IN VITRO Volume14, No. 8, 1978 Allrightsreserved9
H A M S T E R P E R I T O N E A L M A C R O P H A G E S IN V I T R O : SUBSTRATUM ADHESION, SPREADING, PHAGOCYTOSIS AND P H A G O L Y S O S O M E F O R M A T I O N 1 K.-P. CHANG2 Laboratory of Parasitology, The Rockefeller University, New York, New York 10021
SUMMARY A series of manipulations designed to promote cell adhesion and spreading made it possible to maintain satisfactorily hamster peritoneal macrophages in vitro for up to 30 days. The essential requirements for this include in vivo stimulation of the peritoneal cavity, coating of the substratum with polylysine, and the use of HEPES-buffered medium 199 supplemented with horse serum {10% ), fetal bovine serum (10%), and lactalbumin hydrolysate (0.5%). Results with the single deletion of the medium components indicate that serum factors are essential for optimal spreading, and horse serum and lactalbumin hydrolysate for the adhesion of in vivo stimulated macrophages on coated glass surface. The thorotrast-labeling method revealed that secondary lysosomes are especially numerous in cultured cells, which otherwise resemble mouse macrophages in cellular organization, as shown by scanning and transmission electron microscopy. More than 95% of the cultured cells manifested cytochalasin B-sensitive phagocytosis of polystyrene latex spheres which, along with morphologic and ultrastructural evidence, indicate the homogeneity of cell population. Erythrophagocytosis of hamster macrophages was demonstrated by scanning electron microscopy and found higher after opsonization implying the presence of receptors for immune ligands on their cell surface. K e y words: hamster; macrophage; cultivation; adhesion-spreading; phagocytosis; phagolysosome. I NTRODUCTION
ciples established with mouse macrophages but also as a necessity in several areas of research. Notable examples for the latter include aspects of the biology and immunology of cancers and infectious diseases in which host specificity of the causative agents often calls for the use of rodents other than mice as model animals. In vitro assays for the host-parasite interactions and for the cellmediated immunity in these animals frequently require a prolonged cultivation of the macrophages because of their role as the effector cells in immunity or the host cells for intracellular pathogens. Of particular interest to us is the macrophage of hamster, which serves as an excellent model for a great variety of human cancers and infectious agents. The work of Bey and Harrington (3) appears to be the only study which deals specifically 'Supported by NIH Grant No. AI-11916 from with the in vitro maintenance of this cell. Their USPHS. 2Recipient of The Irma T. Hirschl Career Scientist method, however, requires a medium conditioned Award. by fibroblasts or, otherwise, supplemented with 663
A major cell element of the reticuloendothelial system is the macrophage, the very archtype of phagocytes and one of the principal cells concerned in immunity. As a representative of this cell, mouse peritoneal macrophages have been extensively studied due largely to the ease of their maintenance in vitro (IL Undoubtedly, invaluable information has been gained through this in vitro model in understanding the general cell functions and metabolism common to all macrophages. Attempts to maintain similar cells from other small rodents in vitro have encountered enormous difficulties (2), thereby largely precluding their use for general experimentation. There remains, however, considerable interest in using these cells not only for generalizing the basic prin-
664
CHANG
40% of serum, and provides no adequate means to ensure the homogeneity of the cultured cells. Moreover, the ultrastructure and endocytic activity of hamster macrophages in vitro have not been characterized. This paper describes a new method by which hamster peritoneal macrophages can be maintained under satisfactory conditions for up to 1 month by improving their initial substratum adhesion and spreading. Population homogeneity as well as the activities of phagocytosis and phagolysosome formation are demonstrated in these cells indicating that they are adequate not only for the work of interest to us but also for other types of general experiments. MATERIALS AND METHODS
Harvesting cells. Cells were harvested from the peritoneal cavity of male Golden Syrian hamsters (70 to 80 g; Lakeview, N.J.), some of which were prestimulated 2 days in advance by intraperitoneal injection of 4% thioglycollate broth, Hanks' balanced salt solution IHBSS), medium 199 or complete culture medium. Into each etherized animal 20 ml HEPES-huffered medium 199 or HBSS with 10 U per ml of heparin were i.p. injected; the peritoneal fluid was withdrawn subsequently with a 10-ml syringe fitted with a 27G l/~. inch needle without sacrificing the animals. Erythrocytes when present were lysed with 0.87% ammonium chloride in water. Cell numbers were counted in a hemacytometer in pooled samples which then were centrifuged at 900 • g for 10 min. Cell pellets were resuspended in appropriate amounts of culture medium to make a desirable cell density for plating. Cell viability was determined by trypan-hlue dye-exclusion test. Culture procedure and conditions. Cells (1 to 2• in 0.2 to 0.3 ml of medium were plated on flying cover slips (18- or 22-mm 2) placed in tissueculture petri dishes 13- or 6-cm in diameter; Falcon) or in the latter directly. Glass cover slips were cleaned by boiling in 1% " 7 X " detergent for several min; after several rinses in double-distilled water, they then were sterilized by immersion in 70% ethanol and flame-dried before use. Preparations were incubated at 37 ~ -+ 1~ C with 5% CO2 in air for 2 to 4 hr to allow cell adhesion and then flooded with 2 to 3 ml of culture medium. After 24 hr, cell cultures were washed with prewarmed HBSS or medium to remove nonadherent cells. Medium was replenished or changed every other day.
