Gene, 81 (1989) 211-218 Elsevier

GENE 03007

211

Cloning and analysis of structural and regulatory pectate lyase genes of Envinia chysunthemi ENA

(Phasmid vectors; recombinant DNA; E. coli host; gene library construction; protein secretion)

Nick K. Yankovsky, Nick 0. Bukanov, Vita V. Gritzenko, Anatoly N. Evtushenkov, Michaei Yu. Fonstein and Vladimir G. Debabov

Institute for Genetics and Selection of rnd~rn’al M~cr~organ~~, Moscow (U.S.S.R,)

Received by K.F. Chater: 22 October 1988 Revised: 22 February 1989 Accepted: 30 April 1989

SUMMARY

Erwinia ch~~anthemi ENA structural and relator ptl genes, coding for pectate lyase (Ptl) were cloned in Exherichia coli cells. Phage vector JL47.1 and phasmid vector 2pMYF131 were used for constructing libraries of BamHI and EcoRI fragments, respectively, of Er. chrysanthemi chromosomal DNA. Among the 1100 hybrid clones containing BamHI Er. chrysanthemi DNA fragments and 11000 hybrid clones containing EcoRI fragments, six and 45 clones, respectively, were identified as having pectolytic activity. Two different structural genes, designated ptlA and ptlB, have been subcloned on multi-copy plasmids.

Genes ptlA and ptlB are located side by side on the chromosome of Er. chlysanthemi and transcribe in the same direction. Each of the genes has its own promoter. Southern-blot hybridization analysis showed that the cloned ptZ genes shared practically no homology and each of the genes was represented by a single copy on the Er. chrysanthemi chromosome. Other ptl genes capable of expression in E. coli cells were not found in the gene libraries. Negative regulation of the ptlA gene expression by a cloned gene called ptlR was shown. To screen the gene library for the ptlR gene, a specific genetic system was devised. The genes studied are located within an EcoRI chromosomal DNA fragment of 7.3 kb in the order: ptbf-ptlB-ptlR.

INTRODUCTION

Bacteria of the genus Erwinia secrete various enzymes, among them pectate lyase (Ptl; EC4.2.2.2),

Correspondence to: Dr. N.K. Yankovsky, Institute for Genetics and Selection of Industrial Microorganisms, Box 825, Moscow 113545 (U.S.S.R.) Tel. 3153774.

Abbre~atious: Ap, ~pici~n; bp, base pair(s); CAT, Cm acetyi- transferase; cut, gene coding for CAT; Cm, chloramphenicol;

which destroys the primary wall of plant cells ~Rombouts and Pilnik, 1980). The Ptl are produced by different species of the genus and can be accumu- lated in the media in large quantities (Rombouts and

Er, ErwWu; IPTG, isopropyl-~-~~iog~actop~~oside; kb, kilobase or 1000 bp; LB, Luria-Bertani (medium); Ptl, pectate lyase(s) (also F’tl phenotype); ptl, gene(s) coding for ptl and regulating its synthesis; Tc, tetracycline; [ 1, designates plasmid- carrier state; ::, novel joint (fusion).

0378-I 119/89/$03.50 0 1989 Eisevier Science Publishers B.V. (Biomedical Division)

212

Pilnik, 1980); ptl genes can be used for studying the mechanisms of protein secretion of Gram-negative bacteria and for construction of secretory vectors.

In the culture medium of Er. ch~sunthem~ ENA49, it was possible to define at least live pecti- nolytic active proteins (40-50 kDa) with different p1 values (Evtushenkov et al., 1986). This suggests that there are several structural pectate lyase genes in this strain. The purpose of the present work was to clone and survey the structural-functional characteristics of the pectate lyase genes. The experiments provide a working example of the use of a new phasmid- based cloning system (Yankovsky et al., 1989).

