JOURNAL OF BACTERIOLOGY, Mar. 1971, p. 1119-1136Copyright © 1971 American Society for Microbiology

Vol. 105, No. 3Printed in U.S.A.

Formation of Protoplasts from Resting SporesPHILIP C. FITZ-JAMES

Health Sciences Centre, University of Western Ontario, London, Canada

Received for Publication 21 November 1970

Coat-stripped spores suspended in hypertonic solutions and supplied with twoessential cations can be converted into viable protoplasts by lysozyme digestion ofboth cortex and germ cell wall. Calcium ions are necessary to prevent membranerupture, and magnesium ions are necessary for changes indicative of hydration ofthe core, particularily the nuclear mass. Since remnant spore coat covered such pro-toplasts of Bacillus subtilis and the germ cell wall of B. cereus spores is not lyso-zyme digestible, coatless spores of B. megaterium KM were more useful for thesestudies. Lysozyme digestion in cation-free environment produced a peculiar semi-refractile spore core free of a cortex but prone to rapid hydration and lytic changeson the addition of cations. Strontium could replace Ca2+ but Mn2+ could not re-place Mg2+ in these digestions. When added to the spores, dipicolinic acid and otherchelates appeared to compete with the membrane for the calcium needed for stabili-zation during lysozyme conversion to protoplasts. It is argued that calcium couldfunction to stabilize the inner membrane anionic groups over the anhydrous dipico-linic acid-containing core of resting spores.

The observation that spores stripped of theircoats by selective extraction were still viable andheat resistant (1) suggested a method for theformation of resting spore protoplasts. If- thecortex exposed by the coat removal and its innerlayer, the germ cell wall, are both lysozyme di-gestible, such spores suspended in a suitablebuffer should readily be freed from these re-maining layers and the core should be liberatedas a protoplast. This paper describes a number ofstudies with coatless or partly coatless spores ofBacillus megaterium, B. subtilis, and B. cereusundergoing lysozyme treatment in various ionicand osmotic environments. The results indicate arequirement for calcium during the initial stabili-zation of spore protoplast membrane and formagnesium during the uniform conversion of therefractile core into a normal appearing proto-plast.

MATERIALS AND METHODS

The cultures used were B. subtilis 168, B. cereusA(-), a variant of B. thuringiensis var. alesti which haslost the ability to form a parasporal crystal, and B.megaterium KM (Sp+), a sporeforming variant thathas, in three separate instances over the past 12 years,been isolated from aged plate cultures of the asporo-genous B. megaterium KM described by Northrup (16).Its spores, unlike those of other B. megaterium strains,are covered with only a single coat layer (3).

Preparation of spores. Spores were harvested fromeither a fluid or agar form of a BBM-Grelet salts me-dium (1). The agar form was more suitable for largerbatches.

Spores from well lysed cultures were cleaned by re-peated centrifugation, first in 1 M sodium chloride solu-tion and then in 0.14 M NaCl. Occasionally, lysozyme(50 to 100 mg/ml) and deoxyribonuclease (2 to 5 Ag/ml)were used in early washings of B. subtilis and B. mega-terium harvests to get rid of persistent cell debris. Cleanspores could usually be suspended in either distilledwater or 0.14 M NaCl without aggregation.Removal of spore coats. Spore coats were extracted

by the method described by Aronson and Fitz-James (1).In a typical extraction, 109 clean spores in a Corex(Corning) 15-ml centrifuge tube were sedimented andsuspended in 2.0 ml of freshly dissolved 50 to 100 mmdithiothreitol (Cleland's reagent, DTT) plus 0.5% so-dium lauryl sulfate (SLS) in 0.1 M NaCl. The pH levelof the suspension was raised slowly with 2.0 N NaOHand a glass probe combination electrode, to pH 9.5,10.4, or 11.5, depending on the solubility of the coatlayer being extracted. After incubation at room temper-ature or 37 C for 2 to 18 hr, the spore suspension wassedimented and suspended in 1.0 to 2.0 ml of fresh re-agent at the appropriate pH. The suspension wasusually washed four additional times in 0.14 M NaCl.However, since coatless spores tended to aggregate indilute salt or distilled water, saline (0.14 M NaCl) orsucrose (0.3 M) washes at pH 9.0 were often necessaryto reduce clumping. The degree of coat removal wasbest determined by thin-section electron microscopy.Spores with coats partly or completely removed showan increase in density in renografin gradients (Aronsonand Fitz-James, unpublished data), a sensitivity to lyso-zyme but no loss of dipicolinic acid (DPA) or of Ca2 .

Lysozyme digestion of coatless spores. Spores par-tially or completely stripped of their spore coats weresubjected to lysozyme digestion (usually 50 to 100,ug/ml) in a variety of fluids: distilled water, saline (0.14

1119

FITZ-JAMES

t s.

.. -L'I ..,_

,f1.,'

k..

alkd%La

.t

w a

4V42b.';

2a'

'' .* ,_b

-. -T,... 3b

I

4k

Itc.

AA 'A' I

6.,. 0'̂

.~ ~ ~ ~~~~~.

6a

I'm.._,>_.AR-w_

eq49

4b

I-.ci

.4". ..

: C..... .

106

FIG. 1-10. Phase-contrast (Fig. 1-6) photomicrographs and air-mounted, dried nigrosin smears (Fig. 7-10) orspores of B. megaterium KM. Magnification is indicated by the S gm marker in Fig. 1. Fig. 1. Untreated sporepreparation showing normal phase refractility. Coatless spores of this species were identical in appearance by phasemicroscopy (see Fig. 3). Fig. 2. Group of coatless spores treated with lysozyme in sucrose (0.3 M) without cations

1120 J. BACTERIOL.

fp f. I.t

I

11

CIA"*.FM

Add&

4w

SPORE PROTOPLASTS

M NaCI), and tricine buffer (10 mM, pH 7.6) plus su-crose (0.3 M) or disodium succinate (0.45 M). Divalentcation concentrations were achieved by the addition ofappropriate amounts of sterile I M solutions ofMgSO4, MnSO4, CaCl2, and SrCl2. In a typical incuba-tion, 0.2 ml of a thick (109/ml) suspension of coatlessspores was added to a previously warmed (37 C) mix-ture in a 15-ml Cortex centrifuge tube containing 5.0 mlof sucrose (0.5 M), 4.6 ml of tricine buffer (20 mm, pH7.6), and 100 to 300 ,liters of a 1 M solution of diva-lent cation or other agent under study. For 2 to 5 minbefore and again immediately after the addition of lyso-zyme (1 mg in 100 jliters), the optical density wasmeasured in a Coleman model 14 spectrophotometer at650 nm against an appropriate blank. Optical densityreadings (usually measured every 0.5 min) were plottedfrom the time of lysozyme addition.

