SEPARATION OF MURINE MACROPHAGES BY ADHERENCE TO SOLID SUBSTRATES
Donald E. Mosier
I. INTRODUCTION
Physical separation of macrophage and lymphoid populations has been essential in demonstrating principles of cell cooper- ation in immune responses. Evidence now exists that substrate adherent macrophage populations either are required for most forms of lymphocyte activation or significantly enhance lympho- cyte transformation (1 - 12). This chapter deals with separa- tion of macrophage populations on the basis of their ability to adhere actively to a solid substrate such as plastic, glass, or collagen. Such techniques are useful as a rapid and simple method of macrophage purification (in the adherent fraction) and depletion (in the nonadherent fraction), although more rigorous macrophage depletion is best accomplished by a combi- nation of these techniques with Sephadex G-10 column passage
[(13) and Chapter 25 this volume]. It is important to note from the outset that the population of cells which adheres to
METHODS FOR STUDYING Copyright © 1981 by Academic Press, Inc.
MONONUCLEAR PHAGOCYTES 1 7 9 All rights of reproduction in any form reserved.
ISBN 0-12-044220-5
solid substrates is heterogenous and includes in addition to classic phagocytic mononuclear cells the Ia+ antigen-presenting cells (14) and the dendritic cells described by Steinman et al.
(15). Moreover, the fraction of adherent cells that function in specific antigen presentation to lymphocytes has been esti- mated by limiting dilution techniques to be as low as 1 per 10,000 (12). Any conclusions on the functional properties of adherence separated cell populations should take into account the possibility that a very small distinct subpopulation is en- tirely responsible for the observed results.
Separation of macrophages by adherence requires active cell metabolism, and attachment is temperature dependent (16, 17). The presence of serum is necessary for attachment, and it is likely that proteins such as fibronectin play an impor- tant role in substrate adherence. Cells that are in late G2 or metaphase are less adherent to substrates and may detach during mitosis, as may cells that have "overindulged" in pha- gocytosis. The physiological basis for cell adhesiveness is not yet well understood, but these techniques should apply to all species.
II. REAGENTS AND MATERIALS
The hallmark of adherence techniques is their simplicity.
Tissue culture grade plasticware or glass dishes or beads acid washed to tissue culture specifications [e.g. (18)] are re- quired. There is some variation from lot to lot in 35- or 60-mm tissue culture dishes, and plasticware sterilized by gamma irradiation seems to be less variable than that steri- lized with ethylene oxide. Most standard tissue culture media
[e.g., modified MEM (1), RPMI 1640, M199] will support cell adherence. Medium designed for suspension culture has reduced Ca and M g2 + concentrations and is not suitable for adherence techniques. Serum is required for active adherence; 10% fetal calf serum is adequate, although higher serum concentrations promote cell spreading following initial adhesion (17).
The Shortman (18 - 21) glass bead column technique re- quires acid washed glass beads (about 450 ym in diameter, available from a variety of sources including 3-M, Minneapolis, Minnesota) and a 1.5 - 2.0 x 15 - 20 cm glass column. The glass beads should be siliconized before use (e.g., by treat- ment with Siliclad, available from Clay-Adams).
If recovery of adherent cells is a prime consideration, cells may be eluted from solid, substrates by using tissue cul- ture medium without serum and C a2 + or M g2 + salts and supple- mented with 30 mM EDTA. Alternatively, cells may be adhered
to a collagen monolayer and eluted by collagenase treatment (see below). Calf skin collagen (LS 0001660, Worthington) and type-1 collagenase (LS 0004194, Worthington) are adequate for this purpose. The merthiolate added as a preservative to the collagen should be removed shortly before use by dialysis against pH 4.4f 0.075 M sodium citrate.
III. PROCEDURES
A. Adherence to Plastic Culture Dishes
A single-cell suspension of murine spleen cells is prepared by teasing the spleens with fine forceps in cold Hanks' bal- anced salt solution (HBSS). Large cell clumps are allowed to settle 3 - 5 min, and the remaining cells in suspension are transferred to a new tube and sedimented by centrifugation.
