JOURNAL OF BACTERIOLOGY, Sept., 1965Copyright © 1965 American Society for Microbiology

Vol. 90, No. 3Printed in U.S.A.

Nature and Development of Membrane Systems inFood Vacuoles of Cellular Slime Molds

Predatory upon BacteriaHANS R. HOHL

Department of Microbiology, University of Hawaii, Honolulu, Hawaii

Received for publication 14 April 1965

ABSTRACT

HOHL, HANS R. (University of Hawaii, Honolulu). Nature and development ofmembrane systems in food vacuoles of cellular slime molds predatory upon bacteria.J. Bacteriol. 90:755-765. 1965.-During the digestion of bacteria by the myxamoebaeof cellular slime molds, systems of concentric lamellae begin to appear within the foodvacuoles. Each constituent lamella is a unit membrane of 75 to 85 A thickness. A studyof these lamellae in Dictyostelium discoideum and Polysphondylium pallidum revealsthat most of them do not represent original membranes of the ingested bacteria butare formed mainly in two ways. (i) After swelling and partial digestion of the bacteria,the first membranes appear adjacent to pre-existing membranes, e.g., the membranelining the food vacuole and the cytoplasmic membrane surrounding the bacterium.Progressive addition of lamellae leads to the formation of the systems of concentriclamellae. (ii) After digestion of the bacteria has proceeded to a high degree, the con-

centric lamellae are formed spontaneously from clouds of amorphous material throughcondensation and orientation of precursor material. The study shows that, in biologicalsystems, unit membranes may be formed from amorphous material through templateaction of pre-existing membranes, and does not necessarily involve fusion of membrane-bound vesicles.

In the food vacuoles of myxamoebae of cellularslime molds growing on living bacteria, numeroussystems of tightly packed membranes are fre-quently observed (Gezelius, 1959, 1961; Mercerand Shaffer, 1960). It is reasonable to assume thatthe existence of these membranes is closely relatedto the digestion of the bacteria. However, finalproof of this had not previously been presented,since until recently the myxamoebae could not becultivated in the absence of bacteria (Hohl andRaper, 1963b; Sussman, 1963).

Since the ingested bacteria contain much lessmembranous material than these lamellar sys-tems, their presence in such large quantities mustbe accounted for. We hope that an elucidation ofthe mechanism of membrane formation in thefood vacuoles of these organisms might throwsome light on possible membrane-generating proc-esses in the living cell in general.

This report describes the nature of the individ-ual membranes and their configuration in themembranous systems and presents their forma-tion from the beginning of the digestive processtogether with a plausible explanation of the factsobserved. In addition, some other related aspectsof cellular digestion in the cellular slime moldsare considered.

MATERIALS AND METHODSDictyostelium discoideum strain NC-4(S2) and

Polysphondylium pallidum strain FR-47 wereused. The latter strain is able to grow on a solublemedium in the absence of any bacteria or bacterialcomponents (Hohl and Raper, 1963c) and, there-fore, was particularly useful in the present study.

Escherichia coli strain B/r was used as the foodorganism. The bacteria were grown on a solidmedium consisting of 1% lactose, 1% peptone,and 1.5% agar. They were suspended in 0.016 Mphosphate buffer at pH 6.25, centrifuged, resus-pended, and their concentration was adjusted to1010 bacteria per milliliter.The myxamoebae were grown in 5-ml samples

of the bacterial suspension in test tubes rotatingat 250 rev/min at 24 to 26 C; details of the tech-nique have been described elsewhere (Hohl andRaper, 1963a, b). The inoculum was adjusted to104 spores per milliliter of bacterial suspension.

P. pallidum, strain FR-47, also was cultivatedon the complex medium of Hohl and Raper (1963c)consisting of 4% bovine serum albumin and 2%tryptose in an inorganic salt solution.

In some experiments, bacterial suspensionswere added to myxamoebae growing on the com-plex medium; in others, ferritin or colloidal goldsolutions were incorporated in the medium or thebacterial suspensions, usually 4 hr prior to the

755

on April 10, 2019 by guest

http://jb.asm.org/

Dow

nloaded from

J. BACTERIOL.

time of fixation. Ferritin from horse spleen (Pen-tex Inc., Kankakee, Ill.) was used in a final con-centration of up to 25 mg/ml of medium. Colloidalgold solution (Magar Chemicals, Inc., CornwallLanding, N.Y.) of undetermined concentrationwas employed in concentrations of up to 1 part ofgold solution in 2 parts of medium.

