Vol. 174, No. 3JOURNAL OF BACTERIOLOGY, Feb. 1992, p. 682-6860021-9193/92/030682-05$02.00/0Copyright © 1992, American Society for Microbiology

Membrane Properties of a Plant-PathogenicMycoplasmalike Organism

PYUNG-OK LIM,' BARBARA B. SEARS,".2* AND KAREN L. KLOMPARENS2'3Genetics Program,' Department ofBotany and Plant Pathology,2* and Department ofEntomology, Center for Electron

Optics, and Pesticide Research Center,3 Michigan State University, East Lansing, Michigan 48824

Received 26 July 1991/Accepted 13 November 1991

In terms of biosystematics, the plant-pathogenic mycoplasmalike organisms (MLOs) have been tentativelyplaced into the class MoUlcutes. Certain physiological tests have been used to distinguish families within thisclass: the sterol-nonrequiringAcholeplasmataceae differ from the sterol-requiring Mycoplasmataceae in that theformer are more resistant to lysis by digitonin and more sensitive to lysis in hypotonic salt solutions. To testMLOs for these membrane properties and thus assist in their definitive classification, a dot-blot microassayprocedure was used to detect nucleic acids released from lysed cells. The results show that MLOs resembleacholeplasmas grown in the absence of sterols in that they are resistant to digitonin and sensitive to hypotonicsalt solutions. The MLOs can be differentiated from acholeplasmas grown without sterols by their greaterresistance to lysis in hypotonic sucrose solutions.

The nonculturable plant-pathogenic mycoplasmalike or-ganisms (MLOs) have been tentatively classified with theMollicutes because they are small, pleomorphic prokaryoteslacking a cell wall, and thus their morphology resembles thatof animal mycoplasmas in this class. Recent sequence datafrom the 16S rRNA gene allow the MLOs to be classifiedwith the Mollicutes and indicate that MLOs are more closelyrelated to Acholeplasma laidlawii than to any of the animalmycoplasmas (6). However, genome size estimations indi-cate that MLOs contain genomes that are significantlysmaller than those of acholeplasmas (7). To supplementthese studies, other features that differentiate animal myco-plasmas and acholeplasmas need to be examined in theMLOs.

In the Mollicutes, sterol-requiring mycoplasmas are moresusceptible to lysis by digitonin, amphotericin, and lysoleci-thin and more resistant to lysis in hypotonic salt solutionsthan sterol-nonrequiring acholeplasmas (3, 10, 12, 22, 23).Thus, the aim of the present investigation was to test MLOsfor these membrane properties.

In previous studies, lysis of cells in hypotonic salt solu-tions and digitonin was measured by changes in opticaldensity at several wavelengths, resulting from a decrease inthe turbidity of the suspension due to cell lysis and release ofnucleic acids into the medium (12, 23). Since our MLOs areisolated along with plant mitochondria and host cell debris,changes in optical density will be due to lysis of mitochon-dria and vesicles as well as MLOs. To overcome thisproblem, we developed a dot-blot microassay to allow us todetect specifically the lysis of MLOs in a mixed suspension,using an MLO-specific plasmid DNA as a probe.

In this report, we show that MLOs resemble acholeplas-mas in their membrane properties. These results support ourDNA sequence data that indicate a close relationship to A.laidlawii (6).

MATERIALS AND METHODSBacterial strains and growth conditions. A. laidlawii and

Mycoplasma gallisepticum were kindly provided by Terry

* Corresponding author.

Moser (Michigan State University Veterinary Clinic). A.laidlawii and M. gallisepticum were grown at 37°C in 10 mlof a serum-containing medium consisting of 1.5% PPLObroth (Difco), 20% (vol/vol) heat-inactivated pig serum (Dif-co), 10o (vol/vol) of 25% (wt/vol) extracts of active driedyeast (pH 8.0), 0.005% thallium acetate, 1,000 U of penicillinG per ml (U.S. Biochemicals), 0.1% glucose, and 0.005%phenol red. A. laidlawii was also grown in a partially definedmedium in which the pig serum was replaced by 1% bovineserum albumin (17). The two organisms were harvested atthe log phase by centrifugation at 25,000 x g for 25 min.They were washed once with 1 ml of 250 mM NaCl or 500mM sucrose and resuspended in 150 ,ul of the same solution.

