BACTERIOLOGICAL REVIEWS, Dec. 1970, p. 365-377 Vol. 34, No. 4Copyright © 1970 American Society for Microbiology Printed in U.S.A.

Bacterial GlycolipidsNORMAN SHAW

Microbiological Chemistry Research Laboratory, Department of Organic Chemistry,University of Newcastle upon Tyne, England

INTRODUCTION............................................................. 365ISOLATION AND PURIFICATION................................... 365CHEMISTRY OF GLYCOLIPIDS................................... 366

Glycosyl Diglycerides................................... 366Acylated Sugar Derivatives................................................... 369

BIOSYNTHESIS................................... 370DISTRIBUTION AND TAXONOMY................................... 371CELLULAR LOCATION AND FUNCTION................................... 372LITERATURE CITED................................... 375

INTRODUCTIONIn a comprehensive review on bacterial lipids

published in 1964, Kates concluded that "the ac-cumulation of knowledge concerning fatty acidcomposition, phosphatide composition and cellu-lar distribution of bacterial lipids is still proceed-ing at an exponential rate" (37). The notableabsence of the phrase glycolipid composition isindicative of the scant references to bacterial gly-colipids in the literature at that time. Althoughthe complex glycolipids of mycobacteria hadreceived considerable attention (44), there werefew reports on the occurrence of glycolipids inother bacteria. The ensuing years have produceda dramatic change in our knowledge of bacterialglycolipids, and their widespread distribution isnow firmly established. The purpose of this reviewis to discuss this development. The term glyco-lipids is interpreted as those lipids which are com-posed of carbohydrates in combination with long-chain aliphatic acids or alcohols and which arereadily extracted from bacteria into organic sol-vents without the prior use of hydrolytic proce-dures. It does not include those phospholipidscontaining carbohydrate residues for which theterm phosphoglycolipid would seem more appro-priate. The complex glycolipids of mycobacteriaand related organisms will not be discussed asthey have been previously extensively reviewed(44). The bacterial glycolipids so far describedmay conveniently be divided into two categories:(i) glycosyl diglycerides and (ii) acylated sugar de-rivatives. The glycosyl diglycerides are structurallyanalogous to the phosphoglycerides as they arecomposed of carbohydrate residues glycosidicallybound to the 3-position of a sn-1,2-diglyceride.The term acylated sugar is used to describe thoseglycolipids which do not contain glycerol butwhich have fatty acid residues bound directlyto the carbohydrate residue.

ISOLATION AND PURIFICATION

Glycolipids are readily extracted from bacterialcells, together with the phospholipids, by stirringwith chloroform-methanol mixtures. They may bepreferentially removed to some extent by extrac-tion with acetone (59), but the rate of extraction iscomparatively slow and since subsequent frac-tionation is usually required to remove traces ofphospholipids, no saving in time over the moreconventional methods is usually possible. More-over, the more hydrophilic glycolipids (e.g., tri-and tetraglycosyl diglycerides) are the least solu-ble in acetone and are not efficiently extracted withthis solvent. A cautionary note should also bemade concerning the use of the Folch procedurefor the purification of crude lipid extracts (28).We have found that considerable quantities of themore hydrophilic glycolipids are found in theaqueous extract and may be inadvertently dis-carded. The procedure using Sephadex G-25 forthe removal of nonlipid contaminants (81) is verysatisfactory and does not result in any significantlosses. Several reagents are available for thespecific detection of glycolipids on thin-layerchromatograms (78), but the periodate-Schiffreagent, which was first applied to the detection ofcarbohydrates and other a-glycols on paperchromatograms, can also be used on thin-layerchromatograms (64), and this has proved to be avery sensitive method for the detection of thesmall amounts of glycolipids usually present inlipid extracts. However, this method is not suit-able for detecting completely acylated sugar deriv-atives as these lipids do not possess an a-glycolunit susceptible to periodate oxidation. Vorbeckand Marinetti have reported a very efficientmethod for the purification of glycolipids fromtotal lipid extracts by silicic acid chromatography(76). After elution of the neutral lipids withchloroform, the glycolipids may be eluted from

365

on May 19, 2020 by guest

http://mm

br.asm.org/

Dow

nloaded from

BACTERIOL. REV.

the column with solvent mixtures containing in-creasing concentrations of acetone in chloroform,culminating with pure acetone. The phospholipidsmay then be eluted with chloroform-methanolmixtures as usual. Small amounts of phospho-lipids are usually present in the glycolipid frac-tions and may be removed by repeating the frac-tionation procedure. In my experience, the morehydrophilic glycolipids are only eluted from silicicacid columns very slowly with acetone, and largevolumes of the eluant must be passed through thecolumn to ensure the maximal recovery. Even so,small amounts may be subsequently eluted in thephospholipid fraction and could lead to theerroneous identification of phosphoglycolipids inthe extract. The fractionation procedure utilizingdiethylaminoethyl cellulose, which has previouslybeen used for the separation of mammalian glyco-lipids and phospholipids (62), can also be appliedto bacterial lipids (13).

CHEMISTRY OF GLYCOLIPIDSGlycosyl Diglycerides

The first isolations of glycosyl diglycerides frombacterial lipids were reported by Macfarlane whofound a mannosyl diglyceride in Micrococcuslysodeikticus (47) and a glucosyl diglyceride inStaphylococcus aureus (48). Subsequent investiga-tions by Polonoviski, Wald, and Paysant-Diament(58) showed that the glycolipid from S. aureuswas a diglucosyl diglyceride. The first completestructure was determined by Brundish, Shaw,and Baddiley (13) for the glycolipid from a Pneu-mococcus type I, and independently Kaufmanet al. (40) proposed an identical structure, 3-[O - a - D - galactopyranosyl - (1 - 2) - 0 - a - D -glucopyranosyl]-sn-1 ,2-diglyceride, for the glyco-lipid from a Pnewnococcus type XIV. The pres-ence of carbohydrate in the lipids of several lacticacid bacteria (33) suggested the presence of glyco-lipids, and this was confirmed by Brundish, Shaw,and Baddiley (14) who found diglycosyl diglyc-erides in these and other gram-positive bacteria.The widespread distribution of this type of glyco-lipid has now been firmly established (Table 1).The principle glycolipid in nearly all of the organ-isms examined is a diglycosyl diglyceride, andfive major structural types have been charac-terized, depending upon the nature of the disac-charide glycosidically bound to the 3-position of asn-1,2-diglyceride (Fig. 1). In addition to thediglucosyl and galactosylglucosyl diglyceridesalready mentioned, the glycolipid from M. lyso-deikticus has been shown to be a dimannosyldiglyceride (46), and a digalactosyl diglyceridehas been isolated from Arthrobacter globiformis(79). The glycolipid of Streptococcus faecalis is adifferent diglucosyl diglyceride from that in

Staphylococcus aureus (14). The complete struc-ture determination of this type of lipid requiresinformation on (i) the nature of constituent sugars,(ii) linkage between sugar residues, (iii) linkagebetween disaccharide and glycerol, (iv) identifica-tion and location of fatty acid residues, and (v)stereochemistry of sugars and glycerol. Themethods used for complete structure determina-tion are well documented, and the reader is re-ferred to the original publications for a full dis-cussion. The identification of very small amountsof the glycosides produced by deacylation of thediglycosyl diglycerides is possible by gas-liquidchromatography (12), and the structures of two ofthe glycosides have been confirmed by chemicalsynthesis (12, 15). The five major types of diglyco-syl diglyceride(a-galactosylglucosyl, a-diglucosyl-,(3-diglucosyl-, a-dimannosyl-, and (3-digalactosyldiglycerides) have now been isolated from manydifferent bacteria (Table 1), but where completestructures have been elucidated the constituentglycosides within each class (e.g., galactosylgluco-sylglycerols) all have identical structures. An in-teresting recent development is the isolation ofglycolipids containing uronic acids. Pseudomonasdminuta contains (86) a glucosylglucuronosyldiglyceride (Fig. 2a), and an acyl-glucosyl-galac-turonosyl diglyceride (Fig. 2b) has been isolated(5) from a Streptomyces. It is possible that theadditional acyl residue may have been introducedduring isolation. The 6-0-acyl galactosyl diglyc-eride isolated from spinach leaves is probablyformed by acyl transfer to a galactosyl diglycerideduring cell disruption (31). The modification oflipids during extraction and subsequent work-upis a well-documented hazard (50). In view of theincreasing number of phosphoglycerides nowbeing isolated containing three or four fatty acidresidues instead of the expected two (8, 52), it isdesirable wherever possible that the number andlocation of the fatty acid residues in glycolipids beconclusively established.Although monoglycosyl diglycerides are known

