CALIFORNIA STATE UNIVERSITY, NORTHRIDGE A FINE STRUCTURAL AN~~YSIS OF SURFACE ~' INTERACTIONS OF AGGLUTINATED SARCOMA,l80 CELLS A thesis submitted in partial satisfaction of the requirements for the degree of Master of Science in Biology by Rudolf Carl Kuppers, Jr. January, 1976 The thesis of Rudolf Carl Kuppers, Jr. is approved: California State University, Northridge December, 1975 -__________.JI ii ,------------------------------------------------------------- ----------------- ------------------------------------------ ----l I I ACKNOWLEDGEMENT I would like to acknowledge my committee for their patience before and during the writing of this thesis. In particular, Dr. Edward Pollock for his continued inspiration and friendship; Dr. Steven Oppenheimer for his unceasing energy; and Dr. Marvin Cantor for his continued friendship. In addition, I \vould like to thank Hr. Geoff Brennerman, Miss Miriam Burton, and Miss Frederica Witzke for their help in this \vork. iii ,~ --- .. ··-·-·--- ·-- - - --·-·-- ~---------- ------~ --------------~-----~---------------------1 I iii I II iv I1 Table of Contents !I Acknmv-ledgment .I I I I Table of Contents Abstract v Introduction l Methods and Materials 7 Maintenance and preparation of Sarcoma 180 7 Viability Test 7 Electron Micros6opy 7 Lanthanium Oxide Staining 8 Ruthenium Red Staining 9 Concanavalin A Agglutination 9 Factor Agglutination 10 Manganese Ion Agglutination 11 12 Results Fine Structure of Sarcoma 180 12 Concanavalin A Agglutination 15 Factor Agglutination 19 Manganese Ion Agglutination 21 Discussion 24 :Bibliography 37 Table 44 Plates 45 ; I I L_____ _ --- -- ... ___ _J iv r·--~- . --- - - - - - - - - - - - - -------------~-------- -------------------------.---------- - - - - - - - - - - - - - -------- --~~~ ~- - ~-~ II ABSTRACT I l AN ULTRASTRUCTURAL ANALYSIS OF THE FINE STRUCTURE AND SURFACE INTERACTIONS OF AGGLUTINATED SARCOMA 180 CELLS by Rudolf Carl Kuppers, Jr. Master of Science in Biology January, 1976 Sarcoma 180, a mouse ascites tumor, was examined ultrastructurally using transmission electron microscopy. Its general fine struc- ture, and the cell surface interactions occurring as a result of agglutination were examined. Three types of agglutinins were used, Concanavalin A (Con A), an ascites "factor", and manganese ion. Ultrastructurally, Sarcoma 180 had the features of many tumors: A large heterochromatic, multilobate nucleus with numerous nucleoli, reduced amounts of endoplasmic reticulum, and numerous microvilli on the cell surface. The mitochondria were somewhat pleomorphic. Large lipid droplets were seen in addition to several other large, darkly stained bodies in the cytoplasm. Cell surface interactions between untreated cells were not commonly seen. Three surface interactions were seen after agglutination: 1) Microvillous associations characterized by the interaction of villi v r~-------···---·------------- --~--·--· -------------·----------·---------------.. ----------------------1 I ! I I I ! I !I I I I with adjacent cell surfaces; 2) intermediate associations in which ., adjacent cell membranes were approximately parallel, crosslinked by I patches of lightly stained material, and separated by greater than I ! 0 I I I . I lOOA; and 3) close association in which the cells were separated by 1 I 0 0 less than lOOA, commonly 20A, and resembled gap junctions. All three types of associations were seen after Con A agglutination. The intermediate association may represent actual cross- linking by Con A aggregates. The close associations, it was argued, resulted from interactions between cell surface components. Micro- villous associations were most commonly seen, though their significance was not clear. The ascites "factor" at high concentration induced pinocytosis and vacuolization. Agglutination resulted from large aggregates of flocculent material crosslinking the cells. At lower concentration, intermediate associations were seen, small aggregates of "factor" possibly crosslinked the cells together. Manganese ion produced a very rapid agglutination mediated totally by villous interactions. Points of interaction were very tight, with no obvious gap bet·ween membranes. It was thought that ionic bridging, or denaturation of surface components was involved. ·- L ..,--.--·- vi l --~~--~-·-J Introduction In recent years several lines of investigation have shown that the surfaces of transformed and malignant cells are different from those 6f "normal'' cells. These differences include biochemical al- terations, changes in cell-cell adhesion and of cell surface charge, and agglutinability, Gasic and Gasic (1963, 1962) provide the first convincing evi- I deuce of a carbohydrate-rich layer external to the plasma membrane of ' mammalian cells. the 1 This cell coat was conceptualized for all cells as glycocalyx 1 by Bennett (1963). The work of several investiga- tors seems to confirm that the cell coat. of neoplastic cells differs from that found on normal cells. Both increases and decreases of various constituents in the glycocalyx have been reported (Buck, et al., 1970; Brady, et al., 1969; Martinez-Paloma, et al., 1969; Meezan, et al., 1969; Mora, et al., 1969; Wu, et al., 1969; Morgan, 1968; Ohta, et al., 1968; Kraemer, 1966; Purdom, et al., 1958). Coman (1944) demonstrated a general decrease in adhesiveness bet\veen epithelial tumor cells in comparison with normal epithelium. This reduction in adhesion has been postulated to account for the I change in "social" behavior demonstrated by transformed cells .. Trans-\ I formed cells show a loss of contact inhibition of growth and mobility! ' (Dulbecco, 1970; Abercrombie and Ambrose, 1962; Vogt and Dulbecco, . - - ···-· ,-~---·-- ------. ---- --· ~~--- --·-- ---- --- .. --- --·. --~----------------------------·-. 1 -·· ---- -------------- --· - -- ------------------ -----~-·--· --~-----·. I -----l r-·----------------· ----·------···- - - - --·----------·-· ----------------·-· . --------------·---.. ------------------------. I This change in adhesion is not a simple change reflected in i 1962). ! Ian overall decrease II specifically in adhesion to other cells and substrates, but cell-cell adhesion. Indeed, it can be shown that some I i transformed cells stick to other cell types and substrates quite well; 1 for example glass, plastic, or millipore filters (Ambrose, 1967). I I This change in adhesion may be the cause of metastatic behavior in I I tumors ... ! It has been suggested that the change in cell surface components Ii leads to a greater number of negative charges and therefore greater Irepulsion between tumor cells. This is coupled with decreased adhesI ion (Abercrombie and Ambrose, 1962). 1 I (1958) have shown that Isublines of MCIM mouse ! so 1 id Purdom, Ambrose, and Klein the potential at the cell surface of different sarcoma increased as they progressed from the tumor to the ascites form. \Vhile these studies are i I I adhesion, intensive research lI mechanism(s) suggestive in the understanding of cell has failed to reveal the nature of the involved in adhesion and their relationship to tumor be- l lhavior and growth. ! One of the most exciting areas in membrane research in recent I !years has involved the use of plant lectins or agglutinins to probe l i the structure of the cell surface. Lectins are proteins or glycoI l ' j proteins with highly specific, differential saccharide-binding capa- I I I II i l' bilities. i f Exampl~s are Concanavalin A (Con A) which binds a-D- I imannopyranosyl, a-D-glucopyranosyl and sterically related terminal I l I isugar residues (Agrmval and Goldstein, 1967), and wheat germ agglutinid ...... ----------·•"""- ··---···------ ........J 3 ,---------------·--------------· ---------·-····-----------------------------·--------------"---------··--- ------1 ! ! I II (WGA) which binds N-acetyl-D-glucosamine and similar groups (Burger I !and Goldberg, 1967). Il creased I I cells The work of Burger and Goldberg, shmving in- agglutinability by transformed cells as compared with normal - using WGA, provided a stimulus for the use of lectins in the ! !vestigation of tumor cell surfaces. I in- I In general, it has been shown I that transformed cells agglutinate with much lmver concentrations of I lectin than needed for the corresponding ! al., 1971; Inbar and Sachs, 1969; Burger normal cell line (Sela, et and Goldberg, 1967). I I It was originally thought that the increased agglutinability I demonstrated by I bers of, or the transformed cells with Con A was due to increased num-1 i unmasking of "cryptic", Con-A-binding sites. A large I j amount of evidence now suggests these interpretations are not correct i I and, in fact, nearly the same amount of Con A \vill bind to normal ' i ! cells as will bind transformed cells of a particular cell line (Arndt! ! IJavin and Berg, 1971; Cline and Livingston, 1971; Ozanne and Sambrook, I 11971; Sela, et al., 1971). ! I The relative agglutinability of a cell has now been correlated with the topological distribution of Con A bound to the cell surface. l In normal cells the Con A binding sites appear to be randomly dis- ! persed \vhile these sites are clustered on the transformed cell (Nicol- I I son, I 1971). The concentration of binding sites in these clusters ap- , 11 i : pears to promote crosslinking betHeen cells by the tetravalent Con A re~·ulting in cell agglutination. Apparently, Con A binding sites · have an increased freedom of movement in transformed cells allowing these sites to cluster under the influence of Con A ............ __ ..... . ............... - -- ............................ ------ ·- .... ----. -- . . (Nicol~on, -----· --.------- - 1973; 1 I !i i .. -- .. --· ....... --_I 4 Rosenblith, et al., 1973). Agglutination can be induced by other mechanisms as well. For example, there have been some studies in which poorly defined "factors" from the ascites fluid of tumor bearing mice have been used to bring about agglutination. In the case of a mouse teratoma, this factor was shown to be possibly a glycoprotein and to possess some I gree of specificity for the cell line from which it was isolated (Oppenheimer and Humphreys~ 1971). That these "factors" have any im- portance in cell-cell adhesion is unknown. Specific metal ions such as manganese and lanthanium have been shown to increase cellular adhesion to substrates and to cause cell agglutination. ! I de-l But again, the mechanism of their action or its bio- logical significance has yet to be demonstrated (Rabinovitch and De I I I l I . Stefano, 1973.; Levinson, et al., 1972; Oppenheimer, personal communi- 1!! cation). There are several structural specializations which can occur normally between cells and hence are implicated in cell adhesion. These associations are called junctional complexes. They involve in- tercellular, plasma membrane, and cytoplasmic specializations. These II ! ! structures are found universally throughout the animal kingdom. Hence, I they appear to have an important role in cell-cell associations, I though what this role is has not been completely elucidated (McNutt and Weinstein, 1973). Junctional complexes are classified into four major types: 1) the! l zo ~~-~~ ·- ~~~_:_r_~~~---~2- __:~~~~~-~-~~-~-=~~~--_?