Protoplast growth. Viability of protoplasts was as-sessed by following their growth with aeration in a pro-toplast medium (2). Viable protoplasts from sporesusually showed the increase in size and the rise in op-tical density (at 650 nm) typical of growing protoplastswithin 40 to 60 min.

Electron microscopy. Samples (2.0 to 5.0 ml) of sus-pension were prefixed by the addition of either a 4%solution of OSO4 to a final concentration of 0.5%, or an8% solution (pH 7.0) of glutaraldehyde (PolysciencesInc., Warrington, Pa.) to a final concentration of 2 to3%. Osmium-prefixed samples were sedimented imme-diately, whereas the glutaraldehyde-fixed samples werekept 3 to 5 hr at 2 C and then washed four to six timeswith phosphate or cacodylate buffer (0.1 M, pH 6.8)with centrifugation. Both types of cells were suspendedin melted, warm agar (2%) and immediately sedimented(3 to 5 min at 10,000 x g) while cooling. The resultingagar-embedded pellets were trimmed, diced, and placedin 1% osmium containing tryptone. Fixation dehydra-tion and embedding in Vestopal were completed by theKellenberger and Ryter method (12). Sections cut withglass knives were mounted on carbon-coated grids andstained with uranyl acetate (12) and lead citrate (20).These were examined in a Philips EM 200 at 60 kv andphotographed at x 1 1,000 to 19,000.

Light microscopy: phase contrast. Phase-contrastsmears were made by the usual method of placing asmall drop of suspension between a grease-free glasscover slip and slide. These were examined with oil im-mersion optics with Zeiss phase attachment at x 1,500.

Photographs were taken with a Leitz MA4B type ofbellows camera onto sheet film [3.25 x 4.25 inches(-8.3 x 10.8 cm)] at x 1,950 and enlarged twice whenprinted.

Nigrosin smears. Nigrosin smears were made by atechnique devised by C. F. Robinow in 1953 (personalcommunication), wherein test material is spread with adrop of 6 to 8% Nigrosin in water on a grease-freecover slip. When air-dried, the smears were placedsmear slide down on a dry slide and cemented with waxat the corners. Bright-field oil immersion optics andKdhler illumination were used, and photographs weretaken as described for phase-contrast micrographs.

RESULTSOf the three Bacillus species used, the spores of

the B. megaterium KM proved, for two reasons.most useful for these studies. First, their coatscould be readily removed. Overnight extractionof spores with Cleland's reagent plus SLS at pH10.5 (1) stripped the coat from most spores. Sec-ondly, both the cortex and germ cell wall of B.megaterium were readily digested by lysozyme.Although B. cereus spores could be stripped ofspore coats (1; Fig. 11), the germ cell wall, unlikethe cortex of this species but like the cell wall ofthe vegetative form, was lysozyme resistant (22).Hence, in this case lysozyme simply induced arapid germination (Fig. 12). This effect (in spiteof the apparently absent coats) is probably sim-ilar to that obtained by Gould and Hitchins (8)with lysozyme treatment of B. cereus sporeswhose coats, although still present, were alteredby disulfide bond-rupturing agents at low pH (7).With B. subtilis, complete removal of the sporecoats was seldom achieved with these reagents(Fig. 13). Hence, although the cell wall andcortex were both lysozyme digestible, the re-sulting ptotoplasts were usually covered with, andprotected by, a persistent layer of spore coat (Fig.14). Because of these restrictions, the bulk of thework presented here was done with the spores ofB. megaterium KM.

Phase-contrast microscopy. Phase-contrast

showing the variation in degree ofphase-refractility encountered by 5 min (a) and 10 to 20 min (b). On one spore(arrow) the cortical layer can be seen partly removed from the semirefractile core. Fig. 3. Spores of B. megateriumKM suspended in distilled water (a) before and (b) 5 to 10 min after lysozyme in distilled water was run under thecover slip. Refractility by phase microscopy is partly preserved (see also Fig. 32). Fig. 4. Water-lysozyme treatedsmear of coatless spores like those of Fig. 3b (a) before and (b) 5 to 10 min afterflushing 10 mm CaCI2 under thecover slip. Rapid loss of the partial refractility has occured. In the electron microscope these appeared like thoseshown in Fig. 17b. Fig. 5. Preparation similar to that shown in Fig. 4a, photographed 2 to 5 min after running 10mM MgSO4 under the cover slip. A rapid lysis occurs initially more complete than that achieved by the Ca2+treatment (Fig. 4b). Fig. 6. Protoplasts of resting spores of B. megaterium KMformedfrom coatless spores treatedwith lysozyme for 10 min in 0.3 M sucrose containing 10 mM MgSO4 and 20 mm CaCl2 (a) and 50 mM CaCl2 (b).Fig. 7. Control spores of B. megaterium KM showing the normal bright-field refractility. Fig. 8. Coat-extractedspores show the same high degree of refractility. Fig. 9. Coatless spores digested with lysozyme for 5 min in dis-tilled water. The refractility ofphase-contrast optics (Fig. 3b and Fig. 4a) does not pass this more critical and truetest of spore refractility. Some peripheral refractility in the cortical region survives. Fig. 10. Later period of thesame preparation in Fig. 9. Partially digested and somewhat refractile cortical pieces were found thro*ghout thesmear and partly on the spore cores (compare with Fig. 2, 4a, 30, and 32).

