- Email: [email protected]
Use of a Multiwell Assembly for Chemotaxis and Evaluation by Enzyme-Linked Immunosorbent Assay (ELISA)
Immunobiol., vol. 162, pp. 192-198 (1982)
Abteilung fur Experimentelle Dermatologie, Universitats-Hautklinik, Munster, Federal Republic of Germany
Use of a Multiwell Assembly for Chemotaxis and Evaluation by Enzyme-Linked Immunosorbent Assay (ELISA) C. SORG, CHRISTIANE RAMB, B. OVERWIEN, and U. FEIGE Received February 12, 1982 . Accepted in Revised Form April 28, 1982
Multiwell Assembly for Chemotaxis
Abstract A multiwell chamber assembly for chemotaxis tests was designed, which integrates the established microtiter system. A microtiter plate is covered with a plastic plate containing up to 96 holes of the diameter of the microtiter wells. Between the plates, a Nucleopore filter sheet (5 lAm) and a silicon rubber gasket is placed. As a model system, human monocytes and lymphocyte-derived chemotactic factors were used. As it was observed that monocytes migrate through the membrane and settle on the bottom of the micro titer wells, an ELISA was adapted for quantitation of cells. After washing and incubation with a xenoantiserum against human monocytes, the bound antibody was quantitated using protein-A-conjugated alkaline phosphatase and p-nitrophenyl phosphate as detection system. The plates were read in a multichannel photometer. Cell numbers were determined directly from a calibration curve established before with varying numbers of monocytes. Current experience allows the following conclusions: The chemotaxis test in microtiter plates is simpler, faster and uses less material than conventional Boyden chambers. Evaluation by ELISA is much faster and more accurate than by microscopy.
Introduction Chemotaxis assays are usually performed in Boyden chambers, which require excessive handling and relatively large amounts of material (1). Evaluation of cells migrating into or through a micropore filter is done by staining and counting cells under a microscope, which is tedious and timeconsuming (2). The original Boyden chamber, which consists of an upper and a lower well, separated by a membrane filter, has been modified by many investigators (3, 4, 5, 6).One of the biggest drawbacks of the method is its capacity and speed of evaluation. An improvement towards increased capacity has been described by several authors who used multiwell chambers which can be handled as a single unit (7, 8,9). However, evaluation by counting cells in the microscope is still a bottle-neck prohibiting large-scale testing. More recently, very expensive automated image analyzers (9) for counting cells have come into use, whose price, however, is prohibitive for
Multiwell Assembly for Chemotaxis
193
most laboratories. In order to overcome the problem of limited capacity both for setting up chemotaxis chambers per working day and microscopic evaluation, we designed a multiwell chamber, which uses commercial 96well microtiter plates as bottom chamber and 5 micrometer Nucleopore filters to separate the bottom from the top chambers. The cells which migrate through the filter settle in the bottom chamber, where they are fixed and quantitated by an enzyme-linked immunosorbent assay.
Materials and Methods Sera reagents
Tissue culture media and sera were purchased from Seromed, Munich. All chemicals, if not indicated otherwise, were purchased from Serva, Heidelberg. A rabbit anti-human-leukocyte serum was prepared by injection of mononuclear leukocytes in complete Freund's adjuvant and a booster shot after 4 weeks. The serum was used without absorption. Cells
Peripheral blood leukocytes were gained by leukapheresis (10) either from individual donors or from pooled buffy coats. Mononuclear cells were purified by centrifugation on FicollPaque (Pharmacia) (11). Monocytes were prepared by further centrifugation on Percoll (Pharmacia) following the manufacturer's instructions. Briefly, Percoll stock solutions were made isotonic by dilution with 10-fold concentrated minimum essential medium (MEM) and adjusted with phosphate-buffered saline (PBS) to a density of 1.07 g/mL 38 ml of this solution were filled into a 50-ml centrifuge tube and spun at 18,000 rpm at 4°C for 16 min in a Sorvall centrifuge (Sorvall RC-SB, Du Pont Instruments, Newtown/Conn., USA), using the rotor SS 34. 3 ml of cell concentrate (3 X 108 cells) were then placed on top of the gradient and centrifuged for 15 min at 4°C at 400 X g. A band which was highly enriched for monocytes at a density of 1.055 g/ml was recovered. The monocytes were thoroughly washed, resuspended in McCoy's medium supplemented with 20 % autologous serum at a density of 2 X 106 cells per ml and placed into a Teflon membrane bag, which then was sealed (12). When monocytes from pooled buffy coats were used, 10 % fetal calf serum (FCS) was added to the medium. After 3-5 days of cultivation, the cells were used in the chemotaxis assay. Chemotactic factor
