Vol. 155, No. 1JOURNAL OF BACTERIOLOGY, July 1983, P. 357-3660021-9193/83/070357-10$02.00/0Copyright C 1983, American Society for Microbiology

Sporulation of Streptomyces griseus in Submerged CultureKATHLEEN E. KENDRICKt* AND JERALD C. ENSIGN

Department of Bacteriology, University of Wisconsin, Madison, Wisconsin 53706

Received 8 November 1982/Accepted 12 April 1983

A wild-type strain of Streptomyces griseus forms spores both on solid media(aerial spores) and in liquid culture (submerged spores). Both spore types arehighly resistant to sonication, but only aerial spores are resistant to lysozymedigestion. Electron micrographs suggest that lysozyme sensitivity may result fromthe thinner walls of the submerged spores. Studies of the life cycle indicate thatneither streptomycin excretion nor extracellular protease activity is required forsporulation; the analysis of mutants, however, suggests that antibiotic productionmay be correlated with the ability to sporulate. A method was devised to inducethe rapid sporulation of S. griseus in a submerged culture. This method, whichdepends on nutrient deprivation, was used to determine that either ammonia orphosphate starvation can trigger sporulation and that the enzyme glutaminesynthetase may be useful as a sporulation marker after phosphate deprivation.

The life cycle of streptomycetes is uniqueamong bacteria. Under appropriate culture con-ditions, a single spore germinates, grows vegeta-tively as a substrate (primary) mycelium, andultimately segments into chains of spores. Thedeveloping spores are typically enveloped by asheath (3). The thick-walled spores are physio-logically dormant when mature and are relative-ly resistant to lytic enzymes, low temperature,and osmotic extremes (3).The sporulation process has been well docu-

mented with respect to the ultrastructure ofcultures grown on solid media (5, 9, 17, 22).Physiological studies have been hampered bythe inability to obtain uniform sporulation of theorganism in a liquid culture. We have observed,however, that some streptomycetes can be in-duced to sporulate in a liquid culture whencritical nutritional and environmental conditionsare met. In this paper, we report the results ofexperiments conducted with a wild-type strainof Streptomyces griseus, which sporulates wellwhen cultivated in liquid media. A method forthe induction of rapid sporulation is describedand used to characterize the physiology of spor-ulation with respect to nutritional requirements,enzyme behavior, and time course of differentia-tion.

METHODSOrganisms and growth conditions. Table 1 lists the

strains used in this study. The strains were maintainedon slants of medium A composed of casein hydroly-sate (1 g/liter), yeast extract (5 g/liter), NaCl (5 g/liter),

t Present address: Department of Biochemistry, Universityof Wisconsin, Madison, WI 53706.

glucose (5 g/liter), and agar (15 g/liter), buffered with10 mM 3-(N-morpholino)-propanesulfonic acid(MOPS; Sigma Chemical Co., St. Louis, Mo.)-KOH,pH 7.2. The solid defined minimal medium B con-tained ferric citrate (0.05 g/liter), 1 mM Na-K2-PO4(pH 7.2), 2 mM MgCl2, 0.03 mM CaCl2, 0.5 mMK2SO4, 30 mM MOPS-KOH (pH 7.2), 10 mM glucose,1 mM L-asparagine, agar (15 g/liter), and trace salts (1ml/liter). Medium C, the basal medium that was usedfor liquid cultures, contained 0.1 mM FeCl3 (preparedat 10 mM in 50 mM sodium nitrilotriacetate, pH 7), 0.5mM K2SO4, 2 mM MgCl2, 0.03 mM CaCl2, and tracesalts (1 ml/liter). In most experiments, medium C wasbuffered with 50 mM Na-K2-PO4, pH 7.2. In media inwhich the phosphate concentration was at or below 1mM, 50 mM MOPS-KOH (pH 7.2) or 50 mM N-tris(hydroxymethyl)methyl-2-aminoethanesulfonicacid (TES; Sigma)-KOH (pH 7.2) was used as thebuffer. Trace salts contained 0.6 mM ZnCl2, 0.12 mMCuCl2, 0.10 mM MnCl2, 0.05 mM Na2B407, and 0.016mM (NH4)6Mo7024. Filter-sterilized carbon and nitro-gen sources were added to autoclaved basal media. A100-fold-concentrated solution of MgCl2 and CaCl2was added to the remainder of the medium afterseparate autoclaving. All liquid cultures contained acoiled spring and 5% polyethylene glycol (molecularweight, 6,000; added as a 10% [wt/vol] solution aftergermination had occurred) to improve the dispersionof mycelia (10).When auxotrophs were grown, serine was provided

at 4 mM, and tryptophan and nicotinamide wereprovided at 0.1 mM, as required. Thiostrepton (E. R.Squibb & Sons, Princeton, N.J.) was included at 1 ,ug/ml, rifampin (Sigma) was included at 10 ,ug/ml, andneomycin sulfate (Pfizer, Inc., New York, N.Y.) wasincluded at 20 ,ug/ml in cultures of the appropriatedrug-resistant mutants. The supplements were filtersterilized as concentrated stock solutions.