TABLE I

Bacterial strains and vectors

Strains Genotype or phenotype Reference

Bacteria: Er. chrysanthemi

ENA 49 wild type

E. coli LE392 F-, hsdR514 (r- m+),

supE44, supF58, lacY1, galK2,

galT22, met3 1

TGI [F’ tiaD36, praA +B + , lad, IacZAMlS], Ala&-pro, this, supE,

r-m+

Vectors (pbage aad plasmids): x47.1 imm434cV

IpMYF131 ApR, imm2, cIts857

puc19 ApR

pEMBL8( + ) ApR pPD620 TcR, oriV, otiT

pCM 1 ApR, CmR

pEAN4 ApR @A

pEMBLprl- 1 ApR ptl3

pANC2B ApR ptLA : : cat

pPC623 pPD620 : :@A

~PC624 pPD620::ptlA::cat

~PC625 ~PD620 : :pt&f

Bukanov et al. (1985)

Maniatis et al. (1982)

Yankovsky et al. (1989)

Loenen and Brammar (1980)

Yankovsky et al. (1989)

Vieira and Messing (1982)

Dente et al. (1983) Dobrovoiski et al.

(1985) Close and Rodriguez

(1982) this work, RE-

SULTS, section a this work, RE-

SULTS, section e this work, RESULTS,

section c this work,RESULTS,

section h this work, RE-

SULTS, section h this work,

RESULTS, section h

MATERIALS AND METHODS

(a) Bacterial strains and vectors

The bacterial strains and vectors used are listed in Table I.

(b) Media and growth conditions

Er. chrysanthemi and E. co/i were grown at 30°C and 37”C, respectively. LB-rich medium and M9 minimal medium were prepared according to Miller (1972). Antibiotics were used in selective media at the following concentrations: Ap, 5O,ug/ml, Tc, 10 pg/ml, Cm, 100 fig/ml.

(c) DNA isolation and manipulation

Er. chrysanthemi chromosomal DNA was pre- pared by the method of Matthysse (1983). Plasmid DNA was isolated by the method of Holmes and Quigley (198 1) The conditions for tr~sfo~ation, restriction mapping and subcloning were described by Maniatis et al. (1982).

(d) Genomic library construction

BarnHI and EcoRI fragments of Er. chrysanthemi

DNA (average size, 15 kb) were cloned on JL47.1 phage vector and ilpMYF131 phasmid vector, using methods described in Yankovsky et al. (1989).

(e) DNA hybridization

Gel electrophoresis and DNA transfers were carried out by standard methods (Smith and Sum- mers, 1980). Nick translation of plasmid DNA with [ a-32P]dCTP, hybridization and autoradiography were carried out as described (M~iatis et al., 1982).

(f) Detection of pectinolytic activities

A plate assay was used to screen Ptl’ clones. Phage clones from ENA libraries were plated with polypectate gel as a top layer on agar medium as bottom layer. Ptl+ clones made visible wells in the top layer, Ptl - clones made Rat plaques only. Quan- titative assay of pectinolytic activity was carried out in bacterial cultures at ASg, nm = 0.5 spectrophoto-

213

metrically at 235 nm at 20°C. Reaction mixtures contained 0.1 ml of appropriately diluted enzymes in 2.9 ml reaction buffer (50 mM Tris * HCl pH 8.7/0.1% sodium polygalacturonate). One unit of pectinolytic activity was taken as an increase of absorbance of 1.6A unitslmin at 235 nm in 1 ml.

RESULTS

(a) Cloning of pfl genes in phage vector 1L47.1

BumHI-fragments of Er. chrysanthemi ENA chromosomal DNA were cloned in phage vector K47.1. Among 1100 hybrid phages obtained six Ptl’ clones were identified. The production of Ptl was defined by a zone of polypectate gel degradation surrounding the Ptl + plaques. Restriction analysis of Ptl’ phage DNA showed that all of them had a common 4.0-kb BumHI fragment of Er. chrysun-

themi DNA. For further studies of ptl genes the BamHI frag-

ment (4.0 kb) was re-cloned in both orientations on multi-copy plasmid pUC19. The E. coli cells containing the hybrid plasmids expressed pectinoly- tic activity and were designated pEAN4 and pEAN41. Thus, the expression of ptl gene within an E. coli cell is achieved from its own promoter.