Preparation of MLOs from infected plant tissues. TheMLOs were isolated from infected Oenothera leaftip cul-tures by the method of Sears et al. (25), using about 2 g ofplant material for each preparation. Differential centrifuga-tion pelleted the MLOs together with plant mitochondria.The pellets were then resuspended with 150 ,ul of 250 mMNaCl or 500 mM sucrose.

Preparation of total chromosomal DNA and MLO-specificplasmid. Total cellular DNA was isolated from M. gallisep-ticum and A. laidlawii by a method described previously(29), except that the lysozyme treatment was omitted and thepronase was replaced by proteinase K. A clone of theMLO-specific plasmid (pMP 9) was provided by Neta Hol-land.

Assessment of sensitivity to digitonin and osmotic lysis. Totest the sensitivity of the organisms to digitonin, cells weresuspended in 250 mM NaCl, and 10-,ul aliquots were addedto 400 RI1 of 250 mM NaCl containing different concentra-tions of digitonin. The tubes were incubated at 37°C for 30min to allow time for cell lysis, and then the suspensionswere centrifuged at 15,000 x g at 4°C for 30 min. Thesupernatant was removed from each tube. After boiling for 5min, 200 ,u1 of the supernatant was loaded onto a nylonmembrane by using a Bio-Rad vacuum blotter. A subset ofthe cell pellets was examined by transmission electronmicroscopy (TEM).To test the sensitivity to osmotic lysis, we used a similar

procedure. Aliquots (10 ,ul) of cell suspension were added to400 ,ul of serial twofold dilutions of 250 mM NaCl or 500 mM

682

on March 12, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

MEMBRANE PROPERTIES OF MYCOPLASMALIKE ORGANISM 683

sucrose, as well as distilled water, in microcentrifuge tubes,with incubation for 30 min at room temperature. The dotblots were hybridized in solutions lacking formamide, andthe filters were washed at low stringency, according to theprocedure described by Maniatis et al. (9). Nick translation(9) was used to radioactively label the chromosomal DNAsisolated from A. laidlawii and M. gallisepticum and theMLO-specific plasmid DNA, for individual use as probesagainst the dot blots containing the relevant samples.TEM. Cells were fixed by resuspending them in 2%

glutaraldehyde in 100 mM phosphate buffer with 2% sucrosefor 20 to 30 min. Cells were then centrifuged at 15,000 x gfor 30 min, and the pellet was embedded in agar. The agarpellet was briefly fixed again for 15 min. Following threewashes with buffer, the samples were postfixed in buffered1% OS04. Samples were washed twice in buffer and threetimes in distilled water to remove sucrose, infiltrated, andembedded in a mixture of Epon-Araldite-Spurs epoxy resin(4). Ultrathin sections (-80 nm) were cut with a diamondknife and stained with saturated aqueous uranyl acetate andlead citrate before being viewed on a JEOL 100CXII TEMoperated at 100 kV.

RESULTS

Sensitivities of organisms to digitonin and osmotic lysis. Totest the sensitivity of MLOs to digitonin, we used a dot-blotassay procedure with an MLO-specific plasmid probe todetect nucleic acids released from lysed cells after cells wereincubated in different concentrations of digitonin. The sen-sitivities of M. gallisepticum and A. laidlawii were deter-mined under the same conditions. For these organisms, totalcellular DNA isolated from each organism was used as aprobe. As shown in Fig. 1A and Table 1, M. gallisepticumlysed in the presence of digitonin. Similarly, when A. laid-lawii was grown in serum-containing medium, the cells weresensitive to digitonin. However, A. laidlawii grown in se-rum-free medium and plant-pathogenic MLOs were com-pletely resistant.To test the sensitivity of MLOs to lysis in serial NaCl or

sucrose solutions, we used the same dot-blot microassayprocedure. M. gallisepticum was resistant to hypotonic saltsolutions (Fig. 1B). For the A. laidlawii samples, regardlessof whether they had been grown in the presence or absenceof serum, lysis occurred readily in low-tonicity solutions(s31 mM NaCl) (Fig. 1B; Table 1). Under the same testconditions, the sensitivity of the plant-pathogenic MLO wassimilar to that of the acholeplasma. The lysis of A. Iaidlawiiand the MLO in hypotonic salt solutions was completelyprevented by including 10 mM MgCl2 in the NaCl solutions(data not shown), as described previously for other Molli-cutes (23).Because the MLOs were isolated from plants, in which