to be the biosynthetic precursors of the di-glycosyl diglycerides (see below), they do notusually accumulate in significant amounts. A feworganisms, however, do contain monoglycosyldiglycerides as major components (Table 1). Withthe exception of monomannosyl diglyceride andgalacturonosyl diglyceride, the monoglycosyldiglycerides corresponding to the other five digly-cosyl diglycerides have been isolated in sufficientamounts for characterization. The isolation ofO-(3-D-galactofuranosyl diglyceride from Myco-plasma mycoides represents the second report ofthe occurrence of galactofuranose in bacteriallipids. Reeves, Latour, and Lousteau (61) isolated0-fl-D-galactofuranosylglycerol from alkaline

366 SHAW

on May 19, 2020 by guest

http://mm

br.asm.org/

Dow

nloaded from

TABLE 1. Distribution of glycosyl diglycerides in bacteria

Organism Constituent sugars Structure!% References

Chromatium strain DPseudomonas diminutaMicrococcus lysodeikticusStaphylococcus lactis 13S. lactis NCTC 9744S. saprophyticus I2S. aureusPneumococcus type 1P. type XIVStreptococcus faecalis NCIB

8191S. faecalis ATCC 9790S. pyogenesS. lactisS. hemolyticus D58S. MGLactobacillus caseiL. buchneriL. plantarumL. helveticusL. acidophilusL. fermentiListeria monocytogenesBifidobacterium bifidumMicrobacterium lacticumM. thermosphactumArthrobacter globiformisA. pascensA. crystallopoietesBacillus subtilisB. cereusStreptomyces LA 7017Mycoplasma laidlawiiM. pneumoniae

Pseudomonas diminutaP. rubescensA. globiformisA. pascensA. crystallopoietesM. laidlawiiM. mycoides

Chloropseudomonas ethylicum

Chromatium strain DStreptococcus hemolyticus D58L. caseiL. plantarumL. helveticusL. acidophilusBifidobacterium bifidumHalobacterium cutirubrum

L. helveticusL. acidophilus

Diglycosyl diglyceridesMannose, glucoseGlucose, glucuronic acidMannoseGlucoseGlucoseGlucoseGlucoseGalactose, glucoseGalactose, glucoseGlucose

GlucoseGlucoseGlucoseGlucoseGlucoseGalactose, glucoseGalactose, glucoseGalactose, glucoseGalactose, glucoseGalactose, glucoseGalactose, glucoseGalactose, glucoseGalactoseMannoseMannoseMannose, galactoseMannose, galactoseMannose, galactoseGlucoseGlucosebGlucose, glucuronic acidGlucoseGalactose, glucose

Monoglycosyl diglyceridescGlucuronic acidGlucuronic acidGalactoseGalactoseGalactoseGlucoseGalactose

Triglycosyl diglyceridesdGalactose, rhamnose, unidenti-

fied sugarMannose, glucose (2:1)GlucoseGlucose, galactose (2:1)Glucose, galactose (2:1)Glucose, galactose (2:1)Glucose, galactose (2:1)GalactoseGalactose sulphate, glucose,mannose

Tetraglycosyl diglyceridesdGlucose, galactose (3:1)Glucose, galactose (3:1)

2albIdIdididlalaIC

lc

lcIc

lalalalala

lb

lb, lelb, lelb, leid

2bIc

7586461612, 1412, 1412, 14, 58, 80134014

542027355767

Hunter, Le Roux,Minnikin, and Shaw,unpublished data

841926637079

}Shaw and Stead,)unpublished data

6, 12, 1443S

6856

868579

Shaw and Stead,Junpublished data

6855

3c3a

3a3a

3b

21

753567

Hunter, Le-Roux,Minnikin, and Shaw,unpublished data

2638

Hunter, Le Roux,'Minnikin and Shaw,Junpublished data

a Entries in this column refer to the figures. Absence of an entry indicates complete structure is un-known except for the monoglycosyl diglycerides whose structures are not illustrated.

' Unidentified sugar also present in smaller amounts.c These organisms contain monoglycosyl diglycerides as principal components.d Numbers in parentheses are the ratios of constituent sugars.

367

on May 19, 2020 by guest

http://mm

br.asm.org/

Dow

nloaded from

BACrERIOL. REV.

CH2.OH

0CHOH OH

HO 0 H -C H2OH C0H.O.CO.R

umCH OCOROH (a)

CHOH0

CH2.OH H

0 H10 -CH0 HO CHO2OH HO CH.C

HO0

(b)(-N

C02H

0

CHOOHOH7 0 O~~r-CH20\H)|OH

CH.O.CO.R

HO CH0QCQR

OH

(a)

)CO.R

DO.CQR

CH OH

0CH70H OH

0 HO -CH2

HO 0 CH.O.COQRCH IO.CQR

OHd 2

HO

R.CO, H

O-CH2O H2

CH.O.CO.R

CH2.O.CO.R

(c)CH7OH

O O-~CH2

HOH / / o0 O-CH2

HO OH CH.O.CO.R

OH

(d)CH OH

HO 0 O CH2HO HO 0 O-CH2

OH C H.O.CO.ROH

| CHC.0HCO.ROH 2

(0)

FIG. 1. Structures of the five major types ofdiglycosyl diglyceride isolated from gram-positivebacteria. (a) 3-[0-a-D-galactopyranosyl-(1 -- 2)-0-a-D-glucopyranosyl]-sn-1,2-diglyceride; (b) 3-[O-a-

(b)FIG. 2. Structures of (a) the glucuronic acid-con-

taining glycolipid from Pseudomonas diminuta and(b) the galacturonic acid-containing glycolipid fromStreptomyces LA 7017.

hydrolysates of the total lipids from Bacteriodessymbiosus, but in this instance it was not possibleto determine whether the glycoside was a degrada-tion product of a glycolipid or a more complexphospholipid as no attempts were made to frac-tionate the lipids.

In their original survey of gram-positive bac-teria, Brundish, Shaw, and Baddiley found adiglycosyl diglyceride in every organism theyexamined except Lactobacillus plantarum 17-5which contained a glycolipid with the propertiesof a glucosylgalactosylglucosyl diglyceride (14).

D-mannopyranosyl-(i -* 3)-O-o-D-mannopyranosyl]-sn-i ,2-diglyceride; (c) 3-[O-a-D-glucopyranosyl-(] --

2)-O-a-D-glucopyranosyl]-sn-1,2-diglyceride; (d) 3-[O-p-D-glucopyranosyl-(] -- 6)-C-fl-D-glucopyranosylJ-sn-i ,2-diglyceride; (e) 3-[0-jp-D-galactopyranosyl-(l -*6)-0-jS-D-galactopyranosyl]-sn-1, 2-diglyceride.