In most cases, liquid cultures were inoculated withspore suspensions harvested from medium B by themethod of Hirsch and Ensign (7), except that the

357

on May 26, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

358 KENDRICK AND ENSIGN

spores were suspended and stored in water. Thespores were activated by heating in 50 mM Trizmabase (Sigma; pH 9.5) at 45°C for 15 min. The activatedspores were collected by centrifugation at room tem-perature, suspended in water, and used as inoculum ata final concentration of 5 x 106 spores per ml. As aresult of this treatment, more than 90% of the sporesformed germ tubes by 5 h in complex medium or by 8 hin glucose-ammonia minimal medium. For those ex-periments involving nonsporulating strains, mycelialfragments served as the inoculum. Solid media wereinoculated with either spores or mycelia. In mostexperiments, the cultures were incubated at 33°C. Thecultures that were being tested for the production ofstreptomycin were incubated at 26°C, since in ourhands this antibiotic was not produced at temperaturesin excess of 28°C. Liquid cultures were shaken at 250rpm in gyratory shakers.Comparison of spore properties. The organisms were

grown in medium C containing 20 mM glucose, 1 mMNH4Cl, 50 mM phosphate buffer, and casein hydroly-sate (vitamin and salt free [1 g/liter]; ICN-NutritionalBiochemicals, Cleveland), designated mediumC+CAA, or on medium C+CAA solidified with agar(15 g/liter). The spores were harvested after incubationat 33°C for 7 days. Aerial spores (from solid cultures)were harvested by the glass bead technique (7); sub-merged spores (from liquid cultures) were harvestedby centrifugation. Microscopic observation revealedthat the liquid culture consisted of more than 95%spores; therefore, no precautions were needed toseparate the spores from the mycelia. These sporepreparations were washed twice in 1 M KCI to removeexocellular proteases (10), followed by four washes indistilled water. The suspensions were used immediate-ly for physiological measurements; samples were ly-ophilized for electron microscopy.Mycelia were obtained by harvesting cultures grow-

ing exponentially in medium C+CAA. All steps werecarried out at room temperature, because the mycelialysed when held at 4°C for even short lengths of time.The culture was washed twice in 50 mM K-PO4 (pH7.2) and used immediately.Germination was measured by inoculating a heat-

activated aqueous spore suspension in 50 mM K-PO4buffer (pH 7.2) containing 1.0 g of yeast extract perliter; after 4 h, the percentage of spores that hadformed germ tubes was counted by examination with aphase-contrast microscope.Endogenous respiration was measured by adding

concentrated spores or mycelia to 3.0 ml of 50 mM K-P04 buffer (pH 7.2), equilibrated to 33°C. An oxygenelectrode (Yellow Springs Instruments, YellowSprings, Ohio), was used to measure oxygen uptake.The results are given as the oxygen quotient (microli-ters of 02 per hour per milligram [dry weight]).Trehalose was determined in aqueous extracts of

spores or mycelia prepared by boiling for 30 min. To0.3 ml of extraction fluid was added 0.6 ml of 100 mMimidazole-hydrochloride (pH 7.3) and 0.1 ml of a crudepreparation of trehalase from an extract of a soilbacterium that had been isolated by enrichment cul-ture. This amount of trehalase was sufficient to effectcomplete hydrolysis of 1 mM trehalose under thereaction conditions used and did not hydrolyze su-crose, cellobiose, glycogen, lactose, or maltose. Themixture was incubated for 1 h at 36°C, and the reaction

was stopped by boiling for 3 min. One half of thisreaction mixture was combined with 0.5 ml of Stat-zyme reagent (Worthington Diagnostics, Freehold,N.J.; prepared at twice the concentration suggested bythe manufacturer), and the absorbance was read at 500nm.

Sonic resistance was assayed by subjecting sporesor mycelia to 10 min of sonication, at 50% pulse, at thehighest output that did not cause cavitation (BransonSonifier, model 350). The spores were held at 4°Cthroughout the procedure, whereas the mycelia werekept at 20°C. Samples were removed, diluted, andplated on medium A for viable counts. Lysozymeresistance was assayed by measuring the percentsurvivors after incubation at 37°C for 1 h in 20 Fxg oflysozyme per ml in 20 mM TES buffer, pH 7.2.For transmission electron microscopy, spore sam-

ples were fixed in 2.5% (vol/vol) glutaraldehyde-7 mMsodium cacodylate (pH 7.3) for 1 h at room tempera-ture. The spores were homogeneously suspended in 15g of Tonagar (Oxoid Ltd., London, England) per liter,and small agar cubes were postfixed in 0.85% (wt/vol)osmium tetroxide in 7 mM sodium cacodylate, pH 7.3.Fixed spores were dehydrated in ethanol and embed-ded in resin. Thin sections were stained in uranylacetate-lead citrate and viewed with a Hitachi HUllEmicroscope.Two-dimensional polyacrylamide gel electrophore-

sis was carried out by the method of Roberts et al.(18). A spore suspension in lysis buffer (18) wascombined with 5-,um-diameter glass beads (Ultrason-ics, Plainview, N.Y.), at a ratio of 2 ml of spores per gof beads, in a dental amalgamator capsule (Caulk Vari-Mei II-M; L. D. Caulk Co., Milford, Del.). The chilledcapsule was shaken at 4°C in the amalgamator for atotal of 6 min, using 30-s bursts of the highest powerseparated by 30 s of cooling. Protein analysis (6)indicated that 87% of the total soluble spore proteinwas released by this method.