sucrose is the primary osmoticum, we conducted an exper-iment to test the tendency of cells to lyse in serial sucrosedilutions. M. gallisepticum was resistant to the osmoticchanges (Fig. 1C), which is identical to its response in theNaCl solution. But for A. Iaidlawii and the MLO, thesensitivity to lysis in the diluted sucrose solutions differedfrom their response to the NaCI solutions. When A. laidlawiiwas grown in serum-containing medium, it was resistant tothe hypotonic sucrose solution. In contrast, when A. laid-lawii was grown in serum-free medium, cells lysed afterbeing placed in a solution with less than 8 mM sucrose (Table1; Fig. 1C). Plant-pathogenic MLOs showed resistance to

ADigitonin (pg/ml)

Serum 40 20 10 0

M. gallisepticum + * * * *

A. laidlawii + ,

MLO NA

B0.25M

Serum NaCI 2 fold dilutions DW

M. gallisepticum +

A. laidlawii + * 0

MLO NA * *

C0.5M

Serum Sucrose 2 fold dilutions DW

M. gallisepticum +

A. laidlawii +

MLO NA *

FIG. 1. Sensitivities of organisms to digitonin and osmotic lysis.(A) Dot-blot hybridization of nucleic acids released by cells incu-bated in different concentrations (40, 20, 10, 0 ,ug/ml) of digitonin.Cells were grown in the presence (+) or absence (-) of serum,although this is not applicable (NA) to the MLOs, which wereisolated from plants. Their respective chromosomal DNAs wereused as probes for M. gallisepticum and A. laidlawii, and theMLO-specific plasmid was used to detect lysis of the MLO. (B plusC) Dot-blot hybridization of nucleic acids released from cells aftertransfer to twofold dilutions of 250 mM NaCl (B) or 500 mM sucrose(C). Conditions of growth and probes were as in panel A. DW,distilled water.

the entire sucrose dilution series, even when they weretransferred from 500 mM sucrose to distilled water.

Examination of cell lysis by TEM. To verify that thehybridization signals were due to cell lysis, we examined cellpellets of M. gallisepticum and A. Iaidlawii by TEM aftertreatments with several concentrations of NaCl or sucroseranging from 250 mM to distilled water. Figure 2 showsmicrographs of cell pellets from the extremes of the NaCldilution series. All of the cell pellets of M. gallisepticumcontained intact cells, although cells incubated in distilledwater appeared swollen (Fig. 2B). In contrast to this, the A.Iaidlawii pellet contained broken cells after placement indistilled water (Fig. 2D). In the pellet of A. laidlawii incu-bated in distilled water containing 10 mM MgCl2, only intactcells were seen (data not shown).When A. laidlawii was incubated in a sucrose dilution

series, convoluted, but intact, cells were seen throughout therange of conditions examined (Fig. 3A and B).

Since our isolation procedure yields MLOs heavily con-

VOL. 174, 1992

on March 12, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

684 LIM ET AL.

TABLE 1. Conditions resulting in lysis of Mollicutes cells

Lysis test'Mollicutestype ~~SerumindMotlicutes type Smediuma NaCIc Sucrosec Digitonind

(mM) (mM) (>Lg/ml)

M. gallisepticum + - -.210A. laidlawii + <31 - -10A. laidlawii -<31 -8MLO NA -<31 -

a Cells were grown in the presence (+) or absence (-) of serum. It is not applicable (NA) to the MLOs, which were grown in plant leaftip cultures.I Numerical values indicate the concentrations at which significant lysis was observed, while - indicates that no lysis was observed in any concentration of

a solution.c Cells were resuspended in 250 mM NaCI or 500 mM sucrose, and aliquots were transferred to a dilution series and distilled water to test for lysis due to osmotic

stress.d Cells were resuspended in 250 mM NaCl, and aliquots were transferred to a solution containing 0 to 40 p.g of digitonin per ml in 250 mM NaCl.

taminated with mitochondria and host cell debris, it was notpossible to verify MLO lysis with TEM.