368 SHAW

on May 19, 2020 by guest

http://mm

br.asm.org/

Dow

nloaded from

BACTERIAL GLYCOLIPIDS

CH2.OH

OO\ CH OH

~~~~~~~0H O -CH2

OH OH HO 0-CH2

CH.O.COR

I CCH QCOROH

(a)

(b)

CH.O.COR

CH2CQR

(c)

FIG. 3. Structures of (a) the glucosylgalactosyl-glucosyl diglyceride from Lactobacillus casei, (b) thediglucosvlgalactosylglucosyl diglyceride from L. aci-dophilus, and (c) the triglucosyl diglyceride from Strep-tococcus hemolyticus.

Tri- and tetraglycosyl diglycerides have since beenisolated from several bacteria (Table 1). Struc-tures have been determined for the glucosylgalac-tosylglucosyl diglyceride (Fig. 3a) from L. casei(67), the diglucosylgalactosylglucosyl diglyceride(Fig. 3b) from L. acidophilus (Shaw and Hunter,unpublished data), and the triglucosyl diglyceride(Fig. 3c) from Streptococcus hemolyticus (35).An unusual glycolipid has been isolated (38) fromHalobacterium cutirubrun, and in common withthe phospholipids of this organisms (39) it is aphytanyl diether glyceride. The sugar constituentsare glucose, mannose, and galactose sulphate. Noglycosyl diglycerides with more than four sugarresidues have yet been detected. We have triedvarying the extraction conditions by using more

hydrophilic solvents, as it was anticipated thatthese larger glycosyl diglycerides might be in-creasingly less soluble in lipid solvents, but no newlarger glycolipids could be detected.Some photosynthetic bacteria contain the same

digalactosyl diglyceride found as an essential partof the chloroplast in the plant cell (4) but it has a

different structure from the digalactosyl diglyc-eride from Arthrobacter species. However, twophotosynthetic bacteria, Chloropseudomonas ethy-licum and a Chromatium strain D, have both beenshown to contain unusual and as yet poorly char-acterized glycosyl diglycerides. The glycolipidfrom C. ethylicum may be a triglycosyl diglyceridecomposed of galactose, rhamnose, and an uniden-tified sugar (21). Chromatium strain D contains atriglycosyl diglyceride composed of two mannoseresidues and glucose, together with a (mannosyl-glucosyl) diglyceride and a glucosyl diglyceride(75).Analysis of the fatty acid components of gly-

cosyl diglycerides has usually shown a composi-tion similar to that found in the phospholipids ofthe same organism. However, a preferential con-centration of a particular fatty acid has been ob-served (16) in the diglucosyl diglyceride ofStaphylococcus lactis.

Acylated Sugar DerivativesThe first glycolipid of this type to be charac-

terized was the rhamnolipid isolated from theculture filtrates of P. aeruginosa. The major struc-tural features were outlined by Jarvis and Johnson(36), and the complete structure (Fig. 4) wasdescribed by Edwards and Hayashi (24). Manyyears elapsed before other examples of acylatedsugars were discovered. During an investigationinto the diglucosyl diglyceride of Streptococcusfaecalis, another glycolipid was observed whichdid not contain glycerol and which gave glucoseboth on acid hydrolysis and deacylation. Mainlyon the basis of mass spectrometric studies, a tetra-acylglucose structure (Fig. 5a) was proposed (82).Similar glycolipids were found in Aerobacter aero-genes, P. fluorescens, and Escherichia coli, andtraces of acylated di- and trisaccharides weredetected but the small quantities present pre-

4-0 O-CH.CH .C.OCH.CH C02H

(CH2)6 (OH2)61 C

H3 Iu3

H(

FIG. 4. Structure of the rhamnolipid from Pseu-domonas aeruginosa.

VOL. 34, 1970 369

on May 19, 2020 by guest

http://mm

br.asm.org/

Dow

nloaded from

BAVTERIOL. REV.

CH .0.CO.CH30

OCO.CHHOH

CH 00. I

QCO(CH2)1CH3

(a)

CH .0.CGR2

OH

O.CO.RR.CO.O

OH

(b)FIG. 5. Structures of the acylated glucoses from

(a) Streptococcus faecalis and (b) Mycoplasma strainJ.

cluded detailed chemical analysis (82). A tri-acylglucose constitutes the major glycolipid ofMycoplasma strain J, and the full structure (Fig.5b) has been elucidated (74). Brennan, Flynn, andGriffin (10) confirmed the presence of acylglu-coses in E. coli and separated them into two com-

ponents, a tetraacylglucose and another acylatedglucose probably having fewer acyl residues. Al-though a discussion of the complex glycolipids ofmycobacteria and related organisms has been ex-

cluded from this review, it is relevant to record theisolation of acylated glucoses from these orga-

nisms. Corynebacterium diphtheriae, Mycobac-terium smegmatis, M. tuberculosis BCG, andBrevibacterium thiogenitalis all contain a 6-0-mycolylglucose (11, 49). Acylglucoses have alsobeen isolated from Saccharomyces cerevisiae andthe fungus Agaricus bisporus, thereby illustratingtheir ubiquitous occurrence in microorganisms(10).

A different type of acylated sugar has been iso-lated from various propionic acid bacteria andshown to be a diacyl myoinositol mannoside (59,66). Mass spectrometric studies have enabled thedistribution of the fatty acids within the moleculeto be clearly established, and the major compo-nent is l-O-pentadecanoyl-2-0-(6-0-heptadeca-noyl-a-D-mannopyranosyl)-myoinositol (Fig. 6).This structure closely resembles that of 6,6'-diacyl-a, a' -trehalose, cord factor, isolated fromsome mycobacteria and Corynebacteriwn diph-theriae (44).

BIOSYNTHESISThe discovery during studies on the biosynthe-

sis of a Pneumococcus type XIV polysaccharidethat labeled nucleotide sugar precursors werebeing incorporated into lipid fractions led to theisolation of a galactosylglucosyl diglyceride andalso suggested its mode of biosynthesis (22).Further studies (40) established the overall path-way

Uridinediphosphate-

glucoseDiglyceride

g

Glucosyl diglyceride

Urdine diphosphate-galactose

Galactosylglucosyl diglyceride

The rate of synthesis suggested two distinct trans-ferases, and both particulate and soluble enzymepreparations showed activity. This general routehas since been confirmed for three other diglycosyldiglycerides, the dimannosyl diglyceride of Micro-coccus lysodeikticus (46), the diglucosyl diglyc-

OH

OHCH2OR HI 0 R;~.O

OH HO OHHO 0

R = 017

R' C15

FIG. 6. Structure of the diacylfrom some propionic acid bacteria.

inositol mannoside

370 SHAW

on May 19, 2020 by guest

http://mm

br.asm.org/

Dow

nloaded from

BACTERIAL GLYCOLIPIDS

eride of Streptococcus faecalis (54), and the diglu-cosyl diglyceride of Mycoplasma laidlawii (72).A particulate enzyme preparation of Micrococcuslysodeikticus catalyzes the transfer of mannosefrom guanosine diphosphate-mannose to diglyc-eride to form mannosyl diglyceride. The enzymeshows a high specificity for a diglyceride contain-ing branched-chain fatty acids similar to thosefound in vivo and also requires Mg2+ ions and ananionic surface-active agent. The enzyme whichcatalyzes the transfer of the second mannose unitto form dimannosyl diglyceride is present in crudeextracts in soluble form and also requires Mg'+but no surface-active agent (46). The stereo-chemistry of the diglyceride substrate has beeninvestigated in the synthesis of glucosyl diglyc-erides in S. faecalis (54). The synthesis of glu-cosyl diglycerides was significantly stimulated bythe addition of 1 ,2-diacyl-sn-glycerol but not by2, 3-diacyl-sn-glycerol. This observation is consist-ent with the known configuration of the glycerolmoiety in these lipids, a feature shared with mostnaturally occurring phospholipids.The biosynthesis of tri- and tetraglycosyl diglyc-

