JOURNAL OF BACTERIOLOGY, Jan., 1967, p. 399-410Copyright © 1967 American Society for Microbiology

Fine Structure of Sarcina maxima and Sarcinaventriculi

STANLEY C. HOLT AND E. CANALE-PAROLADepartment of Microbiology, University ofMassachusetts, Amherst, Massachusetts

Received for publication August 1966

ABSTRACT

The fine structure of Sarcina maxima and S. ventriculi was studied by electronand phase-contrast microscopy. The two organisms differ mainly with respect totheir cell surface. A thick cellulose layer present on the cell wall of S. ventriculiwas not observed on the surface of S. maxima. Carbon replication indicated thatthe outer surface of S. ventriculi is rough in contour, probably as the result ofthe fibrillar nature of the accumulated cellulose. The cytoplasm of both sarcinaecontains inclusions similar to polysaccharide and polymetaphosphate granules.Mesosomes were observed in cells of S. maxima. Packets of S. ventriculi generallycomprise a larger number of cells and are more irregularly constructed thanthose of S. maxima. Cells in large packets of S. ventriculi assume flattened or other-wise irregular shapes, whereas cells of S. maxima maintain a more uniform ap-

pearance.

Sarcina maxima and S. ventriculi may be differ-entiated on the basis of fermentation characteris-tics, such as the production of butyrate by theformer organism and the production of ethylalcohol by the latter (2, 9, 10, 11).

Morphological differentiation between thesetwo bacteria is not readily achieved since theyresemble each other when viewed by light micros-copy. The appearance of both organisms is quitecharacteristic because their cell and packet size islarger than that of other known sarcinae with thepossible exception of Thiosarcina rosea and S.methanica.

This paper describes the results of a compara-tive study aimed at demonstrating differences or

similarities in the fine structure of S. maxima andof S. ventriculi.

MATERIALS AND MErHODS

Organisms. S. maxima (strain 11) was obtainedfrom Professor H. KndlI of the Institut fur Mikro-biologie und Experimentelle Therapie, Jena, Germany.This strain ferments glucose to butyrate, acetate,C02, and H2 (D. G. Kupfer and E. Canale-Parola,unpublished data).

S. ventriculi (strains AL and CL) were isolated fromCalifornia garden soil by a procedure previously de-scribed (2). Both strains produced ethyl alcohol fromglucose (5). Strain AL exhibited a positive reactionindicative of cellulose production, whereas strain CLgave a negative or weakly positive reaction (1, 10).

Growth conditions. S. maxima was grown in a me-

dium containing the following per 100 ml of distilled

water: glucose and peptone, 1.0 g of each; yeast ex-

tract, 0.5 g; and L-cysteine hydrochloride, 0.05 g. Thegrowth medium for S. ventriculi consisted of 2 g eachof glucose and yeast extract per 100 ml of distilledwater. ThepH of both media was adjusted to approxi-mately 6.0.The organisms were grown at 30 C, either in 1-liter

Erlenmeyer flasks with 800 ml of medium or in 14-liter fermentor vessels (NBS Microferm Fermentor,model MF-214), each containing 10 liters of medium.

Stock cultures of both S. maxima and S. ventriculiwere maintained on the above-mentioned media plus1.5 g of agar per 100 ml at 30 C by means of the agardepression method (2).

Disruption of cells. Cultures were harvested fromthe midexponential growth phase by centrifugationand washed once with 0.2 M phosphate buffer (pH 6.8)containing 0.001 M MgSO4. The pellets were resus-pended to a concentration of 5 mg (dry weight) per mlin 0.2 M phosphate buffer. A 6-ml amount of this sus-pension, together with 3 ml of glass beads (Ballotini,no. lOa), was added to the cup of a Mickle apparatus.The cups were shaken with a 10 mm peak-to-peak dis-placement and were cooled to 4 C every 15 sec, a pro-cedure which maintained the temperature below 20 Cduring the entire treatment. The preparation wasexamined by phase-contrast microscopy to determinethe extent of breakage. Approximately 90% of thecells were broken within 1 to 2 mi.Enzyme treatments. After treatment in the Mickle

apparatus, cell fragments in 0.2 M phosphate bufferwere treated with trypsin (1 mg/mi, pH 8.0) for 5 hrat 37 C or with lysozyme (50 pg/ml, pH 7.0) for 5 hrat 37 C.