DISCUSSION

The main purpose of this study was to compare themembrane properties of mycoplasmas, acholeplasmas, andMLOs. These experiments were designed to determinewhether physiological traits of the MLO plasma membraneresembled those of one of the major Mollicutes families. Weadapted the procedures for examining the membrane prop-erties of other Mollicutes to examine the membrane physi-ology of a nonculturable plant-pathogenic MLO (5).We compared the Oenothera MLO with representatives of

the Mycoplasmataceae and Acholeplasmataceae families ofthe Mollicutes in testing for membrane rupture by digitonin

FIG. 2. Transmission electron micrographs of ultrathin sectionsof M. gallisepticum and A. laidlawii. Cells of M. gallisepticum (Aand B) and A. laidlawii (C and D) were suspended in 250 mM NaCland then transferred to either 250 mM NaCl (A and C) or distilledwater (B and D). The cell suspensions were centrifuged as describedin Materials and Methods, and the pellets were examined byelectron microscopy. Bar = 0.5 n.m.

and osmotic sensitivity to a NaCl dilution series. The dot-blot data (Fig. 1A and B) show that the MLOs are similar toacholeplasmas grown in serum-free medium in that they areresistant to lysis by digitonin and sensitive to lysis inhypotonic salt solutions. The results of our dot blots areconsistent with the extent of cellular integrity observed byTEM (Fig. 2).

Digitonin is a steroid glycoside that forms equimolarinsoluble complexes with cholesterol. Digitonin specificallyperforates the cholesterol-containing cell membranes andmakes the membranes permeable to ions (20), metabolites(28), and proteins (8), possibly leading to cell lysis. There-fore, sensitivity to digitonin has been used as a trait that can

FIG. 3. Transmission electron micrographs of ultrathin sectionsof A. laidlawii and Oenothera MLO. A. laidlawii cells were sus-pended in 500 mM sucrose and then transferred to either 250 mMsucrose (A) or distilled water (B). (C) Ultrathin section of Oenotheravascular tissue, with phloem cell containing MLOs. Bar = 0.5 Rm.

J. BACTERIOL.

on March 12, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

MEMBRANE PROPERTIES OF MYCOPLASMALIKE ORGANISM 685

indicate the presence of sterols in Mollicutes membranes(16, 26).

Sterol-requiring mycoplasmas incorporate high amountsof sterols (about 20% of the total lipids) in their membranes(15). Sterol-nonrequiring acholeplasmas can incorporate ste-rols into their membranes in the presence of serum (sterolswill compose about 7% of the total lipids) (15), or they cangrow without sterols in the absence of serum (19). Ourcontrol experiments with M. gallisepticum and A. laidlawiigrown with and without sterols are in complete agreementwith the concept that the lysis of cells by digitonin is due tothe presence of sterols in their membranes. Our observationthat the MLO is resistant to digitonin treatment indicatesthat the plant pathogen does not contain sterols in itsmembrane. It is unlikely that digitonin was sequestered byother membranes in the MLO-mitochondria fraction, be-cause a dilution series of this fraction showed no differencesin sensitivity when tested against 250 mM NaCl with orwithout 20 pug of digitonin per ml (data not shown).The relative abundance of sterols in the membrane was

also thought to be responsible for the difference in suscep-tibility of Mollicutes in hypotonic salt solutions (12). How-ever, when we and others (18) compared the osmotic sensi-tivity of A. laidlawii grown with and without serum inhypotonic salt solutions, the two samples were equallysensitive to osmotic stress. Other studies (12, 13, 24) alsosupport the conclusion that, in contrast to the sensitivity todigitonin, the resistance of the plasma membrane to osmoticlysis does not depend on the presence of sterols.

Sucrose is the primary osmoticum in the phloem sievetubes of most plants, where MLOs are deposited by theirinsect vectors, and where they reside during the plant-pathogenic stage of their life cycle. For this reason, andbecause sucrose would not have an ionic effect on the cellmembrane (in contrast to NaCl), we decided to test cells fortheir tendency to lyse in a sucrose dilution series. In contrastto the results with hypotonic salt solutions, the MLOs didnot lyse in the hypotonic sucrose solutions (Fig. 1C; Table1). The other Mollicutes tested were also totally resistant toosmotic lysis after resuspension in 500 mM sucrose, with theexception of acholeplasmas grown in serum-free medium,which lysed when transferred to solutions containing lessthan 8 mM sucrose. These results agree with the data ofRazin and Argaman (14) and differ from those of McElhaneyand colleagues (11), although this is probably due to differ-ences in experimental details.The observation that almost none of the Mollicutes sam-