erides has not been established, but their struc-tural features suggest that they are formed by suc-cessive transfers to hexose from the appropriatenucleotide precursor. During studies on the bio-synthesis of glucosyl diglycerides in S. faecalis,Pieringer observed a third lipid component con-taining labeled glucose, and a time-curve experi-ment indicated that it was probably formeddirectly from diglucosyl diglyceride (54). How-ever, the anionic nature of its deacylation producteliminated a triglucosyl diglyceride structure forthe lipid. A structure for this product has recentlybeen suggested (69) and will be discussed later.Lennarz and Talamo (46) also observed the syn-thesis of a third mannolipid by enzyme prepara-tions from M. lysodeikticus which they thoughtmight be a trimannosyl diglyceride, but it has nowbeen identified as mannosyl-l-phosphoryl-undec-aprenol (45).

There is no evidence available on the synthesisof acylated hexoses, although the most reasonableroute would seem to be direct acylation of theappropriate hexose or a phosphorylated deriva-tive. However, with those glycolipids containingtwo or more sugar residues, the problem is morecomplex as acylation could take place either be-fore or after formation of the glycosidic linkage.The biosynthesis of the rhamnolipid of Pseudo-monas aeruginosa occurs by a route analogousto that of the diglycosyl diglycerides, namely thesequential transfer of two L-rhamnose units fromthymidine diphosphate-rhamnose to f3-hydroxy-decanoyl-,B-hydroxydecanoate (17).

DISTRIBUTION AND TAXONOMY

The increasing amount of information availableon the phosphatide and fatty acid composition ofbacteria has enabled certain taxonomic relation-ships to be established, and this field has been ex-tensively reviewed (34, 37). Although informationon glycolipids is still much less complete, signifi-cant relationships are already emerging and theresults obtained suggest that this approach will beof value (65).An examination of the distribution of glycosyl

diglycerides (Table 1) shows that they are mostwidespread in gram-positive bacteria. With theexception of the photosynthetic bacteria and somePseudomonas species, glycosyl diglycerides havenot been found in gram-negative bacteria. Al-though the halophilic bacteria and mycoplasma,both of which possess glycosyl diglycerides, areformally gram-negative, their lack of a normal cellwall places these organisms in a special category,and the presence of glycosyl diglycerides may beof particular significance (see below). The majorglycosyl diglyceride found in most bacteria is adiglycosyl diglyceride, and mono- and trigly-cosyl diglycerides, if present at all, are usually onlyminor components.The most significant feature of their distribution

is the occurrence in members of the same genus ofidentical glycolipids. All of the lactobacilli so farexamined contain a galactosylglucosyl diglyceride.The a-diglucosyl diglyceride has been found inmany streptococci and the ,B-diglucosyl diglyc-eride in staphylococci. These and other examplesare shown in Table 1. Thus, whereas the sameglycolipid can be found in organisms belonging todifferent families, members of the same genus orclosely related species contain identical glyco-lipids. Arthrobacter globiformis contains both adimannosyl and a digalactosyl diglyceride (79), aphenomenon also found in A. pascens and A.crystallopoietes (Shaw and Stead, unpublisheddata). The reason for the occurrence of two un-related diglycosyl diglycerides is unknown and anexamination of the lipids from the two morpho-logical forms, rods and spheres, in which theseorganisms may exist has not revealed any quan-titative or qualitative differences (Shaw and Stead,unpublished data).The isolation of the acylated sugar derivatives is

a comparatively recent development and there isinsufficient information available to enable signifi-cant conclusions to be reached. Their isolationfrom many different organisms suggests they aremore widely distributed than glycosyl diglyceridesalthough the amounts present are usually verysmall. It is probable that the presence of this typeof glycolipid is more dependent upon the age and

371VOL. 34, 1970

on May 19, 2020 by guest

http://mm

br.asm.org/

Dow

nloaded from

BACTERIOL. REV.

composition of the culture than has been observedfor glycosyl diglycerides.

CELLULAR LOCATION AND FUNCTUIONA prerequisite to a discussion of functions for

glycolipids is the establishment of cellular loca-tion. With the exception of the rhamnolipid fromP. aeruginosa, all of the other glycolipids describedin this review are intracellular components. Thecomparative ease with which bacterial lipids maybe specifically extracted from whole cells hasmeant that few studies have been made on cellulardistribution. An examination of cellular prepara-tions from S. faecalis (77), Bacillus subtilis (6), S.pyogenes (20), and Staphylococcus aureus (80)established that the glycolipids are located, to-gether with the phospholipids, in the cytoplasmicmembrane, and it is reasonable to suppose thatthis conclusion is applicable to most gram-positivebacteria whose walls are devoid of lipids. It hasnot been established whether they are locatedspecifically in one area of the membrane or if theyare components of a regular subunit. A lipidanalysis on mesosome preparations would be ofgreat interest. Some preliminary evidence for thespecific location or at least an enhancement inlipid content of newly synthesized membrane hasbeen obtained by Frerman and White (29), whohave measured the changes in lipid compositionof S. aureus membranes upon conversion fromanaerobic to aerobic growth by the addition ofoxygen. The development of the membrane-asso-ciated electron-transport system is accompaniedby a 1.3-fold increase in glycolipid content andalso by a comparable increase in phospholipidcontent. Thus, the electron-transport system couldbe formed either by the addition of subunits ofdiffering composition or the extensive modifica-tion of the basic membrane.The problem of cellular location is more com-

plex in mycobacteria and related organisms and ingram-negative bacteria, in which the differentia-tion between multilayered cell wall and cyto-plasmic membrane is not so well established.Since the acylated sugar derivatives described byWelsh, Shaw, and Baddiley (82) were mostly iso-lated as contaminants in lipopolysaccharide prep-arations, it seems probable that they are locatedin the lipoprotein layer of the cell wall. A similarlocation is also likely for the uronic acid glyco-lipids isolated from Pseudmonas species (85).The ability to expound numerous hypotheses

from the minimum of experimentation is nowhereshown to greater effect than by lipidologists dis-cussing the physiological function of lipids. Tothis the function of bacterial glycolipids is no ex-ception. The discovery of glycosyl diglycerides