Quantification of sporulation. Because spores, butnot mycelia, were resistant to sonic disruption, wemeasured sporulation efficiency (SE) as (colony-form-ing units per milliliter after 4 min of sonication)/(colony-forming units per milliliter in the absence ofsonication). This value represents the fraction of theviable population that is spores. For some experi-ments, the sonic-resistant units per milliliter, the nu-merator of the above equation, are reported below.Enzyme assays. Samples of culture were harvested

by centrifugation and washed twice in 50 mM imidaz-ole-hydrochloride, pH 7.3. The cells were suspendedin 0.05 times the sample volume of imidazole bufferand disrupted with the dental amalgamator as de-scribed above. Cell debris was removed by centrifuga-tion at 4°C, 15,000 x g, for 20 min. The supematantfluid was used as the extract for all of the assaysexcept the protease, for which the culture fluid wasused. Specific activities are expressed as micromoles(substrate converted to product) per minute per milli-gram of protein.

Glucose-6-phosphate dehydrogenase was assayedby combining 0.1 ml of the extract with 100 mMimidazole-hydrochloride (pH 7.3) containing 2 mMglucose 6-phosphate (Sigma), 0.13 mM NADP (Sig-ma), and 1 mM MgCI2 in a final volume of 3 ml. Themixture was incubated at 33°C for 1 h. The change inabsorbance at 340 nm was compared with the extinc-

J. BACTERIOL.

on May 26, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

S. GRISEUS SPORULATION 359

tion coefficient for NADPH of 6,220 liters mol-' cm-'(20).

Alkaline phosphatase activity was assayed with thesubstrate p-nitrophenylphosphate (Sigma). In a finalvolume of 3 ml, 100 mM Tris-hydrochloride (pH 8.5), 1mM MgCl2, and 2 mM p-nitrophenylphosphate weremixed with 20 to 100 ,ul of cell-free extract. Thereaction was monitored spectrophotometrically at 420nm, at which wavelength the extinction coefficient ofp-nitrophenol is 13,200 liters mol-[ cm-' (16).We determined glutamine synthetase by the method

of Kustu and McKereghan (13), using 20 to 50 pL. ofcell-free extract and incubating the reaction at 33°C.Enzyme activity was measured at pH 7.3 in thepresence of manganese.

Extracellular protease was measured by the methodof Ginther (4), with Azocasein (Sigma) as the sub-strate.

Nutrient starvation. Medium C containing 20 mMglucose, 20 mM NH4Cl, 50 mM Na-K2-PO4 (pH 7.2),and 10 g of Bacto-Peptone (Difco Laboratories, De-troit, Mich.) per liter was inoculated with spores ormycelial fragments. When the culture had grown to150 to 200 Klett units (Klett-Summerson photocolori-meter, filter no. 42), the mycelia were harvested bycentrifugation. All steps were performed at roomtemperature. The pellet was washed once with starva-tion medium identical to that described above butlacking peptone and either glucose, ammonia, or phos-phate. Medium lacking phosphate was buffered witheither 50 mM MOPS-KOH (pH 7.2) or 50 mM TES-KOH (pH 7.2). The washed cells were inoculated, to aKlett value of approximately 70, into starvation medi-um pre-equilibrated to 33°C. This time of transfer wastaken as 0 h.Ammonia was measured as described by Kendrick

and Wheelis (10). Glucose was assayed with the Wor-thington Statzyme reagent. Inorganic phosphate wasmeasured colorimetrically, by the procedure of Leloirand Cardini (15). In these assays, MOPS buffer wassubstituted for TES in the medium because the latterinterfered with the ammonia assay, and water wasused instead of polyethylene glycol, which interferedwith the inorganic phosphate reaction. Neither ofthese substitutions had any effect on the time course ofsporulation.

Mutagenesis. All steps were carried out in the dark.A suspension of spores in water was irradiated withmixing under a germicidal lamp having a medianoutput at 260 nm. The spores were irradiated to asurvival of 1%, diluted, plated on medium A, andincubated at room temperature in the dark for 7 days.Auxotrophs were detected by their inability to growwhen replicated to medium B. Strain SKK817 wasdetected by the absence of white, powdery growth onmedium A. All of the drug-resistant mutants arosespontaneously.

Streptomycin bioassay and cross-feeding experi-ments. Streptomyces viridochromogenes NRRL B-1511 and a spontaneously derived streptomycin-resist-ant mutant (resistant to 100 p.g of streptomycin sulfateper ml) were the indicator organisms for the strepto-mycin bioassay by the agar overlay technique. Strep-tomycin assay bottom agar contained 15 g of beefextract per liter, 3 g of yeast extract per liter, 6 g ofpeptone per liter, and 15 g of agar per liter, adjusted to

pH 7.9. Top agar included streptomycin assay bottomagar and 7 g of agar per liter. To measure production ofthe antibiotic by the test organism grown in the liquidculture, we overlaid the plates with top agar containingspores of the indicator strain and filled 6-mm wellswith the culture filtrate; to assay drug production bytest organisms grown on agar surfaces, we placed agarculture plugs of the test producer organism on thesurface of the basal agar and overlaid the seeded topagar. The zones of inhibition were measured afterovernight incubation at 33°C.

