. The above observation was recorded for the protease enzymeproduced by Rhizopus oryzae in solid-state fermentation of wheat bran, (Aikat and Bhattacharyya, 2000).They mentioned some weak ionic binding of the enzyme with either cell or substrate. On the other handWang (1967) reported from the studies on Mucor hiemalis that most of the proteolytic activity is cell surfacebound and could be released by elution with sodium chloride suggesting that the enzyme was probablybound to the cell by weak ionic bonds. Castilho et al. (1999 and 2000) reported that acetate buffergave better results for extraction of pectinases and from the fermented solids by Asp. niger than glycerolor distilled water. The poor performance of glycerol can be explained by its higher viscosity, whichhas a negative effect on mass transfer. Corpe and Winters (1972) found most of the protease activity to becell bound, largely being associated with the cell envelop. They mentioned also that about 50% of thespecific protease activity of the cell envelope could be removed with three washes with sea water, indicatinga fairly weak binding.

The effect of calcium chloride concentration on enzyme extraction was a remarkable observation. Up to0.05% (w/v), extraction increased probably due to the salting-in effect of electrostatic effect of the salt(Aikat and Bhattacharyya, 2000). Beyond the above mentioned concentration a slight salting-out effect dueto hydrophobic effect began to show up resulting in a slight decrease in enzyme activity.

In the present investigation the presence of organic solvent in water was more effective in milk clottingenzyme extraction than tap or distilled water. Tunga et al. (1999) reported that in SSF protease was presentin solid fermented mass due to some binding force. Water has the highest dielectric constant compared withother organic solvents. They mentioned that as the dielectric constant decrease, the force of interactionbetween the enzyme and solvent may increase. Therefore, leaching out of the enzyme by inorganic solventwas more effective.

However, it was noticed that with different concentrations of glycerol, a small increase in enzyme activitywas recorded at higher concentration as shown in Fig. 3. This was due to the protective effect of glycerol onenzyme activity. Scopes (1982) mentioned that glycerol forms strong hydrogen bonds with water, reducingthe water activity. On the other hand, Tunga et al. (1999) reported that protease extraction from SSF wasmaximally achieved using ethanol-glycerol mixture. They suggested that the hydroxyl group of ethanol andglycerol may form hydrogen bonding with protein molecules and this gives the stability of the enzymemolecule. As the literature report (Stryer, 1975, and Bailey and Ollis 1986) the stability of enzymes can beimproved by using sorbitol as solvent.

A contact time of 180 min. at 30°C provided the best enzyme activities for most of the conditions tested.Periods of 15, 30 or 60 min. seem not to be enough for total solubilization of milk-clotting enzyme presentin wheat bran medium. Increasing the time of extraction up to 24 hours has no promising value on extrac-tion of the enzyme. This loss could have been due to the prolonged mechanical or to a greater extraction ofdenaturant agents (Ghildyal et al., 1991). Castilho et al. (2000) investigated the effect of incubation timeon protease extraction in SSF and they found that 30 min. provided the best time for enzyme extraction.Also, Ikasari and Mitchell (1996) studied the effect of incubation time on extraction of protease in SSF.They concluded that maximum enzyme recovery was achieved at 22°C with 60 min. contact time.

According to Aikat and Bhattacharyya (2000) the amount of solute increases with the increase of solventvolume. Our results showed that the level of milk-clotting activity reaches the maximum at 5 ml/g biomassand then decreased. These results can be explained from the calculated specific activity since more solventcause the release of non specific protein. Excessively large volume of extractant used for greater extractionwould also yield enzyme solutions to be too dilute to be profitably utilized.

Agitation of the fermented biomass with the extractant at 30°C and 100 rpm gave appreciable amountof milk-clotting enzyme compared with stationary condition. Agitation helps to reduce enzyme adhesionto cell biomass and also disperses the fermented mass uniformly in the continuous phase of the solvent(Tunga et al., 1999).

Maximum yield of enzyme recovery was obtained at 30°C but at higher temperature it was less. Highertemperatures may have inhibitory effect on the enzyme activity and make it less stable. This fact couldbe explained by the two opposite effects on enzyme extraction. On one hand, greater temperatures increase


the solute solubility and diffusivity, therefore, a higher activity is attained in the extract. On the other hand,enzyme is susceptible to deactivation, which will increase as temperature and contact time increase.

In order to maximize enzyme recovery from solid-state culture, repeated extractions were investigated.It was verified that about 98.7% of milk-clotting activity were recovered in four repeated extractions usingcalcium chloride solution. As the extract obtained from the fifth washing contained about 1.3%, the reco-very of the enzyme was very small. Therefore, four repeated extractions are sufficient, although diluteextracts were obtained.

Literature

A i k a t K. and B.C. B h a t t a c h a r y y a. 2000. Protease extraction in solid-state fermentation of wheat bran by a local strain ofRhizopus oryzae and growth studies by soft gel technique. Process Biochemistry. 35: 907�914.

B a i l e y J.E. and D.E. O l l i s. 1986. Biochemical engineering fundamentals. McGraw Hill, New York.B h u m i b h a m o n O. 1986. Glucoamylase of fungus isolated from rotting cassava tuber by Aspergillus niger. Mircen-J. App.

Microbiol. Biotech. 2: 443�482.B j u r s t r o m E.E. 1985. Biotechnology. Fermentation and down stream processing. Chem. Eng. 92: 120�158.C a l t o n G.J., G.S. C o b b s and J.P. H a m m a n. 1986. Manual of industrial. microbiology and biotechnology; p. 436�445. In:

A.I. Demain and M.A. Solomn (eds), American Society of Microbiology, Washington, USA.C a r l s o n A., C.G. H i l l and N.F. O l s o n. 1985. Improved assay procedure for determination of milk-clotting enzyme.