Mononuclear leukocytes were purified by Picoll centrifugation as described above. The cells were suspended in RPMI 1640 which was supplemented with 1 % Hepes and S X 10- 5 M 2-mercaptoethanol at a concentration of 1-3 X 106 cells per ml with (10 !!g/ml) or without phytohemagglutinin (PHA-P, Difco). Soluble Concanavalin A (Con A, Serva, Heidelberg) was used at a concentration of 15 !!g/ml and insolubilized Con A-Sepharose (Pharmacia) at a concentration of 100 !!g/mL Both concentrations had been determined before as optimal for chemotactic factor production. Two hours after exposure to mitogens, the cells were washed three times with complete RPM I 1640 and then incubated for another 42--48 hrs at 37°C. The supernatants were spun at 48,000 X g for 40 min, aliquoted and stored at - 80°C. Description of chemota.xis chambers
The chemotaxis chambers consist of a bottom plate and a top plate (Fig. 1). A standard 96well microtiter plate is placed between the two plates, covered with a filter sheet (either Nucleopore or Millipore) and a silicone gasket with the same number of holes as the top plate. The top plate has 24, 48, or 96 holes of the size of the microtiter wells. Six threaded nods fitted with wing nuts are spaced along the edges of the apparatus to supply an evenly distributed pressure.
194 . C. SORG, CHRISTIANE RAMB, B. OVERWIEN, and V. FEIGE
Fig. 1. Multiwell assembly for chemotaxis. Right: Bottom plate with standard 96-well microtiter plate, covered with nitrocellulose filter sheet, silicone gasket and top plate (48 wells). Left: Completely assembled unit. Chemotaxis assay
The chemotaxis assay was carried out with two modifications. After filling the microtiter wells with chemotaxins, using the first modification, an 8 ILm cellulose nitrate filter (SMl1301, Sartorius, Giittingen) was placed above the microtiter plate. A 5 /!ffi polycarbonate filter sheet (Nucleopore Cooperation, Pleasanton, Ca.) was laid on tOp. In the second modification, only a 5 [.1m polycarbonate filter was placed between the chamber plates. Before covering the micro titer plate, the filter sheets were wetted with culture medium. Air bubble inclusion in the lower chambers was carefully avoided. After assembly of the chambers, the upper wells were filled with 3 X 105 cells/well, suspended in MEM-E plus 10 % FCS and incubated for 2 hrs at 37°C. The cells in the upper wells were then aspirated, and the chambers were opened. When using the first modification, the cellulose nitrate filter was fixed with isopropanol and stained with Mayer's hemalaun. Filter areas over chambers were cut out and mounted on microscopic slides. Results were expressed as cells per oil immersion field. A minimum of 5 fields was counted. With the second modification, the microtiter plate (V-shaped; Costar No. 3796) was centrifuged at 1200 rpm (PR-6000; International Equipment Company, Nathan Heights, Mass.) for 5 min. The cells were resuspended, washed three times serum-free, and transferred to a Poly-L-Lysine-coated flat-bottomed microtiter plate (flat-bottomed; Costar No. 3596). The microtiter plates had been prepared by incubation of 100 ILl of a Poly-L-Lysine solution (25 [.Ig/ml, No. 1886, Sigma) for 30 min in each well. Immediately before use, the plates were washed three times with PBS. After fixation for 3 min with 0.25 % glutaraldehyde, the cells were stable and could be stored in PBS containing 0.2 % gelatin (E. Merck, Darmstadt) and 0.05 % tween 20 at 4°C for several days. For quantitation of cells by ELISA, 50 [.II of an optimal dilution of a rabbit anti-human leukocyte serum was added to the cells, incubated at 37°C for 30 min, washed and deVeloped with protein-A alkaline phosphatase conjugates as described before (13).