The cells are washed twice more in HBSS and then resuspended in RPMI 1640 + 10% fetal calf serum. A viable cell count is performed, and the cell density is adjusted to 10 - 20 x 106 ml~l. One milliliter per 35-mm dish or 3 - 5 ml per 60-mm dish is dispensed, and the dishes are incubated in a 5%
CO2 atmosphere at 37°C for 1 hr. Rocking the dishes on an os- cillating platform as originally described (1) does not seem to be required for active adherence of macrophages, although it may diminish the number of lymphocytes that bind to macro- phages (21). After 1 hr incubation, the dishes are gently swirled to resuspend settled cells and the nonadherent popula- tion aspirated with a sterile Pasteur pipette. These cells may be transferred to a new set of culture dishes after one wash with HBSS and resuspension to one-half the original volume to correct for the approximately 50% recovery of the starting cell number. The cells remaining in the original culture dish should be washed vigorously three times with cold HBSS. A stream of HBSS should be directed at the entire surface of the dish using a narrow bore Pasteur pipette. Macrophages are firmly attached to the plastic by this point, and it is pro- bably impossible to be too vigorous in this washing step, which removes loosely adherent lymphocytes.
The population of cells originally nonadherent to plastic may be cultured in additional dishes to improve depletion of phagocytic cells. Alternatively, additional purification can be achieved by passing plastic nonadherent cells over Sephadex G-10 columns, a procedure we prefer for more vigorous depletion of macrophages.
At each step, the percentage of macrophages should be moni- tored by assays of phagocytosis and/or nonspecific esterase
staining (see Chapter 28) . Typically, one step of plastic adherence leads to an 80 - 90% decrease in phagocytic cells in the nonadherent population. The adherent population con- tains greater than 90% phagocytic, esterase-positive cells·
Recovery of the adherent population can be achieved by replacing the complete medium with serum-free medium lacking Ca2+ and Mg2+ and containing 30 mM EDTA. Adherent cells then can be scraped free with a rubber policeman. Unfortunately, this procedure leads to reduced viability and cell function so we prefer to use adherent cells in the culture dish in which they were originally prepared. They can be enumerated in the culture dish using an inverted phase microscope. Be- tween 1 and 5% of the initial cell number is recovered as adherent macrophages depending upon the age and strain of mouse used.
B. Adherence to Glass Bead Columns
Several versions of this technique are available, but the one of Shortman et al. (18) seems to work the best for us, al- though there is considerable variability in the number of lym- phocytes retained by the column. Acid-washed, siliconized glass beads (M50 ym diameter) are added to a 2.4 x 14 cm glass column that previously had been plugged with a small pad of acid-washed glasswool and filled with sterile HBSS. The beads are poured rapidly into the column and allowed to settle by gravity. Prior to use, the column should be equilibrated with serum-free medium, e.g., RPMI 1640, at 37°C. Maintenance of pH at the appropriate range of 6.8 - 7.3 can be accomplished by omitting NaHC03 from the medium and adding 25 mM HEPES buf-
fer plus enough 1 N NaOH to achieve a pH of 7.2. This obviates the need to run the column in a CO2 incubator.
The column is loaded with cells suspended in RPMI 1640 + 50% fetal calf serum (FCS). We have not found the use of mouse serum to be essential to the technique. Just prior to column loading, one column volume of RPMI 1640 + 60% FCS
should be added to the beads. From 2 to 10 x 108 spleen cells (prepared as above) are then added in 4 - 5 ml with care taken not to disturb the interface between 50 and 60% FCS. As the cells enter the column, additional medium + 50% FCS should be added and the upper portion of the column bed stirred with the popette tip to dislodge nonspecifically trapped cells. The cells should be allowed to enter the column quite rapidly
(e.g., 3 0 - 6 0 sec), and the column then loaded with one column volume of RPMI 1640 + 25% FCS. The flow rate should be slowed so that the cell band at the 50 - 60% FCS interface takes about 10 min to traverse the column.