Culture medium and variations. During the initial phase of the investigations, a variety of tissueculture media supplemented with 20% to 40% heat-inactivated fetal bovine serum (HIFBS) and antibiotics (100 U per ml of penicillin and 100 t~g per ml of streptomycin) were tested, i.e. NCTC109, CMRL-1066, RPMI-1640, BHK-12, H a m F-12, Eagle's or Dulbecco's M E M , or medium 199 with or without H E P E S (GIBCO). All these media were found unsatisfactory in that most cells became detached in 2 to 3 days a n d / o r were overwhelmed by the growth of flbroblasts. Consequently, various treatments were introduced to avert these unfavorable conditions. Hydroxyurea and bromodeoxyuridine were tested at various concentrations to prevent excessive proliferation of flbroblasts. To promote cell adhesion and spreading, the substratum, glass or plastic, was coated with rat-tail collagen (4), fibrinogen 1250 /~g per 3-cm petri dish), or poly-D, L-lysine (type V or B-VII, Sigma; 100 to 250 ~g per 18- to 22em 2 cover glass); in all cases one side of the cover glass was covered with 0.1 to 0.3 ml of an aqueous solution of these compounds. The cover slips were incubated for 2 to 4 hr at room temperature after which the fluid was removed by vacuum aspiration. HEPES-buffered medium 199 plus antibiotics was chosen as the basal medium and was supplemented at a final concentration of 20% with heat-inactivated FBS or horse serum {HIHS), or a combination of both in equal amounts. Also tested in addition to the above treatments were some factors known to stimulate cell adhesion a n d / o r spreading, e.g. potassium pyruvate at 0.1 to 0.5 mg per ml (5), manganese chloride at 0.001 to 10 mM, dithiothreitol at 0.01 to 2 mM (6), and lactalbumin hydrolysate at 0.5%. Light microscopy. For routine observations, cultures were examined directly under phase contrast or after fixation in methanol and staining with Giemsa. To assess the effects of various treatments, total cell numbers and the percent of spread cells were determined in an area of 0.15 mm 2 under phase contrast as criteria for the adhesion of macrophages to the substratum and their spreading, respectively. Determinations were made from at least five random areas in each of triplicate samples cultured in vitro for 2 to 4 days. Electron microscopy. For transmission electron microscopy, cultures were fixed at 4 ~ C in 2% glutaraldehyde in 0.1 M cacodylate buffer for 2 hr. Cells were removed from culture vessels with the aid of a rubber policeman and centrifuged at 2000 x g. Cell pellets then were postfixed in 1%
HAMSTER PERITONEAL MACROPHAGES osmium tetroxide in the same buffer for 2 hr, stained in-the-block in a saturated aqueous solution of uranyl acetate for 1 hr, dehydrated in graded series of acetone or ethanol, and embedded in Epoxy resin. Thin sections were cut with a diamond knife, stained with lead citrate, and examined with a Phillips E . M . 300 electron microscope. For scanning electron microscopy, macrophages cultured on round cover slips (9-ram diamter) were fixed in glutaraldehyde and dehydrated as mentioned above; specimens then were critical-point dried in amyl acetate or acetone and coated with gold-palladium in a vacuum coater. Preparations were examined at a tilt angle of 45 ~ in a Etec Autoscan scanning electron microscope. Phagocytosis study. Phagocytotic activity of cultured hamster macrophages was assessed by their ability to ingest polystyrene latex heads (2~m diamter; Dow Chemicals), duck and human erythrocytes. All particles were washed three times in phosphate-buffered saline by centrifugation. Erythrocytes were aged at 4 ~ C for several weeks in HBSS before use. For opsonization, 108
665
cells were incubated for 20 min at 37 ~ C in medium 199 containing 1:200 dilution of heat-inactivated rabbit antisermn with an agglutination titer of ]0 -3. Cell clumps were dispersed by extensive vortexing and the remaining aggregated cells were removed by brief centrffugation. Particles suspended in medium 199 with 20% H I F B S were incubated with the macrophages at a particle/cell ratio of 100:1 and 10:1 for latex beads and erythrocytes, respectively. Blockade of particle ingestion by macrophages was determined in experiments with latex beads by using cytochalasin B at a final concentration of 10/~g per ml. Duplicate samples were withdrawn and Giemsa stained at different periods after incubation; the number of particles per cell and the percentage of particlecontaining cells were determined by examining at least 200 cells in each culture. Some preparations with erythrocytes also were processed for scanning-electron-microscope studies. Labeling of secondarylysosomes. To determine the extent of phagolysosome formation in cultured macrophages, the macrophages were labeled with
FIG. l. Monolayer of hamster peritoneal macrophages cultured in vitro for 2 days. Note the presence of round, spindle-shaped and tripolar cells. Giemsa stain, x200. FIG. 2. Monolayer of hamster peritoneal macrophages cultured in vitro for 30 days. Note the voluminous increase of cytoplasm and multipolar appearance of the cells not seen in the younger culture shown in Fig. 1. Giemsa stain, x200.
666
CHANG
667
HAMSTER PERITONEAL MACROPHAGES
FIG. 5. a, Phagocytosis of 2-gm latex beads by 3-day-old hamster peritoneal macrophages for 50 min in vitro. Note the normal appearance of the cells, x400. b, Adherent and ingested beads can be distinguished by the presence of dark "rims" around the latter. Giemsa stain, x900. FIG. 6. Effects of eytochalasin B on the phagocytosis of latex particles by hamster macrophages in vitro. Note the "round-up" appearance of the treated cells, most of which contain no particles. Giemsa stain, x400.
an electron-dense marker, thorotrast, essentially as described by Jones and Hirsch (7). In brief, macrophages cultured in vitro for 1 to 2 days were washed with HBSS and incubated in medium 199 plus 40% HIFBS containing thorotrast at 1:20 or 1:40 dilution for 16 to 20 hr; cells then were rinsed to remove free particles. At different times after the removal of the dense marker, cell cultures were processed by transmission electron microscopy and thin sections were examined for the presence of the label in cytoplasmic vacuoles.
RESULTS
Recovery and composition of peritoneal exudates. From each hamster, approximately threefourths of injected HBSS, or 15 ml, was recovered and contained 12x10" nucleated cells of which macrophages, granulocytes and lymphocytes constituted about 70%, 25% and 5% of the total cell population, respectively. Cell viability was more than 97% at the time of plating as judged by trypan-blue exclusion test.
FIG. 3. The appearance of living hamster macrophages in vitro for 3 days. Note the appearance of spread and nonspread (arrow) cells corresponding to spindle-shaped and round cells seen in Giemsastained materials (Fig. 1k Phase-contrast x700. FIG. 4. The appearance of living hamster macrophages in vitro for 30 days. Note the granular cytoplasm and the occurrence of peripheral ruffled membranes and membranous skirt anchoring cells to the substratum. Phase contrast x700.