Results
Calibration of ELISA Human monocytes were seeded at different densities into a 96-well microtiter plate precoated with Poly-L-Lysine, spun and fixed with
Multiwell Assembly for Chemotaxis . 195
E
1.2 0.8 0.4
105 cells/well Fig. 2. Calibration curve for ELISA. Cell numbers were plotted against extinction at 405 nm after reaction of cells with antibody, protein A-alkaline phosphatase conjugate and chromogenic substrate.
glutaraldehyde. The cells were quantified using a xenoantiserum against human monocytes and protein-A-conjugated alkaline phosphatase with pnitrophenyl phosphate-bis-(2-amino-2-ethyl-l.3-propanediol) salt as substrate. Results are shown in Figure 2, where the extinction at 405 nm was plotted against the cell number. The assay is linear over a wide range of cell numbers and is sensitive enough to detect 500-1000 cells per well. Comparison of chemotactic activity after evaluation by microscopy and
ELISA
The chemotactic activity of a standard preparation of Con-A-stimulated human lymphocyte-culture supernatants was assayed in the multiwell
1:2
1:4 1:8
1:20
1:50 1:100 1:200
1:~
Dilutions
Fig. 3. Comparison of chemotactic activity against Con-A-stimulated lymphocyte supernatant after evaluation by microscopy (e--e) and ELISA (T--T).
196 . C. SORG, CHRISTIANE RAMB, B. OVERWIEN, and U. FEIGE
assembly, and the microscopic evaluation of a cellulose nitrate filter was compared with the quantitation of cells migrated through a nucleopore filter by ELISA. As can be seen in Figure 3, the microscopic evaluation shows a maximal titer of 1 :4, while the enumeration by ELISA yields a titer of 1 :400. It also shows a linear relation between migrated cells and chemotactic activity. In order to make the two methods comparable, the total area of one microtiter well has been reduced to the area of 1 microscopic field, which is approximately 1/ !ODD of a microtiter well.
Effect of serum proteins on chemotactic response of monocytes It has been shown by others (6) that the presence of protein is essential for the chemotactic response of cells. Therefore, as a routine, proteins were added to all media used in chemotactic assays. Since it was intended to use this assay for monitoring chemotactic activity in fractions containing varying amounts of proteins, the influence of human serum albumin or FCS on the chemotactic response was investigated. Figure 4 shows the chemotactic activity of PHA-stimulated lymphocyte supernatants in the presence of 2 % human serum albumin or 10 % FCS. It can be seen that the addition of 10% FCS leads to a more pronounced chemotactic response. In contrast to Con-A-stimulated supernatants, a prozone is regularly observed with PHA-stimulated supernatants.
E
1.4
1.0
0.6 Controls:
0.2
~
HSA
o
FCS
1:3 1:9 1:27 1: 81 1:2431:729 Dilutions Fig. 4. Effect of serum proteins on chemotactic response of monocytes to PHA-stimulated lymphocyte supernatants in presence of 2 % human serum albumin ( . - - . ) or 10 % FCS
(e--e).