Sensitivity to streptomycin was assayed by the well-overlay method described above. Spores or mycelialfragments of the test organism were included in the topagar, and dilutions of streptomycin sulfate were addedto the wells.A similar method was used to assay for the excre-

tion of a substance that stimulated sporulation. Agarplugs, removed at daily intervals from a confluentculture of strain B-2682 growing on medium B, wereplaced on the surface of a fresh medium B plate. Theplate was overlaid with mycelial fragments of strainSKK817 suspended in 7 g of agar per liter. The growthof strain SKK817 in zones surrounding each plug wasobserved during incubation at 33°C for 6 days.

RESULTS

Comparison of aerial and submerged spores.We initially observed that S. griseus B-2682sporulated after growth in a variety of media.Phase-contrast photomicrographs (Fig. 1) of theorganism growing in medium C+CAA show thata 24-h culture consisted of vegetative mycelia.The first indication of sporulation, characterizedby swollen mycelial termini, occurred at approx-imately 36 h. Free spores were microscopicallyvisible 45 h after inoculation. By 72 h, the entireculture consisted of spores.The spores produced in the liquid culture

(submerged spores) were compared with thoseproduced on the solid medium (aerial spores).Their properties are listed in Table 2. In severalrespects, submerged spores resembled aerialspores and were distinct from vegetative myce-lia. Although each type of spore was resistant tosonic disruption, submerged spores were as sen-sitive to lysozyme treatment as were mycelia,whereas aerial spores were resistant.To explore the possibility that this major

difference between submerged and aerial sporeswas due to the sheath (3), we examined bothspore types by electron microscopy. An exami-nation of thin sections by transmission electronmicroscopy indicated that the sheath was rarelypresent on aerial spores, whereas no ensheathedsubmerged spore specimens were seen. Thesubmerged spore wall averaged 22 nm in thick-ness, intermediate in size between the aerialspore wall (37 nm) and that of a mycelium (18nm). No other major differences between themicrographs of submerged spores and aerialspores were found (Fig. 2).

VOL. 155, 1983

on May 26, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

360 KENDRICK AND ENSIGN

TABLE 1. List of S. griseus strains usedStrain Phenotype Source

B-2682 Wild type NRRLaSKK809 Nic- B-2682, FUVbSKK813 Trp- B-2682, FUVSKK815 Ser- B-2682, FUVSKK817 Bld-c B-2682, FUVSKK820 Thsrd B-2682, spontaneous

a NRRL, Northern Regional Research Laboratory.b FUV refers to mutagenesis by UV light at the

median wavelength of 260 nm.c Bld- is characterized by the absence of aerial

mycelia and spores.d Thsr refers to thiostrepton resistance.

We also compared the two spore types withrespect to total spore protein, using two-dimen-sional polyacrylamide gel electrophoresis. Most

of the proteins from the aerial-spore sample hadcounterparts in the submerged-spore sample;however, some proteins were unique to aerialspores.

Figure 3 indicates that both spore typesformed germ tubes. In both cases, germinationoccurred more rapidly after heat activation at45°C for 15 min.

Nutritional aspects of sporulation. Our initialnutritional studies indicated that sporulationof strain B-2682 occurred after growth in allmedia containing single carbon and nitrogensources. Sporulation did not occur in culturesgrown in a rich medium, such as phosphate-buffered medium C plus 10 g of peptone per liter(Table 3). We therefore investigated the timecourse of the life cycle of strain B-2682 growingin the defined medium with single carbon andnitrogen sources (Fig. 4). Under conditions inwhich the initial concentration of glucose was

I I

'at

-1..i

. I

. IJ

l,-'

I..

1. ,.L

-.

A-AI '

*C.

.

.r I

v1 . .

DFIG. 1. Phase-contrast photomicrographs of the life cycle of S. griseus B-2682 growing in medium C+CAA.

(A) 24 h; (B) 36 h; (C) 45 h; (D) 72 h. The bars represent 20 ,um.

J. BACTERIOL.

I

i

tt t

,

on May 26, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

S. GRISEUS SPORULATION 361

TABLE 2. Properties of submerged and aerial spores and mycelia of S. griseus

S. griseus Trehalose Endogenous Germina- Sonic Lysozymestructure (% dry wt) respiration tionb trencs- resistanced

SporesSubmerged 2.7 (0.0)e 9.5 (2.5) 70 75 2.0 (0.85)Aerial 10.2 (4.5) 4.5 (0.4) 90 100 70

Mycelia <0.08 83.6 (13.5) NAf <0.03 1.3a Oxygen quotient (microliters of 02 per hour per milligram [dry weight]) in 50 mM K-PO4, pH 7.2.b Percent germ tube formation after heat activation at 45°C for 15 min and subsequent incubation for 4 h at

32°C in yeast extract (1 g/liter)-50 mM K-PO4, pH 7.2.Percent survivors of 10 min of sonication.

d Percent survivors of 1 h of incubation at 37°C in the presence of 20 p.g of lysozyme per ml.e The values in parentheses indicate the standard deviations.f NA, Not applicable.