VOL. 105, 1971 1121

~12 _FIG. I 1. Cross section of a resting spore of B. cereus A (-) after removal ofspore coats with Cleland's reagentand SLS at pH 10.5. A thin disorganized remnant of coat covers the unaltered cortex. The core is typical of a

resting spore. Magnification is indicated in Fig. 12.FIG. 12. Coatless spores of the same batch shown in Fig. 11, fixed 12 min after the addition of lysozyme (100

1122

SPORE PROTOPLASTS

microscopy of the action of lysozyme on the coat-less spores of B. megaterium KM in a variety ofsupporting media, besides being a useful supple-ment to the optical density changes, was an es-sential prerequisite to electron microscopy.The removal of the coats from the spores of B.

megaterium produced little change in the appear-ance of the spores in phase-contrast microscopyor nitrosin smears (compare Fig. I with 3a andFig. 7 with 8). In solutions of neutral pH and lowionic strength, the clumping of coatless sporescould interfere with photomicroscopy. In theelectron microscope the removal of the spore coatis the only change on the majority of spores (Fig;15 and 16). The characteristic appearance of thespore core is unaltered, resulting in the persistentphase-contrast (Fig. 1 and 3) and nigrosin smear(Fig. 7 and 8) refractility. It is of particular in-terest that many of the coatless spores showedconsiderable damage to, and thinning of, theircortices without apparent damage to the sporecores (Fig. 16). Thus, spore coats both preventclumping of spores and protect the underlyingcortical structure from physical damage. In somepreparations, various amounts of residual coatprotein persisted on the spores; although theyappear prominent in some sections, they failed toinfluence the action of lysozyme.

Electron microscopy of lysozyme-digested coat-less spores. When coatless spores of B. mega-terium KM were treated with lysozyme in theusual buffered sucrose protoplasting mediumsupplemented with magnesium ions to stabilizethe membranes (23), a rapid loss of refractilityand concurrent fall in density occurred. Such pro-toplasting medium failed to prevent an eventuallysis and clumping (Fig. 5). The thin sections ofspores fixed 5 min after the addition of lysozymeindicated that a partial hydration of the core cy-toplasm and nuclear mass was accompanied by arupturing of the plasma membrane (Fig. 17a). By30 min the leaching and lysis were complete (Fig.17b). Repeated studies with this medium revealedthe presence of ruptured membranes by 2 min(Fig. 18).The failure of sucrose plus Mg2+ to support

spore protoplasts was initially puzzling. Suchmedia readily support the membrane integrity ofB. megaterium vegetative cell protoplasts (22).Moreover, sporulating protoplasts function wellin sucrose-Mg2+ (M. R. J. Salton, J. Gen. Micro-

biol., vol. 13, abstract iv). However, since sporesare rich in calcium, it was reasoned that forstructural integrity the membranes of restingspores, unlike those of vegetative protoplasts,which show a magnesium dependence, might ini-tially have a calcium requirement. The additionof Ca2+ (10 to 50 mM) in place of Mg2" to su-crose-buffer-protoplast medium did indeed pre-vent membrane rupture, and led to the slow for-mation of protoplasts with intact membranes, butwith nuclear bodies mostly in a dense fibrousform. Although partly altered from the semi-crystalline appearance of resting spore nuclearmaterial (Fig. 15 and 16), most of these had notyet assumed the netlike hydrated form of typicalbacterial nucleoids. Figures 19 and 21 show suchspores 5 min after the addition of lysozyme. Thelong fibrous bands of deoxyribonucleic acid(DNA) were also seen 15 and 30 min after theaddition of lysozyme, although in many cells thedense DNA material now appeared in smalleraggregates (Fig. 20). When magnesium waspresent with calcium, this dense fibrous form ofDNA was found only in the early minutes afterlysozyme addition (Fig. 22-24). The presence ofboth cations thus permitted the formation by ly-sozyme digestion of a good yield of normal ap-pearing protoplasts from resting spores (Fig. 6).

These findings prompted a fuller structuralcomparison of the effects of magnesium and cal-cium in buffered sucrose (0.3 M) and in sodiumsuccinate (0.45 M) protoplast stabilizing buffer.Two series of experiments were run in an attemptto determine whether protection of the membranecould be achieved by delayed additions of cal-cium to coatless spores undergoing lysozymedigestion in sucrose-tris(hydroxymethyl)amino-methane (Tris)-Mg2+ (or sucrose-Tris withoutmetals) supporting media. These studies indicatedthat, for full protection of membranes or goodyields of protoplasts, calcium should be presentwhen lysozyme was added. For example, the ad-dition of calcium (20 mM, final concentration) 1min after lysozyme addition led to a yield of 5 to10% healthy protoplasts. A 30-sec lag in the addi-tion of Ca2+ only produced a few protoplasts.

Optical density analysis changes of lysozyme-treated coatless spores. The optical densitychanges of coatless B. megaterium spores un-dergoing lysozyme treatment in sucrose withmagnesium and with calcium are compared in

,ug/ml) to spores suspended to sucrose. The germ cell wall resists lysozyme. Bar indicates 0.5 gim.FIG. 13. Resting spore of B. subtilis 168 extracted overnight in Cleland's reagent-SLS (pH 10.5 to 11.0). The

usual heavy multilayered spore coat has been reduced to a few persistent layers. Magnification as indicated in Fig.12.

FIG. 14. Coatless spore of B. subtilis 168 from the same lot shown in Fig. 13 after 30 min digestion with lyso-zyme in Tris-sucrose-Ca2+-Mg2+. The stable protoplast is enclosed by the persistent spore coat. Magnification asindicated in Fig. 12.

1123VOL. 105, 1971

4-:

FIG. 15. Intact resting spore of B. megaterium KM. The spore layers and core components are: spore coat (Ct),outer membrane remnant (OM), cortex (Ctx), germ cell wall (CW), plasma or inner membrane (PM) cytoplasm(cyt), and nuclear body (N). Bar is 0.1 yi.

FIG. 16. Coatless resting spores of B. megaterium KM. Coat protein is largely removed by the Cleland's re-agent plus SLS extraction. A small percentage of spores show local damage to their exposed cortices (arrow) suf-fered during manipulation in the coatless state. The spore cores are unchanged. Bar is 0.5 ,im.