Suspensions of myxamoebae were fixed by addi-tion of an equal volume of 2% phosphate-buffered(pH 7.4) glutaraldehyde; the suspension was im-mediately centrifuged and the myxamoebae wereresuspended for 2 hr in fresh fixative. After re-peated washings in phosphate buffer (0.016 M,pH 6.25) the cells were postfixed in 1% Veronalacetate-buffered osmium tetroxide at a pH of 7.4for another 2 hr. Stepwise dehydration in acetonewas followed by propylene-oxide treatment andsubsequent embedding in Maraglas. Polymeriza-tion took place at 60 C for at least 2 days. Thismethod proved to be superior to all other pro-cedures tried, including fixation in potassiumpermanganate; also, addition of CaCl2 to thefixative did not change the quality of the pic-tures appreciably. Embedding in Epon-aralditeor Vestopal is satisfactory, but does not quiteequal that in Maraglas.

Sections were cut with a LKB ultratome or aPorter-Blum MT2 ultramicrotome and wereobserved in a Hitachi HU-11A electron micro-scope or, occasionally, in a Norelco 75 electronmicroscope.

RESULTSLamellar configurations and fine structure of in-

dividual membranes. The membrane complexesare most numerous in the food vacuoles of myx-amoebae actively feeding on bacteria (Fig. 1).They occur less frequently in aggregated myx-amoebae, but can be found even in the culmina-tion stage. They also have been observed outsidethe cells either closely attached to the plasmamembrane, or between aggregated cells, or freein the medium, as has been described by Gezelius(1961). This strongly suggests that they are ex-pelled -from the cells.The great variety of arrangements of lamellae

found can be satisfactorily grouped into a fewtypes. Yet, although this grouping accounts formany observations, it is only an approximationand does not necessarily include all possible ar-rangements.The most prominent type consists of a set of

closely packed, more or less spherical lamellaewhose outside diameter approximates the size ofthe food bacteria (Fig. 1 and 2). The number ofmembranes per set varies considerably, but isoften about a dozen. These sets of lamellae havebeen described by Mercer and Shaffer (1960) andGezelius (1961), but these authors did not makea decision as to whether the lamellae are concen-

tric or form tight spirals. From the observationspresented here, it is clear that they consist of con-centric lamellae (Fig. 2, 3). Spirals have been seenoccasionally, but under different circumstances:if a myxamoeba engulfs another myxamoeba, ashappens not infrequently (Huffman and Olive,1964), the plasmalemma might break and roll upto a certain degree (as in Fig. 4). In most cases,however, these tightly packed lamellae representsets of concentric membranes comparable to thelayers of an onion. The sets may or may not har-bor a core of amorphous material, probably rem-nants of the bacteria.

Often these sets of concentric lamellae occurnot singly, but in clusters, which themselvesmight be surrounded by concentric lamellae (Fig.5-7). Usually there is one cluster to one food vac-uole. In several cases, even more complicatedforms have been observed: lamellae of two setscan cross each other, resulting in a honeycombtype of network which is restricted to a relativelysmall area (Fig. 5-8). The structures resemble thewell-known interference pattern developing on awater surface between neighboring sets of stand-ing waves.Although the membranes are closely and fairly

evenly packed, each individual membrane is stillclearly separated from the other by a space ofapproximately 200 A. In a few instances, how-ever, they have been observed as stacks withoutany space between membranes, thus giving anappearance similar to myelin sheaths (Fig. 9). Itis noteworthy that in this figure the membranesare not closed, but have open ends. Open endsare not rare (Fig. 10), and in some instances maybe due to breakage during fixation or embedding.In other instances, however, the membranes arejoined at their open ends to a mass of amorphousmaterial and give the appearance of incompletespherical membranes (Fig. 11).The individual membrane of the various com-

plexes mentioned is 75 to 85 A thick and showsthe familiar triple-layered structure of a unitmembrane (Fig. 12). The two dark layers enclo-sing the lighter middle layer are of equal thick-ness, thus giving the membrane a symmetricalappearance. This is in contrast with the mem-brane lining the food vauole, which is stronglyasymmetrical, the dark layer on the inside of thefood vacuole being much more electron-densethan is its counterpart bordering the cytoplasm.It can thus be assumed that the unit membranesmaking up the complexes have a component ofpolar lipids and are most probably lipoproteins.The dark layer may exhibit a somewhat beadedappearance (Fig. 12) similar to that described by

756 HOHL

on April 10, 2019 by guest

http://jb.asm.org/

Dow

nloaded from

FIG. 1-3. Sets of concentric membranes in the food vacuoles of myxamoebae of Dictyostelium discoideum.Fig. 1, X 30,400; Fig. 2, X 69,000; Fig. 3, X 99,700.