(b) Localization of the ptl,4 gene within the cloned DNA fragment

To identify the minimal DNA fragment coding for pectinolytic activity, deletion versions of plasmid pEAN4 (Fig. 1) were obtained using restriction en- donucleases. E. coli cells containing the recombinant plasmids pEANlO-pEAN16 were checked for pecti- nolytic activity. Pectinolytic activity was found in clones with hybrid plasmids pEAN10, pEANl1 and pEAN12. These plasmids have a common 2.4-kb KpnI-BumHI Er. chrysunthemi DNA fragment. The E. coli clone carrying the pEAN13 plasmid with a 2.9-kb BumHI-SulGI Er. chrysunthemi DNA frag- ment (Fig. 1) did not exhibit pectinolytic activity. This suggests that a single functional ptl gene designated ptlA cloned in the BumHI (4.0-kb) DNA fragment.

Plasmids pEAN4 and pUC19 were used to direct

r .I

Plasmid Size(kb) Restriction map pllenotw?

0 I 2 3 1 (kb)

pANC2B [Cdf$

pEAN4 6.7 EKBE P S,K ‘..Sf, B P

I I I I -- +

pFAN 10 5.9 c ,P+

pEAN 11 5.6 C” ,K+

pEAN12 6.4 FYE B ’ +

pEAN13 5.6 B 52

1 -

pEAN14 4.3 B ---a

S 1 -

pEAN 15 4.0 C ,sI 1 -

PEANX 3.6 C $2 -

Fig. 1. Restriction and genetic maps of ENA DNA fragments in plasmid pEAN4 and its derivatives. B, BamHI; E, EcoRI; K, KpnI; P, &I; S, SalGI. Open arrow designates cat cartridge and its direction of transcription. Parentheses indicate end positions of deletions. The construction of plasmids pEAN4, pEANlO- pEAN16 and pANC2B is described in RESULTS, sections a, b and c, respectively.

protein synthesis in E. coli minicells and in E. coli

cell-free systems. There was only one band (45 kDa) different between pUC 19 and pEANCdirected pro- teins (data not shown). The 45-kDa band is pro- posed to be the product of the Er. chrysunthemiptkd

gene.

(c) Determination of gene ptlA orientation

A cut gene transcriptional probe cassette (Close and Rodriguez, 1982) was used for determining the orientation of gene p&l.

The cassette was inserted into the SulGI-2 site of pEAN4, which appeared to be located in the ptlA

structural gene based on the Ptl’ phenotype con- ferred by pEAN13 and pEAN16 (Fig. 1). Cm resis- tance was conferred only when the cassette was inserted so as to read from left to right on pANC2B plasmid as shown by mapping with EcoRI (Fig. 1).

This allows to presume that the promoter region of ptlA is located on the KpnI-SulGI-2 1.2-kb sub- fragment of pEAN4.

(d) Determination of gene ptlA copy number on the Envinia chrysanthemi ENA chromosome

Southern hybridization using pEAN4 as probe was performed to determine the number of loci

214

containing p&i on the Er. ch~sa~themi chromo- some. Analysis of BumHI and SulGI digests of the chromosome showed that only one pt.iA gene was present.

(e) Cloning of gene ptlB

Multiple forms of pectinolytic proteins of strain ENA suggest that there should be more than one ptl gene in the genome of this strain. For cloning other ptf genes EcoRI DNA fragments of ENA strain were cloned on phage vector RL47.1. Among the 1500 recombinant clones arisen on polypectate indicator gel two clones were defined as pectinolytic- positive (Ptl + ). Restriction analysis showed that the EcoRI fragments inserted are identical. One of the EcoRI fragments of the hybrid phages, when re-clon- ed into plasmid pEMBL8( + ), conferred pectinolytic activity on its EcoRI host. This plasmid, pEMBLpt1, contains an approx. 7-kb DNA insert. The plasmid was not stable. In passages the sizes of wells (lytic zones) made by Ptl+ colonies on polypectate indi- cator gel gradually decreased. One of the second-

passage Ptl + clones, designated E. coli TGl[pEMBLptl-11, was studied in detail.

The 3zP-labelled pEMBLptl-1 DNA was used as a probe in Southern blotting against plasmids pEAN4 and pEAN41 cleaved by BarnHI + SalGI.