-~--~t!Xt~ ~-'----~ ~---~~-=--? =-~~?. s ~~-= -'---~~-~---~ 5 4) the tight junction or zonula occludens (see McNutt and Weinstein, 1973 for a complete review). There is some evidence to suggest these structures are involved in adhesion between cells. The use of hy- pertonic media (Brightman and Reese, 1969), or the incubation of cardiac muscle in Ca++-free media (Muir, 1967), reveals tight june- l ! tions, gap junctions and/or desmosomes which remain after these treat-1 ments, keeping the cells attached. The gap junction has been impli- I cated as· the pathway for electrical coupling and the passage way of small molecules between cells. the 11 It has been suggested that some of social 11 behavior of cells may be mediated by communication through this type of junctional complex. (Bennett, 1973; Lowenstein, 1973). In general, a relationship appears to exist between the number of junctional complexes and cell normalcy, tumor cells having fe>ver junctions or none at all (McNutt and 1-leinstein, 1973; Martinez1 Palomo, 1971, 1970; McNutt, et al., 1971; Beneditti and Emmelot, 1968; Emmelot and Beneditti, 1967). The fact that solid tumor cells have reduced numbers of junetions suggests a possible change in the plasma membrane which could i.n turn be responsible for differences in the adhesion and behavior of tumor cells. 11 soci.al 11 One could depict the tumor cell surface as being modified in such a way that the components which mediate the formation of a particular type of junction are missing, in which case one would not expect junctional formation under any set of conditions unless these components were regenerated at the cell surface. i I I ---···-··· ...... ...J 6 r------------------···------------------------·-·-----------·-···-------·---------------~--- ------------------~ I lon the other hand, these components could be present within the cell lbut are not able to interact because of any of several reasons, such I as greater repulsive forces between the cells, or cryptic burying of the components involved. 1 I1 ditions Ishorter 1 Obviously, if the proper environmental con- are met, the time required for formation of junctions would be 1 in the latter case since the components are present and do not l lneed to.be regenerated. I I 11 In vitro'' agglutination appeared to be a useful approach to I Iprobe the phenomenon of junctional complex formation in tumor cells; Iagglutination acting to bring cells in close contact in the presence Iof an exogenous component. I· Ln . t erac t.Lons In this study, the problem of membrane . a mouse ascL. t es t umor 1'Lne, Ln s arcoma 180 (S - 180) , was examined over short and extended periods in culture using thin section j and transmission electron microscopy (TEM) after agglutination by I Con A, an ascites "factor", and manganese ion. I I I,----- ----·-··-··-- ·-·-··---------·------------ ......- ..·--· ·-· -------···----·-------·-· ~-~----~ i Methods and Materials I Maintenance and ~- preparation of S-180. . The ascites form of S-180 was originally obtained by Dr. Steven I • /I Oppenheimer from Dr. Melvin Cohen of the Salk Institute and from Mr. I I Samuel M. Porley of the National Cancer Institute. ~ I by I intr·aperitoneal transfer every 4-6 days in Swiss··Webster, young, male mice. I ascites The abdominal skin \vas opened and the fluid was aspirated with a syringe bearing a 19 gauge needle 1 I I 1 I· '~i; Tumor bearing mice, inoculated 4-6 days prior, were kill-. ed by cervical dislocation. 1 It was maintained or by cutting open the peritoneum and pouring the ascites fluid dir- ectly into a plastic petri dish. I ! i j The cells were harvested and washed tHice by low speed centrifugation in Hanks' Balanced Salt Solution ,1 I I' (BSS) before use at an appropriate dilution. I l Viability Test. Viability was determined using the Nigrosin dye-exclusion test. The test \vas performed by mixing a sample of cells with an equal volume of a 1% solution of Nigrosin in Hanks' BSS. The cells \vere examined after five minutes under a light microscope using low power. Il I I Stained cells were considered dead. The percentage of dead cells was determined by counting at least 100 cells. Electron Microscopy. Fixation of cells was carried out in either a 1:1 dilution of Karnovsky's fixative (Karnovsky, 1965) Hith 0.1 M sodium cacodylate ____ j 7 l 8 ..·---·····-· r·~ ··--~ ................. ---~--.-- .........------------·-·. ·-·----·-·------·--···-··-----· .... ......____1 ·~--- il I I I I I buffer pH 7. 4 (Shea, 1971), (referred to as "modified Karnovsky 1 s 11 } ; 1 1 I i times I or with 2.5% glutaraldehyde in 0.06 M sodium cacodylate buffer at pH 7. 2-7 .4. Fixation \vas at room temperature for 2-4 hours and at continued overnight at 4°C. The cells were washed three times in I I O.lM cacodylate buffer pH 7.4 then postfixed in 2% osmium tetroxide in 0.1 M cacodylate buffer pH 7.4 for two hours. The cells ~vere then washed twice in cacodylat·e buffer and dehydrated in a graded series of acetone. Blocks were sectioned on a Porter-Blum MT2 Sor- vall ultramicrotome and stained with 2'7o uranyl acetate in 50% acetone and Reynold's lead citrate (Reynolds, 1963). Sections ~vere examined ! I I added to enhance staining of the glycocalyx. ! I 1 I Lanthanium Oxide Staining. Only the Cells were fixed in modified Karnovsky's fixative with 0.1% Alcian blue I II ! Con A-treated cells were examined using this staining method. I ! with a Zeiss EM 9S-2 electron microscope. Lanthanium staining was as described by Shea (1971). I Il The cells were embedded in an Epon-Araldite mixt:ure (Mollenhauer, 1964). i l Fixation was carried I I l l I out at room temperature for two hours. Postfixation was carried out in 1% osmium tetroxide and 1% lanthanium nitrate in 0.1 M cacodylate buffer. The final pH of the fix- ative was adjusted by gradually adding sodium hydroxide (0.1 M) to the solution while on a magnetic stirrer. Sodium hydroxide ·was added just until a small amount of precipitate was formed (pH about 7.4). Fixation was carried out for t\vO hours at room temperature. I 1 iI ! l' Dehydra- i . . . . . ···-------.. . . . -.. . . . . . . . . . . . . . . . ._. . . . . . . . . . . . . . . . . . . . . . ----··· . . . -.. . . . . . . . . . . . -.. . . . . . . . . . . . . . . . . . . __~. . . . . . . . . . . ..J 9 tion was carried out in a graded acetone series, 1% lanthanium nitrate being added to the lower dilutions in the series. Embedding was as before. Ruthenium Red Staining. Ruthenium Red (RR) staining was performed as by Luft (1971). Only the manganese-treated cells and untreated cells \vere stained in order to enhance their glycocalyx. A stock solution of RR was made by adding RR to distilled water (lOmg/ml) and heating to a boil. This solution was cooled, placed in a centrifuge tube and spun in a clinical centrifuge for ten minutes at full speed. Supernatant was removed and used as a stock solution. Staining was accomplished in the fixative by using a 1% solution of RR stock in 2.5% glutaraldehyde in 0.06 M sodium cacodylate buffer pH 7.3 The cells were fixed and stained for 4 hours at room temper- ature, then washed in 0.1 M cacodylate buffer followed by postfixation in 2% osmium tetroxide for 2 hours. bedding was as before. Dehydration and em- Sections were examined without any further staining. Concanavalin A Agglutination. Con A agglutination was performed according to the method of Oppenheimer and Odencrantz (1972). Equal numbers of washed cells were agglutinated with either a 500 or 1000 pg/ml solution of Con A in Hanks' BSS. The cells were suspended in 2 mls of the Con A solu- tion or in Hanks' BSS, in screw-capped scintillation vials. The I ·- --· ·-- - - ··- -~··---...J 10 i--·········-··-------~-----·-···---·····-·------···----------·--------------···-·--------·····--------····----, ! !vials were immediately placed on a rotary shaker having a 4-5/8" dia- l !meter of gyration and rotated at 65 rpm in a 37°C incubator for 15 - li60 minutes. At the end of this period, the cells were washed with several milliliters of Hanks' BSS and collected by slow centrifugaltion. The pellets were resuspended in a 1:1 mixture of Hanks' BSS !containing 5 pg/ml of deoxyribonuclease (DNase), and Eagles Minimal !Essential Media (MEM) without glutamine and buffered with 0.01 M IHepes, pH 7.4 The resuspended cells were then incubated in 25 mm !plastic tissue culture dishes for one and four hours at 37°C. The ! !cells were collected and fixed for electron microscopy with modified II !Karnovsky's fixative. I i iFactor Agglutination. I Ascites fluid from tumor bearing mice was collected as before. I i !Cells were removed by centrifugation and the supernate was used as a I crude "factor" preparation. I ! _Ill Hashed cells were agglutinated in a 10% or 37% mixture of the crude factor preparation diluted with Hanks 1 BSS with DNase (5pg/ml) I added and using a rotary shaker as with the Con A agglutination (Oppen-! jheimer and Humphreys, 1971). After 15-60 minutes, the cells were I !washed with Hanks' BSS with DNase and pelleted by centrifugation or by settling. I The pellets were resuspended in a 2:3 mixture of Hanks' BSS with DNase, and Eagles ~~M without glutamine, incubated in 25 mm I plastic tissue culture dishes for one or four r.ours, and fixed with I , modified Karnovsky's. ' L .. · - -··-----·--·-··--····-·· ··-- . ···---· ----·····-----·--··----·-· ··-···---~--·-- - ·-- ·- -....... _. ----··· .. -·-----···-······ ... -- -- .. J I ll r·--·----------·-----·--·-··---·--------------·-··.....------·---------·---·---.. ~---·---------·--------·----------- -----~1 I I I I ----................._.......J r------ -----·-----· -----·-·---------------------------------·--·--------- --·----------------------------------·. ·----------------.., I ! Results i I I Fine structure of S-180. The diameter of S-180 cells is generally 15-20~ but a variable percentage (about 5 - 15%) of the population is found to be 25-3~ 1 1 (Plates lA & B). Plate 2 shows S-180's general appearance. Structurally the most prominent features are a large, highly fissured, multi lobate nucleus I (Plates lA & 2) and a very active cell surface with numerous villus · projections (Plate 2). I The nucleus characteristically has many nucleoli. Most are as- l sociated with the nuclear membrane at the distal ends of intranuclear! I inuclear and closely associated with the nuclear membrane. Many fis- I I sures or deep invaginations of cytoplasm course through the nucleus caniculi (Plates 3A & B). (Plates 2 & 3). inclusions. Heterochromatic regions are normally per- In addition, one occasionally sees several nuclear In Plate 3 for instance, one sees clusters of granules, ! I 1 possibly viral cores, and what appears to be membrane bound vesicles. I The nucleoli are large. They distinctively have a large, light- I staining pars amorpha, and extensive, dark-staining, granular pars granulosa. The pars fibrosa is small but characteristically is I found between the.pars granulosa and amorpha (Plate 3A) (Goessens and Lepoint, 1974). A region of chromatin (with caniculi) can be seen associated with the pars granulosa (Plate 3A). The nuclear envelop is similar to that of normal cells. The ... ---·-··- ····----- -- ·········----------··· . -----.1 12 13 ,--··"----..---·--·--··· .......,... ,- ··-···--- ___,_..... -· ..... ·---------- "' ............