668
CHANG
HAMSTER PERITONEAL MACROPHAGES
Adhesion and spreading of macrophages. The loss of cells alter incubation in vitro was significantly minimized by two essential treatments: stimulation of the peritoneal cavity 1.5 to 2 days before harvest, and coating of substratum with substances promoting cell adhesion. Stimulation of the peritoneal cavity also markedly increased the cell yield, and the complete culture medium (see below) was found generally superior to thioglycollate broth or HBSS for this purpose. Essentially identical results were obtained with surfaces coated with fibrinogen, rat-tail collagen or polylysine; the last compound was routinely used for convenience. Under the conditions mentioned above, optimal cell adhesion and spreading were obtained in the complete culture medium consisting of HEPES-buffered medium 199 supplemerited with 10% HIFBS, 10% HIHS, 0.5% lactalbum~n hydrolysate, 0.5% glucose, hydroxyurea (0.2 mM), and antibiotics (100/~g per ml of streptomycin and 100 U per ml penicillin). The last two components were incorporated to limit the growth of fibroblasts and to prevent bacterial contamination, respectively. Bromodeoxyuridine proved to be toxic to the macrophages. When nonstimulated macrophages were cultured in this medium on uncoated surface, there was substantial cell detachment. However, the total cell number that became detached during the first 3 days in vitro was calculated to be 15% in this medium in contrast to 60% in medium 199 with 20% HIFBS as the only additive. Macrophages remained as monolayers at least up to 30 days in vitro under optimal conditions indicating that the loss of cells was insignificant although this was not determined beyond day 4. The factors in the complete medium contributing to the adhesion and spreading of stimulated cells on glass or plastic surfaces were determined by the single-deletion method. In complete medium for up to 4 days in vitro, the cell density, as a measure of cell adhesion, was about 350 cells per 0.15 mm 2 of which approximately 50% was well spread. Neither spreading nor adhesion could be increased by raising the serum level in the complete medium. The appearance of spread and nonspread cells, as well as the normal cell density of such preparations, is shown in Figs. 1 and 3. Omission of single components from the culture medium showed the following effects: all protein substances (HIFBS, HIHS and
669
lactalbumin hydrolysate) appeared to contribute to cell spreading; HIHS and lactalbumin hydrolysate were essential for cell adhesion; and additional glucose had little, if any, beneficial effect. No improvement was noted with the inclusion of other factors tested, e.g. potassium pyruvate, manganese chloride or dithiothreitol. All macrophages in complete medium spread well several weeks after the initial plating (Figs. 2, 4). Morphology. The appearance of hamster peritoneal macrophages changed markedly with time of incubation in vitro. During the first several days, they were a mixture of round, spindleshaped or tripolar cells in both phase-contrast (Fig. 3~ and Giemsa-stained materials (Fig. 1L The nuclei were oval dense bodies always located near one pole of the cells and filled about one-half to one-third of the total cell volume (Figs. 1, 3, 5a). The cytoplasm was very granular (Figs. 4, 5bL After 2 weeks or longer in vitro, macrophages became multipolar cells and were much larger in size with a cytoplasmic volume, which was now sufficient to accommodate 10 to 20 nuclei tFigs. 2, 4). These "matured" macrophages appeared to be tightly anchored to the substratum by a thin peripheral membranous skirt and had elevated cell body filled with dense granules and laced with ruffled membranes at the periphery ~Fig. 4). Ultrastructure. Scanning electron microscopy at low magnification showed that hamster peritoneal macrophages consisted of round and well spread cells in agreement with the lightmicroscope observation; in addition, both cell types had many surface membrane appendages indicating the "healthy" state of these cells maintained in vitro ~Fig. 7). Higher magnification further revealed numerous surface microvilli, filopodia and membrane veils; the latter two apparently anchored the cells to the substratum (Figs. 8, 9). Transmission electron microscopy showed that the macrophages possessed surface cytoplasmic extensions, corresponding to filopodia or microvilli seen by scanning electron microscopy, and the usual complement of cell organeUes, including nuclei, mitochondria, rough endoplasmic reticulum, Golgi apparatus, eentrioles, microtubules and numerous cytoplasmic vacuoles IFigs. 10, 11). The normal appearance of these cell organdies suggests that the cells are physiologically active. The ultrastructural features de-
F IG.7. Surveyscanning-electronmicrographof hamster peritoneal macrophagescultured in vitro for 2 days. Note the abundance of surface protrusions on spread cells and the occurrence of variations in cell shapes, x1200.
670
CHANG
HAMSTER PERITONEAL MACROPHAGES scribed above were representative of all preparations studied at different times after in vitro cultivation. No granulocytes or lymphocytes were seen; fibroblasts were rare and had a different ultrastructural characteristic. Phagoeytosis and phagolycosome formation. In vitro cultured hamster macrophages showed high activity of phagocytosis toward latex beads added to the culture medium. During a 2-hr incubation period with the particles, cells did not change their normal appearance {Fig. 5) and ingested the particles at a high rate as determined by the increase with time of particle numbers per cell and the percentage of particle-containing cells. During the same period of time, in the presence of cytochalasin B (10 pg per ml}, cells became rounded {Fig. 6) and contained negligible numbers of particles without appreciable increase after 30 min. The kinetics of particle ingestion was about 1.5 particle per cell per l0 min and more than 95% of the cells contained latex beads after 1 hr incubation. In vitro cultured cells also ingested human and duck erythrocytes. Quantitative determination during 1-hr incubation period showed that 60% of the cells ingested about five opsonized erythrocytes per cell, whereas only 25% took in two "aged" erythrocytes per cell. Thus opsonization significantly increased the rate of erythrophagocytosis. Scanning-electron-microscope study showed that erythrocytes were initially trapped by filopodia (Fig. 8) and eventually became buried completely under a thin veil of the cytoplasmic membrane of the macrophages (Fig. 9). In specimens prepared for determining the phagolysosome formation, thorotrast granules were found to accumulate readily and profusely in numerous cytoplasmic vacuoles indicating high pinocytotic activity and abundance of secondary lysosomes in the hamster macrophages IFig. 10). Upon further incubation for 7 days after the removal of the dense label, thorotrast granules appeared to persist in all vacuoles, most of which, however, were only lightly labeled (Fig. 11 ).
671
DISCUSSION The results of the present study provide a reasonably simple procedure for maintaining an adequate and homogeneous population of hamster peritoneal macrophages in vitro on a long-term basis. The method described requires less animal serum and ensures a longer cell survival time under satis|actory conditions than previously offered (3). Several difficulties encountered earlier are overcome by a number of manipulations introduced in the present study. Proliferation of fibroblasts that tends to overwhelm the macrophages is prevented by the use of a DNA biosynthesis inhibitor, hydroxyurea, which as expected, has no adverse effcet on the nondividing macrophages. Extensive loss of cells from the substratum is significantly minimized by the treatments known to increase cell adhesion and spreading. The factors that promote macrophage adherence and spreading include stimulation of the peritoneal cavity, which is known to enhance the spreading of mouse macrophages (1), and coating of substratum with polylysine, fibrinogen or collagen. While all three compounds are equally effective, the underlying mechanisms whereby they facilitate cell attachment and spreading may be different. The increase of cell adhesion mediated by fibrinogen and collagen is not unexpected in view of the presence of fibrin receptors on the surface of the macrophages (8) and the normal role of collagen in intercellular adhesion of normal cells (9). By virtue of its highly positive electrostatic charge, polylysine may effect a similar result by interacting directly with negatively charged cell surface components (10-12) or by increasing the deposition onto the substratum of spreading/adhesion factors present in the culture medium. The latter possibility is very likely as the spreading of hamster macrophages is minimal even on a coated surface in the absence of sera which have been shown to contain cell-sprcading factor (13). Results with the single-deletion experiments indicate that all proteinaceous components in the culture medium contribute, in vary-
FIG. 8. Scanning-electron micrographs showing the engulfment of a human red blood cell by pseudopodia of a hamster macrophage at the initial phase of erythrophagocytosis. Macrophage membrane appears to "flow" over the surface of the red cell closely following its contour at places. Note the presence of numerous surface mierovilli as well as filopodia and rolled membrane skirt that anchors the cell to the substratum, xl0,000. F1c. 9. Phagocytosis of human erythrocytes by a hamster peritoneal macrophage. One red blood cell is fully covered with a microvillus-studded thin veit of the macrophage membrane suggesting coalescence of the pseudopodia trapping the red cell during the initial enguffment phase. •
672
CHANG
HAMSTER PERITONEAL MACROPHAGES ing degrees, to cell spreading a n d / o r adhesion. Particularly striking is the marked increase of cell density at 20% H I H S in the absence of H I F B S {not shown) suggesting that the former may play a more crucial role in the adhesion of hamster macrophages to the substratum. This effect of H I H S has not been noticed in previous studies with mouse peritoneal macrophages {14, 15). Also of interest is the finding that the maximal spreading of hamster macrophages requires the simultaneous presence of H I H S and H I F B S , and thus differs from that of mouse macrophages or other cell types for which H I F B S alone is sufficient {13, 16). The lack of noticeable effects of manganese ions and dithiothreitol on hamster macrophages further distinguishes them from mouse macrophages whose spreading is markedly enhanced by these substances (5). Clearly, the various treatments introduced into the culture system produce favorable conditions for hamster macrophages to remain as a monolayer of adherent cells. Further detailed analysis is needed to determine the precise mechanisms whereby these treatments may induce these effects. Light- and electron-microscope studies showed that all cultured cells have an identical staining property and cellular organization, thus providing evidence for the homogeneity of the cell population. In general, they resemble mouse macrophages in morphology and uhrastructure, except for a more elevated cell body and perhaps more abundant microviUi and filopodia on the cell surface. These features may simply reflect a difference in the degree of spreading between these two kinds of cells. Indeed, hamster macrophages never become completely flattened to the extent of lacking surface ridges and folds as do the mouse macrophages (17). The vitality of the in vitro cultured hamster maerophages is indicated not only by their normal appearance and ultrastructural integrity but also by the vigorous pinoeytosis of thorotrast granules. Transmission electron microscopy showed that the dense marker accumulates at a very high density in numerous cytoplasmic vacuoles of hamster macrophages.