757

on April 10, 2019 by guest

http://jb.asm.org/

Dow

nloaded from

(rerp ..e

$i /'.

FIG. 4. Food vacutole of a ?nyxamoeba of Dictyostelium discoideum containing reninants of anotheramoeba. Encircled area show.' rolled-up membrane. X 28,200.

FIG. 5. Complex toenmbranrc systems of Dictyosteliuzm discoideumnl with honeycotmb -pattern encircled.X 63,000.

.47 0.

'4

,M1.,... 4

IP-,'ai.

.... .!,..1. *:. roo

.,P, "mW.

on April 10, 2019 by guest

http://jb.asm.org/

Dow

nloaded from

FIG. 6. Complex membrane systems of Dictyostelium discoideum with homeycomb-pattern encircled.X 51,000.

FIG. 7. Detail of Fig. 6. X 102,500.FIG. 8. Membrane crossing other membranes (in circle). X 43,500.FIG. 9. Myelinoid structure in food vacuole of Polysphondylium pallidumi. X 103,100.FIG. 10. Parallel set of membranes with open ends, Polysphondylium pallidum. X 49,600.FIG. 11. Set of lamellae with open ends connected to amorphous material, Dictyostelium discoideum.

X 29,000.

759

on April 10, 2019 by guest

http://jb.asm.org/

Dow

nloaded from

i

FIG. 12. Part of Fig. S at higher magnification. X 290,000.FIG. 13. Food vacuole of Polysphondylium pallidum lined with an additional membrane. X S2,800.FIG. 14. Part of myxamoeba of Polysphondylium pallidum with food vacuoles containing one or more

bacteria. X 46,400.FIG. 15. Food vacuole of Polysphondylium pallidum containing half-digested bacterium surrounded by

several membranes. X 46,400.

760

1H

on April 10, 2019 by guest

http://jb.asm.org/

Dow

nloaded from

L99EMBRANE SYSTEMS IN CELLULAR SLIME MOLDS

Slautterback (1965) for mitochondrial membranesof heart muscle cells.

Development of lamellar systems. To study theinitial formation of the lamellae, myxamoebae ofP. pallidum were cultivated on the soluble me-

dium. When the cell density reached 106 to 2 X106 myxamoebae per milliliter, a heavy suspen-sion of E. coli was added. At different time inter-vals thereafter, samples of myxamoebae were

prepared for electron microscopy.In myxamoebae grown on the medium alone,

very few membranes were found in the food vac-

uoles. These had a tendency to lie close to thevacuolar membrane. Since cannibalism is not in-frequent in these organisms (Huffman and Olive,1964), the occurrence of these lamellae is not sur-

prising.

Engulfment of bacteria occurs singly or ingroups (Fig. 14). The bacteria are either tightlyenclosed by the membrane of the food vacuole or

are more or less free within a rounded food vac-

uole. As the first sign of digestion, the bacteriaswell up (Fig. 10 and 14). Lysis of the cells is ac-

companied by a gradual loss of electron-densematerial, such as ribosomes. Plasma membraneand cell wall are much more resistant than thecytoplasmic contents and do not seem to be de-graded appreciably. During this time of cell deg-radation, the membranous systems begin toform. Judging from a large number of pictures,the assembling of the membrane material occurs

mainly in three ways: (i) formation adjacent tothe membrane of food vacuoles; (ii) formationadjacent to bacterial envelopes; or (iii) spontane-ous formation from amorphous pools of precui-soimaterial.