----·-·---------------'"""-'""""'"''""··1 ' l double membrane being typically interrupted by nuclear pores (Plate 3A). The cytoplasm tends to be dense with large numbers of free ribo- II somes characteristic of tumor cells, and large numbers of mitochondria. The mitochondria are quite pleomorphic. They are most common- 1 ly tubular with a moderately dense and granular matrix, with typical cristae,- though they may be club shaped or S\.;rollen in appearance (Plate 4). Less commonly, the mitochondria may have a vesiculated matrix having a lighter appearance but w·ith typical cristae (Plate 2), and occasionally mitochondria may have a dense granular matrix and vesiculated cristae (Plates SD & 7B). due to the energy state of the cell. types reflecting the 11 orthodo~' This pleomorphism is likely The first and maybe the second or resting mitochondrion and the third type, with dense matrix and vesiculated cristae, reflecting the "condensed" or rapidly respiring mitochondrion. Rehn, 1970). (Hackenbrock and The mitochondria are generally distributed uniformly throughout the cytoplasm though it is not uncommon for them to be found somewhat more peripherally. The centriole and Golgi complexes are found in the centrosphere (Plate 4). One to three Golgi complexes are commonly seen in a sec- tion, always associated with a large number of vesicles and vacuoles suggesting activity (Plate SA). Often seen in the region of the Golgi are various darkly stained bodies (Plate SA). The amount of endoplasmic reticulum (ER) is somewhat varible, I l but usually is not extensive in these cells. 1..~-"·-~··'" I I Some smooth ER is seen, 1 --·---·--• ~--·-.-·- -·•- --~···--·• -··~------~···-- ••••·-- .. . .rl 14 rough ERpredominates. The ER is usually filled with a moderately j staining material (Plate SA). I Annulate lamellae are seen but rarely. Various types of lysosomal vacuoles are seen in these cells. !Plate SB shows what appears to be the fusion of residual bodies with ! Ia digestive vacuole. Autophagic vacuoles can likewise be seen I ii (Plate 12D) . In the presence of high concentrations of exogenous Iproteins -(in this case ascitic fluid) Iseen (Plates 12B & C). numerous endocytic vesicles are These probably fuse with numerous lysosomal i I vesicles II can in this region, forming the numerous residual bodies which be recognized by the halo between the contents of the residual I I body and the delimiting membrane (Novikoff, 1973). I t was not un- common to see large myelinoid-type residual bodies in these cells. 1 II These myelinoid figures appear to be readily exocytosed by the cells I (Plate 12D). i I I Many large, spherical, lightly-stained bodies can be seen in these cells. They do not appear to have a delimiting membrane, and i appear to be readily exocytosed (Plates 5C & D). II lipid droplets as they appear similar to those described in other These are probably tumor cells (Chambers and Weiser, 1964; Bergstrand and Ringertz, 1960; I ! Epstein, ~ ~· 1 II 1957). I In addition, there were other large spherical bodies with greater j electron density. Some stained irregularly (Plate SE) and others 1,. appeared to have a halo of densely stained particles surrounding them ' (Plate SF). figures. !' The latter type was frequently associated with myelin-likel No delimiting membrane could be seen around these bodies, 15 ~-------. --- -----------------·------ -·-- -- ----· ---- -----·--- --- --- ----- --·--------------- ·---------------------, I I Iibut I the membrane may have been obscured by the density of the stain. lit was not uncommon to see the first type in clusters of two and I ~three. Apparently they coalesce together in the cytoplasm. Rarely, one sees aggregates which appear to be viral capsids in j ! l the cytoplasm (Plate 6A), and individual capsids in the ER. j The cap- I Isids are about 50 m~ II size and ·symmetry i I et al., 1973). in diameter with isosahedral symmetry. Their suggests they belong to the Papovavirus group (Davis, II i The cell surface is quite variable. Nost commonly the surface is i i ! covered I ' i I with numerous microvilli of varying size, shape and number. I l These may appear thin and finger-like or thick and club shaped (Plate I i I 6B). The distribution of microvilli on the cell is not always uniformJ ! i Nembrane loops are common, and endocytosis can be seen. I I Concanavalin A agglutination. ! After removal from shaker treatment with 500 or 1000 Con A, cell aggregates were seen. ~g/ml of I These ranged from a few millimeters! ! across, down to two and three cell clumps in addition to unagglutinated, single cells (Plates lC & D). Hithin approximately four hours, the large clumps had broken down into smaller aggregates of several cells. Longer incubation (20 hours) resulted in low viability of the cells. The decrease in aggregate size is probably not affected by the dead or dying cells, .as they \vere tightly associated with viable cells (Plate 6C). The agglutinated cells ~;.;rere examined ultrastructurally at one, J i 16 ,-------- ------------·----·--··---------------------------------------------·-·-----·---· ' -------· --·-·····---------" __. . ---------------1 II four I and t\venty hours after treatment. In general, the viability of the one and four material, as determined by dye exclusion, was high, j but I~ I consistently too low in the twenty hour material to be meaningful. This loss in viability could have been due to the cytot6xicity of Con A, especially since it was used at possibly toxic levels in order · I to maximize agglutination for electron microscopy. Also growth con- ) ditions may not have been optimal, as later work suggest: these cells require a rich medium initially after transfer to 11 in vitro" cultures.! Three types of cell-cell interactions were seen between the Con A agglutinated cells. I I! The interactions after four hours \vere similar to those seen after one hour. The interaction could be classified into: 1) Those mediated by microvilli-microvilli and microvilli-cryptic l surface interactions; 2) intermediate associations usually of the focal type, in which the membranes are separated by about 100-250 A and between which a lightly stained material is usually seen; and 3) close associations in :hich the membrane of adjacent cells are separ- ~;: ated by less than 100 A. _ Plate 7A demonstrates the first type of interaction and compares Con A treated and untreated cell surfaces (Plate 7B). treated cells seem to have few projecting villi. The Con A Many villi which can be seen in cross sect:ion apparently have been agglutinated by the Con A to the parent cell surface and to the villi and cryptic surface of the adjacent cells (Plate 7A). Membrane loops can be seen on both Con A and untreated cells, but are accented on the Con A cells by the lack of projecting villi. The obvious projection of microvilli on the l l I . ·-----------.. -----...- -................ ____ -- ................. ,...... J 17 ,---------- --------·· ·-------··---·-·------------.--------·-----·-------- ... _________,__ -·- ------·------------.. . . .l I I ~-. !control cells with few villi-villi or villi-cryptic surface inter!actions is quite apparent; in most instances no associations are seen I ~between i I I untreated cells (Plates lA & B, 2). How stable the microvilli mediated attachments are is unknown, !but similar associations are seen in cells cultured four hours {Plate j8A). Plate 8B shows how extensive these associations can be, some I I v 1."11"1. actually seem to be pushing_into the adjacent cell producing a very tight fit between surfaces. This situation 'vas not seen very of- ten, and only in the four hour preparations, though it may have been present at other times and not observed. The villi normally appear I ~.! i I to I ' be lying flat, apparently agglutinated to the surface of one or both I adjacent cells, and many times lying between cells that are possibly ! being held together by areas of closer association {Plates SA & C). The intermediate associations are very common and characterized by the adjacent membranes being separated by greater than 100 ~- I i They I usually have a lightly stained material in the intercellular space, and do not appear to be extensive in area (a focal type association) (Plate 9A). The adjacent membranes are approximately parallel. This type of association is most commonly seen in areas which hold the cells together in clumps (Plate 8A). Plates 9B and C shmv that multi- ple points of interaction can occur between adjacent cells. The com- mon situation consists of large regions in which focal intermediate associations appear to be holding the cells together. I I I ' In the inter- vals bet,veen these associations, the membrane appears irregular, commanly with villi projecting into the open spaces. l i The cell surface is 1 ______ ..., ............. -- ---------------·" ......... -· .......... .J 18 r-------·---------····-·-- ···--···--·-·--~---------y Ii Ifrequently characterized by amorphous debris possibly derived from I I I !the media. ! There is no type of cytoplasmic specialization found in the intermediate type of association comparable to the classical zonula adherens type of junctional complex. But these may represent a modifi- !cation of this junction, and it may form as a result of the cells be- l ing brought into close contact by the lectin during agglutination. ! I It is not knmvn if Con A is found between the cells in areas of these I Iassociations. I The last type of association seen, the close association (Plates I10 A-D), I separated J is characterized by the membranes of adjacent cells being o by less than 100 A. Imembranes can be estimated II thickness I 1 lOA is about 70 R. In Plate lOA, the gap between adjacent to be about 20 Xassuming the membrane These may appear as either focal (Plates & B) or zonula (Plates lOG & D) types of interactions. No cyto- i !plasmic specialization is seen, and the outer leaflets of the opposing ~membranes I' ly 1 i' do not fuse at any point. common; especially the focal type. teristi~ of the gap junction. This type of association is fairIts appearance is very charac- Attempts to use the characteristic lan- lthanium oxide staining property of gap junctions failed to answer this 1 question. ilremaLn . Ln. ! 1 In It was felt this resulted from the failure of the stain to the loose cell aggregates during washing and dehydration. no case were tight junctions observed in \vhich the outer leaflets 1 lof the adjacent membranes would have fused, nor were desmosomes with !i their characteristic cytoplasmic tonofilaments seen in any of these L____ ---- ---··-· - -·------------- ---· ·- · · . . · ---- - - - - - - - - · - - - --·-- -·------ -· - - -· ---------.. - - - - - - - - - - - - - - - _________j 19 . r···~----...-..-- ---~----··· --~" --------~------------- - ----- ~-·------ ---- ---------·--- - -------·-· -·- ··--·- -- --------·-- --- -------, materials. I Factor agglutination. Two dilutions of "factor" (Oppenheimer and Humphreys, 1971) were used: 10% and 37% in Eagles MEM. the concentration used. The results were dependent upon At the higher concentration after one hour (Plate 11) the cells actively phagocytosed a large amount of flocculent material from the medium. This material also appeared to be ad- hering to the cell surface (Plate 12A). Using high factor concentrations, the cells aggregated to form large clumps, but there appeared to be no type of close cell associa- I I I I tions. Moreover, it appeared that agglutination in this case ~vas l I II i i due to large amounts of material attached to the cell surface, which crosslinked the cells together (Plates 12B & C). Hmvever, it should be noted that the factor used in these experiments was relatively inactive, requiring high concentrations for cell binding activity. The cells appear very vesiculated as a result of pinocytosis. Many lysosomal vesicles were obvious and the rough ER had become branched and slightly swollen with a moderately stained material (Plates 12B & C). The number of villi had decreased, possibly in response to the large amount of membrane involved in pinocytosis. At the lmver concentration of "factor" a much different situation was seen (Plate 13). The cells appeared normal with large numbers of villi and reduced amounts of roughER. In this case intermediate type associations similar to those I 20 r~--··-------- . --·-··------------------........ · -· --------..·--------·-·----·-----.. --·-·----..--------------- ·-·-------- . ----------l Iobserved with Con A were seen between the cells (Plates l4A-C). A I lightly stained material was obvious in some patchy areas (Plate l4D),'. ,I I and in many cases villi were absent from the intercellular spaces. No gap or tight junctions were apparent in these preparations. The material in these associations could be the component(s) of the ascites which \vas responsible for agglutination and may be crosslinking t~e cells together. Shepard and Oppenheimer (unpublished ob- servation) have shown some degree of specifity of the "factor" for· these cells which may involve terminal galactose resides on the factor! binding to cell surface receptor sites (Oppenheimer, 1975; Oppen- ' heimer, personal communication). With time in culture, the cell aggregates tend to breakdmvn. Af- i ter four hours the number of associations decreased. ·Most cells appeared singly and looked like untreated cells with numerous villi and lobated surfaces. It could be that the components responsible for agglutination are removed from the medium and cell surface with time, possibly by pinocytosis and phagocytosis. Factor may have to be re- plenished to maintain the effect. I I Plate 12D shows two cells, one of which is apparently exocytosing! a large myelin figure. The complementarity of the cells' surfaces suggests they are being forced apart by the myelin figure. I I This could suggest that the strength of the association is not very great ; and can be broken by any projection from the cell surface. Also, since the cells have remained nearly complementary, cytoplasmic stabilization of surface shape may be an important aspect of adhesion. L._.............-----·- ------------·- -------------- . --- --------- ________. ___ ------------------------ ----------~------------------ --------- - - - . .... ____ _} 21 ~---··«•·······--· ·-··-----·· ----------·· --·······---·---·--·-·········------···-- --------------- .... ----·------------- .... ---------· ---·-------~ , II jManganese ion agglutination. Manganese ion produced a very rapid agglutination of S-180 into large clumps. I! immediate I utes. The initial events in agglutination were examined by fixation followed by additional fixation at 3 and 30 min- Light micrographs show clumping of the cells after fixation 1 Iand infiltration with I \ I plastic (Plates 15 A-C). Ultr~structurally Ibetween the i minute no differences were seen in the association initial (Plates 16 A & B), and 3 (Plates 16 C & D) and 30 (Plate 17A) preparations. There did not appear to be any dif- 1 ferences in the Ruthenium red staining properties of areas of inter- i action and noninteraction (Plate 17B). 1,11_ Plates 17 C & Dare high mag- 1 , j nification pictures of the interactions occurring between adjacent i ! cell surfaces. The distance separating the membranes is very small, I a ! less than 60 A and in many cases no distinct separation can be seen. I \ While this suggests some type of primary interaction between the mem- l brane surfaces, I ionic bridging between components on the surface of the villi seems likely. I change is I cells. These interactions appear static since no seen in them between the initial, 3 minute, and 30 minute The strength of this interaction is not known, but probably requires the continuing presence of Nn -1+ . After four hours in 20 mM Mn-1+ in serum containing medium, the cell aggregates were dissociated (Plate lSD). I Most cells seem swollen! } (Plate l8A) and large numbers of vacuoles and invaginations are appar- 1 . ent. i Plate 19A shows how deep the invaginations of the cells can be. The surfaces of the cells were quite variable. '----- ... ··-- ..... -- ---·-··· --------------. - -······ ---------- The cells appeared I ! l i .......... ----- .............. -- ······.- -- .................. ------ ...........! 2.2 IIcompletely smooth /villi (Plate 18C). (Plate 18B), lobated (not shown), or with numerous The cytoplasm is denser around the nucleus in ! many cells and the mitochondria are condensed, deeply stained, and with s1;.mllen cristae. The breakdown of the aggregates may, in part, be due to the loss of villi from the cell surface, but the morphology of these cells is quite unusual and may reflect a pathological situation as a result of the relatively high level of manganese ion in the medium. The associations that occur with Mn++, while similar in some II ways to those seen after Con A agglutination (Plate 7A), I i I to lead to flattening of the cells' surfaces nor to more extensive I interactions Ia between cells. situation unsatisfactory for further involvement by the cell sur- 1 face. Ii The relative staining intensity of the microfilaments within the I microvilli is greater after Mn++ treatment (Plate 17D). 1 gests This sug- Mn++ may be interacting with microfilament components. I It is difficult to quantify the occurrence of each type of association in this study. I I l However, using micrographs of situations which were judged to be typical, Table 1 was constructed to indicate I the relative occurrence of each general type of association after treatment with each agglutinin. I Use of the terms 'close', 'inter- mediate', and 'microvillous' does not imply that the agglutinins act in similar Hays or that the fine structure of each is exactly identical. t ~~--~---~------- -···· ·- ....,~--···· -·-------···-------~ -----. ·-··---· ··--·--··· ···--~ -------·-·- ------- -----·-------·· ~ ----- --- I I I __________ __~ 23 i ------- -· ---~·------------ --------·-·------------ ----------------------------------------------------- - ------------------ "--- -------1 I I I Each association may occur alone or in combination with the II I ::::: :::e:~di:::: ::g:::·o::::•:y::.h::ea:::::::::::·.:·:::i::::::e I I I in Plate 14A where two microvilli appear to be associated with the adjacent cell by an intermediate association. In the case of mangan- ! ese) villi interact by associations which appear closer than the close associations seen with Con A. l i ! l I ........ ------.. --- . . ---------------·---·- .... -___ ...l r------·-··. ------------------ - ·- ------- -· -------·----------· --- ---------· ------ -.---------- -------------..·--------------------·-------··: l I I Discussion ' The fine structure of S-180 has been reported previously by Molnar and Bekesi (1972). This work expands on their observations and reports some differences. S-180 shows the general structural characteristics of most tumor, cells' The large, irregular 1 y shaped, heterochromatic nucleus; the I large number of surface villi; the large number of free ribosomes, I and decreased amounts of ER (Molnar and Bekesi, 1972; Chambers and Weiser, 1964; Oberling and Bernhard, 1961; Bergstrand and Ring·ertz, 1960; Bernhard, 1958; Epstein, 1957; Dalton and Felix, 1956). In addition, it shares some structural characteristics of other I ! ascites tumors. !I Large lipid droplets \vere described in Sarcoma 1, Sarcoma 37 and MClM tumors (Chambers and Weiser, 1964; Bergstrand and Ringertz, 1960; Epstein, 1957. by membrane. These do not appear to be bound They tend to show a light, uniform staining, and may also show a granular darkly stained periphery. Similar lipid bodies were seen in S-180 (Plates 5C & D). As seen in Plate 5D these lipid bodies could be readily exocytosed. In addition to the bodies described above, there \vere other large, darkly stained bodies (Plates 5E & F). These did not appear to be mem- brane bound, but the membrane could have been obscured by the density of the stain. Whether these represent pleomorphic forms of the form- er type of lipid body, or a different type of vacuole was not determined. The large size and apparent lack of a delimiting membrane 24 25 suggest they are lipid bodies. I type of residual body. I I type The association of myelin figures \vith one (Plate SF) may be suggestive of this, but residual bodies are !bound by a tripartate membrane (Novikoff, 1973). A cytochemical an- ;alysis will be needed to characterize these bodies. I! Viral capsids occasionally occurred in the cytoplasm of these cells. Eowever, virus particles were not seen budding off the cell membrane and this suggests they are not an enveloped virus. Their size, symmetry and probable lack of envelop suggests they belong to the papovavirus group. Members of this group are commonly seen in a small percentage of cultured mouse cells and some tumor cell populations (Davis, et al., 1973; Sanders, 1973) and similar virus particles have been seen previously in S-180 (Molnar and Bekesi, 1972). C-type viruses were not seen. The large number of microvilli on the cell surface is a common feature to ascites and also some solid tumors (Chambers and Weiser, 1964; Mercer and Easty, 1961; Bergstrand and Ringertz, 1960; Wessel and Bernhard, 1957). number and shape. S-180 shmvs numerous villi of various lengths, They could be irregularly, or uniformly distrib- uted on the cell surface. While these appeared to be important in the cell-cell interactions described, regions of extensive interaction were not seen between untreated cells. The S-180 cells used in this study characteristically have few microfilaments. On the other hand, bundles of fine filaments have been described in Sarcoma l, Sarcoma 37, :tviClM; and also in fibro- I l ......,. ---------- .......... J 26 r--·---------------- ------· . ----· -------- ·---- --------- ·---- ... ---------- --------· .. --------------------------· -------------- -- . -- ---· ------ ---------1 i blasts and as the tonofilaments of epithelial cells (Molnar and Bekesi, 1972; Chambers and Weiser, 1964; Journey and Amos, 1962; Ross and Benditt, 1961; Bergstrand and Ringertz, 1960; Epstein, 1957; Selby, 1955). The lack of cytoplasmic filaments in this study is in contrast to that of Molnar and Bekesi who observed them in the S-180 they used. This difference may relate to the particular line of S- The ultrastructural features of the agglutinated S-180 cells differ depending on the agglutinin and this in turn on the action of the agglutinin on the cell surface. In the Con A system, the cell-cell interactions were of three types: 1) Microvilli-mediated, 2) intermediate, and 3) close asso- ciations. These were seen after both one and four hours. It should ' be noted that aggregates, formed initially after treatment with Con A, were relatively large when examined by eye and these dissociated into smaller aggregates with further incubation. This would suggest that many of the interactions are \veak, reversible, or subject to degradation possibly of the Con A or its receptor. In addition, further dissociations occurred as a result of the preparative steps for electron microscopy. Glutaraldehyde fixation did not stabilize the interactions between the cells. The smaller aggregates may rep- resent only the stronger or more extensive associations that form and which are able to survive the technique. Several workers have shown Con A receptor mobility (clustering) to be related to agglutination (Rutishauser and Sachs, 1974; Rosen........ --···--·· -'- ---------------- __________j 7!.7 r--·---···----~----··-------~~--- -- -------- -·- -· --· - ----·--·-··--------· ----------·------.... ----·-· - . ------------··--·. -- ··-·-l i blith, et al., 1973; Nicolson, 1972; Nicolson, 1971). The types of associations seen here may represent clustering of Con A receptors Ii resulting in crosslinking of the cells, or some type of Con A initi- I• ated interaction between the cells. Host of the work on Con A recep- II i tor mobility and clustering involved examination of free surfaces of cells and not the interaction between cells. There does not appear son (1972) has examined, by thin section, the agglutination of 3T3 cells after proteolysis, while paying heed to mobility of receptors Proteolysis stimulates agglu- tination between normal cells which agglutinate poorly at high lectin concentration. His work shows a greater concentration of ferritin conjugated Con A between the protease-treated 3T3 cells at the points of contact than on the free surfaces. Moreover, the ferritin Con A appeared in the region of contact to be clustered into patches crosslinking the cells. The agglutinated cell surfaces were separ- o ated by 200 - 400 A. Cell aggregation could be reversed in this case by a-methyl-mannoside when added shortly after agglutination. Lee and Feldman (1964) reported a similar gap between human red cells agglutinated by ferritin conjugated group A antibody. The first interaction, the microvilli-mediated association, could represent initial contacts betHeen cells \vhich then develop into the stronger types of associations. and, therefore, can dissociate. IL._.that I Nicol~ to be any work examining the interactions between tumor cells. on the free surface of these cells. I ' They may also be reversible There has been some work showing the microvillus surface is chemically different from the inter- ; 28 r~----------- --------------. I' villal membrane (\veiss and Subjeck, 1974). These differences may be reflected here but they are difficult to assess. The suggestion of villus interactions leading to the more extensive types of associations is difficult to reconcile with what is observed here. The villus interactions are seen in both the one and four hour material. It would be expected that many of these would have progressed into the other types of associations by four hours. Intermediate steps leading from villus interaction to the extensive close associations are not obvious in this material. The villi interact with cryptic cell surfaces and other villus tips by both intermediate and close associations. This may indicate that the interactions of the villous! and cryptic cell surfaces are comparable to the interactions that occur in regions where membrane flattening has occurred. There did I I I not appear to be any cytoplasmic or intercellular specializations oc-! I curring at the site of interaction. An increased staining density was observed at the site of microvilli interaction in Sarcoma 1, "in 'I I vivo" (Chamber and Weiser, 1964). One would expect the villi to inter-1 act extensively after treatment because of the large number of villi II on S-180, and also because they would be involved in the initial con- I tact s be tween the cells. It is d iff icu 1 t to discern whether this I first type of association is independent in the agglutination process l or plays a transistory role in the formation of the other types of I' i Hssociations. I I The intermediate associations seen in this work resemble the in- ! I teractions described by Nicolson between 3T3 cells. t~---·~·- The associations j I ·-··· .. ---·---·~-- ~- .. ------··-----·- ···-··--··- .............! 29" r···---···- ------· "--- . .. ~-- '_1 ~·-- '----------·---··--·-¥--~~------· --- ---------------~- --··-~----·---~---··----------·~-------. . ---~-- .. --------~------~~-- .l in this study are characterized by patches of a lightly stained, mem- 1 l ! brane associated material crosslinking the S-180 cells. I very \vell be aggregates This may of Con A (Agrmval and Goldstein, 1968) agglu- tinating the cells together. The patchy nature of the interaction does suggest clustering of Con A receptors as a requirement for agglutination in this situation as in the 3T3 study. Another interpretation of the intermediate association is that it represents some type of junctional complex which develops secondarily as a result of primary Con A agglutination. While these asso- ciations do appear similar in some \vays to the zonula adherens junetions of normal cells, no cytoplasmic specialization is seen in S-180.! The junctions would normally have actin-like fibers associated in a j mat \vithin the cortical cytoplasm under the plasma membrane. I The ab- 1 sence of fibers and the general patchiness of juxtapositioning does I suggest this association is not a zonula adherens, but represents the 1 crosslinking of the cells by Con A. I! ruled out by this work (McNutt and Heinstein, 1973; Pinto-daSilva and Atypical junctions cannot be j Gilula, 1972). l ' Close associations were not reported by Nicolson between agglutinated 3T3 cells. As seen in this study, these may represent regions.,. of Con A crosslinking dissimilar to those described by Nicolson, but it is not likely that the close associations, especially the extensive ones, (Plate lOC), would represent areas held by Con A. is a polymer in solution. Con A The size of the basic protomer as detero mined by Edelman et al., (1972) is 42X40X39 A. Above pH 7, Con A is I 30 r·---·--·······------- --------·---·----········-··- ........ -------- ---- ... ----······ ...... ---------·--------·- ------- -------·----------- ----~ I found in the form of tetrameric and larger polymeric aggregates of I the protomer (McKenzie, et al., 1972; Agra\val and Goldstein, 1968; Kalb and Lustig, 1968). 0 20 A. The gap in the close association is about Thus, the dimensions of the aggregated Con A agglutinating the cells (Con A tetramers and larger polymers) would be greater than that of the gap. This would argue against the direct involvement of Con A in the close association. Rutishauser and Sachs (1974) reported some work using Con A ; derivatived nylon fibers which shows the cell-cell interactions between two lymphoma tumor cells after Con A agglutination is not reversible by a-methyl-mannoside. It was suggested by the authors that Con A acts to bring the cells into close contact at which time cellular components can interact to produce a mannoside-insensitive binding. These interactions were not examined ultrastructurally by the authors. The work here is suggestive of a comparable situation, though reversibility was not examined in this study. The extensive regions of close contact could represent regions in which strong cellular interactions have occurred; these associations being induced in some way by agglutination, but \•lith cell surface components inter·· acting. The appearance and dimensions of the close association very near-· ly resembles the gap junction or nexus. An attempt to use lanthanium oxide staining of this association failed to produce any staining. It is possible the aggregates were too loosely associated to allow· the stain to remain betHeen the cells (Revel and Karnovsky, 1967). iL_ ____ .. This 31 ,-~---------···-----· I I ! --··--··--·------·--·-·-·-------------- ···- --·----··--·-···---·-- --·-· -·· ·-··- --·--·--···-·······-···. --- ·-------------------, 1 ~- question could be approached by freeze-cleavage techniques which ,I should demonstrate the characteristic hexagonal array of membrane l 1 I components in the gap junction. Whatever the nature of the component(s) involved in the close I association, they are present or inserted very quickly after treatment with Con A. f This can be implied by the fact that close associ- ations ·are seen after just one hour. If these components were being synthesized "de novo" one would expect the close association to develop later, possibly after four hours. interact 11 in vivo" is not known. Hhy these components do not Con A may allmv the uncovering of cryptic sites, or the overcoming of physical restrictions. A relationship has been demonstrated between lectin binding and cell behavior. Burger and Noonan (1970) used Con A to inhibit trans- formed cell growth, and Friberg and his group (1971) \vere able to shmv inhibition of migration. This type of "social" behavior is thought to be a form of cell-cell communication mediated by gap junetions (Bennett, 1973; Lowenstein, 1973). It might be speculated that what \vas seen by these groups represents a Con A induced coupling of cells through the gap junction-like close associations seen here, I allowing for cell-cell communication. Agglutination by the crude ascites "factor" at high concentratio~ l resulted in large amounts of amorphous material adhering to the cell surface resulting in clumping. The interaction of this material with the cell surface is probably nonspecific, binding in many areas of the membrane inducing pinocytosis of the material. The flocculent •••-•-• • - - - - · - • • " c"-'•• I -~-J 32 --- ···---····--··· ... . . .•..•..... ··-·-- ······- --······ .... ·•····· .... ·-·-···-· ··------·-·.. - .... . -····· -·-· ........ ·------ ... - ..... ., -- --·······--. ·····----------·· ., l l material probably includes, in addition to "factor", other ascites proteins such as fibrin. The method of obtaining "factor" may in fact activate some of the clotting proteins resulting in their deposition on the cell surface. l I One would not expect to see large amounts of proteins adhering to the cell surface "in vivo". In fact, 1 this material is not seen on cells fixed in ascites fluid immediately I after removal from the peritoneum. I At lower concentration, the large amounts of flocculent material I were not present and a situation develops similar to the intermediate association seen with Con A. These interactions appear to be medi- ated by small areas of cell surface in \vhich a lightly stained material is obvious and, presumably, is the "factor" from the ascites. The patchy nature of these interactions and the lack of surface interactions in the regions without this material could suggest the involvement of membrane components or receptors. These may be asso- ciated as patches in a mosaic on the cell surface. Otherwise, the factor may promote the clustering of surface receptors in a manner similar to Con A and other lectins. The factor isolated from a teratoma cell line by Oppenheimer and Humphreys (1971) appears to be a glycoprotein, possibly synthesized by the mouse, the tumor cell, or both, and appears to possess terminal galactose residues which bind to the cell surface (Oppenheimer, 1975). Prefixing cells with 4% formaldehyde failed to inhibit agglutination suggesting that with teratoma "factor", active movement of membrane components is not I necessary for agglutination to occur (Oppenheimer and Humphreys,l97l)J . ---. · · · · -· · · . . . --· ........ ···-·-----····--··--J 33 ··---------·····-··-··-·-·-····-·-··-··~-----···-· • • « • ·---~ I The S-180 ascites factor used in this study appears to possess different charge and size properties from the factor isolated from teratoma. There also appears to be a difference between the single cell form and the embryoid body form of the teratoma 11 l Whether the teratoma and S-180 "factors" are mechanistically similar is not knm-rn. I Ij factor 11 (Oppen- i heimer and Humphreys, 1971; Oppenheimer and Connally, unpublished results; Haskett and Oppenheimer, unpublished results). I I I 1 ! I If they are, the clustering of receptors by "factor" is prob-! ably not important in agglutination. II The patchy nature of the interactions at low "factor" concentra-1 tion can be explained by the presence of "factor" aggregates cross- I linking the cells, these aggregates being observed as the lightly I I stained, amorphous material bet\-reen the cells. Perhaps these aggre- I gates are large enough to interact with several receptors without the need for clustering. Recent work (Oppenheimer and Connally, unpublished observation) has shown that the ascites "factors" are inactivated by phosphate which was present in the Hanks' BSS. This may explain the relative inactivity of the material used in this study. Preparations have been partially purified in phosphate-free media which are stable and highly active at 0.5% v/v. Thus the results of this study can be considered only preliminary at this time. Until the active com- ponent of this factor is isolated and purified, it will be difficult to select a likely mechanism to explain these observations. I' l II_________ -. The I last system studied involves the use of the divalent cation . ..... ---·- ·------ ----····-··--··· ·········-··- ····-··· .. J 34 manganese. It has been reported that this ion stimulated the adhes- ion of Sarcoma 1 to serum coated glass, Mn·!+ having high specificity i i in this system (Rabinovitch and DeStefano, 1973). Its mode of action is not understood, but it was suggested that Mn++ increases deformability of the cell. More cellular projections were seen; veils, villi, and cell flattening. II , I It must be noted these cellular changes were seen at 1 mM Mn++ and in fact 20 roM Mn++ as used here was less efficient in stimulating adhesion to treated glas,s. It was suggested that microfilaments and microtubules may be involved since adhesion was sensitive to cytochalasin B and colchicine. What was seen in this study did not involve flattening of the cells nor formation of veils. In fact it appeared as if a stabilization occurred in the microvilli. Cell-cell interaction was not dependent on serum and interactions appeared to be mediated by the microvilli. Microfila- ment staining appeared to be enhanced some\vhat by Mn++. This was obvious at high magnification (Plate l7D). Interactions were very close and in some cases no gap was discernible between the cell surfaces. Ruthenium red sho\vS these interactions involved the carbo- hydrate-rich glycocalyx of the cells and may reflect some type of ionic bridging or perturbation of the normal ionic environment in and around the plasma membrane. It has been reported that lanthanium ion can induce agglutination of Ehrlich ascites tumor and also results in K+ efflux from the cell (Levinson, et al., 1972). Lanthanium can also be used to inhibit clustering of cell surface receptors (Sachs, et al., 1974). I L---····-··· - · - - - · · · - - - ·-- .. Zinc ion has been used to 11 fix" or stabilize mem- 35 r---. .. ·-·····- ·---· -·--··-----·-·-· ....... ······-···-, i ; lbranes (Brunette and Till, 1971). has shmvn Zn ++ ! Oppenheimer (personal communication) I ++ and Mn to have a high specificity in the S-180 ag- glutination system, and perhaps the ions are interacting with surface glyco-protein and lipids producing a denaturation or ionic bridging between adjacent cells. out an 11 est in ~etermining Mn++ agglutination may be a phenomenon with- in vivo'' relevance though its mode of action may be of interthe nature of the cell surface. Junctional complexes are found between many types of tumor cells. It is difficult to assess whether or not these are functional. As mentioned previously, a positive correlation can be made between greater numbers of junctions being found bet\.Jeen normal tissues than between tumor tissues. Also, the number of junctions correlates >vith the form of the tumor: for example, solid versus ascites. All types of classical junctions can be found in neoplasms, but tight junctions and gap junctions are usually not as common in comparison with normal tissues, (Trembley and Babai, 1972; Clarks, 1970; Martinez-Paloma, et al., 1969; McNutt and Weinstein, 1969). The existence of junctions in solid forms of S-180 has been reported (Sheridan, 1970) and these appear to be able to transfer low molecular weight dyes and are electrically coupled. While the im- portance of the transfer of small molecules and electrical coupling has not been demonstrated, it has been suggested that these play a role in the 1973). 11 social" behavior of cells (Bennett, 1973; Lowenstein, Sheridan's work also demonstrated that the junctions involved in this coupling, usually considered to be the gap junction, \¥ere I 36 ··""--~·~--- ··-----~---·---~ . ~- ·------~-----·- ~-····- . -· ---·------~--·----~---"-··· -· .. ··-·-~-----------·- -------- -·-- ;•··--~--------~----~---- -~--~------~ ... -. -------~-·~-1 ! II able to form bet~·Jeen S-180 cells. Heaysman and Pegrum (1973) examined' the interactions of S-180 (an adherent explant of a solid form) with chick heart fibroblasts. Their results indicate their line of S-180 does not form specialized contacts with chick heart fibroblasts. One might ask whether such interactions would be expected to occur between mouse and chick cells. 1,1 I I The S-180's used in these two studies were different from that used here, and one must question whether these observations are applicable in this study. The work presented here suggests the possibility that Con A is able to induce cell-cell associations which appear similar to gap I junctions. I with normal function, these associations could, nonetheless explain Hhile these were not shmvn to be normal gap junctions I the return of "social" behavior of transformed cells after treatment I with lectin. I Ii I "Factor" produced interactions very similar to those reported by Nicolson between 3T3 cells with Con A after proteolysis, and similar to the intermediate association seen here with Con A. This suggests the possibility that "factor" ~vorks in a to Con A; aggregates of factor crosslinking the cells. ~vay similar The mechan- ism of action of Mn++ is difficult to assess in this work. It pro- duces very tight associations via the interaction of microvilli. The significance of this in physiological terms is not clear. i ; i ------~J r--------·-----~---·---------------------- ........ ··-··· ..... ·--·--· .... •· ··-·-- ····· ... --------------·-- .··----· --- .·----· --~--- -----~------1 ! Ii Bibliography I l ! i I ;Abercrombie, M. and Ambrose, E.J. (1962) The surface properties of i cancer cells: A review. Cancer Res. 22:525. I 1 I I lAgra~ml, B. B. and Goldstein, J. (1968)-Protein-carbohydrate inter-l ' actions. VII Physical and chemical studies on concanavalin A, the! I. hemagglutinin of the Jack bean. !' Arch. Biochem. Biophy. J24;218. 1 E. J. (1967) I!Ambrose, membranes. Can. Biochemical and biophysical properties of cell Cancer Con£. 7:247. 125 I jArndt-Jovin, D. J. and Berg, P. (1971) Quantitative binding of I concanavalin A to normal and transformed cells. J. Viral. .§_:716. ! !Benedetti, E. L. and Emmelot, P. (1968) The membranes. In Ultrastructure in Biological Systems. Vol. 4. (ed. Dalton, A. J. and ! Haguenau, F.) Academic Press, New York, p 33££. J iBennett, H. S. (1963) Morphology of extracellular polysaccharides. J. Histochem. Cytochem. Jl:l4. Bennett, M.V.L. (1973) and adult tissue. Function of electronic junctions in embryonic Federation Proc. 32:65. Bergstrand, A. and Ringertz, N. (1960) Electron microscopic examination of the MClM tumor. 1. The tumor in ascites form. J. Natl. Cancer Inst. 25:501. Bernhard, W. (1958) Electron microscopy of tumor cells and tumor viruses: A review. Cancer Res. 18:491. Brady, R. 0., Borek, L. and Bradley, R. M. (1969) Composition and synthesis of gangliosides in rat hepatocyte and hepatoma cell lines. J. Biol. Chern. 244:6552. Brightman, M. W. and Reese, T. S. (1969) Junctions between intimately apposed cell membranes in the vertebrate brain. J. Cell Biol. 40:648. Brunette, D. M. and Till, J. E. (1971) A rapid method for the isolation of L-cell surface membranes using an aqueous two phase polymer system. J. Membrane Biol. 2:215. Buck, C. A., Glici:::, M. c. and Warren, L. (1970) A comparative study of glycoproteins from the surface of control and Rous sarcoma virus transformed hamster cells. Biochemistry 2:4567. < 37 • ' -·-·~~-~ -·-~ 38 !Burger, M. M. and Goldberg, A. R. (1967) Identification of a tumor specific_determinant on neoplastic cell surfaces. Proc. Nat. . Acad. ScL USA 57:359. I i Burger, M. M. and Noonan, K. D. (1970) Restoration of normal growth by covering agglutinin sites on tumor cell surface. Nature 228: 512. j : !l Ceccarini, i I Chambers, V. C. and Weiser, R. S. (1964) Sa~coma 1 in a homologous host. I. tumor. Cancer Res. 24:693. M. A. (1970) Specialized ! Clarks,cells. _An EM study of mouse l j I1 I L. and Eagle, H. (1971) pH as a determinant of cellular growth and contact inhibition. Proc. Nat. Acad. Sci. USA 68:229.! Ij l I Electron microscope study ofl The cells of the grmving I intercellular junctions in tumor sarcoma cells. Anat. Rec. 166:199. 3 Cline, M. J. and Livingston, D. C. (1971) Binding of H-concanavalin A by normal and transformed cells. Nature, New Biol. 232:155. Coman, D. R. (1944) Decreased mutual adhesiveness, a property of cells from squamous cell carcinomas. Cancer Res. ~:625. Dalton, A. J. and Felix, M. D. (1956) The electron microscopy of normal and malignant cells. Ann. N.Y. Acad. Sci. 63:1117. Davis, B. D., Dulbecco, R., Eisen, H. N., Ginsberg, H. S. and Wood, W. B. (1973) Microbiology 2d ed., Harper and Row, Inc. N.Y. Chapts. 44, 63. Dulbecco, R. (1970) Topoinhibition and serum requirements of transformed and untransformed cells. Nature 227:334. Edelman, G. M., Cunningham, B. A., Reeke, G. N., Jr., Becker, J. W., Waxdal, M. J., and Wang, J. L. (1972) The covalent and three dimensional structure of concanavalin A. Proc. Nat. Acad. Sci. USA 69:2580. I Emmelot, P. and Benedetti, E. L. (1967) In Carcinogenesis: ~Broad ! Critique, (ed. Comley, R. \v.) M. D. Anderson Hospital Symposium. I Williams and Wilkins, Baltimore. p 471££. Epstein, M. A. (1957) The fine structure of the cells in Mouse sarcoma 37 ascites fluid. J. Biophy. Biochem. Cyt. 1:567. Friberg, S., Cochran, A. J., and Golub, S. H. (1971) Concanavalin A inhibits tumor cell migration. Nature, New Biol. 232:121. 39 - -- ---~--- -- - --~- <'-- -- -------- -- -- -----·. ------------~---,.-- - - - - - - - ------ - .... --------------- --· .......... --------- - - ------· -----l I! l! Gasic, G. and Gasic, T. (1962) Removal of sialic acid from the cell coat in tumor cells and vascular endothelium, and its effect on metastasis. Proc. Nat. Acad. Sci .. USA 48:1172. I i l ' (1963) Removal of PAS positive surface sugars in tumor cells by glycosidases. Proc. Soc. Exp. Biol. Med. 114:660. j Gasic, G. and Gasic, T. I I I I Goessens, i G. and Lepoint, A. (1974) The fine structure of the nucle- 1 olus during interphase and mitosis in Ehrlich ascites tumor cells! cultivated 11 in vitro". Exp. Cell Res. 87:63. l i Ii C. R. and Rehn, T. G. (1970) Rapid ultrastructural II Hackenb~ock, transformation of mitochondria in the Ehrlich Ascites tumor cells ! related to energy state. In Microscopic electronique. Vol III (ed. Farrard, P.) Societe Francaise de Microscopie Electronique. p 143. I 1 !Heaysman, J. E. M. and Pegrum, S. M. (1973) Early contacts between normal fibroblasts and mouse sarcoma cells. Exp. Cell Res. 78: i i 479. I Inbar, M., I Ben-Bassat, H. and Sachs, L. (1972) Inhibition of ascites tumor development by concanavalin A. Int. J. Cancer. 