673
These vacuoles are interpreted to be secondary lysosomes in accordance with the general assumption that exogenous particulate matters are shuttied into these cell organelles as a consequence of pinosome or phagosome-lysosome fusions (7, 18). That the cultured ceils are phagocytes is indicated by their ability to ingest latex beads and erythrocytes as well as by the inhibition of this activity by cytochalasin B. Scanning electron microscopy of erythrophagocytosis by hamster macrophages provides evidence supporting the contention that the ingestion is accomplished by a coalescence of the cell-surface pseudopodia wrapping the erythrocytes (19). The abundance of surface microvilli and filopodia in these cells might favor this type of phagocytosis instead of the alternative means by which erythrocytes are interiorized by "sinking into" spherical craters formed on the surface of fully spread cells {17). The extent of erythrophagocytosis of hamster macrophages increases significantly after opsonization suggesting that they possess surface receptors for immune ligands for recognition. Preliminary studies, indeed, indicated that hamster macrophages have both Fc and C3 receptors comparable to those found in mouse macrophages {Michl and Chang, unpublished data). Further investigations to characterize these receptors are currently in progress. REFERENCES 1. Gordon, S., and Z. A. Cohn. 1973. The macrophage. Int. Rev. Cytol. 36: 171-215. 2. Cohn, Z. A. 1974. The isolation and cultivation of mononuclear phagocytes. In: S. Fleischer, and L. Packer {Eds.), Methods in Enzymology. Academic Press, New York, pp. 758-764. 3. Bey, E., and J. S. Harrington. 1971. Cytotoxic effects of some mineral dusts on Syrian hamster peritoneal macrophages. J. Exp. Med. 133: 1149-1169. 4. Bornstein, M. B. 1958. Reconstituted rat-tail collagen used as substrate for tissue cukures on coverslips in Maximore slides and roller tubes. Lab, Invest. 7: 134-137. 5. Pickart, L. R., and M. M. Thaler. 1974. Pyruvate induced cellular flattening and pseudopodia formation blocked by cytochalasin B. In:
FIG. 10. Thin section of a hamster macrophage in vitro for 3 days revealing a eentropedaUy located nucleus {N) lying near the flattened cell surface against the substratum and an elevated cell body with surface extensions and filled with cell organdies. The marker, thorotrast, densely packed some more than 100 vacuoles, presumably secondary lysosomes {$1 ). Also visible are newly formed pinosomes in which the dense marker is sparse, x10,300. FIG. 11. Portion of a hamster macrophage labeled with thorotrast in vitro for 7 days showing eentrioles {Ct) closely associated with Golgi apparatus {G) and mitochondria IM) interspersed among lucent vacuoles with varying amounts of the dense marker, x23,600.
674
CHANG E. T. C. Lee, and E. E. Smith (Eds.), Miami
Winwr Symposium--Biology and Chemistry of Eucaryotic Cell Surface. Academic Press, New York, p. 365. 6. Rabinovitch, M., and M.J. DeStafano. 1973. Macrophage spreading in vitro. II. Manganese and other metals as inducers or as co-factors for induced spreading. Exp. Cell Res. 79: 423-430. 7. Jones, T. C., and J. G. Hirsch. 1972. The interaction between Toxoplasma gondii and mammalian cells. II. The absence of lysosomal fusion with phagocytic vacuoles containing living parasites. J. Exp. Med. 136: 1173-1194. 8. Sherman, L. A., and J. Lee. 1977. Specific binding of soluble fibrin to macrophages. J. Exp. Med. 145: 76-85. 9. Klebe, R.J., P.G. Rosenberger, S. L. Naylor, R. L. Burns, R. Novak, and H. Kieinman. 1977. Cell attachment to collagen. Isolation of a cell attachment mutant. Exp. Cell Res. 104: 119-125. 10. Mazia, D., G. Schatten, and W. Sale. 1975. Adhesion of cells to surface coated with polylysin. Applications to electron microscopy. J. Cell Biol. 66: 198-199. 11. McKeehan, W. L., and R. G. Ham. 1976. Stimulation of clonal growth of normal fibroblasts with substrata coated with basic polymers. J. Cell Biol. 71: 727-734. 12. Tsutsui, K., H. Kumon, H. Ichikawa, and J. Tawara. 1976. Preparative method for suspended
13. 14.
15.
16.
17.
18.
19.
biological materials for SEM by using of polycationic substance layer. J. Electron Microsc. 25: 163-168. Grinnell, F. 1976. Cell spreading factor. Occurrence and specificity of action. Exp. Cell Res. 102: 51-62. Chang, Y. T. 1964. Long-term cultivation of mouse peritoneal macrophages. J. Nat. Cancer Inst. 32: 19-35. Bennett, B. 1966. Isolation and cultivation in vitro of macrophages from various sources in the mouse. Am. J. Pathol. 48: 165-182. Leonard, E. J., and A. Skeel. 1976. A serum protein that stimulates maerophage movement, chemotaxis and spreading. Exp. Cell Res. 102: 434-438. Polliack, A., and S. Gordon. 1975. Scanning electron microscopyof routine macrophages. Surface characteristics during maturation, activation and phagocytosis. Lab. Invest. 33: 469-477. Cohn, Z. A., and M. E. Fedorko. ]969. The formation and fate of lysosomes. In: J. T. Dingle, and H.B. Fell (Eds.), Lysosomes in Biology and Pathology. Vol. 1. North-Holland, Amsterdam, pp. 43-63. Orenstein, J. M., and E. Shehon. 1977. Membrane phenomena accompanying erythrophagoeytosis. A scanning electron microscope study. Lab. Invest. 36: 363-374.
I wish to thank Drs. S. Silverstein and W. Trager for advice and reviewing the manuscript; Drs. J. Michl and D. M. Dwyer for helpful suggestions; and Mrs. D. F. Greene for her skillful preparation of this manuscript.