Carbon replication. The preshadowed carbon replica

399

Vol. 93, No. IPrinted in U.S.A.

HOLT AND CANALE-PAROLA

technique of Dalitz (3) and DeBoer and Spit (4) wasused. A few drops of a washed suspension of S.maxima or S. ventriculi in distilled water were placedon the surface of freshly cleaved mica. After standingovernight for complete drying, the mica was shadowedwith palladium-gold or platinum at an angle of 200.After shadowing, carbon was evaporated onto themica to a thickness of approximately 250 A, as deter-mined by interference color on gold-coated platinumfoil. With a razor blade, the carbon film was scratchedinto squares of grid size, the mica was immersed inglass-distilled water, and the squares of carbon werefloated off. The squares were then transferred to thesurface of a commercial chromerge solution (Mano-stat Corp., New York, N.Y.), and the cytoplasmiccontents were allowed to oxidize for 1 hr. The squareswere then transferred to glass-distilled water forwashing and picked up onto specimen grids.

Electron microscopy. Samples were fixed and de-hydrated as described by Kellenberger, Ryter, andSechaud (7), with some modifications as follows.The agar blocks were dehydrated in a graded aqueousethyl alcohol series according to the following sched-ule: 20, 50, 70% ethyl alcohol, 10 min each; 95 and100% ethyl alcohol, 15 min each; and one furtherdehydration in absolute ethyl alcohol for 30 min.Alcohol was replaced with propylene oxide, and thesamples were embedded in the epoxy resin Epon 812according to the procedure of Luft (8). Sections werecut on a Porter-Blum MT-2 ultramicrotome with adiamond knife, were mounted on Formvar-coated 200-or 300-mesh copper grids, and stained with lead hy-droxide according to the procedure of Karnovsky (6).

All samples for electron microscopy were examinedin a Philips EM 200 electron microscope equippedwith a 30-,u objective aperture operating at 60 kv.

REsuLTs

Intact cells. Phase-contrast microscopy showedthat packets of S. ventriculi generally comprise agreater number of cells and are more irregularlyconstructed than packets of S. maxima (Fig. 1and 2). However, packet size and shape tend tovary in both sarcinae depending on the strain andon the growth conditions. Only the top layer ofcells is clearly evident in Fig. 1 and 2, but thecubical nature of the packets was readily appar-ent in wet-mount preparations when water cur-rents caused the packets to rotate on their axes.Many of the cells within the packets in Fig. 1 and2 were beginning to undergo binary fission, asevidenced by the formation of transverse wallswhich appeared as lines of decreased density.Packets of S. ventriculi frequently presented arough and uneven outer surface (Fig. 2), prob-ably due to peripheral accumulation of celluloseby this organism (1).

Thin sections of intact cells. The appearance ofwhole cells in sections (Fig. 3 to 6) confirmed themorphological differences observed with thephase-contrast microscope.

The cytoplasm of both organisms was verydense, presumably the result of the presence ofclosely packed ribosomes. This tended to obscurethe nuclear material which appeared scatteredthroughout the cytoplasm of the cell.

Cytoplasmic regions of low electron density,with circular profiles and diameters of 400 mju(Fig. 5, 7, 9), were observed in S. ventriculi. Thelow density of these regions suggests that they arethe sites of polysaccharide storage granules. Thesegranules did not appear to have limiting mem-branes and were present with some regularity incells of S. ventriculi but only occasionally in S.maxima. Smaller granules of similar electron den-sity were observed by R. G. E. Murray and C. F.Robinow (Abstr. Intern. Congr. Microbiol., 7th,p. 427, 1958) in a tetrad-forming coccus and byThornley, Horne, and Glauert (13) in Micrococcusradiodurans.The former authors reported that these in-

clusions contained periodic acid-Schiff-positivematerial.Dense bodies, similar in appearance to in-

clusions which other authors (13) have identifiedas polymetaphosphate granules, were presentin some sections (Fig. 5, 6, 15). These bodiesprobably corresponded to the refractile granulesvisible by phase-contrast microscopy (Fig. 2).As a result of the high density of the cyto-