TABLE II

The size of Erwinia c~~s~nt~e~i ENA cbromosomdl DNA restriction fragments which hybridize with plasmids pEAN4 and

pEMBLptl-1

[32P]DNA” Size of restriction fragments (kb) b

PSI KpnI EcoRI BamHI SalGI

pEAN4 x2 4.5 7.3 4.0 9.0 2.8 B 1.7 u

1.3 pEMBL@l-I 5,2 a u 5.0 Q

1.6 5.4

a pEAN4 and pEMBLptl-1 genetic and physical structures are shown in Fig. 1; for description of their construction see RE- SULTS, sections a and e, respectivefy. 32P-labelling of the plas- mids was done as described in Maniatis et al. (1982). b The fragments which share homology with both of the probes are underlined.

There was no hyb~disation with the BamHI DNA fragment cloned in plasmid pEAN4. Hence, the ptl gene of plasmid pEMBLptl-1 is not homologous to gene p&4. The second gene was designated ptlB.

The pattern of hybridisation of piasmids pEMBLpt1 and pEAN4 with fragments of ENA chromosomal DNA indicates that genes p&4 and ptlB are located side by side on the chromosome of the strain ENA49. The conclusion is based on the fact that PstI (5.2-kb), KpnI (3.8-kb), EcoRI (7.3-kb) and SalGI (5.2-kb) fragments hybridise with both of the probes, pEMBLpt1 and pEAN4 (Table II).

(f) Cloning of ptZ genes in the phasmid vector ApMYFWl

Detailed analysis of the Er. chrysanthemi chromo- some region carrying the ptlA and ptl3 genes was performed using a library of EcoRI chromosomal DNA fragments of E. chrysanthemi ENA on the phasmid vector pMYF131. Ligated vector and E. ch~sunt~emi DNA was packaged and plated on a bacterial lawn (E. coli LE392) in polypectate gel instead of top agar. Among 11000 plaques screened, 45 were Ptl + . Individu~ plaques were resuspended in LB broth. The suspensions of recombinant phas- mids were used for transduction (at 30’ C) of E. coli TGl cells to Ap resistance. Each transductant proved to be Ptl+ . The phasmid DNA was extracted in plasmid form. A common 7.3-kb E. coli fragment and a common 1.3-kb SulGI fragment were found among twelve ptl phasmids chosen at random. Re- striction maps of phasmids IpMK2, ApMK3, IpMK4, IlpMK8, ApMK9,lpMKll and lpMK12 were made to cover a 15-kb region of the ENA chromosome. IpMK2 and IlpMK4 contain this region of chromosome in opposite orientations (Fig. 2). The phasmids were used for further studies.

We compared restriction and Southern-blot data of p&X- and ptl~-cont~n~g fragments from pEAN4 and pEMBLptl-1 with the restriction maps of the phasmids. The resulting genetic and restriction maps are shown in Fig. 2.

The SulGI DNA fragment carrying a part of the structural ptlA gene was 32P-labelled by nick-trans- lation and hybridized in situ with 45 Ptl’ clones identified in the library. All of the 45 clones hybridized with the probe, i.e., contained the same locus with ptl genes. Probably within the phasmid

E K P E BEP SK S B K PKP E SE B

x~MKS I ---

PL ti ptlA ptlB ptl R

pMKP1

pMKZ-2

pMKz-3

Fig. 2. Restriction and genetic maps of pd gene-containing phasmids and their derivatives. Single line, Er. chrysanthemi DNA, double line - phage DNA, open rectangle is pUC19 plasmid, the triangle (arrowhead) is ZucZ promoter (arrowheads indicate direction of transcription), pk and pL are phage 1 late and early (rightward and leftward) promoters, respectively. For further explanation see Fig. 1. The maps are drawn to scale.

collection there are no other ptl genes capable of polypectate gel degradation areas in passages of the expression in E. coli cells. strain.

We subcloned the EcoRI-SaZGI fragment (7.8 kb) of phasmid IpMK4 (adjacent to the left of ptlA ) to give plasmid pMKC1 (Fig. 2). E. coli clones with pMKC1 do not reveal pectinolytic activity. Hence there is no functional ptl gene(s) in this region.

The ptl gene in pMK2-1 uses its own promoter, since its expression was unaffected by reversing the cloned fragment to give pMK2-2 (Fig. 2). The shortening of plasmids pMK2-1 and pMK2-2 with PstI gave plasmids pMK2-3 and pMK2-4. Only plasmid pMK2-3, but not pMK2-4, provided pec- tinolytic activity to E. coli cells. Plasmid pMK2-3 (g) Subcloning of ptlB gene on the multi-copy plas-

mid pUC19

For subcloning the ptlB gene on plasmid pUC19, hybrid phasmid LpMK2 (Fig. 2) was used.