during investigations on pneumococcal polysac-charide biosynthesis led to suggestions that theymight be involved in transfer of sugar residues topolysaccharide chains (22). In many instances, acomparison of the sugar components of variousbacterial polymers with those of the respectiveglycolipid is particularly striking. The type XIVpneumococcal polysaccharide (3), the membrane-associated polysaccharides of M. lysodeikticus(30) and S. lactis (2), and the galactan of Myco-plasma mycoides (18) all contain similar sugarresidues to those found in the respective glycosyldiglycerides. The intracellular teichoic acids ofStreptococcus faecalis (83) and Staphylococcusaureus (60) both contain disaccharide residues,kojibiose and gentiobiose respectively, which arealso present in the respective glycosyl diglycerides.Thus, in these two examples transfer of the com-plete disaccharide residue could have taken place,and in those polysaccharides containing sugarlinkages unlike those present in the glycosyldiglycerides, only the terminal sugar residue mayhave been transferred. The utilization of glycosyldiglycerides as polysaccharide intermediateswould mean continual turnover of these glyco-lipids within the cell. Evidence to support such aturnover of these components has not been re-ported. The metabolism of the diglucosyl diglyc-eride in M. laidlawfi has been studied but noturnover of glucose residues could be detected(71). Unfortunately, in this instance, as the organ-ism does not possess any suitable polysaccharide(68), the result is not pertinent to the problem.The results of experiments specifically designed todemonstrate the incorporation of sugar residuesfrom labeled glycolipids into polysaccharides havenot so far been reported. Recent developments,however, suggest that this hypothesis should bediscarded. Lipid intermediates have now been iso-lated in which the sugar residues are boundthrough a phosphodiester or pyrophosphate link-age to a C55 isoprenoid alcohol. These lipid inter-mediates have been demonstrated in the biosyn-thesis of wall polysaccharides (87), peptidoglycan(32), teichoic acids (23), and intracellular polysac-charides (45). There is still one instance in which aglycolipid may be involved in polysaccharide syn-thesis. The production of cellulose by variousAcetobacter species, involves a nonphosphate,glucose-containing lipid whose chromatographicproperties and lability to alkali are similar to thoseof glycosyl diglycerides and in direct contrast tothose of the isoprenoid lipid intermediates (41).The production of acylated glucoses by coryne-

bacteria is dependent upon the presence of glucosein the culture medium (11). These glycolipids dis-appear when the glucose is replaced by glycerol.

372 SHAW

on May 19, 2020 by guest

http://mm

br.asm.org/

Dow

nloaded from

BACTERIAL GLYCOLIPIDS

These comparatively simple glycolipids may there-fore be either carbohydrate reservoirs or amedium for transport of glucose across the mem-brane. The effect of culture conditions upon bac-terial composition is well established, and the useof synchronous cultures under various conditionsof substrate limitation has led to dramatic changesin cell wall composition (25). The results of similarstudies on lipid composition may, well produceequally dramatic results. Indeed, the productionof the uronic acid glycolipid by P. rubescens andP. diminuta may well be a result of substratelimitation (85, 86). These unusual glycolipids areonly produced when the organisms are grown onagar slopes; they are completely absent from thelipids of organisms grown in liquid culture. More-over, the organisms grown on solid media are verylow in phospholipid content. Growth in solidmedia is probably an effective way of simulatingphosphate-limiting growth conditions. After thefirst few cell divisions, the supply of phosphate forphospholipid synthesis may well be exhausted andthe organism replaces the essential anionic phos-pholipids by similarly charged glycolipids. Thisprocess is directly analogous to the replacementof teichoic acids in the wall of B. subtilis byteichuronic acids when grown under phosphate-limiting conditions (25).The importance of phospholipids in maintain-

ing the structural integrity of the membrane haslong been recognized, and a structural functionhas been proposed for glycosyl diglycerides. Froman examination of the molecular shape of digly-cosyl diglycerides, Brundish, Shaw, and Baddileysuggested that these glycolipids, irrespective of thenature of the disaccharide, can adopt a conforma-tion (Fig. 7) in which all of the hydroxyl groupslie on one side of the molecule and the lipophiliccomponents (i.e., fatty acids, ring oxygen ofsugars, and glycosidic oxygens) lie on the otherside (16). The hydrophilic regions of severalmolecules could come together to form pores inthe membrane through which small moleculesmay pass. The presence in some organisms oflarge tri- and tetraglycosyl diglycerides may rep-resent an attempt to regulate the size of thesepores. The location of these pores may be withinthe membrane or even on the surface where someinvolvement in binding or anchoring intracellularcomponents might be possible. Intracellular ormembrane teichoic acids are common constituentsof gram-positive bacteria and are probably lo-cated in or on the outer surface of the protoplastmembrane (1). The chemical nature of this asso-ciation between teichoic acid and membrane hasnot been clearly defined, but recent studies byWicken and Knox (84) suggest the involvement of

FIG. 7. Model ofdiglycosyl diglyceride. For clarity,hydrogen atoms directly attached to carbon have beenomitted. The hydroxyl groups are seen along the right-hand edge of the molecule. (Reproducedfrom Biochem.J. 105:888.)

glycolipids. The intracellular teichoic acid of L.fermenti has been isolated by a phenol extractionprocedure as a lipoteichoic acid complex. Thelipid component is not removed by extractionwith lipid solvents, and on alkaline hydrolysis agalactosylglucosyl glycerol was isolated, similarin structure to the glycoside obtained by deacyla-tion of the diglycosyl diglyceride present in totallipid extracts of this organism. Such a glycosidecould not have been produced from the teichoicacid, and therefore, by implication was probablyderived from the bound lipid component. Furtherwork will be necessary to establish conclusivelythe presence of bound diglycosyl diglyceride inthis lipoteichoic acid, but in any event this glyco-lipid only represents a small fraction of the totalglycolipid present in the membrane, the majorportion being unassociated and freely extractable.The amounts of glycolipid present in bacterial

lipids are usually quite small although some orga-nisms e.g. Microbacterium lacticum (63), Myco-plasma laidlawii (68), and Pneumococcus type I(13) contain appreciable quantities. Until moreinformation is available on the rate of accumula-tion, turnover, and effects of culture medium, it isdifficult to draw useful conclusions from thesefigures. In M. laidlawii, although the total glyco-lipid concentration remains approximately con-stant throughout the culture period, the ratio ofmonoglucosyl diglyceride to diglucosyl diglyc-eride steadily increases to a maximum of 2.6:1(68). This unexpected result cannot be explainedby the rapid metabolism of the terminal glucose in

373VOL. 34X 1970

on May 19, 2020 by guest

http://mm

br.asm.org/

Dow

nloaded from

BACTERIOL. REV-

TABLE 2. Lipid composition of membranes fromStaphylococcus aureus, Streptococcus pyogenes,

and their derived L-forms compared withthat of Mycoplasma laidlawii

Peet Per centOrganism lipid in glycolipid Refer-

membrane in total encelipid

S. aureus 23 10 80L-form 30 32

S. pyogenes 15 11 20L-form 36 22

M. laidlawii 36 43 68

the diglucosyl diglyceride since, as already dis-cussed, labeling experiments have shown that theterminal glucose does not turn over. The myco-plasma are a group of organisms devoid of thenormal rigid cell wall, and the presence of rela-tively large amounts of cholesterol in the parasiticspecies may be partly responsible for maintainingthe structural integrity of the cell (71). The largeamount of glycolipids present in M. laidlawiiwhen grown in the absence of cholesterol may besynthesized as a substitute for cholesterol, al-though the addition of cholesterol to the growthmedium does not significantly effect the glyco-lipid concentration. The halophilic H. cutirubrum,which also lacks a cell wall, contains a glycosyldiglyceride which, like the major phospholipidof this organism, phosphatidylglycerophosphate(39), also possess an overall anionic charge inthe form of a sulphate residue (38). A compari-son of the lipid composition of various gram-positive bacteria and their derived L-forms (Table2) reveals that the L-forms contain a greatly in-creased glycolipid content. Thus, a characteristicfeature of many organisms lacking a rigid cellwall is not only enhanced total lipid content butalso a proportionately higher concentration ofglycolipids.