plasm, the plasma membrane was only rarelyobserved. In some sections (Fig. 4, 7, 8), itappeared as a "unit membrane," 75 to 100 Athick, and was seen at the surface of the cell andat both sides of a developing cross wall (Fig. 4).The membranous bodies, or mesosomes, whichare associated with the plasma membrane ofboth gram-positive and gram-negative bacteria,were also visible in S. maxima (Fig. 3, 4). Inthis organism, these membranous structures wereassociated with the ingrowing cross walls. Similarstructures were scattered throughout the cyto-plasm.The cell wall of both sarcinae was similar in

appearance to that of other gram-positive bac-teria, except for a thick layer outside the cellwall of S. ventriculi (Fig. 5, 6). This outer layerwas previously observed in S. ventriculi and wasshown to consist of cellulose fibrils (1).A cell wall of medium density and measuring

40 m,u in thickness was present at the cell surfaceof S. maxima (Fig. 3, 4). In S. ventriculi the cellwall was more difficult to identify owing to thethickness of the cellulose layer. In S. ventriculiCL, which deposits little cellulose at the outersurfaces, a cell wall 40 m,u thick was readily ap-parent (Fig. 9). This strain was found to havemany more granules of low electron densityscattered throughout the cytoplasm. When S.

400 J. BACrERIOL.

FINE STRUCTrURE OF ANAEROBIC SARCINAE

FIG. 1. Photomicrograph of living cells ofSarcina maxima, strain 11. Phase contrast.FIG. 2. Photomicrograph of living cells ofSarcina ventriculi, strain AL. Phase contrast.

401VOL. 93, 1967

402 HOLT AND CANALE-PAROLA J. BACrERIOL.

My,F: ........~~.4

c

w...:E. ........

i;; : 0 1t. .';'j''~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~.......

.A~~~~~4~

...Pv

E~~~:YE;0 0 0 I,N 00

FINE STRUCTURE OF ANAEROBIC SARCINAE

ventriculi was treated by ballistic disintegration,the cellulose layer became "loosened" and thecell wall was readily observed (Fig. 10).The surface of both sarcinae presented a fine

structure similar to that reported for M. radio-durans (13), with a basic pattern consisting of anarray of fibers, where the long axis of the fibersruns perpendicular to the cytoplasmic membrane.The fine structure of the cell wall of S. maximaand of the cellulose layer of S. ventriculi wassimilar to that of tangential sections through theperiphery of M. radiodurans in which there existsa pattern of light "holes" in a roughly hexagonalarray (Fig. 7, 8).

Disrupted cells. After treatment with the Mickledisintegrator, many of the cells were broken andlost most of their cytoplasmic contents (Fig. 10to 12). The packets of S. maxima generally brokeinto smaller units of one or two cells (Fig. 11, 12),whereas S. ventriculi maintained its packet con-figuration even after the cytoplasmic contents hadbeen removed (Fig. 10). Thin sections of brokencells showed the plasma membrane still attachedto the cell wall of S. maxima (Fig. 11 to 13). InS. ventriculi the plasma membrane has been re-moved and the outer layer has lost much of itsdensity (Fig. 10), a possible result of treatment inthe disintegrator. Underlying this area of lowelectron density was a region of high density,40 m,u thick, which presumably corresponded tothe cell wall.Enzyme digestion of broken cells. Trypsin ap-

parently had little effect on the rigidity of thedisrupted cells, and the outlines of the packetswere clearly preserved (Fig. 13). The plasma mem-brane took on an unusually dense appearanceafter trypsin treatment. The enzyme usually af-fected the outer surface of the cell where it causeda general disorganization of the structure.The entire cell wall was removed by treatment

with lysozyme. Some of the disrupted cells losttheir rigidity, whereas others retained their orig-inal shape (Fig. 14, 15). The plasma membraneenclosed a very complex membranous interior.

Carbon replication. Figure 16 is an electronmicrograph of a carbon replica of S. maxima. Thecells are in a packet configuration of eight cells(only the top four cells are visible), and the sur-face consists of a nonstriated, smooth structure.There is no evidence of any fine structure in the

outer surface. Surface replication of S. ventriculi(Fig. 17) shows the cells in their configuration oflarge packets. The outer surface is rough in con-tour, probably as the result of cellulose deposi-tion.