high, and that of ammonia was low (Fig. 4A),growth proceeded until 30 h, when all of theammonia was utilized. At this point, the numberof spores began to increase, reaching a maxi-mum at 50 h. The glucose concentration de-creased slowly during the incubation period. Inanother experiment, in which the initial concen-tration of glucose was low, and that of ammoniawas high, an increase in spore concentration wasnot detected until after growth had stopped (Fig.4B). The pH of both cultures remained between7.2 and 7.6. No extracellular protease activitywas detected, nor was any streptomycin evidentin the culture supernatant under these experi-mental conditions (data not shown).The preceding experiments suggested that ei-

ther carbon or nitrogen limitation results in thesporulation of S. griseus. A major disadvantageto this study was that the point of initiation of

sporulation could not be ascertained. We there-fore turned to nutritional downshift studies todetermine (i) whether sporulation could be in-duced in streptomycetes by incubation understarvation conditions and (ii) the time course ofsporulation.The morphological changes associated with

sporulation after ammonia or phosphate deple-tion closely paralleled those observed duringsporulation of the organism in complete medium(Fig. 1). A phosphate-depleted culture existed astypical vegetative mycelia through the first 4 hafter downshift. At 4 h, the mycelial tips wereslightly swollen. This swelling was accentuatedthrough 8 h, when the swollen zones wereelongated and bulging. By 12 h, free spores wereevident, as were spore chains. The entire proc-ess was similar, but less pronounced, and wasdelayed for approximately 4 h in the ammonia-

r

IA

FIG. 2. Transmission electron micrographs of submerged and aerial spores harvested after 7 days of growthon medium C+CAA. (A) Aerial spores; (B) submerged spores. The bars represent 0.5 j±m.

VOL. 155, 1983

, .....

on May 26, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

362 KENDRICK AND ENSIGN

.9

FIG. 3. Phase-contrast photomicrographs of germinated spores after 4 h of incubation in 50 mM K-PO4 (pH7.2) containing 1 g of yeast extract per liter. (A) Aerial spores; (B) submerged spores. The bars represent 10 ,um.

depleted culture. These results are depictedgraphically in Fig. 5. After ammonia depletion,the culture underwent a slight increase in turbid-ity before the appearance of free spores. Theincrease in sonic resistance preceded the micro-scopic appearance of spores by approximately 2h. The ammonia concentration in the culturefluid was less than 1 ,uM throughout the experi-ment, whereas the phosphate concentration re-mained at approximately 35 mM. The glucose

TABLE 3. SEs of S. griseus strains after nutritionaldownshift

Strain Limiting Addition SEnutrienta

B-2682 None Peptoneb 0.02 (30)CNH3 d 2.7 (34)Pi 2.3 (26)

SKK813 Pi Trp 1.3 (24)Pi 0.01 (24)

SKK815 Pi Ser 0.87 (20)Pi 4.8 (20)

SKK817 NH3 0.07 (54)SKK820 NH3 Thiostrepton 1.2 (30)

Pi Thiostrepton 1.9 (30)

a NH3 indicates ammonia starvation; Pi indicatesphosphate starvation.

b Ten grams per liter.c The numbers in parentheses indicate the time

(hours) of the determination of the SE; the SEs did notchange significantly after 20 h of phosphate starvationor after 30 h of ammonia starvation.

d , None.

concentration fell gradually, reaching 15 mM by32 h.The phosphate-depleted culture (Fig. 5)

showed a dramatic increase in turbidity. This

5

4 - * 8 20

A

2 /KL~~ETT A 1

0 20 406

B

TIME ( HR )

FIG. 4. Graphic account of the life cycle of S.griseus B-2682 in submerged culture. In (A) and (B),growth is represented by Klett units, and the sporesare represented by sonic-resistant units (SRU). Thecultures were grown in medium C containing (A) 25mM glucose and 1 mM ammonium chloride or (B) 15mM ammonium chloride and 2 mM glucose.

J. BACTERIOL.

on May 26, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

S. GRISEUS SPORULATION 363

z 43i *'

VD 3 Pj- DELT D:°. .

- 630O 20-Q 20 NHDEPLETED:,

Pi- DEPLETED:

Ar 2

A

30

z

20 +, GLUCOSE

1-

z

W 201NH,

z

GLUCOSE

10 I

0 20 40 60

TIME (HR)FIG. 5. Culture parameters after nutritional down-

shift of S. griseus B-2682. (A) Growth (Klett units) andspores (sonic-resistant units [SRU]) during either am-

monia or phosphate deprivation are shown; (B) exoge-nous glucose and inorganic phosphate concentrationsafter ammonia deprivation are given; (C) exogenousglucose and ammonia concentrations after phosphatedeprivation are given.

was due primarily to an increase in dry weight,but also to the development of refractile spores(data not shown). Again, the increase in sonicresistance slightly preceded the appearance ofspores. The culture supernatant contained lessthan 1 ,uM inorganic phosphate throughout theexperiment. The ammonia concentration de-creased to 16 mM, and the glucose concentra-tion decreased to 12 mM, by 32 h, at which pointspores constituted more than 90% of the culture.When depleted for glucose (data not shown),

at least 90% of the mycelia lysed within 20 h.The remaining mycelia did eventually sporulate.The ability of both ammonia- and phosphate-

starved cultures to sporulate suggested that asignificant amount of the sporulation-specificmetabolism utilized endogenous precursors.This hypothesis was tested by subjecting a tryp-tophan auxotroph (SKK813) to starvation condi-tions. The results (Table 3) indicate that therewas insufficient intracellular tryptophan to sup-port the sporulation of this strain. A serineauxotroph (SKK815), however, contained suffi-cient amounts of glycine or serine to effectsporulation in the absence of either amino acidin the phosphate-depleted medium (Table 3).