Evidence is now accumulating that many bac-teria contain phosphoglycolipids with structuresrelated to glycolipids present in the same orga-nisms, and this has renewed speculations concern-ing glycolipids as biosynthetic intermediates. Thelipids of propionic acid bacteria contain smallamounts of a phosphatidylmyoinositol mannoside(9, 42), and although its structure has not beenrigorously established the present evidence sug-gests it is a monomannoside derivative of phos-phatidylmyoinositol. The isolation of diacyl inosi-tol monomannoside from the same organisms(59, 66) led to suggestions of an alternative routefor the biosynthesis of the mannophosphoinosi-

tide (66), namely the direct transfer of a phospha-tidic acid residue from cytidine diphosphate-diglyceride to the glycolipid (Fig. 8). This routewould give an initial product with four acyl resi-dues, and it is significant that in their biosyntheticstudies Brennan and Ballou (9) observed a labeledproduct formed by Propionibacterium shermaniiwith the endogenous acceptor that they thoughtmay have been a more highly acylated phosphati-dylmyoinositol monomannoside. Similar highlyacylated phospholipids have been found incorynebacteria (8) and mycobacteria (52). How-ever, addition of exogenous phosphatidylmyo-inositol and labeled guanosine diphosphate-man-nose to an enzyme preparation from P. shermaniidid result in the synthesis of a small amount ofphosphatidylmyoinositol monomannoside (9).Further acylation followed by enzymatic cleavageof the phosphate-inositol linkage would be analternative route to diacyl myoinositol mono-mannoside (59). However, it has not yet beenshown whether propionic acid bacteria possessany phospholipase activity which this route wouldrequire, or indeed whether this reaction can becarried out in vitro. The other product of thisreaction, phosphatidic acid, is not present in thelipids of these organisms (Shaw and Dinglinger,unpublished data).The unusual glucose-containing phospholipid

isolated from M. laidlawii and thought to be aphosphatidylglucose (73) has now been shown tohave a glycerylphosphoryldiglucosyl diglyceridestructure (69). The glycerophosphate residue islocated on one of the 6-hydroxyl groups of thetwo glucose residues (Fig. 9), and its structuralrelationship to a diglucosyl diglyceride is imme-diately apparent. Vigorous alkaline hydrolysis ofthe lipid yielded a diglucosylglycerol of identicalstructure to that derived from the glycolipid of

HO

OH GDP-MmosOH

HO

OH OH

HOOH

CH OH2 CDP-Do-xc>aD

R'O

H ,,OHO~~~~~~~~~IPIHO O-CH2

CH.OR

CH2OR

HO

OHCH2fHO

0 HOH

Oo OHHO0

R- Cot

R-CoA

HO

OHCH2O OH

ORHOHH H

HO

FIG. 8. Possible route for the biosynthesis of thediacyl inositol monomannoside and its utilization asan intermediate in the biosynthesis of diacyl phospha-tidyl-myoinositol monomannoside.

374 SHAW

on May 19, 2020 by guest

http://mm

br.asm.org/

Dow

nloaded from

BACTERIAL GLYCOLIPIDS

CH2O .PO (O H).O -, OH

HO.H2C

O-CH12

1-CH.CO.CR

FIG 9. Structure of the glucose-containing phos-pholipidfrom Mycoplasma laidlawii.

this organism (68). The most likely biosyntheticroute would involve transfer of glycerophosphateto a diglucosyl diglyceride. It seems very probablethat Pieringer has already observed the biosynthe-sis of a similar phosphoglycolipid in Streptococcusfaecalis. While studying the biosynthesis of theglycosyl diglycerides in this organism, he foundthat three glucose-containing lipids were formedsequentially from uridine diphosphate-glucoseand diglyceride (54). The first two were identifiedas monoglucosyl diglyceride and diglucosyl diglyc-eride, but the third lipid which was apparentlyformed from diglucosyl diglyceride was not iden-tified. It was not a triglucosyl diglyceride, but ondeacylation yielded an anionic water-solublehydrolysis product the nature of which was notdetermined; a glycerylphosphoryldiglucosyl di-glyceride structure for this unknown lipid seemsvery likely. Two other reports of the occurrence ofglucose-containing phospholipids in streptococcihave appeared. Both Fischer and Seyferth (27),when examining the lipids of S. faecalis and S.lactis, and Ishizuka and Yamakawa (35), whenexamining the lipids of S. hemolyticus, isolated aphosphoglycolipid for which they suggested adiglucosylphosphatidylglycerol structure analo-gous to the glucosaminylphosphatidylglycerolpresent in B. megaterium (51) and Pseudomonasovalis (53). However, Shaw, Smith, and Verheij(69) pointed out that the structural evidence ob-tained by both groups could apply equally as wellto the alternative glycerylphosphoryldiglucosyldiglyceride structure, and in view of Pieringer'sobservations (54) the latter structure seems more

likely. The two structures may be immediately dis-tinguished by their reaction with periodate. Thelipid from M. laidlawai, on oxidation with peri-odate, liberates one mole proportion of formal-dehyde produced from the terminal glycol residueof the glycerol moiety. The alternative diglucosyl-phosphatidylglycerol structure does not possess

this structural feature and will only liberate form-aldehyde after removal of the two fatty acid resi-dues. Studies currently in progress in this labora-tory indicate that glycerylphosphoryldiglycosyldiglycerides are present in Leuconostoc mesen-teroides and Listeria monocytogenes (Shaw, Hun-ter, and Stead, unpublished data). In these twoorganisms, the constituent sugars are galactoseand glucose, which is consistent with the presenceof galactosylglucosyl diglycerides in these organ-isms (Table 1). This new type of phosphoglyco-lipid may occur as widely as the glycosyl diglyc-erides. It is beyond the scope of this review tospeculate on their function other than to observethat their production and possible subsequentmetabolism may be an important primary func-tion of glycosyl diglycerides.The possibility that glycolipids may have im-

munological significance has received little atten-tion. The immunological properties of the cera-mide hexosides of animal cells has been studied indetail (7), and it has been suggested that glycosyldiglycerides may be their bacterial counterparts(56). In M. pneumoniae, an organism lacking acell wall in which immunological determinantsare usually located, hapten activity measured incomplement fixation tests with both rabbit andhuman antisera was found to be associated withthe glycolipid fraction (56). The observed sero-logical activity of the preparations may have beendue to small amounts of unknown substances ac-companying the glycolipids during purification,but inactivation by periodate and by carbohy-drase preparations suggested that glycolipids wereresponsible. Moreover, cross-reactivity was ob-served between antisera to M. pneumoniae andpurified glycolipids from other sources, includingthe diglucosyl diglycerides of a Streptococcus (57).

LITERATURE CITED

1. Archibald, A. R., and J. Baddiley. 1966. The teichoic acids,p. 323-375. In M. L. Wolfrom and R. S. Tipson (ed.),Advances in carbohydrate chemistry, vol. 21. AcademicPress Inc., New York.

2. Archibald, A. R., J. Baddiley, and D. Button. 1968. Themembrane teichoic acid of Staphylococcus lactis 13. Bio-chem. J. 110:559-563.

3. Barker, S. A., M. C. Keith, and M. Stacey. 1961. Mono-saccharide sequence in Pneumococcus type XIV poly-saccharide. Nature (London) 189:746-747.

4. Benson, A. A. 1964. Plant membrane lipids. Annu. Rev.Plant Physiol. 15:1-16.