ples lysed or even became swollen in the sucrose dilutionseries was surprising. There are several possible explana-tions for this observation. Although it may seem unlikely,the simplest explanation is that the membranes are renderedimpermeable to water by the sucrose treatment. Biologicalprecedence for this can be found in the renal epithelial cells,which allow passage of water only through membrane pores(21). Alternatively, perhaps the sucrose in the resuspensionmedium stabilized the membranes to the point of rigidity.This does not seem very likely, although sugars can stabilizemembranes through hydrogen bonding with the lipid headgroups (2). A third possibility is that the hypertonic sucrosesolution rendered the membranes leaky to ions (27). If thecells were fully permeable to ions, when they were trans-ferred into the hypotonic sucrose solutions and distilledwater, ion efflux may have eliminated the osmotic differ-ences across the membrane.The cellular morphology shown in Fig. 3A and B is

reminiscent of the natural appearance of MLOs in phloem of

infected plants (Fig. 3C). We believe this observation pointsto the high concentration of sucrose in the sieve tubes asbeing responsible for the distortions of MLOs in situ. Thefact that the MLO seems to be more tolerant to variations insucrose levels than A. laidlawii grown without sterol sug-gests that MLOs have another means of maintaining mem-brane integrity. MLO tolerance of a wide range of sucroseconcentrations is consistent with their adaptation to plantphloem, where sucrose levels may fluctuate from 150 to 600mM, depending on the season (1).

In conclusion, the membrane properties of the MLOresemble those of an acholeplasma grown in serum-freemedium. These results support the conclusions derived fromour previous molecular data of the 16S rRNA gene sequencethat indicate a close relationship between A. laidlawii andthis representative of the MLOs. The information that ste-rols are probably absent from the MLO membrane may beuseful for establishing an in vitro culture of the MLOs, whichwill allow a more precise determination of the nutritionalrequirement of these pathogens.

ACKNOWLEDGMENTS

We gratefully acknowledge the critical and helpful comments ofR. Hollingsworth, T. Beveridge, H. Bertrand, M. Thomashow,R. N. McElhaney, and A. Wieslander. J. I. Wood, H. Y. Yang, andA. Taggie are acknowledged for their skilled technical assistance.

This research was supported by National Science Foundationgrant BSR-89068279 and U.S. Department of Agriculture grant91-37303-6388 and by the Michigan State Agricultural ExperimentStation.

REFERENCES1. Crafts, A. S., and C. E. Crisp. 1971. Phloem transport in plants,

p. 481. W. H. Freeman & Co., San Francisco.2. Crowe, J. H., L. M. Crowe, J. F. Carpenter, A. S. Rudolph,

C. A. Wistrom, B. T. Sparogo, and T. J. Anchordoguy. 1988.Interactions of sugars with membranes. Biochim. Biophys. Acta947:367-384.

3. Grabowski, M. W., S. Rottem, and M. F. Barile. 1976. Choles-terol requirement of mycoplasmas as determined by a microtitertest using polyene antibiotics. J. Clin. Microbiol. 3:110-112.

4. Klomparens, K. L., S. L. Flegler, and G. R. Hopper. 1986.Procedures for transmission and scanning electron microscopyfor biological and medical science: a laboratory manual. LaddResearch Industries, Burlington, Vt.

5. Lee, I.-M., and R. E. Davis. 1986. Prospects for in vitro cultureof plant-pathogenic mycoplasmalike organisms. Annu. Rev.Phytopathol. 24:339-354.

6. Lim, P.-O., and B. B. Sears. 1989. 16S rRNA sequence indicatesthat plant-pathogenic mycoplasmalike organisms are evolution-arily distinct from animal mycoplasmas. J. Bacteriol. 171:5901-5906.

7. Lim, P.-O., and B. B. Sears. 1991. Genome size of a plant-pathogenic mycoplasmalike organism resembles those of animalmycoplasmas. J. Bacteriol. 173:2128-2130.

8. MacKall, J., M. Meredith, and M. D. Lane. 1979. A mildprocedure for the rapid release of cytoplasmic enzyme fromcultured animal cells. Anal. Biochem. 95:270-274.