Commitment to sporulate. Because strain B-2682 did not sporulate in rich medium, we coulddetermine whether an induced culture reaches acritical point, beyond which the culture sporu-lates in spite of exposure to inhibitory condi-tions. We therefore subjected either phosphate-or ammonia-depleted cultures to nutrient-richconditions at various times after nutrient deple-tion. The cultures were incubated for a total of30 h, at which point the sonic-resistant units permilliliter was determined for each sample. Theresults (Table 4) indicate that ammonia-depletedcultures became committed to sonic resistanceby 9 h after downshift. The phosphate-depletedculture behaved similarly, reaching commitmentby 6 h after downshift. In both experiments,microscopic examination at 30 h indicated thatfree spores were present in all cultures thatshowed an increase in the sonic-resistant unitsper milliliter.Enzyme activities during sporulation. One of

our goals in devising a sporulation system was toidentify either a sporulation-specific enzyme or

an enzyme activity that fluctuated characteristi-cally during the transformation from a vegeta-tive to a sporulating culture. In preliminaryexperiments, we observed that the activities ofglutamine synthetase, alkaline phosphatase, andglucose-6-phosphate dehydrogenase fluctuatedat various stages of the life cycle. We thereforeassayed these enzymes in cultures subjected tonutritional downshift. Under ammonia starva-tion conditions, alkaline phosphatase activityremained at the low initial level, whereas thespecific activity of glutamine synthetase in-creased at a rate of 0.23 h-1 until reaching a

plateau at 9 h (data not shown). During phos-phate limitation (Fig. 6), alkaline phosphataseactivity did not increase until at least 12 h after

TABLE 4. Time required for commitment of S.griseus B-2682 to sporulate after ammonia or

phosphate depletion

Time (h) before Relative SRU/ml in:nutrient Ammonia-starved Phosphate-starved

replenishment culture culture

0 0.03 <0.013 NDb <0.014 0.01 ND5 0.01 ND6 ND 0.178 0.06 ND9 0.29 0.2312 0.17 0.3730 1.0 1.0

a Sonic-resistant units (SRU) at 30 h relative to thesample that was starved for 30 h.

b ND, Not determined.

VOL. 155, 1983

on May 26, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

364 KENDRICK AND ENSIGN

Cf)

A_vW_

Z- 5 -0o2 0 4

urn ~~~~~(1)z41

Z Ill/KLETT GLUTAMINE 01.YNTHETASE a0.

ALKALINE

POSPHATASE _

CD

-1 0 5 10 15 20 25

TIME (HR)FIG. 6. Specific activities of glutamine synthetase

and alkaline phosphatase after phosphate deprivationof strain B-2682. The open symbols indicate the Klettunits, glutamine synthetase activity, and alkalinephosphatase activity of the exponentially growingstarter culture 1 h before phosphate starvation. Thearrow indicates the first microscopic observation offree spores.

downshift. Under these same conditions, gluta-mine synthetase activity displayed an initial rap-id increase, followed by an intermediate de-crease. By 30 h, glutamine synthetase activityhad regained its high level. No glucose-6-phos-phate dehydrogenase activity could be detectedduring either starvation period, although thisenzyme was active in cultures growing exponen-tially with glucose as the sole source of carbon.Mutant studies. Several thiostrepton-, rifam-

pin-, or neomycin-resistant spontaneous mu-tants were isolated. None of these mutantsshowed any defect in sporulation on solid media,in the presence or absence of the drug, and athigh (37°C) or low (28°C) temperature. Thethiostrepton-resistant strain SKK820 was unim-paired in its sporulation properties after nutri-tional downshift in the presence of 1 ,g of thio-strepton per ml, whether ammonia or phosphatedepleted (Table 3).Most of the auxotrophs that were isolated

after UV mutagenesis (e.g., Trp- Ade-, Met-)sporulated normally. However, a strain thatrequired nicotinamide (SKK809) and one thatrequired either serine or glycine (SKK815) spor-ulated poorly on medium A or on supplementedmedium B. Neither of these strains excretedstreptomycin under conditions in which thewild-type strain produced approximately 7 ,ug/ml. An antibiotic inhibitory to both streptomy-cin-resistant and streptomycin-sensitive indica-tor strains was produced by both auxotrophs.The serine auxotroph reverted to serine inde-pendence at a frequency of 3 x 10-6 and simul-taneously regained the ability to sporulate well

on medium B. Strain SKK809 reverted to nico-tinamide independence at a frequency of 2.4 x10-4; of these revertants, 3% sporulated normal-ly.One bald mutant, strain SKK817, defective in

aerial spore formation, was isolated after UVmutagenesis. This strain reverted to normalsporulation at a frequency of less than 1 x 10-9.Although strain SKK817 produced no strepto-mycin, at least one substance inhibitory to bothindicator strains was excreted. The sensitivity ofstrain SKK817 to streptomycin (inhibition bymore than 10 ,ug of streptomycin sulfate per ml)was identical to that of the wild-type strain.Strain SKK817 did not sporulate on medium Aor in medium C containing various carbon andnitrogen sources. Ammonia depletion conditionsresulted in a sporulation efficiency of less than0.07 (Table 3). The wild-type strain of S. griseusdid not excrete any substance that inducedstrain SKK817 to sporulate in either solid orliquid culture, nor did strain SKK817 produce anagent that inhibited the sporulation of strain B-2682.