(i) Whereas in myxamoebae feeding on the sol-uble medium the membrane of the food vacuoleis clearly a single asymmetric unit membrane,upon the addition of bacteria it seems often toconsist of two or even more membranes (Fig. 13,16, and 17). Gezelius (1959) mentioned that thefood vacuoles are sometimes lined with a doublemembrane, sometimes with a single unit mem-

brane. All our observations indicate that theseadditional membranes are formed on the mem-

brane lining the food vacuole during the digestionof bacteria. The use of ferritin or colloidal gold as

electron-dense markers l)rovides some additionalevidence: the marker, added to the bacterial sus-

pension upon which the myxamoebae feed, isfound preferentially towards the center of thelamellar sets (Fig. 18), indicating that at leastsome of the lamellae oriiginate from the peripheryof the vacuole. The possibility that these mem-

branes originate from repeated shedding of themembrane of the food vacuole can almost cer-

tainly be excluded, since the additional mem-branes are symmetric, in contract with the mem-brane of the food vacuole, and the same mecha-nism should work when the myxamoebae grow onthe soluble medium. (However, this is not thecase, since, under those conditions, membranesare found only rarely.)

(ii) Membranes also appear adjacent to themembranes surrounding the bacterium. Since thecell wall of E. coli appears in the electron micro-scope as a triple-layered structure similar to a unitmembrane, it cannot be decided whether the la-mellae form initially inside the plasma membraneor on the outside of the cell wall. However, thefact that the sets of concentric lamellae are nor-mally the size of a bacterium indicates that mostof the lamellae are produced inside the plasma-lemma. Figure 15 shows an early stage of mem-brane formation. As lysis of the cells proceeds,progressively more lamellae are formed, even-tually replacing the whole bacterial cytoplasm.It seems as if each lamella formed mediates theformation of the next lamella.

(iii) Several observations indicate an additionalmechanism of membrane formation. First, areaswithin the food vacuoles give the appearance ofintermediate stages in membrane formation. Thewhole area consists of a mosaic pattern of smallpieces of lamellae intermingled with amorphousmaterial. The short pieces of lamellae are ar-ranged in a way already suggesting the outline ofthe future lamellae (Fig. 3, 19, and 21). The pointhere is that a whole area of precursor materialseems to be transformed spontaneously and si-multaneously into a set of lamellae. This wouldbe in contrast with the previous two cases, wherethe lamellae are formed progressively or stepwise.The second observation, pointing in the same

direction, refers to the honeycomb pattern de-scribed earlier. If, on a water surface, standingwaves from two different sources meet, an inter-ference pattern develops that resembles the ar-rangement observed in Fig. 5, 6, and 7. Thiswould indicate, in our case, that the impulse forlamella formation has started at two points andthat the pool of precursor material is under theinfluence of the two sources while the membranesare being formed. In other words, the moleculesin the interference area arrange themselves ac-cording to the pattern of forces created by thetwo sources, the resulting membranes thus beingtheir visible expression. A mechanism assuminga stepwise formation of lamellae could hardly ac-count for a phenomenon of this type.

Figure 23 summarizes the development of themembrane complexes as suggested by our obser-vations. The diagram does not include the pos-sibility of growth of these membranes by intus-

761VOL. 90,7 1965

on April 10, 2019 by guest

http://jb.asm.org/

Dow

nloaded from

II& i.:n-

;.

:.... f

"i' b ,.,s,,1iks w o+\. ig*t

'-. , .>y^s,: .A! d-:t* Si i i a B siF

i 's *;.9F S .. w.,

W 16 it

0

I.4.4" A..

44k

o-. rr..!. P-. ..e

.1

I

FIG. 16-17. One to several membranes lining food vacuoles of Polysphondylium pallidum containingingested bacteria. X 46,400 and X 20,800, respectively.

FIG. 18. Food vacuoles of Dictyostelium discoideum containing membranes and ingestedferritin. X 58,400.FIG. 19. Set of incomplete membranes in food vacuole of Dictyostelium discoideum, interpreted as inter-

mediate stage of membrane formation. X 72,500.FIG. 20. Cytoplasmic protrusion into food vacuole of Polysphondylium pallidum. X 43,500.FIG. 21. Structures similar to those in Fig. 19. X 116,000.FIG. 22. Lamellar protrusion into food vacuole of Polysphondylium pallidum. X 24,500.

762

::.?i. jpf%.,tp

;4,k :.-

s: 1. .1, ., 0zi,i

Xllii.