J. Dairy Sci. 68: 290�299.C a s t i l h o L.R., T.L.M. A l v e s and R.A. M e d r o n h o. 1999. Recovery of pectolytic enzymes produced by solid-state

culture of A. niger. Process Biochemistry. 34:181�186.C a s t i l h o L.R., R.A. M e d o n h o and T.I. A l v e s. 2000. Production of pectinases obtained by solid-state fermentation of

agro-industrial residues with Aspergillus niger. Bioresource Technology 71: 45�50.C o r p e W.A. and H. W i n t e r s. 1972. Hydrolytic enzymes of some periphytic marine bacteria. Can. J. Microbiol. 18:

1483�1490.G h i l d y a l N.P., M. Ramakrishna, B.K. I o n s a n e, N.G. K a r a n t h and M.M. K r i s h n a i a h. 1993. Temperature varia-

tions and amyloglucosidase levels at different bed depths in a solid-state fermentation system. Chem. Eng. J. 51: 17�23.G h i l d y a l N.P., M. R a m a k r i s h n a, B.K. L o n s a n e and N.G. K a r a n t h. 1991. Efficient and simple extraction of mouldy

bran in a pulsed column extractor for recovery of amyloglucosidase in concentrated form. Process Biochemistry 26: 235�241.I k a s a r i L. and D.A. M i t c h e l l. 1996. Leaching and characterization of Rhizopus oligosporus acid protease from solid-state

fermentation. Enzyme and Microbial Technology 19: 171�175.L o n s a n e B.K. and M.M. K r i s h n a i a h. 1992. Leaching of the product and further down stream processing. In: solid-

substrate cultivation, Doelle, H.S.; Mitchell, D.A. and C.E. Rolz. (eds), Elsevier Science Publishers, Essex, U.K.M a l a t h i S. and R. C h a k r a b o r t y. 1991. Production of alkaline protease by a new Aspergillus flavus isolate under solid-

substrate fermentation conditions for use as a depilation agent. Applied and Environmental Microbiology. 57: 712�716.O h a n i s t t i S.T. and J.K. B a r. 1978. A simplified method of quantitating protein. The buriet and phenol reagents. Anal.

Biochem. 86: 193�200.O t a n i M., M. I w a g a k i and A. M o n s o n o. 1991. The screening trees having milk-clotting activity. Animal. Sci. Technol.

(Jap). 62: 417� 423.R i v e r a - M u n o z G., J.R. T i n o c o - Va b n i c, S. S a n c h e z and A. F a r r e s. 1991. Production of microbial lipase in

a solid-state fermentation system. Biotech. Lett. 13: 274�280.S c o p c s R.K. 1982. Protein purification: Principles and Practice. Springer. New York.S h a t a M.A.H. 1999. Ph.D. Thesis Microbiology Cairo University, Egypt.S i l m a n R.W. 1980. Enzyme formation during solid-substrate fermentation in rotating vessels. Biotech. Bioeng. 22: 411�420.S t r y e r L. 1975. Biochemistry; W. H. Freeman and company; New York; 2nd Edition.T u n g a R., R. B a n e r j e e and C.B. B h a t t a c h a r y y a. 1999. Some studies on optimization of extraction process for pro-

tease production in SSF. Bioprocess Engineering 20: 485�489.T u n g a R., R. B a n e r j e e and B.C. B h a t t a c h a r y y a. 1998. Optimizing some factors affecting protease production under

solid-state fermentation. Bioprocess Engineering 19: 187� 190.W a n g H.L. 1967. Release of proteinase from mycelium of Mucor hiemalis. J. Bact. 93: 1794�1799.W a n g H.L., E.W. S w a i n and C.W. H e s s e l t i n e. 1984. Isolation, purification and characterization Glucoamylase of

Amylomyces rouxil J. Food. Sci. 49: 1210�1211.Ya n g S.S. and W.F. C h i u. 1987. Protease production with starchy agricultural wastes by solid-state fermentation. Microbe.

283. 86 Int. Cong.. Microbiol, 14 Meet.Ya n o T., S. A s h i d a, T. T a c h i k i, H. K u m a g a i and T. T o c h i k u r a. 1991. Development of a soft gel cultivation

method. Agric. Biol. Chem. 55: 379�385.

248 Shata H.M.A. 3


Cloning and Preliminary Characterizationof a GATC-specific $2-class DNA:m6A methyltransferase Encoded

by Transposon Tn1549 from Enterococcus spp.

MONIKA RADLIÑSKA1,*, ANDRZEJ PIEKAROWICZ1, MARC GALIMAND2

and JANUSZ M. BUJNICKI3

1 Institute of Microbiology, University of Warsaw, ul. Miecznikowa 1, 02-096 Warszawa, Poland2 Unité des Agents Antibactériens, Institut Pasteur, 75724 Paris Cedex 15, France

3 Laboratory of Bioinformatics and Protein Engineering, International Institute of Molecularand Cell Biology, Trojdena 4, 02-109 Warsaw, Poland

Received 13 April 2005, received in revised form 8 June 2005, accepted 10 June 2005

A b s t r a c t

A recent study revealed a subfamily of N6-adenine (m6A) methyltransferases that comprises a few functionally studiedeukaryotic members acting on mRNA and prokaryotic members acting on DNA as well as numerous uncharacterizedopen reading frames. Here, we report cloning and functional characterization of a prokaryotic member of this familyencoded by transposon Tn1549 from Enterococcus spp.