1124

a S 17a

.. 4

FIG. 17-18. Lysozyme treatment of coatless spores of B. megaterium KM in sucrose-Mg2+-phosphate. Fig. 17a.After 5 min, lysozyme digestion refractility has completely disappeared, some expansion (hydration) of the cyto-plasm has occurred, and the nuclear zones have changed from the compact nonstaining stage of resting spores tothe peculiar dense masses ofcompact strands. However, without added calcium, the membranes appeared rupturedand coiled (arrow). Bar represents 0.5 fim. Fig. 17b. By 30 min, in spite of the osmotic support and adequate Mg2+,the preparation had undergone severe lysis. Magnification as indicated in Fig. 17a. Fig. 18. Segments of two adja-cent spores at higher magnification 2 min after the addition of lysozyme in sucrose-Mg2+-phosphate, showing thebroken membranous ends and the densely staining fibrous nature of early Mg2--treated nuclear material. Barmarker represents 0.1 uim.

1125

Ii

I

Ct

'I

20)*~!13~ - .:. ;I ", A- -_

FIG. 19-21. Coatless spores of B. megaterium KM treated with lysozyme in sucrose (0.3 M) plus CaCl2 (20 mM)but no added Mg2+. Bar marker represents 0.1 Am. Fig. 19 and 21. Fifteen minutes after lysozyme addition, thecortical digestion is complete, and the majority of spores appear as partially hydrated protoplasts. Fig. 20. By 30min, the conversion of the nuclear regions to typical netlike structures is delayed by the absence of added Mg2+.Various amounts of remnant coat protein (Ct) cover the spore protoplasts. Fig. 21. At higher magnification, rheclose packing of the fibrous elements in the nuclear bodies is more clearly evident.

1126

4L.., :. 4"ltrw. .. 0 4fl.l "

*%%l.A Aj,.r. 1.0.,

1"'. -.. .. L' posaw-li..

t-

I.4;