2:143. II Inbar, M. and Sachs, L. (1969) Interaction of the carbohydrate binding protein concanavalin A with normal and transformed cells. Proc. Nat. Acad. Sci. USA 63:1418. l Journey, L. J. and Amos, D. B. (1962) An electron microscope study of! histiocyte response to ascites tumor homografts. Cancer Res. 22: - 998. i Kalb, , A. J. and Lustig, A. (1968) The molecular \•might of Concanavalin A. Biochem, Biophy. Acta 168:366. I Karnavosky, I I M. J. (1965) A formaldehyde-glutaraldehyde fixative of high osmolality for use in electron microscopy. J. Cell Biol. 27:131 A (Abstr) I I Kraemer, · P. M. (1966) Physiol. 67:23. Sialic acid of mammalian cell lines. J. Cell i i !Lee, R. E. and Feldman, J.D. (1964) Visualization of antigenic sites of human erythrocytes with ferritin-antibody conjugates. J. Cell Biol. 23:396. )Levinson, C., M.ikiten, T. M., and Smith, T. C. (1972) Lanthanium in. duced alteration in cellular electrolytes and membrane potential in Ehrlich ascites ~-~~~-~--~e-7~-~-~---~-: Cell Physiol_-_ -~~-:-~~~· ______ _j 40 ~<--~ <~-~ ~---~ ---,~~~~~--~~~-~---1 l !Lowenste1.n, \v. R. (1973) Membrane junctions in groHth and differen1 tiation. Federation Proc. 32:60. I • I !Luft, J. H. (1971) Ruthenium Red and violet. II. Fine structural localization in animal tissue. Anat. Rec. 171:369. l I !McKenzie, G. H., Sawyer, W. H. and Nichol, L. W. (1972) The molecular 1 weight and stability of Concanavalin A. Biochem. Biophys. Acta 1 263:283. I)McNutt, I 1 N. S., Hershberg, R. A. and Weinstein, R. S. (1971) Further observations on the occurrence of nexuses in benign and malignant human cervical epithelium. J. Cell Biol. 51:805. !McNutt, N. S. and Weinstein, R. S. (1969) Carcinoma of the cervix: i deficiency of nexus intercellular junctions. Science 16~:597. i N. S. and Weinstein, R. S. (1973) Membrane ultrastructure at !.McNutt,mammalian intercellular junctions. In Progress in Biophysics and 1 ' Molecular Biology. Vol. 26. (ed. Bulter, J.A. and Noble, D.) Pergamon Press, New York. jMartinez-Palomo, A. (1970) Ultrastructural modifications of inter! cellular junctions in some epithelial tumors. Lab. Invest. 22: ! 605. ! !Martinez-Paloma, A. (1971) In Pathobiology Annual (ed. Toachim, H. L.) ; Appleton-Century Crofts, New York, p 261££. ! lMartinez-Palomo, A., Braislovsky, c. and Bernard, W. (1969) Ultrai structural modification of the cell surface and intercellular contacts of some transformed cell strains. Cancer Res. 29:925. I !Meezan, E., Wu, H. C., Black, P. H. and Robbins, P. W. (1969) Comparative studies on the carbohydrate containing membrane components of normal and virus transformed mouse fibroblasts II Separation of glycoproteins and glycopeptides by Sephadex chromatography. Biochemistry .§.: 2518. Mercer, E. H. and Easty, G. C. (1961) tumor. Cancer Res. 21:52. Mollenhauer, H. H. (1964) electron microscopy. The fine cytology of the Walker Plastic embedding mixtures for use in Stain Tech. 39:111. Molnar, Z. and Bekesi, J. G. (1972) Effects of D-glucosamine, Dmannosamine and 2-cleoxy-D-glucose on the ultrastructure of ascites tumor cell "in vitro". Cancer Res. 32:380. 41 .. ; ·~. . -····· ... ·~· ..• . ·-·~--···-- -~· ·-·-~-····-·~----······-·~·- ····--~~., I I I I jMora, P. T., Brady, R. 0., Bradley, R. M. and McFarland, V. W. (1969) 1 Gangliosides in DNA virus-transformed and spontaneously trans! formed tumorigenic mouse cell lines. Proc. Nat. Acad. Sci. USA 1 63:1290. i 'Morgan, H. R. (1968) Ultrastructure of the surfaces of cells infected with avian leukosis-sarcoma virus. J. Virol. £:1133. ll I !Muir, A. R. (1967) The effects of divalent cations on the ultra! structure of the perfused rat heart. J. Anat. (Lond.) 101:239. i Nicolson, G. L. (1971) Differences in topology of normal and tumor cell membranes shown by different surface distributions of I ferritin-conjugated concanavalin A. Nature 233:244. 1 !, I Nicolson, G. L. (1972) Topography of membrane concanavalin A sites modified by proteolysis. Nature, Ne\v Biol. 239:193. I i Novikoff, I i A. B. (1973) Lysosomes: A personal account. In Lysosomes and Storage Diseases. (ed. Hers, H. G. and Van Hoof, F.) Academic Press, N.Y. pp 2-37. I IOberling, i i I Ch and Bernhard W. (1961) The morphology of the cancer cell. In The Cell. Vol. V. (ed. Brachet and Mirsky), Acadamic Press, N.Y. p 405. ( Ohta, R., Pardee, A. B., McAuslan, B. R. and Burger, M. M. (1968) ' Sialic acid contents and controls of normal and malignant cells. Biochem. Biophys. Acta 158:98. Oppenheimer, S. B. (1975) Functional involvement of specific carbohydrate in teratoma cell adhesion factor. Exp. Cell Res. (in press). Oppenheimer, S. B. and Humphreys, T. (1971) Isolation of specific macromolecules required for adhesion of mouse tumor cells. Nature 232:125. Oppenheimer, S. B. and Odencrantz, J. (1972) A quantitative assay for measuring cell agglutination of mouse teratoma and sea urchin embryo cells by Concanavalin A. Exp. Cell Res. 1.]_:475. Ozanne, B. and Sambrook, J. (1971) Binding of radioactively labeled Concanavalin A and wheat germ agglutinins to normal and virus transformed cells. Nature, New Biol. 232:156. Pinta da Silva, P. and Gilula, N. B. (1972) Gap junction in normal and transformed fibroblasts in culture. Exp. Cell Res. 11:393. ! I I I ! r·--·-···· ---·· -------- ··- ··---··--····. --·-····- · · ·-· - ------··- -·· · · --·· ---·------...... ----·--····--·-··-----·· -· · --~-- .· · --·-------------·· · · . . . ··-----~ i I I !Purdom, L., Ambrose, E. J. and Klein, G. (1958) A correlation between · electrical surface charge and some biological characteristics during the stepwise progression of a mouse sarcoma. Nature 181: 1586. Rabinovitch, M. and De Stefano, M. J. (1973) Manganese stimulates adhesion and spreading of mouse sarcoma I ascites cells. J. Cell Biol. 59:165. i i II I ! i 'Revel, J. P. and Karnovsky-, M. J. (1967) Hexagonal array of subunits t in intercellular junctions of the mouse heart and liver. J. Cell Biol. 33:C7. Reynolds, E. S. (1963) The use of lead citrate at high pH as an electron opaque stain in electron microscopy. J. Cell Biol. 17: 208. Rosenblith, J. A., Ukena, T. E., Yin, H. H., Berlin, R. D. and Karnovsky, M. J, (1973) A comparative evaluation of the distribution of Con A binding sites on the surfaces of normal, virally transformed and protease treated fibroblasts. Proc. Nat. Acad. Sci. USA 70:1625. Ross, R. and Benditt, E. P. (1961) Wound healing and colligen formation I. Sequential changes in components of Guinea pig skin wounds observed in the electron microscope. J. Biophys. Biochem. Cyto 1. 11: 6 77 . Rutishauser, U. and Sachs, L. (1974) Receptor mobility and the mechanism of cell-cell binding induced by concanavalin A. Proc. Nat. Acad. Sci. USA 71:2456. Sachs, L., Inbar, M. and Shinitzky, M. (1974) In Control of Prolifer~tion in Animal Cells. (eds Clarkson, B. and Baserga, R.) Cold Spring Harbor Laboratory. Sanders, F. K. (1973) The presence of viruses in uninoculated tissue cultures: Sources and methods of detection. In Contamination in Tis~ Culture. (ed. Foch, J.) Academic Press, N.Y. p 243. i iSela, B., Lis, H., Sharon, N. and Sachs, L. (1971) Quantitation of N-acetyl D-galactosamine like sites on the surface membrane of normal and transformed mammalian cells. Biochem. Biophys. Acta 1_49:564. i 1 Selby, C. C. (1955) An electron microscope study of mammalian skin by thin section. I. Derma-Epidermal junction and basal layer. J. Biophys. Biochem. Cytol. .!.:429. I _j l~3 1-----~----~-------~--~--=----·--·-- . ·······-·-·········-··-----------~---·--·-------····-········-·····--··········-···---- ····------·---·····-···-----················, I Shea, S. M. (1971) Lanthanum staining of the surface coat of cells: Its enhancement by the use of fixatives containing alcian blue or cetylpyridinium chloride. J. Cell Biol. 21:611. I! ) Sheridan, J. D. (1970) Low resistance junctions between cancer cells in various solid tumors. J, Cell Biol. 45:91. Tremblay, G. and Babai, F. (1972) Intercellular junctions between Novikoff hepatoma cells and hepatic cells. Exp. Cell Res. _74: 355. Vogt, M! Delbecco, R. (1962) Studies on cells rendered neoplastic by polyoma virus: The problem of the presence of virus related materials. Virology 16:41. ·weiss, L. and Subjeck, J. R. (1974) The densities of colloidal iron hydroxide particles bound to microvilli and the spaces between them. Studies on glutaraldehyde-fixed Ehrlich ascites tumor cells. J. Cell Sci. 14:215. \vessel, w. and Bernhard, \v. (1957) Vergleichende elektron. mikroskopische untersuchung von Ehrlich-und Yoshida Ascitestumorzellen~ ~ Z. Krebsforsch. 62:140. Wu, H. C., Meezan, E., Black, P. H. and Robbins, P. W. (1969) Comparative studies on the carbohydrate containing membrane components of normal and virus-transformed mouse fibroblasts. I. Glucosamine labeling patterns in 3T3, spontaneously transformed 3T3, and SV40 transformed 3T3 cells. Biochemistry ~:2509. I I ... ! Table l. Relative Occurrence of Each Association After Agglutination. a) Treatmen!:_ .p.p- Relative Percentage of Occurrence of Each Association Microvillous Intermediate Close Con A 1 hour 4 hour 60-80 60-80 20-30 20-30 5-15 5-15 llfactor 11 1 hour 4 hours b) 10-25 10-25b) 75-90 75-90 0 0 l Manganese ion 0' 3' 30 minutes 4 hours Untreated 100 0 0 0 0 0 0-10 0 0 a) The 11Relative Percentage of Occurrence of Each Association" represents a rough estimate which encompasses both the number of times each association was seen and the relative area occupied by that association. b) Microvilli appeared to be held by 'intermediate' type association. i i i j : i___,.__ ·---~-·--·······- ···-··------·· ··----··----- l -------------------··-----------····'·····--------------·-·· -------····-·--·---··----.....J . --- -·-· ___ , ---··-- ·~ ------·· --·-----~---· -- ... -----~- ·-------~- --·· . ---· ---··-- ...., I l Ii I Plates 45 46 Plate 1 A. Light micrographs of untreated S-180 after washing \vith Hanks 1 BSS and fixation with 2.5% glutaraldehyde and Os04. Note the large nucleus (N) and dark staining body (1':) in the cytoplasm. 4688X. Phase. B. Similar to A. Note the lack of cellular aggregates. Cells appear to be mainly single cells with possibly a few pairs (--"-). 710X. Brightfield. c. Light micrograph of Con A agglutinated S-180, unfixed. large size of the aggregate. 3000X. Brightfield. Note the D. Similar to C, but after 4 hours in culture, unfixed. Note that the size of the aggregate has decreased in size compared to Plate lC. 1875X. Brightfield. 47 D 48 Plate 2 Example of an untreated S-180 cell after removal from peritoneum and washing \V"ith Hanks' BSS. Fixation was \V"ith 2.5% glutaraldehyde and Os04. Note the irregular shape of the highly fissured nucleus. Also the apparent emperipolesis (~). 7560X. Stained with lead citrate and uranyl acetate. 49 so Plate 3 A. Cells were treated with Con A for l hour and fixed in modified Karnovsky' s fixative and Os04. Micrographs shmvs a nuclear caniculus (c) associated with the nucleolus. Also note nuclear vesicles (-) and unusual aggregate of tvhat could be viral cores (.-..). Pars amorpha, pa; Pars granulosa, pg; Pars fibrosa, pf; chromatin, Ch; nuclear pore (.-.). 24,600X. Stained \vith lead citrate and uranyl acetate. B. Cells treated as in A. Note two large nucleoli, nuclear caniculi (c), and cytoplasmic channel (~) associated with one of the nucleoli. 17, lOOX. Stained with lead citrate and uranyl acetate. 51 52 Plate 4 Cells treated with "factor" for one hour and then fixed in modified Karnovsky' s and Os04. Plate shmvs centrosphere region, note Golgi apparatus (G), and numerous vesicles and vacuoles in the cytoplasm. The centriole (Ct) is prominent and so is the large nucleolus (n). Nucleus (N), mitochondrion (M). 