K e y w o r d s: DNA methyltransferase, sequence specificity

DNA of prokaryotic and eukaryotic organisms and their viruses is often modified by methylation, carriedout by S-adenosyl-L-methionine (AdoMet)-dependent DNA methyltransferases (MTases). In Eukaryota DNAmethylation plays a role in crucial regulatory processes, such as regulation of gene expression, embryonicdevelopment, genomic imprinting, and carcinogenesis (reviewed by Scarano et al., 2005). Most DNAMTases in prokaryota belong to the restriction-modification (RM) systems, where they serve to protect thehost genome against the cleavage by a cognate restriction endonuclease (REase). However, some prokaryoticMTases (here termed �solitary�) are not associated with REases and are involved in processes distinct fromrestriction, such as DNA mismatch repair, regulation of gene expression, and control of timing of DNAreplication (reviews: Dryden, 1999; Noyer-Weidner and Trautner, 1993). The best studied solitary MTase isDam of Escherichia coli, which specifically methylates adenine residues within the palindromic sequenceGATC, to yield N6-methyladenine (m6A) (Herman and Modrich, 1982). It is noteworthy that GATC methyla-tion serves as a regulator of gene expression including virulence factors. Accordingly, the Dam activity wasfound to be necessary for both in vitro and in vivo virulence of numerous pathogenic bacteria (Chen et al.,2003; Heithoff et al., 1999; Watson et al., 2004).

All known DNA MTases are homologous (i.e. they evolved from one common ancestor by accumulatingdivergent mutations) but their sequences are strongly divergent. One of the characteristic features of DNAMTases is the variability of the linear arrangements of nine common motifs (I�VIII and X). According tothe possible linear arrangements of conserved and variable regions, DNA MTases were subdivided into6 classes: ", $, (, *, g and . (Malone et al., 1995). The majority of DNA MTases fall into the classes ", $,and ( (Bujnicki, 2002; Malone et al., 1995). Recently, a subfamily of $-class enzymes was identified thatinclude structurally unusual DNA:m6A MTases M.MunI and M.AvaI as well as m6A MTases acting onmRNA (homologs of the MT-A70 protein) (Bujnicki et al., 2002; Matveyev et al., 2001). These enzymes

Abbreviations: RM, restriction-modification; MTase, methyltransferase; REase, restriction endonuclease; TRD, target recogni-tion domain; m6A, N6-methyladenine; m4C, N4-methylcytosine; m5C, C5-methylcytosine; AdoMet, S-adenosyl-L-methionine

* to whom correspondence should be addressed, e-mail: [email protected]

250 Radliñska M. et al. 3

lack the large variable region involved in the target sequence recognition, which is present between motifsVIII and X in the �orthodox� $-class MTases (Bujnicki et al., 2002). For clarity, this subclass of MTaseswill be referred to hereafter as $2, as proposed earlier (Matveyev et al., 2001).

Phylogenetic analyses and structure prediction suggested that M.MunI and M.AvaV are related to mRNAMTases more closely than to any other known DNA MTases. Interestingly, these two bacterial MTases arefunctionally dissimilar: M.MunI is a part of a restriction-modification system (Siksnys et al., 1994) whileM.AvaV is a solitary MTase (Matveyev et al., 2001). Their specificity is also different: M.MunI methylatesthe second adenine in the hexanucleotide CAATTG while M.AvaV methylates the GATC tetranucleotide(i.e. exhibits the Dam specificity). Thus, a comprehensive characterization of this intriguing subfamily ofDNA MTases requires functional characterization of other prokaryotic members.

Fig. 1. Sequence alignment of the M.MunI/M.AvaV/M.EfaBMDam family of prokaryotic $2-class MTases, including manyuncharacterized members

Protein names follow the REBASE format, and in the first panel include the Gene Identification numbers (NCBI). N-terminal extensionsof M.MunI (15 aa) and MspMCORF3385P (210 aa, including a predicted ParB-like nuclease domain) have been omitted for clarity. Dashesrepresent insertions or deletions. Identical and conservatively substituted residues are shown on dark background (black and grey, respectively).

Conserved motifs (Malone et al., 1995) are indicated; it is noteworthy that motifs III and IX are missing.

251Short communication3

Among the proteins closely related to M.MunI and M.AvaI we foundan uncharacterized open reading frame (ORF18; dubbed M.EfaBMORFAP)in REBASE (Roberts et al., 2005) encoded by transposon Tn1549,which confers VanB-type resistance in Enterococcus spp. (Garnier et al.,2000); GenBank Acc.no. AF192329. Neither phylogenetic studiesreported earlier (Bujnicki et al., 2002) nor the detailed comparisonof sequence similarities between the ORF18 protein and M.MunI orM.AvaI (Figure 1) could reveal whether ORF18 may share sequencespecificity with any of these MTases. Therefore, we cloned the corre-sponding gene and tested its activity in vivo.

DNA manipulations, general techniques, and standard reactions weredone according to protocols described for E. coli (Sambrook et al., 2002)or following recommendation of the enzyme�s suppliers. The plasmidexpressing the putative M.EfaBMORFAP MTase was constructed as fol-lows: the 444 bp region encompassing open reading frame 18 was ampli-fied by PCR using Tn1549 DNA from Enterococcus faecalis 268�10 andthe following primers: Primer Ofr18Nde:5'-GAAGGAGATATACATATGTTGTTTATTTCAACGTACAACATC-3'corresponds to the 5' terminal part of the orf18 gene and possesses anoptimized Shine-Dalgarno sequence (underlined), the NdeI site (italic)with translation codon ATG (bold). Primer Orf18Mun:5'-CAATTGCTACGTCAGGATAAGGTCACATTCCAC-3' is comple-mentary to the 3' terminal part of the orf18 gene and possesses a stopcodon (bold) and the MunI restriction site (italic). The amplified blunt-ended PCR fragment was inserted in the SmaI and Ecl136II cleavedpBluescript KSII(+) [Stratagene] generating plasmid pEfaORF18KS car-rying the orf18 gene cloned in-frame with the $-galactosidase promoterof the pBluescript vector. The nucleotide sequence of orf18 was verified.