H A M S T E R P E R I T O N E A L M A C R O P H A G E S IN V I T R O : SUBSTRATUM ADHESION, SPREADING, PHAGOCYTOSIS AND P H A G O L Y S O S O M E F O R M A T I O N 1 K.-P. CHANG2 Laboratory of Parasitology, The Rockefeller University, New York, New York 10021
SUMMARY A series of manipulations designed to promote cell adhesion and spreading made it possible to maintain satisfactorily hamster peritoneal macrophages in vitro for up to 30 days. The essential requirements for this include in vivo stimulation of the peritoneal cavity, coating of the substratum with polylysine, and the use of HEPES-buffered medium 199 supplemented with horse serum {10% ), fetal bovine serum (10%), and lactalbumin hydrolysate (0.5%). Results with the single deletion of the medium components indicate that serum factors are essential for optimal spreading, and horse serum and lactalbumin hydrolysate for the adhesion of in vivo stimulated macrophages on coated glass surface. The thorotrast-labeling method revealed that secondary lysosomes are especially numerous in cultured cells, which otherwise resemble mouse macrophages in cellular organization, as shown by scanning and transmission electron microscopy. More than 95% of the cultured cells manifested cytochalasin B-sensitive phagocytosis of polystyrene latex spheres which, along with morphologic and ultrastructural evidence, indicate the homogeneity of cell population. Erythrophagocytosis of hamster macrophages was demonstrated by scanning electron microscopy and found higher after opsonization implying the presence of receptors for immune ligands on their cell surface. K e y words: hamster; macrophage; cultivation; adhesion-spreading; phagocytosis; phagolysosome. I NTRODUCTION
ciples established with mouse macrophages but also as a necessity in several areas of research. Notable examples for the latter include aspects of the biology and immunology of cancers and infectious diseases in which host specificity of the causative agents often calls for the use of rodents other than mice as model animals. In vitro assays for the host-parasite interactions and for the cellmediated immunity in these animals frequently require a prolonged cultivation of the macrophages because of their role as the effector cells in immunity or the host cells for intracellular pathogens. Of particular interest to us is the macrophage of hamster, which serves as an excellent model for a great variety of human cancers and infectious agents. The work of Bey and Harrington (3) appears to be the only study which deals specifically 'Supported by NIH Grant No. AI-11916 from with the in vitro maintenance of this cell. Their USPHS. 2Recipient of The Irma T. Hirschl Career Scientist method, however, requires a medium conditioned Award. by fibroblasts or, otherwise, supplemented with 663
A major cell element of the reticuloendothelial system is the macrophage, the very archtype of phagocytes and one of the principal cells concerned in immunity. As a representative of this cell, mouse peritoneal macrophages have been extensively studied due largely to the ease of their maintenance in vitro (IL Undoubtedly, invaluable information has been gained through this in vitro model in understanding the general cell functions and metabolism common to all macrophages. Attempts to maintain similar cells from other small rodents in vitro have encountered enormous difficulties (2), thereby largely precluding their use for general experimentation. There remains, however, considerable interest in using these cells not only for generalizing the basic prin-
664
CHANG
40% of serum, and provides no adequate means to ensure the homogeneity of the cultured cells. Moreover, the ultrastructure and endocytic activity of hamster macrophages in vitro have not been characterized. This paper describes a new method by which hamster peritoneal macrophages can be maintained under satisfactory conditions for up to 1 month by improving their initial substratum adhesion and spreading. Population homogeneity as well as the activities of phagocytosis and phagolysosome formation are demonstrated in these cells indicating that they are adequate not only for the work of interest to us but also for other types of general experiments. MATERIALS AND METHODS
Harvesting cells. Cells were harvested from the peritoneal cavity of male Golden Syrian hamsters (70 to 80 g; Lakeview, N.J.), some of which were prestimulated 2 days in advance by intraperitoneal injection of 4% thioglycollate broth, Hanks' balanced salt solution IHBSS), medium 199 or complete culture medium. Into each etherized animal 20 ml HEPES-huffered medium 199 or HBSS with 10 U per ml of heparin were i.p. injected; the peritoneal fluid was withdrawn subsequently with a 10-ml syringe fitted with a 27G l/~. inch needle without sacrificing the animals. Erythrocytes when present were lysed with 0.87% ammonium chloride in water. Cell numbers were counted in a hemacytometer in pooled samples which then were centrifuged at 900 • g for 10 min. Cell pellets were resuspended in appropriate amounts of culture medium to make a desirable cell density for plating. Cell viability was determined by trypan-hlue dye-exclusion test. Culture procedure and conditions. Cells (1 to 2• in 0.2 to 0.3 ml of medium were plated on flying cover slips (18- or 22-mm 2) placed in tissueculture petri dishes 13- or 6-cm in diameter; Falcon) or in the latter directly. Glass cover slips were cleaned by boiling in 1% " 7 X " detergent for several min; after several rinses in double-distilled water, they then were sterilized by immersion in 70% ethanol and flame-dried before use. Preparations were incubated at 37 ~ -+ 1~ C with 5% CO2 in air for 2 to 4 hr to allow cell adhesion and then flooded with 2 to 3 ml of culture medium. After 24 hr, cell cultures were washed with prewarmed HBSS or medium to remove nonadherent cells. Medium was replenished or changed every other day.