5. Bergelson, L. D., S. G. Batrakov, and T. V. Pilipenko. 1970.A new glycolipid from Streptomyces. Chem. Phys. Lipids4:181-190.

6. Bishop, D. G., L. Rutberg, and B. Samuelson. 1967. Thechemical composition of the cytoplasmic membrane ofBacillus subtilis. Eur. J. Biochem. 2:448-453.

7. Brady, R. 0. 1966. Immunochemical properties of glycolipids.J. Amer. Oil Chem. Soc. 43:67-69.

8. Brennan, P. J. 1968. Phosphoinositides of Corynebacteriumxerosis. Biochem. J. 109:158-160.

375VOL. 34, 1970

on May 19, 2020 by guest

http://mm

br.asm.org/

Dow

nloaded from

BACTERIOL. REV.

9. Brennan, P. J., and C. E. Ballou. 1968. Phosphatidylmyo-inositol monomannoside in Propionibacterium shermanii.Biochem. Biophys. Res. Commun. 30:69-75.

10. Brennan, P. J., M. P. Flynn, and P. F. S. Griffin. 1970.

Acylglucoses in Escherichia coli, Saccharomyces cereisviaeand Agaricus bisporus. Fed. Eur. Biochem. Soc. Lett8:322-324.

11. Brennan, P. J., D. P. Lehane, and D. W. Thomas. 1970.

Acylglucoses of the Corynebacteria and Mycobacteria.Eur. J. Biochem. 13:117-123.

12. Brundish, D. E., and J. Baddiley. 1968. Synthesis of glucosyl-glycerols and diglucosylglycerols and their identification insmall amounts. Carbohyd. Res. 8:308-316.

13. Brundish, D. E.. N. Shaw, and J. Baddiley. 1965. The glyco-lipids from the non-capsulated strain of Pneumococcus

1-192R, ATCC 12213. Biochem. J. 97:158-165.14. Brundish, D. E., N. Shaw, and J. Baddiley. 1966. Bacterial

glycolipids. Glycosyl diglycerides in gram-positive bac-teria. Biochem. J. 99:546-549.

15. Brundish, D. E., N. Shaw, and J. Baddiley. 1966. Synthesisof O-a-D-galactopyranosyl-(l -. 2)-O-a-D> glucopyranosyl-(I - I)-D-glycerol, a degradation product of the glyco-lipid from a Pneumococcus. J. Chem. Soc. (C):521-523.

16. Brundish, D. E., N. Shaw, and J. Baddiley. 1967. The struc-ture and possible function of the glycolipid from Staphylo-coccus lactis 13. Biochem. J. 105:885-889.

17. Burger, M. M., L. Glaser, and R. M. Burton. 1963. The enzy-matic synthesis of a rhamnose-containing glycolipid byextracts of Pseudomonas aeruginosa. J. Biol. Chem. 238:2595-2602.

18. Buttery, S. H., and P. Plackett. 1960. A specific polysac-charide from Mycoplasma mycoides. J. Gen. Microbiol.23:357-368.

19. Carroll, K. K., J. H. Cutts, and E. G. D. Murray. 1968.The lipids of Listeria monocytogenes. Can. J. Biochem.46:899-904.

20. Cohen, M., and C. Panos. 1966. Membrane lipid compositionof Streptococcus pyogenes and derived L-form. Biochem-

istry 5:2385-2392.

21. Constantopoulos, G., and K. Bloch. 1967. Isolation andcharacterization of glycolipids from some photosyntheticbacteria. J. Bacteriol. 93:1788-1793.

22. Distler, J., and S. Roseman. 1964. Polysaccharide and glyco-lipid synthesis by cell-free preparations from type XIVPneumococcus. Proc. Nat. Acad. Sci. U.S.A. 51:897-905.

23. Douglas, L. J., and J. Baddiley. 1968. A lipid intermediatein the biosynthesis of a teichoic acid. Fed. Eur. Biochem.Soc. Lett 1:114-116.

24. Edwards, J. R., and J. A. Hayashi. 1965. Structure of a

rhamnolipid from Pseudomonas aeruginosa. Arch. Biochem.Biophys. 111:415-421.

25. Ellwood, D. C. 1970. The wall content and composition ofBacillus subtilis var. niger grown in a chemostat. Biochem.J. 118:367-373.

26. Exterkate, F. A., and J. H. Veerkamp. 1969. Biochemicalchanges in Bifidobacterium bifidum var. Pennsylvanicusafter cell wall inhibition. 1. Composition of lipids. Biochim.Biophys. Acta 176:65-77.

27. Fischer, W., and W. Seyferth. 1968. l-[O-a-D-Glucopyrano-syl-(l 2)-O-a-D-glucopyranosyll-glycerin den Glyko-lipiden von Streptococcus faecalis und Streptococcus lactis.Z. Physiol. Chem. 349:1662-1672.

28. Folch, J., M. Lees, and G. H. Sloane Stanley. 1957. A simplemethod for the isolation and purification of total lipidsfrom animal tissues. J. Biol. Chem. 226:497-509.

29. Frerman, F' E., and D. C. White. 1967. Membrane lipidchanges during the formation of a functional electrontransport system in Staphylococcus aureus. J. Bacteriol.94:1868-1874.

30. Gilby, A. R., A. V. Few, and K. McQuillen. 1958. The chemi-cal composition of the protoplast membrane of Micro-coccus lysodeikticus. Biochim. Biophys. Acta 29:21-29.

31. Heinz, E. 1967. Uber die enzymatische Bildung von Acylgala-ktosyldiglyceride. Biochim. Biophys. Acta 144:333-343.

32. Higashi, Y., J. L. Strominger, and C. C. Sweeley. 1967.Structure of a lipid intermediate in cell wall peptidoglycansynthesis: a CG isoprenoid alcohol. Proc. Nat. Acad.Sci. U.S.A. 57:1878-1884.

33. Ikawa, M. 1963. Nature of the lipids of some lactic acidbacteria. J. Bacteriol. 85:772-781.

34. Ikawa. M. 1967. Bacterial phosphatides and natural relation-ships. Bacteriol. Rev. 31:54-64.

35. Ishizuka, I., and T. Yamakawa. 1968. Glycosyl glyceridesfrom Streptococcus hemolyticus strain D 58. J. Biochem.64:13-23.

36. Jarvis, F. G., and M. J. Johnson. 1949. A glycolipid producedby Pseudomonas aeruginosa. J. Amer. Chem. Soc. 71:4124-4126.

37. Kates, M. 1964. Bacterial lipids, p. 17-90. In R. Paoletti andD. Kritchevsky (ed.), Advances in lipid research, vol. 2.Academic Press Inc., New York.

$8. Kates, M., B. Palameta, M. P. Perry, and G. A. Adams. 1967.A new glycolipid sulphate ester in Halobacterium cuti-rubrum. Biochim. Biophys. Acta 137:213-216.

39. Kates, M., L. S. Yengoyan, and P. S. Sastry. 1965. A dietheranalogue of phosphatidylglycerophosphate in Halobac-terium cutirubrum. Biochim. Biophys. Acta 98:252-268.

40. Kaufmann, B., F. D. Kundig, J. Distler, and S. Roseman.1965. Enzymatic synthesis and structure of two glycolipidsfrom type XIV Pneumococcus. Biochem. Biophys. Res.Commun. 18:312-318.

41. Khan, A. W., and J. R. Calvin. 1961. Isolation of the pre-

cursor of bacterial cellulose. J. Polymer Sci. 51:1-9.42. Laneelle, M. A., and J. Asselineau. 1968. Sur les lipides de

Propionibacterium frendenreichii. C. R. Acad. Sci. Paris266D:1901-1903.

43. Lang, D. R., and D. G. Lundgren. 1970. Lipid compositionof Bacillus cereus during growth and sporulation. J. Bac-teriol. 101:483-489.

44. Lederer, E. 1967. Glycolipids of Mycobacteria and relatedorganisms. Chem. Phys. Lipids 1:294-315.

45. Lennarz, W. J., M. Scher, and C. C. Sweeley. 1968. Biosyn-thesis of mannosyl-l-phosphoryl-polyisoprenol in Micro-coccus lysodeikticus and its role in mannan synthesis.Proc. Nat. Acad. Sci. U.S.A. 59:1313-1320.

46. Lennarz, W. J., and B. Talamo. 1966. The chemical char-acterization and enzymatic synthesis of mannolipids inMicrococcus lysodeikticus. J. Biol. Chem. 241:2207-2719.

47. Macfarlane, M. G. 1961. Isolation of a phosphatidylglyceroland a glycolipid from Micrococcus lysodeikticus. Biochem.J. 80:45.

48. Macfarlane, M. G. 1962. Lipid components of Staphylococcusaurelns and Salmonella typhimurium. Biochem. J. 82:40-41.

49. Okazaki, H., H. Sugino, T. Kanzaki, and H. Fukuda. 1969.L-Glutamic acid fermentation. VI. Structure of a sugar

lipid produced by Brevibacterium thiogenitalis. Agr. Biol.Chem. (Tokyo) 33:764-770.

50. O'Leary, W. M. 1967. The chemistry and metabolism ofmicrobial lipids. The World Publishing Co., Cleveland,Ohio.

51. Op den Kamp, J. A. F., P. P. M. Bonsen, and L. L. M. van

Deenen. 1969. Structural investigations on glucosaminylphosphatidylglycerol from Bacillus megaterium. Biochim.Biophys. Acta 176:298-305.

52. Pangborn, M. C., and J. A. McKinney. 1966. Purification ofserologically active phosphoinositides of Mycobacteriumtuberculosis. J. Lipid Res. 7:627-633.

53. Phizackerley, P. J. R., J. C. MacDougall, and M. J. 0.

Francis. 1966. Phosphatidylglycerolglucosamine. Biochem.J. 99.21-22C.

54. Pieringer, R. A. 1968. The metabolism of glyceride glyco-lipids. I. Biosynthesis of monoglucosyl diglyceride anddiglucosyl diglyceride by glucosyltransferase pathways inStreptococcus faecalis. J. Biol. Chem. 243:4894-4903.