Plasmids encoding an active DNA MTase are methylated in vivo andhence resistant to digestion by the corresponding restriction endo-nuclease, while the DNA from cells lacking the MTase activity arecompletely cleaved. The degree of resistance of the plasmid DNA todigestion by the restriction enzyme can be used as an indicator of therelative in vivo activity of the MTase clones. For this assay, pBluescriptplasmid carrying orf18 gene and the empty vector pBluescript weregrown in E. coli GM2163 (dam dcm) cells. Plasmid DNA from over-night cultures was prepared using DNA minipreparation kits [Sigma]according to the instruction of the supplier. The purified DNA was tested

Fig. 2. Cleavage of pEfaORF18KS andpBluescript DNA isolated from E. coli

GM2163 cells.

Aliquotes of 0.4 :g DNA were digested in20 :l reaction volumes with 10 u (25-foldexcess) of enzymes in buffers recom-mended by the manufacturers for 8 hrs at37oC. Lane A: non-digested DNA ofpBluescript; lane B: pBluescript digestedwith MboI; lane C: pBluescript digested withBsp143I; lane D: pEfaORF18KS digestedwith Bsp143I; lane E: pEfaORF18KS di-gested with MboI; lane F: non-digestedDNA of pEfaORF18KS; lane M DNA Lad-

der Mix (GeneRuler� � Fermentas).

A B C M D E F

for sensitivity to cleavage by several restriction enzymes (including MunI and MboI) known to be inhibitedby activity of m6A MTases (data not shown). pEfaORF18KS DNA was completely digested with all en-zymes used except MboI. The digestion reaction for pEfaORF18KS was repeated using isoschizomericrestriction enzymes MboI and Bsp143I, which differ in their sensitivity to Dam methylation. The cleavageby Bsp143I is not affected by adenine methylation in the 5�GATC3� sequence, whereas MboI depends on bythis modification (Hermann and Jeltsch, 2003; McClelland et al., 1994).

The results presented in Figure 2 show that plasmid pEfaORF18KS DNA was totally resistant to cleav-age by MboI but sensitive to Bsp143I. On the other hand, the pBluescript vector DNA was cleaved byMboI and Bsp143I, indicating that the cloned insert of pEfaORF18KS exerts the GATC-specific DNA:m6AMTase activity.

In this study, we have cloned and characterized the in vivo activity of M.EfaBMORFAP a M.MunI/M.AvaVhomolog from Enterococcus spp. transposon Tn1549. Preliminary studies suggest that the product of orf18encodes a DNA:m6A MTase with a Dam-like sequence specificity (GATC) � i.e. like M.AvaV and unlikeM.MunI. Therefore, we suggest to rename M.EfaBMORFAP as M.EfaBMDam. It will be interesting todetermine the function and sequence specificity of other members of the M.MunI/M.AvaV/M.EfaBMDamsubfamily. The identification of a Dam-like MTase on mobile genetic element Tn1549, which confers anti-biotic resistance in clinical isolates of Enterococcus spp., suggests that this important determinant of bacterialvirulence may be transmitted by horizontal gene transfer. It remains to be determined if M.EfaBMDam may

252 Radliñska M. et al. 3

be essential for the virulence of Enterococcus spp. In the light of the fact that Dam MTases are consideredas potential drug targets it is important to note that MTases from different classes, such as the �orthodox�Dam from E. coli and T-even phages � " class (Herman and Modrich, 1982), Dam MTases from phagessuch as HP1, VT-2 or T1 (Bujnicki et al., 2001; Piekarowicz and Bujnicki, 1999; Radliñska and Bujnicki,2001) � ( class and M.EfaBMDam and M.AvaV � $2 class (Matveyev et al., 2001) are structurally verydivergent, and therefore may require the development of different specific inhibitors.

Acknowledgments. This work was supported by the Polish State Committee for Scientific Research (grant 2P04B00827)

Literature

B u j n i c k i J.M. 2002. Sequence permutations in the molecular evolution of DNA methyltransferases. BMC Evol. Biol. 2: 3.B u j n i c k i J.M., M. F e d e r, M. R a d l i ñ s k a and R.M. B l u m e n t h a l. 2002. Structure prediction and phylogenetic analysis

of a functionally diverse family of proteins homologous to the MT-A70 subunit of the human mRNA: m6A methyltransferase.J. Mol. Evol. 55: 431�44.

B u j n i c k i J.M., M. R a d l i ñ s k a, P. Z a l e s k i and A. P i e k a r o w i c z. 2001. Cloning of the Haemophilus influenzae Dammethyltransferase and analysis of its relationship to the Dam methyltransferase encoded by the HP1 phage. Acta. Biochim.Pol. 48: 969�83.

C h e n L., D.B. P a u l s e n, D.W. S c r u g g s, M.M. B a n e s, B.Y. R e e k s and M.L. L a w r e n c e. 2003. Alteration of DNAadenine methylase (Dam) activity in Pasteurella multocida causes increased spontaneous mutation frequency and attenua-tion in mice. Microbiology 149: 2283�90.

D r y d e n D.T. 1999. Bacterial DNA methyltransferases. p. 283�340. In: Cheng X, Blumenthal RM (eds) S-Adenosylmethionine-dependent methyltransferases: structures and functions. World Scientific Publishing, NJ.

G a r n i e r F., S. T a o u r i t, P. G l a s e r, P. C o u r v a l i n and M. G a l i m a n d. 2000. Characterization of transposon Tn1549,conferring VanB-type resistance in Enterococcus spp. Microbiology 146 1481�9.

H e i t h o f f D.M., R.L. S i n s h e i m e r, D.A. L o w and M.J. M a h a n. 1999. An essential role for DNA adenine methylationin bacterial virulence. Science 284: 967�70.