42,600X. Stained with lead citrate and uranyl acetate. 53 ------~--------------------~-------------·--·--·---~ - 54 Plate 5 A. Cells treated with Con A for one hour and fixed in modified Karnovsky's and Os04. Note the Golgi apparatus (G) with large number of associated vesicles and vacuoles (v), the dense staining body(~), endoplasmic reticulum (E), and autophagic vacuole (av). 25,400X. Stained with lead citrate and uranyl acetate. B. Cell treated as in A. Plate shows the condensation of several residual bodies with an autophagic vacuole. 49,000X. Stained with lead citrate and uranyl acetate. C & D. Untreated cells fixed in modified Karnovsky's and Os04. Shows a large lipid droplet in the cytoplasm and the apparent exocytosis of a lipid droplet. Mitochondrion (H). C. 8560X. D. 8320X. Stained with lead citrate and uranyl acetate. E. Cell treated as in A. Note the large dark and irregularly stained bodies, and the myelin figure (-+). 13650X. Stained with lead citrate and uranyl acetate. F. Cell treated with factor for one hour and fixed with modified Karnovsky's and Os04. Example of another type of dark staining body. 40,800X. Stained with lead citrate and uranyl acetate. I 55 56 Plate 6 A. Cells treated with "factor'' for one hour and fixed in modified Karnovsky's and Os04. High magnification of virus particles in cytoplasm. 114,660X. Stained with lead citrate and uranyl ace-· tate. B. Untreated S-180 cells after \vashing with Hanks' BSS and fixa.., tion with 2.5% glutaraldehyde and Os04. Note the variety of surface projections shmvn by the two cells and the lack of interaction bet\veen the cells. Apparent emperipolesis (\/). 15,700X. Stained with lead citrate and uranyl acetate. C. Cells treated with Con A for one hour then fixed with modified Karnovsky 1 s and Os04. Shmvs the interactions between normal (Y) and an apparently dying cell (Z). The obliqueness of the section does not allow a clear determination of the intercellular area. 30,000X. Stained with lead citrate and uranyl acetate. 57 58 Plate 7 A. Cells treated with Con A for one hour and fixed \vith modified Karnovsky's and Os04. Note the large number of microvilli which have been agglutinated to the surface of the parent cell and also the number of cell-cell interactions bet\veen the microvilli and other membrane projections. 18,100X. Stained with lead citrate and uranyl acetate. B. Untreated cells fixed with modified Karnovsky's and Os04. Compared with A, the microvilli tend to project from the cells and are not flattened to the cell surface; also the number of cellcell contacts are reduced. Mitochondrion (M). 18,100X. Stained with lead citrate and uranyl acetate. 59 60 Plate 8 A. Cells treated with Con A for 4 hours and fixed with modified Karnovsky 1 s and Os04. Plate shows an aggregate of five cells. Several types of associations appear to be involved. Note that the microvilli in the free regions are apparently projecting out from the cell, and are no longer agglutinated to the surface. Intermediate association(-), close association~). 15,650 X. Stained with lead citrate and uranyl acetate. B. Cells treated as in A. An extreme example of the extensive interactions which can occur bet\veen villi of adjacent cells, Note how adjacent cell villi are pushing into the cell surface of the other cell (~) and that some of the villi appear to branch ( +). 35 ,400X. Stained with lead citrate and uranyl acetate. C. Cells treated with Con A for one hour and fixed with modified Karnovsky's and Os04. A region in which villi are pinned between adjacent cell.s. Note some villi are agglutinated to both cells. A close association is also seen in this region (-). 10,300X. Stained with lead citrate and uranyl acetate. 61 62 Plate 9 A. Cells treated with Con A for one hour and fixed in modified Karnovsky's and Os04. Note lightly stained material occurs only in region where intermediate association (--) occur. 68,000X. Stained with lead citrate and uranyl acetate. B. Cells treated with Con A for 4 hours and fixed with modified Karnovsky's and Os04. Region of interaction between two cells in which intermediate associations (bars) and microvilli interactions (-+)occur. 21,900X. Stained ~vith lead citrate and uranyl acetate. C. Cells treated as in B. T~vo cells held by several small intermediate associations (-). 22,300X. Stained with lead citrate and uranyl acetate. 63 Plate 10 A. Cells treated with Con A for one hour and fixed in modified Karnovsky's and Os04. An example of the close association seen after agglutination. 203,000X. Stained with lead citrate and uranyl acetate. B. Cells treated as in A. An example of a common type of association, probably a close association cut obliquely in several regions (--). Note villi between cells. 37,300X. Stained lead citrate and uranyl acetate. C. Cells treated as in A. This shows adjacent cells held by an extensive close association. 68,000X. Stained with lead citrate and uranyl acetate. D. Cells treated as in A. Another example of an extensive close association. 54,600X. Inset illustrates the close apposition of adjacent membranes and also shows an unusual thickening of the membrane which appears to be a third bilayer (-H-). 22l,OOOX. [ ] region enlarged. Stained with lead citrate and uranyl acetate. 65 66 Plate 11 Cells treated \vith a high concentration of 11 factor 11 for one hour, then fixed with modified Karnovsky' s and Os04. Shmvs the extensive pinocytosis and phagocytosis of the flocculent material on the cell surface. About 7500X. Stained with lead citrate and uranyl acetate. 67 .. r . I 68 Plate 12 A. Cells _treated >vith a high concentration of "factor" for one hour then fixed with modified Karnov~ky's and Os04. Higher magni·fication of the flocculent material on the cell surface. 60,000X. Stained with lead citrate and uranyl acetate. B. Cells treated as in A. Note the large amount of material adhering to the cells' surface. Interactions seem to be the result of the material on the cells' surface (~). Also seen are numerous smaller lysosomal vesicles (y) possibly GERL (Novikoff, 1973) which appear to fuse (--)with the larger residual body (Rb). 13,700X. Stained with lead citrate and uranyl acetate. c. Cells treated as in A. Shmvs the large amount of material sticking to the surface of the cells. Interactions appear to be due to the material on the cell surfaces. Again numerous lysosomal vesicles are present (y). Note the swollen ER (E) filled with a moderately stained material. 20,000X. Stained tvith lead citrate and uranyl acetate. D. Cells treated with a low concentration of factor for one hour then fixed \vith modified Karnovsky's and Os04. Shows the exocytosis of two large myelin figures. It appears as if the cells are being forced apart by the vacuoles. Note the surfaces of the adjacent cells appear to be complementary. Several autophagic vacuoles (av) can be seen in the cells. 20,000X. Stained with lead citrate and uranyl acetate. 69 70 Plate 13 Cells treated \vith low concentration of "factor" for one hour then fixed with modified Karnovsky's and Os04. Note the numerous villi and lack of vacuolization as seen at higher concentrations of "factor". Also note viral aggregates (V) in the cytoplasm. About 7000X. Stained with lead citrate and uranyl acetate. 71 72 Plate 14 A. Cells treated for one hour with "factor" at lmv- concentration, then fixed with modified Karnovsky's and Os04. Sho-.;vs region of intermediate association (rt); note microvilli involved(--). 23,600X. Stained with lead citrate and uranyl acetate. B. Cells treated as in A. Sho-.;vs patchy nature of the interaction bet-.;veen these cells. Bars indicate regions in which patchy material occurs. 23,700X. Stained with lead citrate and uranyl acetate. c. Cells treated as in A. Sho-.;vs region of intermediate interaction. Sl,OOOX. Stained -.;vith lead citrate and uranyl acetate. D. Higher magnification of A. Note the interaction appears to be mediated by patches of staining material (-). 65,400X. Stained with lead citrate and uranyl acetate. 73 74 Plate 15 A. Cells were fixed immediately after addition of Mn++ with 2.5% glutaraldehyde. Note the lack of large aggregates. 600X. Phase microscopy. Postfixed with Os04. B. Cells were fixed three minutes after addition of Mn++ with 2.5% glutaraldehyde. Shmv-s a large aggregate of cells. 1500X. Phase microscopy. Postfixed with Os04. C. Cells were fixed 30 minutes after addition of Hn++ with 2.5% glutaraldehyde. High magnification of a small aggregate. Microvilli can be made out in areas. 3750X. Phase microscopy. Postfixed with Os04. D. Cells were fixed four hours after addition of Mn++ with 2.5% glutaraldehyde. Note vacuolization of cells, also no aggregates are seen. 1500X. Phase microscopy. Postfixed \vith Os04. 75 c 76 Plate 16 A. Cells were treated with 20 mM Mn++. 2.5% glutaraldehyde was added immediately after addition of Mn++. Shows the interactions between the cells are mediated by surface projections. Note that interactions are not very extensive. 7000X. Postfixed ~vith Os04. Stained 'vith lead citrate and uranyl acetate. B. Cells treated as in A. Shows interactions can occur between several types of surface projections. 12,500X. Postfixed with Os04. Stained with lead citrate and uranyl acetate. c. Cells were treated with 20 mM Mn++. 2.5% glutaraldehyde 'vas added after three minutes. Shows that the number of villi involved in interaction has increased and some cell flattening has occurred. 14, lOOX. Postfixed 'vith Os04. Stained Hith lead citrate and uranyl acetate. D. Cells treated as in C. Villi interact at point contacts. The cell surface does not smooth out in the regions of interaction. 27,300X. Postfixed with Os04. Stained with lead citrate and uranyl acetate. 77 78 Plate 17 A. Cells treated with 20 l11i'1 Mn++ for 30 minutes then fixed wHh 2.5% glutaraldehyde. After extended incubation, interactions are still similar to those of shorter duration. 17,100X. Post-fixed with Os04. Stained with lead citrate and uranyl acetate. B. Cells treated as in A, but fixed using the ruthenium red staining method. There appears to be no differences in staining properties between those areas of interaction and those not involved in agglutination. 38,200X. c. Cells treated \vith 20 mM Mn++. 2.5% glutaraldehyde Has added immediately after addition of Mn++. Plate shmvs the interactions 168,000X. Postfixed \vith are tight between villus membranes. Os04. Stained with lead citrate and uranyl acetate. D. Cells treated with 20 l11i'1 Hn++ for three minutes then fixed with 2.5% glutaraldehyde. Again villus interactions are very tight~JPqte also obvious microfilaments in the villi (--). 226,000X. Postf~ed with Os04. Stained with lead citrate and uranyl acetate. ?r . , ..,., > 'J 79 80 Plate 18 A. Cells were treated with 20 rru'1 Hn+t- for four hours, then fixed with 2.5% glutaraldehyde. Cells have become very vacuolated. Cytoplasm seems to have two distinct areas, one staining darker (d) than the other (1). Mitochondria stain darkly. Villi are much reduced in number. 7000X. Post fixed \vith Os04. Stained with lead citrate and uranyl acetate. B. Cells treated as in A. Unusual cell with large number of vacuoles and completely smooth surface. 7800X. Postfixed with Os04. Stained with lead citrate and uranyl acetate. C. Cells treated as in A. Cell has villi and is not vacuolated. Cytoplasm appears to have a dark and a light staining region. 3510X. Postfixed with Os04. Stained \vith lead d_trate and uranyl acetate. 81 A .B·