T::

i"'IJ-1'l:....

on April 10, 2019 by guest

http://jb.asm.org/

Dow

nloaded from

0MEMBRANE SYSTEMS IN CELLULAR SLIME MOLDS

FIG. 23. Diagram)?1 of proposed nmentdrane formtaa-tion daring digestion of bacteria in food vacuoles oJ'cellular slime molds. Each line represents a unitmembrane. The seq?ence starts with the presence ofbacteria in the food vacueole (stage A). T'he bacteriaswell up and begin to ityse, releasing breakdownproducts (stage B). Suibsequent membrane formtia-tion, most probably uising the breakdown productsof the bacteria as precarsor nmaterial, proceeds in atleast two ways, depicted as C-D and E-F, respec-tively. In C-D, many lamlellae form spontaneouislyand simultaneously; in E-F, lamellae are fornmedin sequence through temtlplate action of pre-existingmiembranes, as described in the text.

susceltion, that is, enlargemenent by incorporatingmembrane material into existing membranes. Al-though this might occur, wve have no clear indica-tions of it.

Other observations o07 the pro(cess of digestion. heprocess of engulfment and ingestion of food ma-

terial does not seeml to be extremely sl)ecific,since cannibalism is quite frequently observed(Huffman and Olive, 1964). So far, we have con-

sidered only the intake of E. coli. If myxamoebaeare engulfed, the process of digestion is similar atthe beginning. The cytoplasmic matrix with theribosomies is degraded, whereas the membranouselements of the cell, e.g., the l)lasmalemma, themitochondria, and also the concentric rings withinthe food vacuoles of the engulfed myxamoeba, arefairly well preserved (Fig. 4). The sets of concen-

tric lamellae characteristic for the digestion ofthe bacteria have not been observed to form dur-ing digestion of myxamoebae.The subsequent fate of the degraded, ingestible

material in the food vacuole is obscure. Few mom-

phological observations were made that mightprovide some clues.

Sonmetimes an invagination of the cytoplasnminto the food vacuole has been observed. This in-vagination can be either l)lain groundplasm or

mav be more lamellar in nature (Fig. 20, 22). On

other occasions, a zoine of cytoplasm clear of' ribo-somies surrounds the foodl vacuole (Fig. 4).Strands of the endoplasmic reticulum or othermembranes might l)arallel the membrane of thefood vacuole, but this, too, does not occur- fre-quently enough to be significant. There are noobvious group)s of small xesicles around the foodv-acuole to suggest a process of micropinocytosis.Small breaks in the membrane of the food vacuoleand in other membranes, as reported by Gezelitus(1961), were observed blut are considered artifactsof fixation.

I)mSCUSSION

I)igestion of E. (oli by the mnyxamoebae of D).discoideumi and P. pallidumi begins with a swellingof the bacterial cell, followed by gradual lysis ofthe cytoplasmic contents. Fragm-ientation of thecell does not seem to occur in E. coli, but has beenobserved during digestion of the much larger Ba-(illus megaterium by Raper (1937). Although thebacterial cell wall can break open, it seems to bemost resistant to the digestive process and re-tains its morphological distinctiveness. These ob-servations are substantiated by Rosen, Rosen,and Horecker (1965), who found that the l)oly-saccharide component of the cell-wall lipopoly-saceharide of Salmonella typhimulrium is not sig-nificantly altered by P. pallidum and is exeretedquantitatively into the culture medium.Two fractions result from the digestive process:

a digestible and an undigestible one. The subse-quent fate of the digestible material is hard totrace; very few morphological l)eculiarities havebeen observed that might be related to the uptakeof the digested products into the host cytoplasmii.It seems most likely that the iiaterial permeatesthe intact membrane of the food vacuole, sinceop)timally lpreserved cells show few openings inthe membrane, and these are most probably arti-facts. The halo depr-ived of ribosomes observedaround the food vacuole might be due to the largeamount of digested miaterial entering the cellcytoplasm; it could also be due to accumulationof digestive enzymes. The role of the cytoplasmicintrusions into the food vacuoles might be placedin the area of resorl)tion, but here again we knowvery little, since tracers such as ferritin and col-loidal gold could not be folloNed into the cell ev-toplasm.

Of the material that is not digested by the myx-amoebae, the lamellar systems described in detailare by far the most conspicuous ones. Theirpresence in large amounts is surprising, since theyseem to be lost for any nutritional purposes andare exereted into the surrounding medium. Thus,a large amouint of the material taken in by the

1 ()I,. 90, 1965 /7(;3

on April 10, 2019 by guest

http://jb.asm.org/

Dow

nloaded from

J. BACTERIOL.

myxamoebae is wasted, which seems to renderthe digestive process fairly inefficient.These sets of concentric lamellae are not re-

stricted to the cellular slime molds, but have beenobserved in the plasmodial slime molds (Schuster,1964) and a variety of other cells. Focal degrada-tion in mammalian cells, as described by Swiftand Hruban (1964), leads to formation of some-what similar membranous bodies. Thus, the phe-nomenon is widespread among biological systems.