H e r m a n G.E. and P. M o d r i c h. 1982. Escherichia coli dam methylase. Physical and catalytic properties of the homogeneousenzyme. J. Biol. Chem. 257: 2605�2612.

H e r m a n n A. and A. J e l t s c h. 2003. Methylation sensitivity of restriction enzymes interacting with GATC sites. Biotechniques34: 924�6, 928, 930.

M a l o n e T., R.M. B l u m e n t h a l and X. C h e n g. 1995. Structure-guided analysis reveals nine sequence motifs conservedamong DNA amino-methyltransferases, and suggests a catalytic mechanism for these enzymes. J. Mol. Biol. 253: 618�632.

M a t v e y e v A.V., K.T. Yo u n g, A. M e n g and J. E l h a i. 2001. DNA methyltransferases of the cyanobacterium AnabaenaPCC 7120. Nucleic. Acids. Res. 29: 1491�506.

M c C l e l l a n d M., M. N e l s o n and E. R a s c h k e. 1994. Effect of site-specific modification on restriction endonucleasesand DNA modification methyltransferases. Nucleic. Acids. Res. 22: 3640�3659.

N o y e r - W e i d n e r M. and T.A. T r a u t n e r. 1993. Methylation of DNA in prokaryotes. EXS 64: 39�108.P i e k a r o w i c z A. and J.M. B u j n i c k i. 1999. Cloning of the Dam methyltransferase gene from Haemophilus influenzae

bacteriophage HP1. Acta Microbiol. Pol. 48: 123�9.R a d l i ñ s k a M. and J.M. B u j n i c k i. 2001. Cloning of enterohemorrhagic Escherichia coli phage VT-2 dam methyltransferase.

Acta Microbiol. Pol. 50: 161�7.R o b e r t s R.J., T. V i n c z e, J. P o s f a i and D. M a c e l i s. 2005. REBASE-restriction enzymes and DNA methyltransferases.

Nucleic Acids Res. 33 Database Issue: D230�2.S a m b r o o k J., D.W. R u s s e l l and J. S a m b r o o k. 2002. Molecular cloning: a laboratory manual. Cold Spring Harbor

Laboratory Press, Cold Spring Harbor, N.Y.S c a r a n o M.I., M. S t r a z z u l l o, M.R. M a t a r a z z o and M. D� E s p o s i t o. 2005. DNA methylation 40 years later: Its

role in human health and disease. J. Cell. Physiol.S i k s n y s V., N. Z a r e c k a j a, R. Va i s v i l a, A. T i m i n s k a s, P. S t a k e n a s, V. B u t k u s and A. J a n u l a i t i s. 1994.

CAATTG-specific restriction-modification munI genes from Mycoplasma: sequence similarities between R.MunI and R.EcoRI.Gene 142: 1�8.

W a t s o n M.E., Jr., J. J a r i s c h and A.L. S m i t h. 2004. Inactivation of deoxyadenosine methyltransferase (dam) attenuatesHaemophilus influenzae virulence. Mol. Microbiol. 53: 651�64.


Co-occurrence of Urogenital Mycoplasmas and Group B Streptococciwith Chlamydial Cervicitis

DANIELA FRIEDEK1, ALICJA EKIEL1, MA£GORZATA ROMANIK1, ZBIGNIEW CHELMICKI2,BARBARA WIECHULA1, IWONA WILK1, JAROS£AW JÓZWIAK3 and GAYANE MARTIROSIAN1,3*

1 Department of Medical Microbiology, Medical University of Silesia, Katowice, Poland1 Department and Clinic of Gynecology and Endocrinology, Medical University of Silesia, Katowice, Poland

3 Department of Histology and Embryology Center of Biostructure Research,Warsaw Medical University, Warsaw, Poland

Received 8 March 2005, received in revised form 21 June 2005, accepted 23 June 2005

A b s t r a c t

The aim of our study was to evaluate whether in women with chlamydial cervicitis urogenital mycoplasmas and groupB streptococci (GBS) colonization is found more often than among women with non-chlamydial cervicitis. This studyincluded 351 (mean age 31.7 ± 6.82) not pregnant, menstruating, sexually active women. We confirmed a highfrequency (49.3%) of C. trachomatis infection among women with cervicitis. Cervical ectopia was confirmed in26.5% of examined women, in half of them ectopia was associated with chlamydial infection. We did not noticedifferences in frequency of colonization by urogenital mycoplasmas and GBS among women with chlamydial andnon-chlamydial cervicitis.

K e y w o r d s: Chlamydia trachomatis, GBS, urogenital mycoplasmas, cervicitis, ectopia

Urogenital mycoplasmas and group B streptococci (GBS) are microorganisms colonizing female uro-genital tract and playing an important role in the pathology of fetus and newborn. Urogenital mycoplasmasare often isolated, even in 54% of tested sexually active women of childbearing age (Schlicht et al., 2004).

Sexual transmission of GBS is questioned (Honig et al., 2002), while it is recognized in urogenitalmycoplasmal infection (Keane et al., 2000, Nunez-Troconis, 1999).

Today Chlamydia trachomatis is on the first place among sexually-transmitted bacteria (Millman et al.,2004). Cervicitis, often with co-occurring ectopia, is the dominating clinical finding during C. trachomatisinfection (Critchlow et al., 1995, Giedrys-Kalemba et al., 1994). Thus, the aim of our work was to evaluate,whether in women with chlamydial cervicitis urogenital mycoplasmas and GBS colonization is found moreoften than among women with non-chlamydial cervicitis.