NM.. At,'P:.0. - -I'l., -1

--aL'd

II

'4

19

-99

¶tC Tcw

~~~Ct

F-1Qlkbi23 24FIG. 22-24. Coatless spores of B. megaterium KM undergoing digestion with lysozyme for 5 to 10 min in su-

crose (0.3 M) with both MgSO4 (20 mM) and CaCl2 (100 mM). Magnification is indicated by the 0.5-,um bar in Fig.22. Fig. 22. The cortex and cell wall are gone. With the exception ofsome dense masses (arrow), the cytoplasm ispartly converted to that ofa vegetative cell, although the nuclear structures are still seen as long fibrous bands. Fig.23. In many spores, the conversion of the condensed DNA into the more open net is now underway (arrows). Fig.24. In a few sections the conversion of the nuclear (N) material to net appears complete. Occasionally a mesosome(M) can be seen in the extruded state under the still digesting germ cell watt (CW) and the remnant coat (Ct).

1127

II

FITZ-JAMES

Fig. 25. In these digestions the optical densitydecline occurs in two stages. The first, usuallyrapid fall to about half of the original opticaldensity appears in phase optics as the shift from abright refractile to a dull white spore. The secondand more gradual decline accompanies thechange from the dull white state to a protoplastor lysed residue, depending on the osmotic andionic conditions. In Fig. 25 with sucrose-Mg2+,the immediate and rapid drop (30 sec) is followedby a slow lytic decline for 20 to 30 min. Whenmagnesium was replaced by calcium, the rate ofdensity loss was similar, and the optical densitydropped to a lower level but remained constant.However, the onset of the optical density dropwas delayed for 2 min after lysozyme was addedto 20 mm calcium (Fig. 25). Similar effects onboth optical density and fine structure were foundwhen 0.45 M succinate (2) was used instead ofsucrose for osmotic support.The peculiar semihydrated protoplast forms

seen in sucrose-calcium or succinate-calciummedia after 10 to 15 min (Fig. 19 and 21) werestill sensitive to the late addition of magnesiumion (10 mM). Many of these were convertedwithin 2 to 3 min after magnesium addition tonormal appearing protoplasts. Even after pro-longed (I to 2 hr) incubation at 37 C in succinate-Ca2+ or sucrose-Ca2+, a few normal protoplastscould be found. Presumably, the protoplasts had,by this time, picked up some Mg2+ from thelysed ones. However, the less dense partially lyticforms still possess dense nucleoids (Fig. 26).The delay of the rapid fall in optical density

0.5 -

0

Z 0.3-

J

cL

0

0.1

0 2 4 6MI N U TES

8 10 12

FrG. 25. Effect of lysozyme (100 ,g/ml added atzero time) on the optical density of suspended coatlessspores of B. megaterium KM in buffered sucrose con-

taining MgSO4 (10 mM) and with CaCI2 (20 mM).

induced by calcium (Fig. 25) was analyzed fur-ther in both sucrose and succinate stabilizingmedium. With increasing concentrations of Ca2+at a constant level of Mg2+ there was a corre-sponding increase in the delay time for the mainloss of refractility (Fig. 27). In spite of these de-lays in onset of lysozyme action, the higher cal-cium concentration did not appear to alter thephase-contrast appearance of the resulting proto-plasts formed in buffered sucrose (Fig. 6). Again,similar optical density changes were encounteredwith succinate-Mg2+ and various concentrationsof calcium. However, the uniformity of proto-plasting was usually greatly improved in this sup-porting medium. For example, a highly uniformpreparation of protoplasts fixed at 5 min in 0.45M succinate, 10 mM Mg2+, and 30 mM Ca2+ isshown in Fig. 28. In agreement with the results ofFig. 27, the optical density began to drop at 2.5min after the addition of lysozyme. ln Fig. 28,although the cortex is gone, the germ cell wallsare only partially digested and the spore nuclearmaterial is partially in the dense fibrous form.

Semiprotoplasts of coatless spores. Of partic-ular interest in the experiments typified by Fig.27 was the effect of lysozyme on coatless sporessuspended in sucrose without added cations. Apeculiar partial loss of refractility occurred whichwas identical in sucrose and water. These sporeswere more avoid, i.e., still "white" in the phase-contrast microscope (Fig. 2a and b). When ex-amined in nigrosin smears, the spores showed anearly loss of core refractility with some persistenceof refractility in peripheral or cortical remnants(Fig. 9 and 10). In cover slip smears, the loss ofthe spore phase-contrast refractility by lysozymein cation-free media (Fig. 3) and the subsequenteffect of adding calcium (Fig. 4) or magnesium(Fig. 5) could be readily followed and recorded.Optical density studies of lysozyme digestion ofcoatless spore suspensions in water or sucroseare gathered in Fig. 29. Although largely non-viable, the sucrose (alone) and water (alone) di-gested spores still respond (hydrate) rapidly whendivalent cations are added (Fig. 9). The delayingeffect of calcium ions on lysozyme digestion ofsucrose or succinate suspended coatless sporesis also observed with coatless spores suspendedin water. However, when the calcium is addedto the spores pretreated by lysozyme digestionin either water or sucrose, lysis is very rapid(Fig. 29). The water-Ca21 preparations com-pared to sucrose-Ca2+ ones were severely lysed,indicating a requirement for osmotic support.

Uhe semirefractile forms produced by lysozymein water, although nonviable, retained their initialappearance and sensitivity to added cations forsome days at 2 C. Those formed in sucrose (Fig.

Slr.~~~~~~~~~I*S C

SUC ROS E +C o +*

1128 J. BACTERIOL.

'74.

rA. Si

'..,-FIG. 26. Two hours after the addition of lysozyme to coatless spores of B. megaterium KM suspended in so-

dium succinate (0.45 M) containing CaCl2 (25 mM) but no added Mg2+. Early changes were similar to those shownfor sucrose-Ca2+ in Fig. 23 and 25. Half of the spores appear as normal protoplasts. The remaining spores wereeither in a partly converted state (a) or in various states of lysis (b). Magnification is indicated by the 0.5 ,um bar inpart a.

1129

)"WI -

.,

.J

1130

O.4-

0.3-

0.2-

0.1-

I I I I I0 2 4 6 8

MINUTES

FIG. 27. Effect of varying the ionic supplements inthe sucrose stabilizing medium on the action of lyso-

zyme on coat-stripped spores of Bacillus megaterium.With "no addition" lysozyme produces a semirefractilespore (See Fig. 2, 3b, 9, and 10). Magnesium alone(0.01 M) induces a rapid drop of refractility and ulti-mate lysis (see Fig. 5). At a magnesium concentrationof 0.01 M, calcium at increasing concentrations corre-

spondingly delays the initial drop in the optical densityand permits recovery ofprotoplasts (see Fig. 6).

30) had similar keeping qualities but a few couldbe recovered as stable protoplasts if magnesium,but not calcium, ions were added within 20 minof the initial fall in optical density (Fig. 31).

In the electron microscope these semirefractilespores formed in the sucrose and distilled waterpossessed similar condensed masses and vacu-

olated cytoplasmic patterns. In both, the nuclearbodies possessed the crystalline-like packing oftheir DNA similar to that of resting spores. (Fig.30 and 32). However, practically every profile ofthese spores treated with lysozyme in watershowed a broken and coiled plasma membrane.A few breaks were also seen in the membranes ofthose spores treated in sucrose; in many profiles,however, the membrane seemed to be intact. Thispresumably accounts for the recovery of a fewintact protoplasts on adding Mg'+ to the sucrose,but only lysis when cations are added to thewater-lysozyme-treated group. As indicated bylight microscopy (Fig. 10), both types of cation-free digestions showed partial breakdown of thecortex into either various sized chunks (Fig. 30and 32) or a confluent mass sticking the semire-fractile cores together, very much like prepara-tions of isolated cortex (5).

Action of other cations. These observations ofsomewhat specific effects of magnesium and cal-

cium on the lysozyme digestion of coatless sporesprompted a comparison of strontium with cal-cium and of manganese with magnesium, as thedivalent cations affecting lysozyme action oncoat-stripped spores of B. megaterium. The ac-tions of these cations were assessed by observingoptical density and phase-contrast and electronmicroscopic changes.

Strontium. When compared with calcium at 20mm concentration Sr2+ had a similar effect. Therapid lysis induced by the addition of calcium tocoatless spores pretreated with lysozyme in cat-ion-free sucrose (Fig. 29) was identical on theaddition of strontium ions. Moreover, a magne-sium-strontium-supplemented environment ap-peared identical to a magnesium-calcium one,including the delayed onset of optical densitydrop after lysozyme addition (Fig. 27). That is,Sr2+ could replace Ca2+ in maintaining mem-brane integrity and support the formation of in-tact protoplasts. Likewise, lysozyme digestion ofthe cortex of spores suspended in Tris-sucrosewith strontium ion was similar, in respect to op-tical density and structural changes, to that withcalcium as the sole cation (Fig. 19 and 25). Theseresults were not unexpected, since it was pre-viously discovered that strontium can take theplace of calcium both as a factor in spore heatresistance (6) and as a stimulant of DPA synthe-sis (17).Manganese. The manganese cation was unable

to replace magnesium during cortical digestion ofcoatless spores. Manganese-calcium-supple-mented sucrose digestions were very similar tothose in calcium (alone)-supplemented media inrespect to the optical density curve and the poorlyhydrated appearance of the resulting forms. InTris-sucrose supplemented with manganesealone, lysozyme induced an optical density curvevery similar to a cation-free system ("no addi-tion," Fig. 27), except that the optical densitystabilized at a lower level. The resulting semi-refractile spores were slightly denser than thoseof Fig. 2. Manganese ions, as might be expected,were unable to induce the rapid fall in opticaldensity of coatless spore suspensions pretreatedwith lysozyme in cation-free media, as can theions of magnesium and calcium (Fig. 29).

Effect of DPA on the digestion of coatlessspores by lysozyme. The apparent protective ef-fect of calcium on the spore membrane duringearly cortical digestion prompted some compari-sons of protoplast supporting media containingCa2+,Ca-DPA chelates, DPA, ethylenediamine-tetraacetic acid (EDTA), and ethylenebis(oxy-ethyleneitrilo)tetraacetic acid (EGTA).

In his studies of germination with chelates,Riemann (19) found that the equimolar (1: 1) Ca-

E

0

v0

z

0-

0

x No Addition

o Mg++(OO1M)

0+Ca++(0.02M)

00-6A * " " (0.05M)

\ x x - " x

O.,

-

-

I

FITZ-JAMES J. BACTERIOL.

40~ ~ ~ *I~~~~~~~~~I

~~~~~~~-~~ ~ ~ o,4,-6

ccw

f

E CW-P ,

^ t. k

it

FIG. 28. Coatless spores of B. megaterium KM treated with lysozyme for 5 min in 0.45 M sodium succinate con-

taining 10 mm MgSO4 and 30 mm CaCl2. The cortices have been digested off; the germ cell walls (CW) are stilllargely intact but show some destruction. The core cytoplasm is evenly hydrated and contains well preserved ribo-somes. The nuclear structures are as condensed bands opening out in regions into looser fibrillar structures. Insome places, the nuclear fibrils are in contact with the plasma membrane (unlabeled arrows). Bar marker is 0.1 im.