Subcloning of genes from the hybrid IpMYF131 phasmid is simpler than from hybrid phages because the gene of interest is already placed close to the multicopy plasmid pUC19. All that is needed for subcloning is to cut out pUC19 together with the linked fragment. After cutting the phasmid, ligation and transformation of cells to Ap resistance, all other DNA fragments of the insert and vector would be eliminated.

The phasmid IZpMK2 was treated with BamHI and ligase followed by transformation of E. coli TG 1 cells. E. coli clones containing plasmid pMKZ1 revealed pectinolytic activity. Plasmid pMKZ 1 was stable in E. coli TGl cells, as judged by the size of

TABLE III

IPTG induction of the pectate lyase synthesis by Escherichiu cob

TGl cells containing specified plasmids

Plasmid a Pectate lyase activity in culture media b

Without IPTG With IPTG

pMK2-1 0.55 5.6 pMR2-2 0.081 0.071 pMK2-3 0.066 0.024

a The plasmids’ genetic and physical structures are shown in Fig. 1; for description of their construction see RESULTS, section g. b Bacteria were grown in LB broth, 1 mM IPTG (if added) at 37°C; at A = 0.5 the cells were pelleted by centrifugation. For detection of pectinolytic activity see MATERIALS AND METHODS, section f.

contains a 1.8-kb ENA DNA fragment containing gene ptlB. The data confm that gene ptlB is located to the right of gene ptlA on the Er. chlysanthemi

chromosome. To determine the orientation of ptlB we studied

the influence of inducer IPTG on the production of Ptl by E. co& cells containing plasmids pMK2-1, pMK2-2 and pMK2-3, in which ptlB was placed side by side with the 1acZ promoter (Fig. 2).

The level of Ptl production upon addition of 1 mM of IPTG to the growth medium, increased tenfold in the case of plasmid pMK2-1 and decreased about three times in the case of plasmid pMK2-3 com- prising gene pfZB in the opposite direction (Table III). This led to the conclusion that tr~sc~ption of the ptZB, like that of ptL4, is from left to right (Fig. 2).

(h) Cloning of regulatory gene controlling the ex- pression of gene ptlA

The synthesis of Ptl in ENA cells is induced by Na - polypectate (Prokulevich et al., 1984). This sug- gests that the synthesis of Ptl in cells of strain ENA is controlled by a polypectate-sensing regu- latory system. Since the pd.4 promoter cloned in pANC2B conferred constitutive Cm resistance (result not shown), it seemed probable that ptL-l

regulation involves a negatively acting repressor. There is some genetic evidence for Ptl-negative regu- latory genes in Erwiniu ~Kotouj~sky, 1987). To clone the gene(s) for this repressor we subcloned the ptL4 : : cat fusion region of pANC2B into a low-copy- number plasmid (pPD620) to give pPC624 (Fig. 3) and used this as an indicator system in search of cloned Er. chrysanthemi fragments that repressed Cm resistance when present at high copy number in the same cell.

On the principle that regulatory genes are com- monly linked to the genes that they control, hybrid phasmid IpMK2 containing the ptk4.B region (de- scribed above, section f) was introduced into E. coli[pPC624]. This phasmid inhibited the resis- tance of cells to Cm, i.e., made them Cm-sensitive. The regulatory gene was further localized by analysing the Cm resistance of E. co& TGl [ pPC624] cells transformed by plasmids pMK2-1, pMK2-2, pMK2-3 and pMK2-4 (Fig. 2). Only pMK2-3 failed to cause sensitivity to Cm. Thus, the regulatory gene ptlR controlling gene ptL4 expression is located to the right of gene ptlB as seen on Fig. 2.

SalGI

Fig. 3. Construction of test-plasmid pPC624. Note that not all sites for particular restriction enzymes are shown. OH, origin of replication; on’?‘, origin of mobilisation, of pPD620 (Dobrovolski et al., 1985). Ppd4, promoter of@4 gene; ‘ptL4, ptL4’, promoter- distal and proximal portions ofprl4. Curved lines inside the circle show positions of genes. For further explanation see Fig. 1.