This study included 351 (mean age 31.7 ± 6.82) not pregnant, menstruating, sexually active women whoattended the Department and Clinic of Gynecology and Endocrinology, Medical University of Silesiain Katowice between 2001 and 2004. Cytological examination of cervix was performed in each case. Allstudied women had symptoms of cervicitis: mucopurulent endocervical discharge and/or greater or equalto 30 neutrophils per × 1000 field on the cervical Gram stain, and/or bleeding contact.

Patients with gonococcal infection and those receiving antibiotic therapy within the month beforeconsultation were excluded from the study.

Sterile swabs were used to obtain material for testing/culturing of expected microorganisms (Friedeket al., 2004). First swab (no 1) from vaginal fornix for GBS culturing was inoculated on Columbia sheepblood agar plate and incubated aerobically for 24�48 hours at 37°C. Identification of GBS was basedon latex Slidex Streptokit (bioMerieux, France). Susceptibility of isolated GBS to antibiotics (ampicillin,

* Corresponding author: Gayane Martirosian1,3, Department of Medical Microbiology, Medical University of Silesia,18 Medyków str. 40-752 Katowice, Poland, phone/fax: +48 32 262 5075, e-mail: [email protected], [email protected]

254 Friedek D. et al. 3

erythromycin and clindamycin) was examined by disk-diffusion method. Second swab (no 2) from endocer-vical canal for isolation of genital mycoplasmas was inoculated in urea-arginine-broth transport medium(bioMerieux, France). Culturing of genital mycoplasmas was performed by using Mycoplasma IST(bioMerieux, France), according to manufacturer�s instruction. Mycoplasma IST shows good sensitivityand specificity for U. urealyticum (100% and 90%, respectively), and for M. hominis (100% and 85%respectively) (Rastawicki et al., 2004). Third swab (no 3) from endocervical canal for detection ofC. trachomatis was fixed by acetone on a slide. Chlamydia Direct IF � DIF (bioMérieux, France) was usedfor C. trachomatis antigen detection, according to manufacturers instruction. Slides were examined in NikonModel HB � 10101AF fluorescent microscope (x40 objective).

In studied group chlamydial etiology of cervicitis was confirmed in 49.3% (173/351). Genital mycoplas-mas were isolated in 25.9% of women with cervicitis. There were statistically insignificant differencesbetween occurrence of urogenital mycoplasmas in women with and without chlamydial cervical infection.U. urealyticum was a more frequently isolated species than M. hominis (Table I).

ectopia 42 24.3 51 28.7

Ureaplasma urealyticum 36 20.8 31 17.4

Mycoplasma hominis 4 2.3 3 1.7

Ureaplasma urealyticum and Mycoplasma hominis 7 4.0 10 5.6

GBS 23 13.3 26 14.6

Table IPrevalence of cervical ectopia, urogenital mycoplasmas and GBS among women with chlamydial

and non-chlamydial cervicitis

C. trachomatis � positivewomen (n = 173)

C. trachomatis � negativewomen (n = 178)

No % No %

The frequency of GBS isolation was 13.3% in chlamydia-positive and 14.6% in chlamydia-negativewomen. All isolated GBS strains were sensitive to ampicillin, only 9.6% of strains were resistant to erythro-mycin and 7.7% � to clindamycin.

Cervical ectopia was confirmed in 26.5% (93/351) of examined women. In 42 out of them ectopia wasassociated with cervical chlamydial infection.

In regions, where early screening program for detection of C. trachomatis infection was established,percent of infection is very low: in the USA � 4.7%, in Sweden � 5.4%, in Norway � 2.4% (Bakken et al.,2004, Egger et al., 1998, Miller et al., 2004). In Poland, frequency of C. trachomatis infection in studiedgroups of symptomatic and asymptomatic women is around 20�40% (Choroszy-Król et al., 1994, Giedrys-Kalemba et al., 1994, Zbroch et al., 2004). It is a well-known fact that cervicitis may be a predisposingfactor for cervical ectopia (Critchlow et al., 1995, Giedrys-Kalemba et al., 1994). In our study in 45.2% ofwomen with cervical ectopia we showed co-existence of C. trachomatis infection. It is in concordance withthe data of other authors: 46.9% reported by Giedrys-Kalemba et al. (1994) and 39.7% by Barnes et al.(1990). However, when analyzing cervical ectopia rate in women with and without C. trachomatis infec-tion, we obtained similar results (24.3% and 28.7%, respectively).

Urogenital mycoplasmas are frequently isolated from clinical samples. We did not notice differences infrequency of colonization by urogenital mycoplasmas among women with chlamydial and non-chlamydialcervicitis. The ratio was 27.2% and 24.7%, respectively (Table I). Maeda et al. (2004) did not observestatistically significant differences in the frequency of isolation of mycoplasmas among NGU patients with-and without chlamydial infection. U. urealyticum was isolated much more often than M. hominis, whichagrees well with the results of other authors (Keane et al., 2000, Schlicht et al., 2004). Schlicht et al. (2004)showed high prevalence of genital mycoplasmas among sexually active young women with cervicitis (54%for ureaplasmas and 26% for M. hominis). They also observed a high level (16%) colonization of healthyfemale volunteers by mycoplasmas. High level of mycoplasmal colonization in asymptomatic women wasalso reported by Keane et al. (2000): appropriatly 29% for U. urealyticum and 12% for M. hominis.

In our study we demonstrated 14% of GBS-positive swabs obtained from vaginal fornix. We did notobserve any significant correlation between occurrence of C. trachomatis and GBS or urogenital mycoplas-

255Short communication3

mas and GBS. Honig et al. (2002) did not demonstrate any correlation of vaginal colonization with GBSand chlamydial infection or other STIs. Frequency of isolation for these streptococci from the urogenitaltract of healthy women was estimated to be 7% to 34% (Bayo et al., 2002, Manning et al., 2001).