1131

*%.'

IF'o. ; "'kt,'' 'P,.1. *V. f --

-If A, :1. . 'Ail. . 4..

I

FITZ-JAMES

0.8

E WATER

10.-a_SUCROSE M

_0 +Mg+

Co~

0.4-

o

o +ca++

0

SUCROSE +Mgi+

SUCROSE * Ca++

. * . * *0 2 4 6

MINUTES

FIG. 29. Effect of adding cations to suspensions ofcoatless spores of B. megaterium pretreated with lyso-zyme (added at 0 time; 100 ig/ml) in sucrose or waterat 37 C. The addition of Ca2+ (10 mM) to a water-pre-

treated suspension was identical to that shown here forthe sucrose suspension. A sample, lysozyme treated inwater with Ca2+ initially present, is included for com-

parison.

DPA chelate at 40 mm concentration was an ac-

tive inducer of whole-spore germination. The 1 :2Ca-DPA chelate and DPA itself appeared largelyinactive. Lysozyme digestion of coatless B. mega-terium spores in Tris-sucrose-Mg2+ (10 mM) con-taining, instead of 20 mm Ca2+, Ca-DPA (1: 1, 20mM:20 mM) was initially identical to a controlcontaining both Mg2+ and Ca2+. However, by 5min, the optical density of the Ca-DPA sus-pended spores was 64% of the control. By 10 min,the optical density was 59% and the yield of in-tact protoplasts was 50% that of the control. InTris-sucrose-Mg2+ with Ca2+ replaced by Ca-DPA (1:2, 20 mm:40 mM), the optical density ofcoatless spores with the addition of lysozyme fellagain for 2 min, like that of the control. A severelysis then occurred. By 5 min, the optical densitywas only 33% of the control level; by 10 min, itwas 3%, at which time over 99% of the sporeswere only faint membrane ghosts. Such severe

lysis with Ca(DPA)2 was unexpected, in compar-ison to the digestions with Mg2+ alone (Fig. 25

and 27) which took 30 to 60 min to produce asimilar degree of lysis (Fig. 17b).

This marked lysis suggested that the addedDPA was competing with the membrane for theprotective spore Ca2+ during early lysozymedigestion, and it also prompted other compari-sons. Digestion of coatless spores in Tris-sucrose-Mg2+ (20 mM) and DPA at a low concentration(10 mM) without calcium (Mg2-DPA) pro-duced an optical density curve and pellet appear-ance identical to that of lysozyme digestion inTris-sucrose-Mg2+ without DPA. However, inequimolar Mg-DPA (20 mM:20 mM) a more se-vere lytic curve resulted, quite similar to thatfound for 10 mm Mg2+, 20 mm Ca2+, 20 mMDPA. Apparently, externally added DPA above acertain concentration can chelate Ca2+ from itsprotective function in the membrane. It seemsunlikely in these Mg-DPA experiments that theDPA is completely complexing Mg2+. If it were,in the absence of added Ca2 , the semirefractiletype of optical density curve (Fig. 27) and spore(Fig. 3b and 4a) should have resulted. Indeed,when coatless spores were lysozyme digested inTris-sucrose without added cations but containingDPA (20 mM), the optical density did not plateauat the semirefractile stage but fell in a curve verysimilar to that recorded with the Tris-sucrose-Mg2+-suspended spores and likewise producedadherent masses of partly lysed spores.The peculiar requirements for the cations Ca2+

and Mg2+ during lysozyme digestion of coatlessspores were also analyzed with other chelatingagents. When EDTA was added (20 mM) to theTris-sucrose-Mg2+ (10 mM)-Ca2+ (20 mM) sup-porting medium, the ultimate effect of lysozymedigestion of resting spores was very like that oc-curring in a medium without added cations-i.e.,the formation of semirefractile cores. Lysozymedigestion in an excess of EDTA (20 mM EDTA inTris-sucrose-Mg2+, 10 mM, but no Ca2+) pro-duced a pellet of granular membrane ghosts. Thecalcium chelate, EGTA, had an effect on lyso-zyme digestion only somewhat like that of DPA.The completely lysed residue of DPA andCa(DPA)2 digestions did not occur. Instead,lysozyme digestion with EGTA yielded a pellet,which was similar to that obtained in digestionsin Tris-sucrose-Mg2+.Control suspensions of coatless spores in the

Tris-sucrose-M g2+ medium containing CaDPAand Ca(DPA)2 or DPA showed no fall in opticaldensity without the addition of lysozyme. How-ever, coatless spores in EDTA (20 mM) did showa gradual decline in optical density (50% loss in20 min), whereas those in EGTA (20 mM) lostdensity at a lower rate (50% in 30 min). Neitherof these drops interfered with the assessment of

1132 J. BACTERIOL.

lFSi,,f, .

P:N 4\

.>r .,^<.;.F_E4L J

.. ..^. S

'F

;__r

t *.s

FIG. 30. Thin section of coatless spores of B. megaterium KM fixed after I to 2 min of digestion with lysozymein sucrose without added cations (see Fig. 2). (a) Considerable removal of cortical substance which remains as largeaccumulations adjacent to (arrow) and free from the spores. Many of the spores in the preparation were cementedtogether by the partially digested cortex. The poorly fixed cytoplasm appears only partly hydrated, and the nuclearmasses are still in the semicrystalline form of resting spore chromatin. (b) Closely packed repeating structures(arrows) in the DNA-containing masses (at higher magnification than part a). Magnifications are indicated by 0.1-ztm bars.

1133

-',,k7-f .*

Ki ^

vv- .iP.-* :IjI

I-'BMW

FIG. 31. Thirty minutes after adding Mg2+ to a preparation like that shown in Fig. 30. Magnesium was addedafter 20 min of lysozyme digestion in sucrose buffer. About 50% of the spore cores have hydrated normally intoviable protoplasts covered by some remnants ofspore coat. Bar marker is 0.5 Am.