376 SHAW

on May 19, 2020 by guest

http://mm

br.asm.org/

Dow

nloaded from

BACTERIAL GLYCOLIPIDS

55. Plackett, P. 1967. The glycerolipids of Mycoplasma mycoides.Biochemistry 6:2746-2754.

56. Plackett, P., B. P. Marmion, E. J. Shaw, and R. M. Lemcke.1969. Immunochemical analysis of Mycoplasma pneumoniae.3. Separation and chemical identification of serologicallyactive lipids. Aust. J. Exp. Biol. Med. Sci. 47:171-195.

57. Plackett, P., and E. J. Shaw. 1967. Glycolipids from Myco-plasma laidlawii and Streptococcus MG. Biochem. J. 104:61-62C.

58. Polonovski, J., R. Wald, and M. Paysant Diament. 1962.Les lipides de Staphj lococcus aureus. Ann. Inst. Pasteur103:32-41.

59. Prottey, C., and C. E. Ballou. 1968. Diacyl myoinositolmonomannoside from Propionibacterium shermanii. J.Biol. Chem. 243:6196-6201.

60. Rajbhandary, U. L., and J. Baddiley. 1963. The intracellularteichoic acid from Staphylococcus aureus H. Biochem. J.87:429-435.

61. Reeves, R. E., N. G. Latour, and R. J. Lousteau. 1964. Aglycerol galactofuranoside from the lipid of an anearobe

provisionally designated as Bacteriodes symbiosus. Bio-chemistry 3:1248-1249.

62. Rouser, G., A. J. Bauman, G. Kritchevsky, D. Heller, andJ. S. O'Brien. 1961. Quantitative chromatographic frac-tionation of complex lipid mixtures. Brain lipids. J. Amer.Oil Chem. Soc. 38:544-555.

63. Shaw, N. 1968. The lipid composition of Microbacteriumlacticum. Biochim. Biophys. Acta 152:427-428.

64. Shaw, N. 1968. The detection of lipids on thirn-layer chromat-

ograms with- the periodate-Schiff reagents. Biochim.Biophys. Acta 164:435-436.

65. Shaw, N., and J. Baddiley. 1968. Structure and distributionof glycosyl diglycerides in bacteria. Nature (London)217:142-144.

66. Shaw, N., and F. Dinglinger. 1969. The structure of an

acylated inositol mannoside in the lipids of propionic acidbacteria. Biochem. J. 112:769-775.

67. Shaw, N., K. Heatherington, and J. Baddiley. 1968. Theglycolipizs of Lactobacillus casel ATCC 7469. Biochem.J. 107:491-496.

68. Shaw, N., P. F. Smith, and W. L. Koostra. 1968. The lipid

composition of Mycoplasma laidlawii, strain B. Biochem.J. 107:329-333.

69. Shaw, N., P. F. Smith, and H. M. Verheij. 1970. On the

structure of phosphatidylglucose. Biochem. J. 120:439-441.

70. Shaw, N., and D. Stead. 1970. A study of the lipid composi-

tion of Microbacterium thermosphactum as a guide to its

taxonomy. J. Appl. Bacteriol. 33:470-473.

Statement of Ownership, Management and Circulation requiredby the Act of October 23, 1962; Section 4369, Title 39, UnitedStates Code.1. Date of Filing: 30 September 1970.2. Title of Publication: Bacteriological Reviews.3. Frequency of Iseu.: Monthly.4. Location of known Office of Publication: Mount Royal and

Guilford Aves., Baltimore, Maryland 21202.5. Location of the Headquarters or General Business Offices of

Publisher: 4715 Cordeil Ave., Bethesda, Maryland 20014.6. Publisher: American Society for Microbiology, 4715 Cordell

Ave., Bethesda, Maryland 20014.Editor: R. G. E. Murray, Department of Bacteriology,

University of Western Ontario London 72 Ont., Canada.Managing Editor: Robert A. bay, 4011 Rickover Rd., Silver

Spring, Maryland 20902.7. Owners: (If owned by a corporation, its name and address

must be stated and also immediately thereunder the names andaddresses of stockholders owning or holding 1 percent or more oftotal amount of stock. If not owned by a corporation, the namesand addresses of the individual owners must be given. If ownedby a partnership or other unincorporated firm, its name andaddress, as well as that of each individual, must be given.)American Society for Microbiology, 4715 Cordell Ave., Bethesda,Maryland 20014.

8. Known bondholders, mortgagees, and other security holdersowning or holding I percent or more of total amount of bonds, mort-

gages or other securities are: None.

71. Smith, P. F. 1968. The lipids of Mycoplasma, p. 69-105. InR. Paoletti and D. Kritchevsky (ed.), Advances in lipidresearch, vol. 6. Academic Press Inc., New York.

72. Smith, P. F. 1969.. Biosynthesis of glucosyl diglycerides byMycoplasma laidlawii strain B. J. Bacteriol. 99:480-486.

73. Smith, P. F., and C. V. Henrickson. 1965. Glucose-containingphospholipids in Mycoplasma laidlawli, strain B. J. LipidRes. 6:106-11.

74. Smith, P. F., and W. R. Mayberry. 1968. Identification of themajor glycolipid from Mycoplasma sp., strain J as 3,4,6-triacyl #-n-glucopyranose. Biochemistry 7:2706-2710.

75. Steiner, S., S. F. Conti, and R. L. Lester. 1969. Separationand identification of the polar lipids of Chromatlum strainD. J. Bacteriol. 98:10-15.

76. Vorbeck, M. L., and G. V. Marinetti. 1965. Separation ofglycosyl diglycerides from phosphatides using silicic acidchromatography. J. Lipid Res. 6:3-6.

77. Vorbeck, M. L., and G. V. Marinetti. 1965. Intracellulardistribution and characterisation of the lipids of Strepto-coccu. faecalis ATCC 9790. Biochemistry 4:296-305.

78. Wagner, H., L. Horhammer, and P. Wolff. 1961. Dunnschicht-chromatographie von Phosphatiden und Glykolipiden.Biochem. Z. 334:175-184.

79. Walker, R. W., and C. P. Bastl. 1967. The glycolipids ofArthrobacter globiformis. Carbohyd. Res. 4:49-54.

80. Ward, J. B., and H. R. Perkins. 1968. The chemical compo-

position of the membranes of protoplasts and L-forms ofStaphlIococcus aureus. Biochem. J. 106:391-400.

81. Wells, M. A., and J. C. Dittmer. 1963. The use of sephadexfor the removal of non-lipid contaminants from lipid ex-

tracts. Biochemistry 2:1259-1263.82. Welsh, K., N. Shaw, and J. Baddiley. 1968. The occurrence

of acylated sugar derivatives in the lipids of bacteria.Biochem. J. 107:313-314.

83. Wicken, A. J., and J. Baddiley. 1963. Structure of intra-cellular teichoic acids from group D streptococci. Biochem.J. 87:54-62.

84. Wincken, A. J., and K. W. Knox. 1970. Studies on the group

F antigen of lactobacilli: isolation of a teichoic acid-lipidcomplex from Lactobacillus fermenti NCTC 6991. J. Gen.Microbiol. 60:293-301.

85. Wilkinson, S. G. 1968. Glycosyl diglycerides from Pseu-domonas rubescens. Biochim. Biophys. Acta 164:148-156.

86. Wilkinson, S. G. 1969. Lipids of Pseudomonas diminuta.Biochim. Biophys. Acta 187:492-500.

87. Wright, A., M. Dankert, P. Fennessey, and P. W. Robbins.1967. Characterisation of a polyisoprenoid compoundfunctional in O-antigen biosynthesis. Proc. Nat. Acad.Sci. U.S.A. 57:1798-1803.

9. For completion by nonprofit organizations authorized tomail at special rates (section 185.1ff, Postal Manual): The pur-

pose, function, and nonprofit status of this organization andthe exempt status for Federal income tax purposes have notchanged during preceding I1 months.

(a)* (b)t

10. A. Total No. Copies Printed (Net PreRun)................................... 15,170 14,185

B. Paid Circulation1. Sales Through Dealers and Carriers,

Street Vendors and Counter Sales none none

2. Mail Subscriptions................... 13,976 12,415C. Total Paid Circulation 13,976 12,415D. Free Distribution (including samples)

by Mail, Carrier or other means........ 41 43

E. Total Distribution (Sum of C and D).. 14,017 12,458F. Office Use, Left-Over, Unaccounted,

Spoiled after Printing ....... ........... 1,153 1,727G. Total (Sum of E & F-should equal net

press run shown in A).................. 15,170 14,185

* Average number of copies for each issue during preceding12 months.

t Single issue nearest to filing date.I certify that the statements made by me above are correct and

complete. (Signed) Robert A. Day, Managing Editor.

VOL. 34, 1970 377

on May 19, 2020 by guest

http://mm

br.asm.org/

Dow

nloaded from