DISCUSSIONAn analysis of a variety of spore properties

indicates that the submerged spores of S. griseusB-2682 are similar, but not identical, to aerialspores. Both submerged and aerial spores con-tain relatively high levels of trehalose, a sugarvirtually absent from exponentially growing my-celia. The endogenous respiration rate of eachspore type was considerably lower than that ofactively respiring mycelia, consistent with theconclusion that mature streptomycete spores arephysiologically dormant (8). The apparent in-creased oxygen quotient of submerged sporesrelative to aerial spores may be accounted for bya 5% level of contamination by mycelia in thesubmerged spore preparation. The germinationof both submerged and aerial spores respondedto heat activation. The major difference betweensubmerged and aerial spores was the markedsensitivity of the former to lysozyme treatment.This sensitivity appears to be independent of thepresence of a sheath surrounding the sporechain, since we were unable to observe thesheath consistently on either spore type. Thesensitivity to lysozyme may result from thethinness of the submerged spore wall.Our initial studies of the life cycle suggested

that either nitrogen or carbon depletion mayelicit sporulation. It is clear that sporulationoccurs independently of streptomycin excretion,extracellular protease activity, and changes inthe pH of the medium. The nutritional downshiftstudies indicate that nitrogen or phosphate de-privation is an effective nutritional trigger ofsporulation. Phosphate depletion studies suggest

J. BACTERIOL.

on May 26, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

S. GRISEUS SPORULATION 365

that vegetative mycelia possess a large pool ofphosphate available to the sporulating organism.Although we have been unable to identify a

sporulation-specific enzyme the appearance ofwhose activity signals the initiation of sporula-tion, the reproducible fluctuation in glutaminesynthetase activity suggests that this enzymemay be useful as a sporulation marker underphosphate depletion conditions. The rapid in-crease in glutamine synthetase activity afterphosphate deprivation is striking in light of theobservation that glutamine synthetase fromStreptomyces cattleya is adenylylated and,thereby, inactivated in response to a high con-centration of ammonia (21). The continued lowactivity of alkaline phosphatase in the absenceof exogenous phosphate supports the hypothesisthat mycelia contain large pools of inorganicphosphate; these pools apparently must be de-pleted before alkaline phosphatase activity canincrease.Based on results obtained with Bacillus

strains (12, 14), we predicted that drugs thatinterfere with transcription or translation instreptomycetes may select for resistant mutantsthat concomitantly have acquired altered sporu-lation properties. Although our search for suchmutants was not extensive, our initial results donot substantiate this hypothesis. Chater similar-ly reported that several rifampin-resistant mu-tants of Streptomyces coelicolor did not simulta-neously become defective in differentiation (2).Among the auxotrophic mutants isolated were

two that were oligosporogenous, sporulatingpoorly on media that ordinarily supported abun-dant sporulation. Reversion analysis of the ser-ine auxotroph indicated that a single, revertiblemutation was responsible for both the serinerequirement and the oligosporogeny; likewise, asingle mutation appears to have resulted in bothnicotinamide auxotrophy and poor sporulationof strain SKK809. That strain SKK815 doessporulate when subjected to phosphate starva-tion indicates that this strain is a conditionallysporulating mutant. Although data mentionedearlier suggest that streptomycin excretion is notobligatory for sporulation, the fact that neitherthe serine auxotroph nor the nicotinamide auxo-troph produced extracellular streptomycin sug-gests that the biosynthesis of the antibiotic maybe indirectly correlated with the ability to formspores, as other investigators have suggested (4,19). This hypothesis is also supported by theinability of strain SKK817 to produce strepto-mycin. In all cases examined, at least one bioac-tive substance was produced by strains that didnot sporulate, indicating that the correlationbetween sporulation and antibiotic excretionmay be specific for streptomycin.

Unlike strain SKK815, which can sporulate

under certain culture conditions, strain SKK817appears to be asporogenous under all condi-tions. Phase-contrast microscopy revealed nospores when strain SKK817 was subjected tonutritional downshift. We presume that the mea-sured SE (Table 3) reflects a limited number ofmycelial fragments that escaped sonic disrup-tion. Our inability to detect normal sporulationrevertants of strain SKK817 suggests that adeletion or gene rearrangement is responsiblefor the pleiotropic phenotype of this strain. Bir6and colleagues (1) and Khokhlov et al. (11) havereported the production, by sporulating strainsof S. griseus, of two potent factors, each ofwhich promotes the sporulation of certain baldmutants. Our results indicate that strainSKK817 does not sporulate in response to expo-sure to culture fluid from the wild-type strain.Parallel experiments show that spent culturemedium from strain SKK817 does not inhibit thesporulation of strain B-2682.The results in this paper demonstrate that S.

griseus B-2682 can sporulate abundantly in aliquid culture under appropriate nutritional andenvironmental conditions. We have observedthat other strains of S. griseus can also sporulatein a submerged culture and that S. viridochro-mogenes can be induced to sporulate by nitro-gen starvation in a liquid medium (R. Koepseland J. C. Ensign, manuscript in preparation).After nitrogen or phosphate deprivation, S. gri-seus rapidly sporulates. Additionally, C. Hirschand L. Koupal have observed that an unknownstreptomycete soil isolate sporulates well in adefined, phosphate-limited medium (personalcommunication). Streptomycete sporulationmay, therefore, be a feasible model system forthe study of the physiological and molecularevents involved in differentiation, events thatwere previously difficult to study with culturesgrowing on solid media.