FIG. 32. Coatless spores of B. megaterium KM digested with lysozyme in distilled water without added cations.The essential defect in these spores besides the absence of hydration is the rupture and coiling of the plasma mem-brane. They are otherwise quite similar to the cation-free sucrose-lysozyme digestion (Fig. 30). Bar marker is 0.5,um.

1134

Irr. s

IL I;N.,

.; r-.

.4 t.

-t"Js

.;_s8ll sP

SPORE PROTOPLASTS

the more rapid density loss with lysozyme (50%in 1 min).

DISCUSSIONEnzymatic removal of the cortex and the asso-

ciated cell wall primordium (germ cell wall) froma coat-stripped resting spore suspended in anosmotic supporting medium does not yield aviable protoplast. To do so via lysozyme diges-tion, two metal requirements must be satisfied.First, calcium ions are essential to prevent a rup-ture of the spore core plasma membrane; second,magnesium ions are required for the conversionof this initally digested spore into a viable proto-plast. This conversion appears, as observed byelectron microscopy, to be a process of hydrationinvolving the core cytoplasmic and nuclear struc-tures. The latter body is particularly dependenton Mg2+ to achieve the usual appearance ofDNA plasma. In fact, the fine structure studieshere suggest the conversion of spore to vegeta-tive-type DNA is in two stages. The first, achange from the semicrystalline closely packedand poorly staining structure into a denselystaining ropy mass winding through the cell, is arapid event. The second change appears to be asimple expansion of this mass into the more opentype of fibrous nuclear body. These changes areof interest in view of the recent report by Tan-ooka and Terano (21) that the high degree of ra-diation resistance of B. subtilis spore DNA tosingle-strand breaks was maintained in sporespheroplasts but was lost when these were osmot-ically ruptured. Since cation additions were notused, their spheroplasts are presumably similar tothe partly refractile forms described here in detailfor B. megaterium. The spheroplasts are alsoencountered when coatless spores of B. cereusand B. subtilis were lysozyme digested in Tris-sucrose without added cations and which stillcontain DNA in a crystalline-like structure (Fig.30).The requirement for Ca2+ can no longer be

shown once the Mg2+-dependent hydration hasbegun. That is, coat-stripped spores digested withlysozyme for 5 to 10 min in sucrose-Mg2+-Ca2+or in sucrose-Ca2t will, when spun out and sus-pended in sucrose-Mg2+ with additional lyso-zyme, continue towards the formation of stableand viable protoplasts. On the other hand, whensuspended in sucrose-Mg2+ or in sucrose alone,coatless spores must be supplied with Ca2+ within1 min of the addition of lysozyme to prevent ex-tensive lysis. This initial period of calcium de-pendency also corresponds to the period of earlyand rapid release of DPA. Both the cation sup-plemented, typical protoplasts and the semire-fractile spore cores are essentially free of DPA. A

subsequent report now in preparation will de-scribe the escape rates of DPA and calcium fromcoatless spores undergoing lysozyme digestion invarious ionic environments. It is, however, inter-esting to note that the semirefractile forms indi-cate the separability of DPA release and hydra-tion of resting spores. The former process is asso-ciated with a Ca2+ need for membrane integrity,the latter with a Mg2+ requirement.The apparent early dependence of the spore

core membrane on Ca2t coupled with the highcontent of calcium in resting spores suggests aunique role of this ion as the stabilizer of themembrane during the spore state. For such a rolein dormancy, it is first necessary to reconsider theconcept of an anhydrous core (18) and to ques-tion the necessity of DPA as the sole anion linkedto calcium. In 1953 water determinations on ab-solute ethanol washings and disruption exudatesof vegetative cells and resting spores were madeby using the Karl Fischer titration technique (P.C. Fitz-James, Ph.D. Thesis, Health SciencesLibrary, University of Western Ontario, London,Canada, 1953). Vegetative cells are rendered non-viable by the ethanol wash before disruption andhad a water content dependent on the numberof washings. However, the spores survived ethanolwashings; when disrupted (and killing), theyyielded less than 5% of their dry weight as waterto absolute ethanol, a content of water which wasthe lowest limit of the techniques used. Hyatt andLevinson (10) also found a similar low water con-tent for spores by drying at 100 C.

If the core is nearly anhydrous, a majorproblem to such a biological system would bemaintenance of the polarity of the phospholipidsand hence the integrity of the core membrane.These protoplasting studies have shown that cal-cium is required for membrane integrity beforehydration of the core; one might then speculatethat most of the calcium of resting spores may, infact, be involved in rendering the membrane sur-faces nonpolar by bridging the free ionic groupsof the phospholipids.The role of calcium in binding to, and so modi-

fying, the mechanical resistance and tension of awide variety of biological membranes is well doc-umented. It is interesting that the increased re-sistance or stiffness that Ca2t imparts to teleostand amphibia membranes occurs without altera-tions in permeability (14). Hauser and Dawson,using lipid films, found that Ca2+ binding was de-pendent on the net excess negative charge on thelipid molecule (9). At the physiological pH range,membranes containing phosphotidylserine, tri-phosphoinositide, and phosphotidylethanolaminepossess Ca2+ binding groups in large amountsand can show large variations in bound ions withcomparatively small pH changes (1 1).

VOL. I 05, 1 971 1135

FITZ-JAM ES

Confirmation of a calcium binding to sporemembranes was recently obtained by autoradiog-raphy of 45Ca labeled, resting spores in which65% of the activity was located in the region ofthe inner membrane and less than 10% was lo-cated in the core region (unpublished data).A concept of dormancy based on an anhydrous

core and a calcium-stabilized membrane systemruns counter to the usual concept of DPA as thesole calcium chelate in dormant spores. Studieswith actinomycin D showed that core refractilityand heat resistance could be established in theabsence of cortex or coat (4), suggesting a corelocation of DPA. At this location, DPA couldreadily function in the removal of Ca2+ from themembrane in the initial moments of core hydra-tion during germination. Indeed, it has alreadybeen shown that DPA-less mutants which havesome heat resistance (24) cannot germinate unlesssuspended in some calcium binding system suchas DPA, EDTA, or NaHCO3 (H. Orin Halvor-son, personal communication). It is interesting tonote that in the studies of Hauser and Dawson(9), although Mg2+ had a greater relative abilityover Na+ or K+ to displace Ca2+ from varioustypes of lipid films, all such films showed a muchgreater preference for Ca2+. Possibly, a similarpreference occurs in the resting spore, and duringgermination the reduction of the Ca2+ adsorbedto spore membranes cannot be achieved by Mg2+;thus, there is a need for DPA or some similarbinding activity.Calcium has already been implied as a major