9. Maniatis, T., E. F. Fritsch, and J. Sambrook. 1982. Molecularcloning: a laboratory manual. Cold Spring Harbor Laboratory,Cold Spring Harbor, N.Y.

10. Mardh, P.-A., and D. Tayler-Robinson. 1973. The differentialeffect of lysolecithin on mycoplasmas and acholeplasmas. Med.Microbiol. Immunol. 158:219-226.

11. McElhaney, R. N., J. De Gier, and E. C. M. Van Der Neut-Kok.1973. The effect of alterations in fatty acid composition andcholesterol content on the nonelectrolyte permeability of A.Iaidlawii B cells and derived liposomes. Biochim. Biophys. Acta298:500-512.

12. Razin, S. 1963. Osmotic lysis of mycoplasmas. J. Gen. Micro-

VOL. 174, 1992

on March 12, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

686 LIM ET AL.

biol. 33:471-475.

13. Razin, S. 1982. Sterols in mycoplasma membranes. Curr. Top.Membr. Transp. 17:183-205.

14. Razin, S., and M. Argaman. 1963. Lysis of mycoplasma, bac-terial protoplasts, spheroplasts and L-forms by various agents.J. Gen Microbiol. 30:155-172.

15. Razin, S., M. Argaman, and J. Avigan. 1964. Chemical compo-

sition of mycoplasma cells and membranes. J. Gen. Microbiol.33:477-487.

16. Razin, S., and E. A. Freundt. 1984. The mycoplasmas, p.

740-793. In N. R. Kreig and J. G. Holt (ed.), Bergey's manualof systematic bacteriology, vol. 1. The Williams & Wilkins Co.,Baltimore.

17. Razin, S., and B. C. J. G. Knight. 1960. A partially definedmedium for growth of mycoplasmas. J. Gen. Microbiol. 22:492-503.

18. Razin, S., M. E. Tourtellotte, R. N. McElhaney, and J. D.Poliack. 1966. Influence of lipid components of M. laidlawiimembranes on osmotic fragility of cells. J. Bacteriol. 91:609-616.

19. Razin, S., and J. G. Tully. 1970. Cholesterol requirement ofmycoplasmas. J. Bacteriol. 102:306-310.

20. Ronning, S. A., G. A. Geatley, and T. F. S. Martin. 1982.Tryptophan-releasing hormone mobilizes Ca"+ from endoplas-mic recticulum and mitochondria of GH3 pituitary cells: char-acterization of cellular Ca"+ pools by a method based on

digitonin permeabilization. Proc. Nati. Acad. Sci. USA 79:6294-6298.

21. Rose, B. D. 1989. Clinical physiology of acid-base and electro-lyte disorders. McGraw-Hill Book Co., New York.

22. Rottem, S. 1972. Differentiation of sterol-requiring from sterol-nonrequiring mycoplasmas by amphotericin B. Appl. Microbiol.23:659-660.

23. Rottem, S., and S. Razin. 1972. Isolation of mycoplasma mem-branes by digitonin. J. Bacteriol. 110:699-705.

24. Russell, N. J. 1989. Functions of lipids: structural roles andmembrane functions, p. 279-365. In C. Ratledge and S. G.Wilkinson (ed.), Microbial lipids, vol. II. Academic Press, Inc.,New York.

25. Sears, B. B., P.-O. Lim, N. Holland, B. C. Kirkpatrick, andK. L. Klomparens. 1989. Isolation and characterization ofDNAfrom a mycoplasmalike organism. Mol. Plant Microbe Interact.2:175-180.

26. Subcommittee on the Taxonomy of MoUicutes. 1979. Proposal ofminimal standards for descriptions of new species of the classMollicutes. Int. J. Syst. Bacteriol. 29:172-180.

27. Taupin, C., D. Maya, and S. Claude. 1975. Osmotic pressureinduced pores in phospholipid vesicles. Biochemistry 14:4771-4775.

28. Ucker, D. S., and K. R. Yamamoto. 1984. Early event in thestimulation of mammary tumor virus RNA synthesis by gluco-corticoids. J. Biol. Chem. 259:7416-7420.

29. Yee, T., and M. Inouye. 1981. Reexamination of the genome sizeof myxobacteria, including the use of a new method for genomesize analysis. J. Bacteriol. 145:1257-1265.

J. BACTERIOL.

on March 12, 2020 by guest

http://jb.asm.org/

Dow

nloaded from