The effects of gene ptlR products on the expres- sion of p&A and ptls genes were studied quantita- tively (Table IV) by determining the production of Ptl in E. coli cells, carrying plasmid pMKZ4 with gene

TABLE IV

Effects of ptlR on the expression of prlA and ptlB genes

E. coli[plasmids] a ptl genes Pectate lyase activityb (units)

In culture media

In cells

TGI[pPC623/pUC19] PrW - 0.011 0.15 TGI[pPC623/pMK2-41 pdA/ptlR 0.001 0.026 TGI[pPC625/pUC19] ptlB/ - 0.018 0.26 TGI[pPC62S~pM~-41 ptlBlptlR 0.020 0.31

a See Table I. b See footnote b to TABLE III.

217

ptlR and hybrid plasmids pPC623 and pPC625 with genes ptL4 and ptlB, respectively (see Figs. 1 and 2, for constructions).

The presence of ptlR led to a six- to tenfold de- crease of ptL4 expression, confirming the role of ptlR in the negative regulation of ptlA. The presence of pMK2-4 did not reduce ptlB expression, indicating that ptlR specifies a negative regulator only for gene ptlA and not for gene ptlB.

(i) Conclusions and discussion

In the present work we describe the cloning of two structural genes ptL4 and ptlB, and a regulatory gene ptlR, which determine the synthesis of Ptl in Er. chry- santhemi ENA49. We have shown that genes ptlA

and ptlB are located side by side on the chromo- some. Other loci with ptl structural genes capable of expression in E. coli were not found in the gene libraries constructed. However, the data obtained are not sufficient to provide evidence for the absence of other ptl genes within the ENA genome. Data from the literature (Van Gijsegem et al., 1985; Reverchon et al., 1985) indicate that the presence of some ptl genes is lethal if they are represented in many copies in E. coli cells. In the present work we used phage vector IL47.1 and phasmid vector ,JpMY F 13 1 for cloning. During lytic development of hybrid phages and phasmids the genes cloned in these vectors are represented in about 100 copies per cell, which could enhance the lethal effect and hinder the cloning of some ENA ptl genes. Taking into account the lethal action of some ptl genes in E. coli cells, it should be helpful to clone these genes on low-copy-number vectors.

A number of investigations deal with cloning of ptl genes of different Erwinia strains (see Kotoujansky, 1987, for references). Comparison of restriction maps of genes ptlA and ptlB of strain ENA cloned by us with those of pectate lyase genes of other strains indicates that the studied Erwinia strains differ in the number of ptl genes and in their organisation. For screening for the regulator gene ptlR we developed a relatively simple plate test that

allows to analyse a large number of clones. In this case the inhibitory action of the regulatory gene is estimated indirectly, by the decrease in the level of resistance of cells to Cm and not by decrease in pectinolytic activity.

The proposed scheme of regulatory ptl gene cloning can also be used for cloning regulatory genes of other enzymes.

Gene ptlR does not seem to influence the expres- sion of ptlB in the plasmids studied. However, it is possible that a regulatory region of gene ptlB recognised by the ptlR gene product was not included in the ptlB-containing fragment tested. The influence of the gene ptlR product on the expression of gene ptlB requires further study.

The advantages of the phasmid vector ApMYF13 1 for constructing libraries are described in Yankovsky et al. (1989). Here we describe its advantages for analysis of gene libraries.

The results of plate assay based on expression of the cloned genes is more distinct in phage plaques where cells are lysed than in bacterial colonies. But it is simpler and quicker to purify recombinant DNA as a multi-copy plasmid from bacterial cells. The phasmids were easily converted after screening from phage to plasmid form just by transduction. Shorten- ing of inserts is simpler with phasmids than with phages: part of the insert is simply cut out from the recombinant molecule along with the plasmid repli- con.

Transduction without helper phage followed by effective lysogenization is a unique feature of phas- mid vectors. This method was used for the screening of the ptlR gene in a ENA gene library. The phasmids are very useful for harvesting the products of the cloned genes. Thermoinduction of E. coli HBlOl containing JpMK2 or 2pMK4 results ac- cordingly in nine and 20 times increase of the Ptl level. IpMK2 and IZpMK4 contain the insert of ENA DNA in opposite orientations (Fig. 2). Hence the method can be used with either orientation of a cloned fragment in the phasmid vector, possibly because the insert can be transcribed both early after induction in one direction from phage pL promoter and late after induction in the opposite direction from phage pk promoter.