In spite of long-time using of penicillins in the treatment of streptococcal infections GBS are still sensi-tive to this group of antibiotics. The sensitivity of GBS to penicillins and percentage of resistance to erythro-mycin (9.6%) and clindamycin (7.7%) in our study was similar to that reported by others (Stiller et al.,2003, Weisner et al., 2004).

Our study confirms high frequency of C. trachomatis infection among women with cervicitis in theregion of Upper Silesia. However we demonstrated that C. trachomatis infection does not influence uro-genital colonization by mycoplasmas and GBS.

Literature

B a k k e n I.J., F.E. S k j e l d e s t a d, T. O v r e n e s s, S.A. N o r d b o and G. S t o r v o l d. 2004. Prevalence of Chlamydiatrachomatis by sexual behaviour among young women. Tidsskr. Nor. Laegeforen. 124: 1633�1635.

B a r n e s R.C., B.P. K a t z, R.T. R o l f s, B. B a t t e i g e r, V. C a i n e and R.B. J o n e s. 1990. Quantitative culture of endo-cervical Chlamydia trachomatis. J. Clin. Microbiol. 28: 774�780.

B a y o M., M. B e r l a n g a and M. A g u t. 2002. Vaginal microbiota in healthy pregnant women and prenatal screening ofgroup B streptococci (GBS). It. Microbiol. 5: 87�90.

C h o r o s z y - K r ó l I. and J. R u c z k o w s k a. 1994. The prevalence of C. trachomatis genitourinary tract infection 1986�1993.Przegl. Epidemiol. 48:261�264.

C r i t c h l o w C., P. W o l n e r - H a n s s e n, D. E s c h e n b a c h, N. K i v i a t, L. K o u t s k y, C. S t e v e n s and K. H o l m e s.1995. Gynecology. Determinants of cervical ectopia and cervicitis: Age, oral contraception, specific cervical infection, smokingand douching. Am. J. Obstet. Gynecol. 173: 534�543.

E g g e r M., N. L o w, G.D. S m i t h, B. L i n d b l o m and B. H e r r m a n n. 1998. Screening for chlamydial infections and therisk of ectopic pregnancy in a county in Sweden: ecological analysis. BMJ. 316: 1776�1780.

F r i e d e k D., A. E k i e l, Z. C h e l m i c k i and M. R o m a n i k. 2004. HPV, Chlamydia trachomatis and genital mycoplasmasinfections in women with Low Grade Squamous Intraepithelial Lesions (LSIL). Ginekol. Pol. 75: 457�463.

G i e d r y s - K a l e m b a S., E. W y d r a, J. H a ³ a s a and I. B i l s k a. 1994. The biocenosis of the vagina and infection byChlamydia trachomatis. Med. Do�w. Mikrobiol. 46: 73�77.

H o n i g E., J.W. M o u t o n and v a n d e r W.I. M e i d e m. 2002. The epidemiology of vaginal colonisation with group Bstreptococci in a sexually transmitted disease clinic. Eur. J. Obstet. Gynecol. Reprod. Biol. 105: 177�180.

K e a n e F.E., B.J. T h o m a s, C.B. G i l r o y, A. R e n t o n and D. T a y l o r - R o b i n s o n. 2000. The association of Myco-plasma hominis, Ureaplasma urealyticum and Mycoplasma genitalium with bacterial vaginosis: observations on heterosexualwomen and their male partners. Int. J. STD AIDS. 11: 356�360.

M a e d a S.I., T. D e g u c h i, H. I s h i k o, T. M a t s u m o t o, S. N a i t o, H. K u m o n, T. T s u k a m o t o, O. S h o u i c h iand S. K a m i d o n o. 2004. Detection of Mycoplasma genitalium, Mycoplasma hominis, Ureaplasma parvum (biovar 1)and Ureaplasma urealyticum (biovar 2) in patients with non-gonococcal urethritis using polymerase chain reaction-microtiterplate hybridization. Int. J. Urol. 11: 750�775.

M a n n i n g S.D., M.D. P e a r l m a n, P. T a l l m a n, C.L. P i e r s o n and B. F o x m a n. 2001. Frequency of antibiotic resis-tance among group G Streptococcus isolated from healthy college students. Clin. Infect. Dis. 33: 137�139.

M i l l e r W.C., C.A. F o r d and M. M o r r i s. 2004. Prevalence of chlamydial and gonococcal infections among young adults inthe United States. JAMA. 291: 2229�2236.

M i l l m a n K., C.M. B l a c k, R.E. J o h n s o n, W.E. S t a m m, R.B. J o n e s, E. H o o k, D.H. M a r t i r n, G. B o l a n,S. T a v a r e and D. D e a n. 2004. Population-based genetic and evolutionary analysis of Chlamydia trachomatis urogenitalstrain variation in the United States. J. Bacteriol. 186: 2457�2465.

N u n e z - T r o c o n i s J.T. 1999. Mycoplasma hominis and Ureaplasma urealyticum in different gynecologic diseases. Invest.Clin. 40: 9�24.

R a s t a w i c k i W., H. K a l o t a, M. J a g i e l s k i and R. G i e r c z y ñ s k i. 2004. Comparison of Polymerase Chain ReactionAssay and Mycoplasma IST 2Test with culture for detection of Infections caused by Ureaplasma urealyticum and Myco-plasma hominis. Med. Dosw. Mikrobiol. 56: 99�108.

S c h l i c h t M.J., S.D. L o v r i c h, S. S a r t i n, P. K a r p i n s k y, S.M. C a l l i s t e r and W.A. A g g e r. 2004. High preva-lence of genital mycoplasmas among sexually active young adults with urethritis or cervicitis. Symptoms in La Crosse,Wisconsin. J. Clin. Microbiol. 42: 4636�4640.