agent of heat resistance from studies of sporesloaded with Ca2+ (13). Of more interest to thispresent consideration of a possible change inmembrane requirement from Ca2+ to Mg2+ onprotoplasting spores are the interesting findingsof Murrell and Warth (15) that heat resistancecan be closely related to the Mg-Ca ratio of thespore; the higher the Mg2+ the less the resistance.Possibly, the membranes of those spores withlower resistance have insufficient Ca2+ to achievefull stability and are partially complexed withMg2+.

ACKNOWLEDGMENTSI am grateful to Doryth Loewy and Leah Mitchell for their

technieal assistance and to the Medical Research Council ofCanada for continued support. The Karl Fischer water titra-tions were done by Hugh Vance.

LITERATURE CITED

1. Aronson, A. I., and P. C. Fitz-James. 1968. Biosynthesis ofbacterial spore coats. J. Mol. Biol. 33:199-212.

2. Fitz-James, P. C. 1958. Cytological and chemical studies ofthe growth of protoplasts of Bacillus megaterium. J.Biophys. Biochem. Cytol. 4:257-266.

3. Fitz-James, P. C., and 1. E. Young. 1959. Cytologicalcomparison of spores of different strains of Bacillusmegaterium. J. Bacteriol. 78:755-764.

4. Fitz-James, P. C. 1965. Spore formation in wild and mu-

tant strains of B. cereus and some effects of inhibitors, p.529-544. In M.J-C Senex (ed.), Regulations chez lesmicroorganisme. Centre National de la RechercheScientifique, Paris.

5. Fitz-James, P. C., and1. E. Young. 1969. Morphology ofsporulation, p. 39-72. In G. W. Gould and A. Hurst(ed.), The bacterial spore. Academic Press Inc., NewYork.

6. Foerster, H. F., and J. W. Foster. 1966. Endotrophic cal-cium, strontium and barium spores of Bacillus mega-terium and Bacillus cereus. J. Bacteriol. 91:1333-1345.

7. Gould,G. W., and A. D. Hitchins. 1963. Sensitization ofbacterial spores to lysozyme and hydrogen peroxide withreagents which rupture disulphide bonds. J. Gen. Micro-biol. 33:413-428.

8. Gould, G. W., and A. D. Hitchins. 1965. Germination ofspores with Strange and Dark's spore lytic enzyme, p.213-221. In L. L. Campbell and H. 0. Halvorson (ed.),Spores 111. American Society for Microbiology, AnnArbor, Mich.

9. Hauser, H., and R. M. C. Dawson. 1967. The binding ofcalcium at lipid-water interfaces. Eur. J. Biochem. 1:61-69.

10. Hyatt, M. T., and H. S. Levinson. 1968. Water vapor,aqueous ethyl alcohol, and heat activation of Bacillusmegaterium spore germination. J. Bacteriol. 95:2090-2101.

11. Joos, R. W., and C. W. Carr. 1967. The binding of calciumin mixtures of phospholipids. Proc. Soc. Exp. Biol. Med.124:1268-1272.

12. Kellenberger, E., A. Ryter, and J. Sechaud. 1958. Electronmicroscope study of DNA-containing plasms. 11.Vegeta-tive and mature phage DNA as compared with normalbacterial nucleoids in different physiological states. J.Biophys. Biochem. Cytol. 4:671-678.

13. Levinson, H. S., and M. T. Hyatt. 1969. Activation of Ba-cillus megaterium spore germination, p. 262-275. In L.L. Campbell (ed.), Spores IV. American Society forMicrobiology, Bethesda, Md.

14. Manery, J. F. 1969. Calcium and membranes, p. 405-452.In C. L. Comar and F. Bronner (ed.), Mineral metabo-lism, vol. 111. Academic Press Inc., New York.

15. Murrell, W. G., and A. D. Warth. 1965. Composition andheat resistance of bacterial spores, p. 1-24. In L. L.Campbell and H. 0. Halvorson (ed.), Spores 111. Amer-ican Society for Microbiology, Ann Arbor, Mich.

16. Northrop, J. H. 1951. Growth and phage production of ly-sogenic B. megaterium. J. Gen. Physiol. 34:715-735.

17. Pelcher, E. A., H. P. Fleming, and Z. J. Ordal. 1963. Somecharacteristics of spores of Bacillus cereus produced by areplacement technique. Can. J. Microbiol. 9:251-258.

18. Powell, J. F., and J. R. Hunter. 1956. Adenosine deaminaseand ribosidase in spores of Bacillus cereus. Biochem. J.62:381-387.

19. Reimann, H. 1961. Germination of bacteria by chelatingagents, p. 24-48. In H. 0. Halvorson (ed.), Spores 11.Burgess Publishing Co., Minneapolis, Minn.

20. Reynolds, E. S. 1963. The use of lead citrate at high pH asan electron-opaque stain for electron microscopy. J. Cell.Biol. 17:208-212.

21. Tanooka, H., and H. Terano. 1970. Resistance of DNAagainst radiation-induced strand breakage in bacterialspores. Radiat. Res. 43:613-626.

22. Warth, A. D., D. F. Ohye, and W. G. Murrell. 1963. Loca-tion and composition of spore mucopeptide in Bacillusspecies. J. Cell. Biol. 16:593-609.

23. Weibull, C. J. 1956. Bacterial protoplasts their formationand characteristics, p. 111-126. In E. T. C. Spooner andB. A. D. Stocker (ed.), Bacterial Anatomy VIth Sym-posium Soc. Gen. Microbiol. Cambridge UniversityPress, Cambridge.

24. Wise, J., A. Swanson, and H. 0. Halvorson. 1967. Dipico-linic acid-less mutants of Bacillus cereus. J. Bacteriol.94:2075-2076.

1136 J. BACTERIOL.