Culture medium and variations. During the initial phase of the investigations, a variety of tissueculture media supplemented with 20% to 40% heat-inactivated fetal bovine serum (HIFBS) and antibiotics (100 U per ml of penicillin and 100 t~g per ml of streptomycin) were tested, i.e. NCTC109, CMRL-1066, RPMI-1640, BHK-12, H a m F-12, Eagle's or Dulbecco's M E M , or medium 199 with or without H E P E S (GIBCO). All these media were found unsatisfactory in that most cells became detached in 2 to 3 days a n d / o r were overwhelmed by the growth of flbroblasts. Consequently, various treatments were introduced to avert these unfavorable conditions. Hydroxyurea and bromodeoxyuridine were tested at various concentrations to prevent excessive proliferation of flbroblasts. To promote cell adhesion and spreading, the substratum, glass or plastic, was coated with rat-tail collagen (4), fibrinogen 1250 /~g per 3-cm petri dish), or poly-D, L-lysine (type V or B-VII, Sigma; 100 to 250 ~g per 18- to 22em 2 cover glass); in all cases one side of the cover glass was covered with 0.1 to 0.3 ml of an aqueous solution of these compounds. The cover slips were incubated for 2 to 4 hr at room temperature after which the fluid was removed by vacuum aspiration. HEPES-buffered medium 199 plus antibiotics was chosen as the basal medium and was supplemented at a final concentration of 20% with heat-inactivated FBS or horse serum {HIHS), or a combination of both in equal amounts. Also tested in addition to the above treatments were some factors known to stimulate cell adhesion a n d / o r spreading, e.g. potassium pyruvate at 0.1 to 0.5 mg per ml (5), manganese chloride at 0.001 to 10 mM, dithiothreitol at 0.01 to 2 mM (6), and lactalbumin hydrolysate at 0.5%. Light microscopy. For routine observations, cultures were examined directly under phase contrast or after fixation in methanol and staining with Giemsa. To assess the effects of various treatments, total cell numbers and the percent of spread cells were determined in an area of 0.15 mm 2 under phase contrast as criteria for the adhesion of macrophages to the substratum and their spreading, respectively. Determinations were made from at least five random areas in each of triplicate samples cultured in vitro for 2 to 4 days. Electron microscopy. For transmission electron microscopy, cultures were fixed at 4 ~ C in 2% glutaraldehyde in 0.1 M cacodylate buffer for 2 hr. Cells were removed from culture vessels with the aid of a rubber policeman and centrifuged at 2000 x g. Cell pellets then were postfixed in 1%
HAMSTER PERITONEAL MACROPHAGES osmium tetroxide in the same buffer for 2 hr, stained in-the-block in a saturated aqueous solution of uranyl acetate for 1 hr, dehydrated in graded series of acetone or ethanol, and embedded in Epoxy resin. Thin sections were cut with a diamond knife, stained with lead citrate, and examined with a Phillips E . M . 300 electron microscope. For scanning electron microscopy, macrophages cultured on round cover slips (9-ram diamter) were fixed in glutaraldehyde and dehydrated as mentioned above; specimens then were critical-point dried in amyl acetate or acetone and coated with gold-palladium in a vacuum coater. Preparations were examined at a tilt angle of 45 ~ in a Etec Autoscan scanning electron microscope. Phagocytosis study. Phagocytotic activity of cultured hamster macrophages was assessed by their ability to ingest polystyrene latex heads (2~m diamter; Dow Chemicals), duck and human erythrocytes. All particles were washed three times in phosphate-buffered saline by centrifugation. Erythrocytes were aged at 4 ~ C for several weeks in HBSS before use. For opsonization, 108
665
cells were incubated for 20 min at 37 ~ C in medium 199 containing 1:200 dilution of heat-inactivated rabbit antisermn with an agglutination titer of ]0 -3. Cell clumps were dispersed by extensive vortexing and the remaining aggregated cells were removed by brief centrffugation. Particles suspended in medium 199 with 20% H I F B S were incubated with the macrophages at a particle/cell ratio of 100:1 and 10:1 for latex beads and erythrocytes, respectively. Blockade of particle ingestion by macrophages was determined in experiments with latex beads by using cytochalasin B at a final concentration of 10/~g per ml. Duplicate samples were withdrawn and Giemsa stained at different periods after incubation; the number of particles per cell and the percentage of particlecontaining cells were determined by examining at least 200 cells in each culture. Some preparations with erythrocytes also were processed for scanning-electron-microscope studies. Labeling of secondarylysosomes. To determine the extent of phagolysosome formation in cultured macrophages, the macrophages were labeled with
FIG. l. Monolayer of hamster peritoneal macrophages cultured in vitro for 2 days. Note the presence of round, spindle-shaped and tripolar cells. Giemsa stain, x200. FIG. 2. Monolayer of hamster peritoneal macrophages cultured in vitro for 30 days. Note the voluminous increase of cytoplasm and multipolar appearance of the cells not seen in the younger culture shown in Fig. 1. Giemsa stain, x200.
666
CHANG
667
HAMSTER PERITONEAL MACROPHAGES
FIG. 5. a, Phagocytosis of 2-gm latex beads by 3-day-old hamster peritoneal macrophages for 50 min in vitro. Note the normal appearance of the cells, x400. b, Adherent and ingested beads can be distinguished by the presence of dark "rims" around the latter. Giemsa stain, x900. FIG. 6. Effects of eytochalasin B on the phagocytosis of latex particles by hamster macrophages in vitro. Note the "round-up" appearance of the treated cells, most of which contain no particles. Giemsa stain, x400.
an electron-dense marker, thorotrast, essentially as described by Jones and Hirsch (7). In brief, macrophages cultured in vitro for 1 to 2 days were washed with HBSS and incubated in medium 199 plus 40% HIFBS containing thorotrast at 1:20 or 1:40 dilution for 16 to 20 hr; cells then were rinsed to remove free particles. At different times after the removal of the dense marker, cell cultures were processed by transmission electron microscopy and thin sections were examined for the presence of the label in cytoplasmic vacuoles.
RESULTS
Recovery and composition of peritoneal exudates. From each hamster, approximately threefourths of injected HBSS, or 15 ml, was recovered and contained 12x10" nucleated cells of which macrophages, granulocytes and lymphocytes constituted about 70%, 25% and 5% of the total cell population, respectively. Cell viability was more than 97% at the time of plating as judged by trypan-blue exclusion test.
FIG. 3. The appearance of living hamster macrophages in vitro for 3 days. Note the appearance of spread and nonspread (arrow) cells corresponding to spindle-shaped and round cells seen in Giemsastained materials (Fig. 1k Phase-contrast x700. FIG. 4. The appearance of living hamster macrophages in vitro for 30 days. Note the granular cytoplasm and the occurrence of peripheral ruffled membranes and membranous skirt anchoring cells to the substratum. Phase contrast x700.