218

REFERENCES

Bukanov, N.O., Fonstein, M.Yu., Evtushenkov, A.N., Svarin- skiy, M.A., Strelchenko, P.P., Yankovsky, N.K., Alikhanian, S.I., Fomichev, Yu.K. and Debabov, V.G.: The cloning and study of the gene for pectate lyase of Envinia carotovora.

Genetics (in Russian) 21 (1985) 1601-1607. Close, T.J. and Rodriguez, R.L.: Construction and characteriza-

tion of the chloramphenicol-resistance gene cartridge: a new approach to the transcriptional mapping of extrachromoso- mal elements. Gene 20 (1982) 305-316.

Dente, L., Cesareni, G. and Cortese, R.: pEMBL: a new family of single-stranded plasmids. Nucleic Acids Res. 11 (1983) 1645-1655.

Dobrovolski, P., Ugarov, V.I., Sakanian, V.A. and Alikhanian, S.I.: Constructionofwide host range vectors. Biol. J. Armenia (in Russian) 38 (1985) 471-477.

Evtushenkov, A.N., Shevchyck, V.E., Popova, L.B. and Fomichev, Yu. K.: Purification and properties of two extra- cellular endopectatelyases of Er. chrysanthemi ENA49. Appl. Biochem. Microbial. (in Russian) 2 (1986) 187-192.

Holmes, D.S. and Quigley, M.: A rapid boiling method for the preparation ofbacterial plasmids. Anal. Biochem. 114 (1981) 193-197.

Kotoujansky, A.: Molecular genetics of pathogenesis by soft-rot Erwinias. Annu. Rev. Phytopathol. 25 (1987) 405-430.

Loenen, W.A.M. and Brammar, W.J.: A bacteriophage lambda vector for cloning large DNA fragments made with several restriction enzymes. Gene 10 (1980) 249-259.

Maniatis, T., Fritsch, E.F. and Sambrook, J.: Molecular Cloning.

A Laboratory Manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY, 1982.

Matthysse, A.G.: Role of bacterial cellulose librils in Agrobac-

terium tumefaciens infection. J. Bacterial. 154 (1983) 906-915. Miller, J.H.: Experiments in Molecular Genetics. Cold Spring

Harbor Laboratory, Cold Spring Harbor, NY, 1972.

Prokulevich, V.A., Evtushenkov, A.N. and Fomitchev, Yu.K.: Regulation of extracellular endopectate lyase synthesis in the bacterium Erwinia chrysanthemi ENA49. Mol. Genet. Mi- crobiol. Viral. (in Russian) 16 (1984) 26-30.

Reverchon, S., Hugouvieux-Cotte-Pattat, N. and Robert- Baudouy, J.: Cloning of genes encoding pectolytic enzymes from a genomic library of the phytopatogenic bacterium, Erwinia chrysanthemi. Gene 35 (1985) 121-130.

Rombouts, F.M. and Pilnik, W.: Pectic enzymes. Microb. Enz. Bioconverse, 5 (1980) 227-282.

Smith, G.E. and Summers, M.D.: The bidirectional transfer of DNA to nitrocellulase or diazobenzyloxymethyl paper. Ana- lyt. Biochem. 109 (1980) 123-129.

Van Gijsegem, F., Toussaint, A. and Schoonejans, E.: In vivo cloning of the pectate lyase and cellulase gene of Erwinia

chrysanfhemi. EMBO J. 4 (1985) 784-792. Vieira, J. and Messing, J.: The pUC plasmids, an Ml3mp7-

derived system for insertion mutagenesis and sequencing with universal primers. Gene 19 (1982) 259-268.

Yankovsky, N.K., Fonstein, M.Yu., Lashina, S.Yu., Bukanov, N.O., Yakubovich, N.V., Ermakova, L.M., Rebentish, B.A., Janulaitis, A.A. and Debabov, V.G.: Phasmids as effective and simple tools for construction and analysis of gene libraries. Gene 81 (1989) 203-210.