S t i l l e r R.J., L. P a d i l l a, R. C h o u d h a r y, T. T i n g h i t e l l a and S. L a i f e r. 2003. Group B streptococcal antibioticresistance patterns in pregnant women. Conn. Med. 67: 323�326.

W e i s n e r A.M., A.P. J o h n s o n, T.L. L a m a g n i, E. A r n o l d, M. W a r n e r, P.T. H e a t h and A. E f s t r a t i o u. 2004.Characterization of group B streptococci recovered from infants with invasive disease in England and Wales. Clin. Infect.Dis. 38: 1203�1208.

Z b r o c h T., P. K n a p p, E. B ³ o n s k a, M. K o b y l e c and P. K n a p p. 2004. Chlamydia trachomatis infection and Bacterialvaginosis in the aspect a life style and their impact on cervical lesions occurrence. Ginek. Pol. 75: 538�544.

256 Friedek D. et al. 3

Polish Journal of Microbiologyformerly Acta Microbiologica Polonica

2005, Vol. 54, No 3

Instructions to authors

I. General information

Polish Journal of Microbiology publishes descriptions of all aspects of basic and applied research thatfocuses on topics of basic research of practical value in microbiology. Topics that are considered includemicrobiology of a genetic and molecular nature, foods, agriculture, industry, biotechnology, microbial eco-logy, public health and basic biological properties of bacteria, viruses, and simple eukaryotic microorganisms.

Submit manuscripts directly to the Editorial Office, Polish Journal of Microbiology. The manuscriptshould be accompanied by a covering letter stating the address, fax number, e-mail of the correspondingauthor and �running head� of the manuscript (no longer than 47 characters). Submit two complete copies ofeach manuscript, including figures and tables. The manuscript should be either the original typescript fromjet or laser printer (not dot matrix).

Accepted papers are copy-edited as word-processor files, so authors are asked to provide their paperin this form on a disk when they submit the revised version. The text should be edited in Word 7 or higher orASCII. Submission of figures in TIF or CorelDraw Format is appreciated.

All manuscripts are subjected to peer review by the editors, by members of the editorial board and byqualified outside reviewers. When a manuscript is returned to the authors for modification, it shouldbe returned to the editor within 2 months; otherwise it may be considered withdrawn.

15 reprints are sent free to the first author.

II. Preparation of the manuscript

The oryginal paper should be divided into the following sections written in sequence: Abstract,Introduction, Experimental: Materials and Methods and Results, Discussion, Acknowledgments, Literature.Type every portion of the manuscript double spaced with left hand margins, including figure legends, tables,table footnotes, and literature cited (type Literature sections on separate pages), and number all pages insequence, including the abstract, tables and figure legends. The literature section must include all citedwork. Arrange the citations in alphabetical order by first authors. Key words (no more than five) and thesuggestion of runing head should be included.

A paper in the form of a short communication must have an abstract of no more than 100 words. Do notuse section headings in the body of the �Communication�; report introduction, methods, results, and discus-sion in a single section. The text should be concise, and the number of figures and tables should be kept toa minimum. Material and methods should be described in the text, not in figure legends or table footnotes.Present acknowledgments as in full-length papers.

Minireviews are published in areas of particular interest and importance. They are usually invited, butauthors wishing to submit a minireview should contact the scientific editor for further information.

Before writing a manuscript authors are advised to consult a current issue of Polish Journal of Microbio-logy and carefully read the detailed �Instruction to authors� printed in number 1 of every volume in order tobe familiar with the literature citations, preparation of figures and tables and the rules concerning chemical,biochemical, genetic etc. nomenclature recommended.

Polish Journal of Microbiology, formerly Acta Microbiologica Polonica, is a broad-based microbiologicaljournal covering: General Microbiology, Bacterial Physiology and Genetics, Virology, Clinical Bacterio-logy, Environmental Microbiology and Biotechnology.

Polish Journal of Microbiology publishes mainly original research articles including short communicationsbut also minireviews, book reviews and press releases. All the papers should be prepared following�Instructions to Authors� which is always included in the first issue of every year. The papers appear usuallywithin 3 months after final acceptance.

Polish Journal of Microbiology is currently indexed by: BIOSIS/Biological Abstracts, EMBASE/ExcerptaMedica, Index Medicus/MEDLINE, Cambridge Scientific Abstracts (series: Biological Sciences, Micro-biology A, B, C), SCOPUS, Index Copernicus.

Submission of manuscripts in various areas of Microbiology is very welcome.

Advertising offer

The editorial office of the Polish Journal of Microbiology encourages our Readers to place advertisementsin our quarterly. Polish Journal of Microbiology reaches scientists working in various research institutionsand Universities and involved in different areas of microbiology, such as General Microbiology, MolecularMicrobiology, Clinical Microbiology, Environmental Microbiology, Biotechnology. We ensure rapid ap-pearance of the ad.

Price of black and white ad on inside cover:half page = 150 PLNfull page = 300 PLN

Price of ad inside an issuehalf page = 75 PLNfull page = 150 PLN

Please send file with graphical layout to the editorial office of the Polish Journal of Microbiology,02-096 Warsaw, Miecznikowa 1, phone 48 (22) 55 41 302

Do cz³onkówPolskiego Towarzystwa Mikrobiologów

Zarz¹d P.T.M. uprzejmie prosi wszystkich cz³onków o wp³atê zaleg³ych i bie¿¹cych sk³adek cz³onkowskich.Zachêcamy tak¿e do zaprenumerowania �Polish Journal of Microbiology�.

Documents

Polish Journal of Microbiology · Polish Journal of Microbiology with the following points: P04 Œ 5, P05 Œ 5, P06 Œ 6, T09 Œ 6, T12 Œ 6 POLISH SOCIETY OF MICROBIOLOGISTS 00-725