In contrast to elution of adherent cells from plastic dishes, the medium recommended by Shortman for elution of macrophages from glass bead columns contains 2% serum, in ad- dition to 30 mM EDTA and no Ca2+ or Mg2 +. The procedure for eluting the cells from the column includes agitating the column while filled with the elution medium. This can be done by inverting the stoppered column or by stirring the beads with a sterile glass rod or pipette. Again, our ex- perience has been that the function of eluted cells is im- paired when compared to undisturbed adherent cells.
The relative yield of adherent and nonadherent cells from the Shortman column is somewhat variable. About a tenfold depletion in macrophages with a 40 - 70% recovery can be ex- pected in the nonadherent fraction, while eluted cells are enriched 5 to 10-fold for macrophages. Again, recovered cells should be evaluated for purity by assays for phagocyto- sis, esterase activity, and function.
C. Adherence of Cells to Collagen Monolayers
The advantage of this technique is that adherent cells can be recovered with good viability and function. The dis- advantage is that the purity of separation is not good, so that the main use of the technique is in recovering enriched populations of viable, functioning macrophages.
Soluble collagen at 1 - 2 mg/ml, pH 4.4, is adjusted to pH 7.0 with Qm5 M sodium carbonate and 1 ml immediately added to a prewarmed 60-mm plastic culture dish. The dish is incu- bated without agitating for 30 min at 37°C during which time the collagen gels. The dish is then washed gently with 37°C RPMI 1640 + 10% FCS twice, and allowed to incubate at 37°C for another 30 min with medium and FCS. At this point, the medium is gently decanted, and 3 ml of a spleen cell suspension con- taining 10 - 20 x 10^ cells/ml is added slowly to prevent dis- ruption of the collagen monolayer. The culture dishes are in- cubated at 37°C for 1 hr, and the nonadherent cells are de- canted. The surface of the collagen is gently flushed with three washes of cold HBSS. The cells adherent to the collagen are recovered by adding 1 mg/ml type I collagenase in warm HBSS. Incubation is continued at 37°C for 1 - 2 hr with gentle swirling of the dish at frequent intervals. The cells are re- covered by decanting and washed in RPMI 1640 + 10% FCS twice.
The recovery is variable with this technique (10 - 20%), but one can achieve an enrichment for phagocytic cells of 3 to 5-fold. Lymphocytes still comprise the majority population, however. Activated macrophages secrete endogenous collagenase
(22), which may mean they will not be recovered by these pro- cedures .
IV. CALCULATION OF DATA
The calculation of percentage cell recovery, cell viability, phagocytic cells, etc., is straightforward provided one has per-
formed cell counts at each step of the procedure.
V. CRITICAL COMMENTS
As emphasized above, none of these techniques yields abso- lutely pure cell populations, and the best strategy may be to combine two or more different procedures. The final criterion for purity of separation should be a functional assay, not morphological scanning. For example, the finding that adherent accessory cells are required for many, if not all, thymic- independent antibody responses in vitro was based upon improved methods for more rigorous depletion of adherent cells (7). An additional functional criterion of effective accessory cell de- pletion in in vitro systems is the inability of 2-mercapto- ethanol (2-ME) to replace macrophage function. In our labora- tory, depletion of adherent cell function for primary in vitro antibody responses routinely is accomplished in medium supple- mented with 2-ME.
All of these techniques show some variability that might be attributed to the biological status of the cell source or to variables in cell culture technique such as serum source or the type (or even age) of tissue culture vessels employed. Conver- sion of monocyte precursors (which might be expected to be non- adherent) to adherent macrophages does not occur with any de- tectable frequency in our experience, but it may occur in species other than the mouse or under different culture condi- tions. Such a differentiation process should be checked in other systems by assessing the number of macrophage-like cells appearing with time.
Finally, until more is known about the heterogeneity of macrophage subpopulations and their cell biology, these ap- proaches remain empirical and should be subject to constant im- provement.
REFERENCES
1. D. E. Mosier. A requirement for two c e l l types for a n t i - body formation In vitro. Science 158: 1573-1575, 1967.
2. C. W. Pierce and B. Benacerraf. Immune response in vitro:
Independence of "activated" lymphoid cells. Science 166:
1002-1004, 1969.
3. R. W. Dutton, M. M. McCarthy, R. I. Mishell, and D. J.
Raidt. Cell components in the immune response. IV.