In our system, a pool of precursor material isformed by the digestion of the bacterial cyto-plasm. This step is followed by the assembling ofthe material into membranes. Membranous ma-terial was only rarely observed in food vacuolesof myxamoebae growing in the absence of bac-teria. Thus the composition of the bacteria, par-ticularly the lipid components, must play an im-portant role. Apart from the bacteria, precursormaterial could possibly be provided by the hostcell, e.g., in the form of digestive enzymes. How-ever, no evidence for this has been obtained sofar.The membranes formed have a three-layered

configuration typical for unit membranes. Fromthis it is concluded that most probably they arebuilt from lipoproteins. They are symmetrical innature, the bimolecular lipid leaflet being lined oneither side by a protein layer of identical thick-ness. Our observations suggest two main ways ofmembrane assembly: (i) the membranes form ad-jacent to pre-existing membranes, such as themembrane lining the food vacuole and the cellwall of the bacterium; and (ii) the membranesform spontaneously from a pool of precursor ma-terial. Spontaneous formation of membranousmaterial from mixtures of lipids or mixtures oflipids and proteins have been studied intensively(Stoeckenius, 1962; Luzzati and Husson, 1962).Little attention has been given to a p)ossible roleof pre-existing structures in determining the con-figuration of newly forming membranes.Our results suggest that, given a membrane and

a pool of precursor material, the new membranetends to be formed through mediation of the pre-formed membrane. In other words, the preformedmeImbrane serves as a pattern or template for theprecursor material, the newly formed membranelying l)arallel to it and taking ul) the same shape.The term "template" is used here in a wide sense,such that the resulting new meimebrane would nothave to be a copy of the old one. That this is thecase in our system is obvious from the fact thatthe membrane lining the food vacuole is stronglyasymmetric, in contrast with the symmetricalmembranes forming upon it. It is reasonable toassume that the rather unspecific template action

is due to preferential adsorption of the hydro-philic parts of the lipoproteins by the proteinlayer of the pre-existing membrane. The hydro-phobic ends of the lipoproteins would thus pointaway from the pre-existing membrane, therebyattracting hydrophobic ends of lipoproteins fromthe pool, and so on. This process, repeating itself,eventually leads to the formation of the lamellarstacks. This mechanism could explain the com-plex patterns observed: the set of concentric la-mellae reflecting the form of the bacterial enve-lope, and the lamellae surrounding several setsreflecting the form of the membrane lining thefood vacuole.Morre and Mollenhauer (1964) found sets of

concentric lamellae in isolated Golgi apparatusfrom plant cells, and Mercer (1962) discussed thepossibility that the Golgi body in the cells of theovotestis of the snail might arise from sets of con-centric lamellae formed from lipoid droplets. Thesets of concentric lamellae would then break up,eventually leading to the well-known arrangementof stacked lamellae in a Golgi body. The relationbetween this example and the mechanism dis-cussed for the formation of the membranes in thefood vacuoles is obvious. The question is, doesthis mechanism have much wider application, andshould it be considered in other cases where stacksof lamellae develop? The idea is not entirely new.Waddington (1962), including earlier suggestionsby Rebhun (1956) and by Swift (1956), discussedthe possibility that stacks of annulate lamellaein various kinds of cells are formed by some typeof copying mechanism from the nuclear envelope,since they bear strong resemblance to it.The idea that membranes can be formed -roi1

a pool of precursor material through template ac-tion by pre-existing membranes might, in somecases, be a fruitful alternative to the prevailingview that membranes form by fusion of vesicleslined with preformed membranes (Blondel andTurian, 1960; Humphreys, 1964; Kessel, 1963;Schuster, 1964).

ACKNOWLEDGMENTSThe assistance of Susan Hamamoto is grate-

fully acknowledged.This investigation was supported by Public

Health Service grant GM 11758-1 from the Divi-sion of General Medical Scienees and by granitGB-1938 from the National Science Founidatioin.It is contribution Ino. 63 from the Pacific Biomedi-cal Ilesearch Ceinter.