ACKNOWLEDGMENTS

This work was supported by NRSA postdoctoral traininggrant F32GM07405 to K.E.K. from the National Institutes ofHealth.We thank Martin Garment for the electron microscopy.

LITERATURE CITED

1. Bir6, S., I. Bekesi, S. Vital1is, and G. Szab6. 1980. Asubstance effecting differentiation in Streptomy'ces gri-seus. Purification and properties. Eur. J. Biochem.103:359-363.

2. Chater, K. F. 1974. Rifampicin-resistant mutants of Strep-tomyces coelicolor A3(2). J. Gen. Microbiol. 80:277-290.

3. Ensign, J. C. 1978. Formation, properties, and germina-tion of actinomycete spores. Annu. Rev. Microbiol.32:185-219.

4. Ginther, C. L. 1979. Sporulation and the production ofserine protease and cephamycin C by Streptomvces lac-tamdurans. Antimicrob. Agents Chemother. 15:522-526.

5. Hardisson, C., and M. B. Manzanal. 1976. Ultrastructuralstudies of sporulation in Streptomvces. J. Bacteriol.127:1443-1454.

VOL. 155, 1983

on May 26, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

366 KENDRICK AND ENSIGN

6. Herbert, D., P. J. Phipps, and R. E. Strange. 1971. Chemi-cal analysis of microbial cells, p. 209-344. In J. R. Norrisand D. W. Ribbons (ed.), Methods in microbiology, vol.SB. Academic Press, Inc., London.

7. Hirsch, C. F., and J. C. Ensign. 1976. Nutritionally de-fined conditions for germination of Streptomyces viri-dochromogenes spores. J. Bacteriol. 126:13-23.

8. Hirsch, C. F., and J. C. Ensign. 1978. Some properties ofStreptomyces virodochromogenes spores. J. Bacteriol.134:1056-1063.

9. Hopwood, D. A., H. Wildermuth, and H. M. Palmer.1970. Mutants of Streptomyces coelicolor defective insporulation. J. Gen. Microbiol. 61:397-408.

10. Kendrick, K. E., and M. L. Wheelis. 1982. Histidine dis-similation in Streptomyces coelicolor. J. Gen. Microbiol.128:2029-2040.

11. Khokhlov, A. S., L. N. Anisova, I. I. Tovarova, E. M.Kleiner, I. V. Kovalenko, 0. I. Krasilnikova, E. Ya. Kor-nitskaya, and S. A. Pliner. 1973. Effect of A-factor on thegrowth of asporogenous mutants of Streptomyces griseus,not producing this factor. Z. AlIg. Mikrobiol. 13:647-655.

12. Kobayashi, H., K. Kobayashi, and Y. Kobayashi. 1977.Isolation and characterization of fusidic acid-resistant,sporulation-defective mutants of Bacillus subtilis. J. Bac-teriol. 132:262-269.

13. Kustu, S. G., and K. McKereghan. 1975. Mutations affect-ing glutamine synthetase activity in Salmonella typhimur-ium. J. Bacteriol. 122:1006-1016.

14. Leighton, T. 1974. Sporulation-specific translational dis-crimination in Bacillus subtilis. J. Mol. Biol. 86:855-863.

15. Leloir, L. F., and C. E. Cardini. 1957. Characterization ofphosphorus compounds by acid lability. Methods Enzy-mol. 3:840-850.

16. Malamy, M., and B. L. Horecker. 1966. Alkaline phospha-tase (crystalline). Methods Enzymol. 9:639-642.

17. McVittie, A. 1974. Ultrastructural studies on sporulationin wild-type and white colony mutants of Streptomycescoelicolor. J. Gen. Microbiol. 81:291-302.

18. Roberts, G. P., T. MacNeil, D. MacNeil, and W. J. Brill.1978. Regulation and characterization of protein productscoded by the nif (nitrogen fixation) genes of Klebsiellapneumoniae. J. Bacteriol. 136:267-279.

19. Schrempf, H., and W. Goebel. 1979. Functions of plasmidgenes in Streptomyces reticuli, p. 259-268. In K. N.Timmis and A. Puhler (ed.), Plasmids of medical, environ-mental, and commercial importance. Elsevier/North-Hol-land Publishing Co., Amsterdam.

20. Segel, I. H. 1968. Biochemical calculations, p. 267. JohnWiley & Sons, Inc., New York.

21. Streicher, S. L., and B. Tyler. 1981. Regulation of gluta-mine synthetase activity by adenylylation in the gram-positive bacterium Streptomyces cattleya. Proc. Nat].Acad. Sci. U.S.A. 78:229-233.

22. Wildermuth, H. 1970. Development and organization ofthe aerial mycelium in Streptomyces coelicolor. J. Gen.Microbiol. 60:43-50.

J. BACTERIOL.

on May 26, 2020 by guest

http://jb.asm.org/

Dow

nloaded from