Relationship and possible interactions. Cell. Immunol. 1:
196-206, 1970.
4. K. Shortman and J. Palmer. The requirement for macro- phages in the in vitro immune response. Cell. Immunol. 2:
399-410, 1971.
5. M. Feldman and A. Basten. The relationship between anti- genic structure and the requirement for thymus-derived cells in the immune response. J. Exp. lied. 134: 103-119, 1971.
6. E. Unanue and D. H. Katz. Immunogenicity of macrophage- bound antigens: The requirement for hapten and carrier determinants to be on the same molecule for T and B lym- phocyte collaboration. Eur. J. Immunol. 3: 559-563, 1973.
7. T. M. Chused, S. S. Kassan, and D. E. Mosier. Macrophage requirement for the in vitro response to TNP Ficoll: A thymic-independent antigen. J. Immunol. 116: 1579-1581, 1976.
8. A. A. Nordin. The in vitro immune response to a T-inde- pendent antigen. I. The effect of macrophages and 2-mer- captoethanol. Eur. J. Immunol. 8: 776-781, 1978.
9. D. L. Rosenstreich and A. S. Rosenthal* Peritoneal exudate lymphocyte. II. In vitro lymphocyte proliferation induced by brief exposure to antigen. J. Immunol. 110:
934-942, 1973.
10. R. C. Seeger and J. J. Oppenheim. Synergistic interaction of macrophages and lymphocytes in antigen-induced trans- formation of lymphocytes. J. Exp. Med. 132: 44-65, 1970.
11. D. S. Nelson (ed.). "Immunobiology of the Macrophage."
Academic Press, New York, 1976.
12. D. E. Mosier and L. W. Coppleson. A three-cell interac- tion required for the induction of the primary immune re- sponse in vitro. Proc. Nat. Acad. Sei. USA 61: 542-547, 1968.
13. I. Ly and R. I. Mishell. Separation of mouse spleen cells by passage through columns of Sephadex G-10. J. Immunol.
Methods 5: 239-248r 1974.
14. C. Cowing, B. D. Schwartz, and H. B. Dick1er. Macrophage la antigens. I. Macrophage populations differ in their expression of la antigens. J. Immunol. 120: 378-384, 1978.
15. R. M. Steinman and Z. A. Conn. Identification of a novel cell type in peripheral lymphoid organs of mice. I.
morphology, quantitation, tissue distribution. J. Exp.
Med. 137: 1142-1162, 1973.
16. S. Gordon and Z. A. Cohn. The macrophage. The macrophage as a tool in cell biology. Int. Rev. Cytol. 36: 171-124, 1973.
17. M. Rabinovitch and M. J. DeStefano. Macrophage spreading in vitro. I. Induces of spreading. Exp. Cell. Res. 77:
323-334, 1973.
18. K. Shortman, N. Williams, H. Jackson, P. Russell, P. Byrt and E. Diener. The separation of different cell classes from lymphoid organs. IV. The separation of lymphocytes from phagocytes on glass bead columns, and its effect on subpopulations of lymphocytes and antibody-forming cells.
J. Cell Biol. 48: 566-579, 1971.
19. K. Shortman. The separation of different cell classes from lymphoid organs. I. The use of glass bead columns to separate small lymphocytes, remove damaged cells, and fractionate cell suspensions. J. Exp. Biol. Med. Sei.
44: 271-284, 1966.
20. Y. Rabinowitz. Separation of lymphocytes, polymorphonu- clear leukocytes, and monocytes on glass columns, includ- ing tissue culture observations. Blood 23: 811-828, 1964.
21. J. Ellner, P. E. Lipsky, and A. S. Rosenthal. Antigen handling by guinea pig macrophages : Further evidence for the sequestration of antigen relevant for activation of primed lymphocytes. J. Immunol. 118: 2053-2057, 1977.
22. L. M. Wahl, S. M. Wahl, S. E. Mergenhagen, and G. M.
Martin. Collagenase production by endotoxin-activated macrophages. Proc. Nat. Acad. Sei. USA 71: 3598-3601, 1974.