LITERATURE CITED

BLONDEL, B., AND G. TURIAN. 1960. Relationt be.tween basophilia and fine structure of cytc.-plasm ill the fungus Allo7ryces macrogynus EnIi.J. Biophys. Biochem. Cytol. 7:127-134.

764 HOHL

on April 10, 2019 by guest

http://jb.asm.org/

Dow

nloaded from

MEMBRANE SYSTEMS IN CELLULAR SLIME MOLDS

GEZELIUS, K. 1959. The ultrastructure of cells andcellulose membranes in Acrasiae. Exptl. CellRes. 18:425-453.

GEZELIUS, K. 1961. Further studies in the ultra-structure of Acrasiae. Exptl. Cell Res. 23:300-310.

HOHL, H. R., AND K. B. RAPER. 1963a. Nutritionof cellular slime molds. I. Growth on living anddead bacteria. J. Bacteriol. 85:191-198.

HOHL, H. R., AND K. B. RAPER. 1963b. Nutritionof cellular slime molds. II. Growth of Poly-sphondylium pallidum in axenic culture. J.Bacteriol. 85:199-206.

HOHL, H. R., AND K. B. RAPER. 1963c. Nutritionof cellular slime molds. III. Specific growthrequirements of Polysphondylium pallidum. J.Bacteriol. 86:1314-1320.

HUFFMAN, D. M., AND L. S. OLIVE. 1964. Engulf-ment and anastomosis in the cellular slimemolds (Acrasiales). Am. J. Botany 51:465-471.

HUMPHREYS, W. J. 1964. Electron microscopestudies of the fertilized egg and the two-cellstage of Mytilus edulis. J. Ultrastruct. Res.10:244-262.

KESSEL, R. G. 1963. Electron microscope studieson the origin of annulate lamellae in oocytes ofNecturus. J. Cell Biol. 19:391-414.

LUZZATI, V., AND F. HUSSON. 1962. The structureof liquid-crystalline phases of lipid-water sys-tems. J. Cell Biol. 12:207-220.

MERCER, E. H. 1962. The evolution of intracellularphospholipid membrane systems, p. 369-384.In R. J. C. Harris [ed.], The interpretation ofultrastructure. Academic Press, Inc., NewYork.

MERCER, E. H., AND B. M. SHAFFER. 1960. Elec-tron microscopy of solitary and aggregatedslime mould cells. J. Biophys. Biochem. Cytol.7 :353-356.

MORRP,, D. J., AND H. H. MOLLENHAUER. 1964.Isolation of the Golgi apparatus from plantcells. J. Cell Biol. 23:295-305.

RAPER, K. B. 1937. Growth and development ofDictyostelium discoideum with different bac-terial associates. J. Agr. Res. 55:289-316.

REBHUN, L. I. 1956. Electron microscopy of baso-philic structures of some invertebrate oocytes,I. and II. J. Biophys. Biochem. Cytol. 2:93-104,159-170.

ROSEN, 0. M., S. M. ROSEN, AND R. L. HORECKER.1965. Fate of the cell wall of Salmonella typhi-murium upon ingestion by the cellular slimemold: Polysphondylium pallidum. Biochem.Biophys. Res. Commun. 18:270-276.

SCHUSTER, F. 1964. Electron microscope observa-tions on spore formation in the true slime mold,Didymium nigripes. J. Protozool. 11:207-216.

SLATUTTERBACK, D. B. 1965. Mitochondria incardiac muscle cells of the canary and someother birds. J. Cell Biol. 24:1-22.

STOECKENIUS, W. 1962. The molecular structureof lipid-water systems and cell membranemodels studied with the electron microscope,p. 349-367. In R. J. C. Harris [ed.], The inter-pretation of ultrastructure. Academic Press,Inc., New York.

SUSSMAN, M. 1963. Growth of the cellular slimemold Polysphondylium pallidum in a simplenutrient medium. Science 139:338-339.

SWIFT, H. 1956. The fine structure of annulatelamellae. J. Biophys. Biochem. Cytol. 2(suppl.):415-418.

SWIFT, H., AND Z. HRUBAN. 1964. Focal degrada-tion as a biological process. Federation Proe.23:1026-1037.

WADDINGTON, C. H. 1962. New patterns in geneticsand development, p. 62-63. Columbia UniversityPress, New York.

VOL. 90, 1965 765

on April 10, 2019 by guest

http://jb.asm.org/

Dow

nloaded from