DISCUSSION

This study of the large sarcinae, S. maxima andS. ventriculi, has shown that the fine structure ofthe cytoplasm of these two organisms is similar,but that the cell surface and gross morphology aredifferent.The most obvious structural difference between

the two sarcinae is the presence of a thick layer ofcellulose deposited on the cell wall of S. ventriculi(1). In contrast, this layer is not observed on thesurface of S. maxima. Packets of S. ventriculigenerally include a larger number of cells thanthose of S. maxima. This difference in packet sizemay be attributed to the accumulation of celluloseat the cell surface of S. ventriculi. It has beenpostulated that the polysaccharide has the struc-tural function of binding the cells tightly intolarge packets (1). As a result of the tight packing,cells in the large packets of S. ventriculi assumeflattened or otherwise irregular shapes, whereascells of S. maxima maintain a more uniform ap-pearance. Generally, flattening of the cells is notobserved in strains of S. ventriculi which producesmall amounts of cellulose. In addition, thesestrains tend to form small, loosely bound packets.

Measurements of whole and sectioned cellsindicate that S. maxima (strain 11) exhibits asomewhat larger cell diameter (about 2.5 ,u) thanS. ventriculi under the growth conditions de-scribed. The cell diameter of S. ventriculi (strainsAL and CL) is about 2 ,u, in agreement withmeasurements reported for another strain (1).These dimensions are smaller than those reportedby Suringar (12) and Smit (10, 11) for S. ventriculiand S. maxima.

ACKNOWLEDGMENrS

This investigation was supported by Public HealthService grant Al 06294 from the National Institute ofAllergy and Infectious Diseases and by Contract 3357from the Office of Naval Research.We thank Eunice Hench for the preparation of the

thin sections for electron microscopy.

FIG. 3. Electron micrograph of thin section of Sarcina maxima, strain 11. The smooth cell wall (CW) obscuresthe unit cytoplasmic membrane (CM). The nuclear material (N) is very diffuse. Main fixation 16 hr; post-stainedwith lead hydroxide.

FIG. 4. Electron micrograph of thin section of Sarcina maxima, strain 11. At the leading portion of the invagi-nzating cell wall and cytoplasmic membrane (CM) is'a mesosomal element (M). Figure 4a (inset) is an enlargedarea ofFig. 4.

403VOL. 93, 1967

404 HOLT AND CANALE-PAROLA J. BACrERIOL.

.. ........ _

'.

FIG. 5 and 6. Electron micrographs of thin sections of Sarcina ventriculi, strain AL. The thick cellulose layer(CL) surrounds the cell walland cytoplasmic membrane. The light area (G) may correspond to the site ofa storagegranule. The very dense bodies (PM) may correspond to polymetaphosphate granules. Note the flattened irregularshape ofsome of the terminal cells of the packet in Fig. 6.

FINE STRUCTURE OF ANAEROBIC SARCINAE

* cw

FIG. 7. High-magnification micrograph of the "convergence point" offour cells of Sarcina ventriculi, strainAL. The cell wall (CW) appears to possess an array of "holes". Compare with Fig. 8. The light areas (G) maycorrespond to sites ofstorage granules. The bar indicates 0.1 As.

FIG. 8. Electron micrograph of a section of the surface of Sarcina ventriculi, strain AL. Note the structuredappearance of the cellulose layer (CL). The bar indicates 0.1 IA.

FIG. 9. Electron micrograph of Sarcina ventriculi, strain CL. Note the absence of a thick cellulose layer. Thisstrain possesses many storage granules,(G).

405VOL. 93, 1967

HOLT -ANID CANALE-PAROLA..~~~~~~~r

FIG. 10. Section of Sarcina ventriculi, strain AL, after ballistic disintegration. The cellulose layer (CL) hasbeen "loosened" and the cell wall (CW) appears very distinct. < I

FIG. 11. Section ofSarcina maxima, strain 11, after ballistic disintegration. The cytoplasmic membrane (CAM)has been pulled away from the cell wall. Note, the-appearance (arrow) ofthe cell wall.