668
CHANG
HAMSTER PERITONEAL MACROPHAGES
Adhesion and spreading of macrophages. The loss of cells alter incubation in vitro was significantly minimized by two essential treatments: stimulation of the peritoneal cavity 1.5 to 2 days before harvest, and coating of substratum with substances promoting cell adhesion. Stimulation of the peritoneal cavity also markedly increased the cell yield, and the complete culture medium (see below) was found generally superior to thioglycollate broth or HBSS for this purpose. Essentially identical results were obtained with surfaces coated with fibrinogen, rat-tail collagen or polylysine; the last compound was routinely used for convenience. Under the conditions mentioned above, optimal cell adhesion and spreading were obtained in the complete culture medium consisting of HEPES-buffered medium 199 supplemerited with 10% HIFBS, 10% HIHS, 0.5% lactalbum~n hydrolysate, 0.5% glucose, hydroxyurea (0.2 mM), and antibiotics (100/~g per ml of streptomycin and 100 U per ml penicillin). The last two components were incorporated to limit the growth of fibroblasts and to prevent bacterial contamination, respectively. Bromodeoxyuridine proved to be toxic to the macrophages. When nonstimulated macrophages were cultured in this medium on uncoated surface, there was substantial cell detachment. However, the total cell number that became detached during the first 3 days in vitro was calculated to be 15% in this medium in contrast to 60% in medium 199 with 20% HIFBS as the only additive. Macrophages remained as monolayers at least up to 30 days in vitro under optimal conditions indicating that the loss of cells was insignificant although this was not determined beyond day 4. The factors in the complete medium contributing to the adhesion and spreading of stimulated cells on glass or plastic surfaces were determined by the single-deletion method. In complete medium for up to 4 days in vitro, the cell density, as a measure of cell adhesion, was about 350 cells per 0.15 mm 2 of which approximately 50% was well spread. Neither spreading nor adhesion could be increased by raising the serum level in the complete medium. The appearance of spread and nonspread cells, as well as the normal cell density of such preparations, is shown in Figs. 1 and 3. Omission of single components from the culture medium showed the following effects: all protein substances (HIFBS, HIHS and
669
lactalbumin hydrolysate) appeared to contribute to cell spreading; HIHS and lactalbumin hydrolysate were essential for cell adhesion; and additional glucose had little, if any, beneficial effect. No improvement was noted with the inclusion of other factors tested, e.g. potassium pyruvate, manganese chloride or dithiothreitol. All macrophages in complete medium spread well several weeks after the initial plating (Figs. 2, 4). Morphology. The appearance of hamster peritoneal macrophages changed markedly with time of incubation in vitro. During the first several days, they were a mixture of round, spindleshaped or tripolar cells in both phase-contrast (Fig. 3~ and Giemsa-stained materials (Fig. 1L The nuclei were oval dense bodies always located near one pole of the cells and filled about one-half to one-third of the total cell volume (Figs. 1, 3, 5a). The cytoplasm was very granular (Figs. 4, 5bL After 2 weeks or longer in vitro, macrophages became multipolar cells and were much larger in size with a cytoplasmic volume, which was now sufficient to accommodate 10 to 20 nuclei tFigs. 2, 4). These "matured" macrophages appeared to be tightly anchored to the substratum by a thin peripheral membranous skirt and had elevated cell body filled with dense granules and laced with ruffled membranes at the periphery ~Fig. 4). Ultrastructure. Scanning electron microscopy at low magnification showed that hamster peritoneal macrophages consisted of round and well spread cells in agreement with the lightmicroscope observation; in addition, both cell types had many surface membrane appendages indicating the "healthy" state of these cells maintained in vitro ~Fig. 7). Higher magnification further revealed numerous surface microvilli, filopodia and membrane veils; the latter two apparently anchored the cells to the substratum (Figs. 8, 9). Transmission electron microscopy showed that the macrophages possessed surface cytoplasmic extensions, corresponding to filopodia or microvilli seen by scanning electron microscopy, and the usual complement of cell organeUes, including nuclei, mitochondria, rough endoplasmic reticulum, Golgi apparatus, eentrioles, microtubules and numerous cytoplasmic vacuoles IFigs. 10, 11). The normal appearance of these cell organdies suggests that the cells are physiologically active. The ultrastructural features de-
F IG.7. Surveyscanning-electronmicrographof hamster peritoneal macrophagescultured in vitro for 2 days. Note the abundance of surface protrusions on spread cells and the occurrence of variations in cell shapes, x1200.
670
CHANG
HAMSTER PERITONEAL MACROPHAGES scribed above were representative of all preparations studied at different times after in vitro cultivation. No granulocytes or lymphocytes were seen; fibroblasts were rare and had a different ultrastructural characteristic. Phagoeytosis and phagolycosome formation. In vitro cultured hamster macrophages showed high activity of phagocytosis toward latex beads added to the culture medium. During a 2-hr incubation period with the particles, cells did not change their normal appearance {Fig. 5) and ingested the particles at a high rate as determined by the increase with time of particle numbers per cell and the percentage of particle-containing cells. During the same period of time, in the presence of cytochalasin B (10 pg per ml}, cells became rounded {Fig. 6) and contained negligible numbers of particles without appreciable increase after 30 min. The kinetics of particle ingestion was about 1.5 particle per cell per l0 min and more than 95% of the cells contained latex beads after 1 hr incubation. In vitro cultured cells also ingested human and duck erythrocytes. Quantitative determination during 1-hr incubation period showed that 60% of the cells ingested about five opsonized erythrocytes per cell, whereas only 25% took in two "aged" erythrocytes per cell. Thus opsonization significantly increased the rate of erythrophagocytosis. Scanning-electron-microscope study showed that erythrocytes were initially trapped by filopodia (Fig. 8) and eventually became buried completely under a thin veil of the cytoplasmic membrane of the macrophages (Fig. 9). In specimens prepared for determining the phagolysosome formation, thorotrast granules were found to accumulate readily and profusely in numerous cytoplasmic vacuoles indicating high pinocytotic activity and abundance of secondary lysosomes in the hamster macrophages IFig. 10). Upon further incubation for 7 days after the removal of the dense label, thorotrast granules appeared to persist in all vacuoles, most of which, however, were only lightly labeled (Fig. 11 ).
671
DISCUSSION The results of the present study provide a reasonably simple procedure for maintaining an adequate and homogeneous population of hamster peritoneal macrophages in vitro on a long-term basis. The method described requires less animal serum and ensures a longer cell survival time under satis|actory conditions than previously offered (3). Several difficulties encountered earlier are overcome by a number of manipulations introduced in the present study. Proliferation of fibroblasts that tends to overwhelm the macrophages is prevented by the use of a DNA biosynthesis inhibitor, hydroxyurea, which as expected, has no adverse effcet on the nondividing macrophages. Extensive loss of cells from the substratum is significantly minimized by the treatments known to increase cell adhesion and spreading. The factors that promote macrophage adherence and spreading include stimulation of the peritoneal cavity, which is known to enhance the spreading of mouse macrophages (1), and coating of substratum with polylysine, fibrinogen or collagen. While all three compounds are equally effective, the underlying mechanisms whereby they facilitate cell attachment and spreading may be different. The increase of cell adhesion mediated by fibrinogen and collagen is not unexpected in view of the presence of fibrin receptors on the surface of the macrophages (8) and the normal role of collagen in intercellular adhesion of normal cells (9). By virtue of its highly positive electrostatic charge, polylysine may effect a similar result by interacting directly with negatively charged cell surface components (10-12) or by increasing the deposition onto the substratum of spreading/adhesion factors present in the culture medium. The latter possibility is very likely as the spreading of hamster macrophages is minimal even on a coated surface in the absence of sera which have been shown to contain cell-sprcading factor (13). Results with the single-deletion experiments indicate that all proteinaceous components in the culture medium contribute, in vary-
FIG. 8. Scanning-electron micrographs showing the engulfment of a human red blood cell by pseudopodia of a hamster macrophage at the initial phase of erythrophagocytosis. Macrophage membrane appears to "flow" over the surface of the red cell closely following its contour at places. Note the presence of numerous surface mierovilli as well as filopodia and rolled membrane skirt that anchors the cell to the substratum, xl0,000. F1c. 9. Phagocytosis of human erythrocytes by a hamster peritoneal macrophage. One red blood cell is fully covered with a microvillus-studded thin veit of the macrophage membrane suggesting coalescence of the pseudopodia trapping the red cell during the initial enguffment phase. •
672
CHANG
HAMSTER PERITONEAL MACROPHAGES ing degrees, to cell spreading a n d / o r adhesion. Particularly striking is the marked increase of cell density at 20% H I H S in the absence of H I F B S {not shown) suggesting that the former may play a more crucial role in the adhesion of hamster macrophages to the substratum. This effect of H I H S has not been noticed in previous studies with mouse peritoneal macrophages {14, 15). Also of interest is the finding that the maximal spreading of hamster macrophages requires the simultaneous presence of H I H S and H I F B S , and thus differs from that of mouse macrophages or other cell types for which H I F B S alone is sufficient {13, 16). The lack of noticeable effects of manganese ions and dithiothreitol on hamster macrophages further distinguishes them from mouse macrophages whose spreading is markedly enhanced by these substances (5). Clearly, the various treatments introduced into the culture system produce favorable conditions for hamster macrophages to remain as a monolayer of adherent cells. Further detailed analysis is needed to determine the precise mechanisms whereby these treatments may induce these effects. Light- and electron-microscope studies showed that all cultured cells have an identical staining property and cellular organization, thus providing evidence for the homogeneity of the cell population. In general, they resemble mouse macrophages in morphology and uhrastructure, except for a more elevated cell body and perhaps more abundant microviUi and filopodia on the cell surface. These features may simply reflect a difference in the degree of spreading between these two kinds of cells. Indeed, hamster macrophages never become completely flattened to the extent of lacking surface ridges and folds as do the mouse macrophages (17). The vitality of the in vitro cultured hamster maerophages is indicated not only by their normal appearance and ultrastructural integrity but also by the vigorous pinoeytosis of thorotrast granules. Transmission electron microscopy showed that the dense marker accumulates at a very high density in numerous cytoplasmic vacuoles of hamster macrophages.