406 J. BkcTWOOL.4kakiii;.

FIG. 12. Electron micrograph of Sarcina maxima, strain

The cytoplasmic membrane has been removed by Mickle tro

brane is still prese-rved (arrow). Main fixation 16 hr.

FIG 13 Electron micrograph of Sarcina maxima, strain

Note the density of the cytoplasmic membrane (CM). Main

407

11, after ballistic disintegration and trypsin treatment.eatment. In the undamaged cell, the cytoplasmic mem-

11, after ballistic disintegration and trypsin treatment.7fixation 16 hr.

408 HOLT AND CANALE-PAROLA J. BAcnuuoL.

M

im

M

hl7qO,

u.

.Ot -WimTNt'

A

M. hmC

js I-M

6W,VIllp i

74 2'wS a

tW 17t.T,4.. V !MiZ iP i--' t.N,

...i_hinl

O.C al'J E-4

gik FKI"RT

Z,iT, f..q, V

J

FAJ i4W U.

tq -,ss r;; u. f, o:

'MP

K 4z;

ARlao

4-4JA

J,

or.

WA

Ffq

AM Pk.:AV

Oliili !Aw

IT

IW..

Cs Z:

AA

Vp.

Ti

Ji

Al

W.0S

lt

rh, ..,t s.:

.M

MI

FIG. 14 and 15. Electron micrographs ofSarcina maxima, strain 11, after lysozyme treatment. The cell wall hasbeen completely removed, leaving only the cytoplasmic membrane (CM). The general configuration of the cells hasbeen retained. A dense body (PM) may bea polymetaphosphate granule.

FINE STRUCIURE OF ANAEROBIC SARCINAE

FIG. 16 and 17. Carbon replicas ofSarcina maxima, strain 11, and S. ventriculi, strain AL. Note size and archi-tecture of the packets ofS. ventriculi AL, as compared to S. maxima 11.

409VOL. 93, 1967

HOTLT AND CANALE-PAROLA

LITERATURE CITED

1. CANALE-PAROLA, E., R. BORASKY, AND R. S.WOLFE. 1961. Studies on Sarcina ventriculi. HI.Localization of cellulose. J. Bacteriol. 81:311-318.

2. CANALE-PAROLA, E., AND R. S. WOLFE. 1960.Studies on Sarcina ventriculi. I. Stock culturemethod. J. Bacteriol. 79:857-859.

3. DALITZ, V. C. 1953. Abdruckverfahren zur Beo-bachtung von Holz im Uebermikroskop. Z.Wiss. Mikroskopie 61:292-293.

4. DEBoER, W. E., Am B. J. SPIT. 1964. A new typeof bacterial cell wall structure revealed byreplica technique. Antonie van LeeuwenhoekJ. Microbiol. Serol. 30:239-249.

5. FEIGL, F. 1960. Spot tests in organic analysis,6th ed., p. 358. Elsevier Publishing Co., Ams-terdam.

6. KARNOVSKY, M. J. 1961. Simple methods forstaining with lead at high pH in electron mi-croscopy. J. Cell Biol. 11:729-732.

7. KELLENBERGER, E., A. RYTER, AND J. SkHAUD.1958. Electron microscope study of DNA-con-taining plasms. II. Vegetative and maturephage DNA compared with normal bacterialnucleoids in different physiological states. J.Biophys. Biochem. Cytol. 4:671-675.

8. LuFr, J. H. 1961. Improvements in epoxy resinembedding methods. J. Biophys. Biochem.Cytol. 9:409-414.

9. MILHAUD, G., J.-P. AUBERT, AND C. B. VANNIEL. 1956. Etude de la glycolyse de Zymosar-cina ventriculi. Ann. Inst. Pasteur 91:363-368.

10. SMIT, J. 1930. Die Garungssarcinen. GustavFischer, Jena.

11. SMrT, J. 1933. The biology of the fermenting sar-cinae. J. Pathol. Bacteriol. 36:455-468.

12. SURINGAR, W. F. R. 1866. La sarcine de l'esto-mac. Arch. Neerl. Sci. 1:209-271.

13. THORNLEY, M. J., R. W. HORNE, AND A. M.GLAUERT. 1965. The fine structure of Micro-coccus radiodurans. Arch. Mikrobiol. 51:261-289.

410 J. BAcrERioL.