673
These vacuoles are interpreted to be secondary lysosomes in accordance with the general assumption that exogenous particulate matters are shuttied into these cell organelles as a consequence of pinosome or phagosome-lysosome fusions (7, 18). That the cultured ceils are phagocytes is indicated by their ability to ingest latex beads and erythrocytes as well as by the inhibition of this activity by cytochalasin B. Scanning electron microscopy of erythrophagocytosis by hamster macrophages provides evidence supporting the contention that the ingestion is accomplished by a coalescence of the cell-surface pseudopodia wrapping the erythrocytes (19). The abundance of surface microvilli and filopodia in these cells might favor this type of phagocytosis instead of the alternative means by which erythrocytes are interiorized by "sinking into" spherical craters formed on the surface of fully spread cells {17). The extent of erythrophagocytosis of hamster macrophages increases significantly after opsonization suggesting that they possess surface receptors for immune ligands for recognition. Preliminary studies, indeed, indicated that hamster macrophages have both Fc and C3 receptors comparable to those found in mouse macrophages {Michl and Chang, unpublished data). Further investigations to characterize these receptors are currently in progress. REFERENCES 1. Gordon, S., and Z. A. Cohn. 1973. The macrophage. Int. Rev. Cytol. 36: 171-215. 2. Cohn, Z. A. 1974. The isolation and cultivation of mononuclear phagocytes. In: S. Fleischer, and L. Packer {Eds.), Methods in Enzymology. Academic Press, New York, pp. 758-764. 3. Bey, E., and J. S. Harrington. 1971. Cytotoxic effects of some mineral dusts on Syrian hamster peritoneal macrophages. J. Exp. Med. 133: 1149-1169. 4. Bornstein, M. B. 1958. Reconstituted rat-tail collagen used as substrate for tissue cukures on coverslips in Maximore slides and roller tubes. Lab, Invest. 7: 134-137. 5. Pickart, L. R., and M. M. Thaler. 1974. Pyruvate induced cellular flattening and pseudopodia formation blocked by cytochalasin B. In:
FIG. 10. Thin section of a hamster macrophage in vitro for 3 days revealing a eentropedaUy located nucleus {N) lying near the flattened cell surface against the substratum and an elevated cell body with surface extensions and filled with cell organdies. The marker, thorotrast, densely packed some more than 100 vacuoles, presumably secondary lysosomes {$1 ). Also visible are newly formed pinosomes in which the dense marker is sparse, x10,300. FIG. 11. Portion of a hamster macrophage labeled with thorotrast in vitro for 7 days showing eentrioles {Ct) closely associated with Golgi apparatus {G) and mitochondria IM) interspersed among lucent vacuoles with varying amounts of the dense marker, x23,600.
674
CHANG E. T. C. Lee, and E. E. Smith (Eds.), Miami
Winwr Symposium--Biology and Chemistry of Eucaryotic Cell Surface. Academic Press, New York, p. 365. 6. Rabinovitch, M., and M.J. DeStafano. 1973. Macrophage spreading in vitro. II. Manganese and other metals as inducers or as co-factors for induced spreading. Exp. Cell Res. 79: 423-430. 7. Jones, T. C., and J. G. Hirsch. 1972. The interaction between Toxoplasma gondii and mammalian cells. II. The absence of lysosomal fusion with phagocytic vacuoles containing living parasites. J. Exp. Med. 136: 1173-1194. 8. Sherman, L. A., and J. Lee. 1977. Specific binding of soluble fibrin to macrophages. J. Exp. Med. 145: 76-85. 9. Klebe, R.J., P.G. Rosenberger, S. L. Naylor, R. L. Burns, R. Novak, and H. Kieinman. 1977. Cell attachment to collagen. Isolation of a cell attachment mutant. Exp. Cell Res. 104: 119-125. 10. Mazia, D., G. Schatten, and W. Sale. 1975. Adhesion of cells to surface coated with polylysin. Applications to electron microscopy. J. Cell Biol. 66: 198-199. 11. McKeehan, W. L., and R. G. Ham. 1976. Stimulation of clonal growth of normal fibroblasts with substrata coated with basic polymers. J. Cell Biol. 71: 727-734. 12. Tsutsui, K., H. Kumon, H. Ichikawa, and J. Tawara. 1976. Preparative method for suspended
13. 14.
15.
16.
17.
18.
19.
biological materials for SEM by using of polycationic substance layer. J. Electron Microsc. 25: 163-168. Grinnell, F. 1976. Cell spreading factor. Occurrence and specificity of action. Exp. Cell Res. 102: 51-62. Chang, Y. T. 1964. Long-term cultivation of mouse peritoneal macrophages. J. Nat. Cancer Inst. 32: 19-35. Bennett, B. 1966. Isolation and cultivation in vitro of macrophages from various sources in the mouse. Am. J. Pathol. 48: 165-182. Leonard, E. J., and A. Skeel. 1976. A serum protein that stimulates maerophage movement, chemotaxis and spreading. Exp. Cell Res. 102: 434-438. Polliack, A., and S. Gordon. 1975. Scanning electron microscopyof routine macrophages. Surface characteristics during maturation, activation and phagocytosis. Lab. Invest. 33: 469-477. Cohn, Z. A., and M. E. Fedorko. ]969. The formation and fate of lysosomes. In: J. T. Dingle, and H.B. Fell (Eds.), Lysosomes in Biology and Pathology. Vol. 1. North-Holland, Amsterdam, pp. 43-63. Orenstein, J. M., and E. Shehon. 1977. Membrane phenomena accompanying erythrophagoeytosis. A scanning electron microscope study. Lab. Invest. 36: 363-374.
I wish to thank Drs. S. Silverstein and W. Trager for advice and reviewing the manuscript; Drs. J. Michl and D. M. Dwyer for helpful suggestions; and Mrs. D. F. Greene for her skillful preparation of this manuscript.