Multiwell Assembly for Chemotaxis . 197
Discussion
The application of the Boyden chamber technique to large-scale testing of chemotactic activity has been limited by the capacity of this technique. This has been overcome in part by the design of various multiwell chemotaxis chamber assemblies (7, 8, 9). Here we describe another version of a multiwell chemotaxis chamber, which uses a commercial 96-well microtiter plate as bottom wells. The assembly can be used as a 24, 48, or 96-well device. The time needed for handling two plates until the point where the cells are glutaraldehyde-fixed and may be stored for several days is 4 hrs. This would mean that eight plates (48 wells per plate) or approximately 400 samples may be handled per person per working day. Quantitation of cells by ELISA can easily keep up with the capacity of the test. Using multichannel micropipettes and multichannel photometers for reading the ELISA, 400 samples can be assayed within 3 hrs. The sensitivity of enumeration by ELISA is approximately IOO-fold higher than that by microscopy The evaluation by ELISA takes advantage of the fact that monocytes migrate through the Nucleopore membrane, detach and settle to the bottom of the lower chamber. The question was whether these cells were representative in number and type, or whether the bulk of cells was adherent to the lower site of the filter. When the filter was stained and the lower site was examined microscopically, only few cells were adherent to the filter. Thus it seems reasonably safe to assume that the amount of cells passing and settling to the bottom chamber is directly proportional to the chemotactic activity. Similar observations have been described by KELLER et al. (1972) (2) of polymorphonuclear leukocytes. The quantitation of cells that migrate through the filter poses less problems than the quantitation of cells migrating into a cellulose nitrate filter. It has been suggested by others to determine only the cells of the leading front, neglecting all the other cells that were chemotactically responsive (14). The adaptation of a commercial 96-well microtiter plate into the chemotactic assembly opens some novel experimental perspectives. It has been described that the response to a certain chemotactic stimulus is not uniformly expressed in all types of macrophages but is a phenotypic trait (NEUMANN and SORG, 1980). With the microtiter plate those cells that have responded to a chemotactic factor can be collected and characterized further. Under the described conditions, 4-6 X 104 cells/chamber may migrate into the bottom well within 2 hrs. Further cultivation of these cells is possible. They might be used in further studies to elucidate the phenotypic changes of monocytes or macrophages that have undergone chemotactic activation. Acknowledgement We thank Ms. MARION SCHULZ for skilful technical assistance and Ms. BRUNHILDE SCHEIBEL for typing the manuscript.
198 . C. SORG, CHRISTIANE RAMB, B. OVERWIEN, and U. FEIGE
References 1. BOYDEN, S. V. 1962. The chemotactic effect of mixtures of antibody and antigen on polymorphonuclear leucocytes. J. Exp. Med. 115: 453. 2. KELLER, H. U., J. F. BOREL, B. C. WILKINSON, H. N. HESS, and H. COTTIER. 1972. Reassessment of Boyden's technique. for measuring chemotaxis. J. Immunol. Methods 1: 165. 3. KELLER, H. U., and E. SORKIN. 1968. Chemotaxis of leukocytes. Experientia 24: 641. 4. TEMPEL, T. R., R. SNYDERMAN, H. V. JORDAN, and S. E. MERGENHAGEN. 1970. Factors from Salliva and other bacteria chemotactic for polymorphonuclear leukocytes: Their possible role in gingival inflammation. J. Periodont. 41: 71. 5. BOUMSELL, L., and M. S. MELTZER. 1975. Mouse mononuclear cell chemotaxis. I. Differential response of monocytes and macrophages. J. Immunol. Methods 115: 1746. 6. WILKINSON, P. c., and R. B. ALLAN. 1978. Assay systems for measuring leukocyte locomotion: an overview. In: Leukocyte Chemotaxis, eds. J. I. Gallin and P. G. Quie (Raven Press, New York), p. 1. 7. SWANSON, M. J. 1977. A simple multiple chamber apparatus for measuring chemotaxis of polymorphonuclear leukocytes utilizing centrifugation of the chambers before incubation. J. Immunol. Methods 16: 387. 8. ARVILOMMI, H., and A. LAATIKAINEN. 1977. A simple technique for measuring leucocyte chemotaxis in reversible Boyden chambers. Z. Immun. Forsch. 153: 179. 9. FALK, W., R. H. GOODWIN, Jr., and E. J. LEONARD. 1980. A 48-well micro chemotaxis assembly for rapid and accurate measurement of leucocyte migration. J. Immunol. Methods 33: 239. 10. FEIGE, U., J. KNOP, and C. SORGo 1980. Large scale production of human monocytes by leukapheresis. Immunobiology 157: 217. 11. BOYUM, A. 1968. Isolation of mononuclear cells and granulocytes from human blood. Scand. J. Clin. Lab. Invest. Vol. 21, Supp!. 97, 77: 89. 12. NEUMANN, c., and C. SORGo 1980. Sequential expression of functions during macrophage differentiation in murine bone marrow liquid culture. Eur. J. Immunol. 10: 834. 13. SUTER, L., J. BROGGEN, and C. SORGo 1980. Use of an enzyme-linked immunosorbent assay (ELISA) for screening of hybridoma antibodies against cell surface antigens. J. Immunol. Methods 39: 407. 14. ZIGMOND, S. H., and J. G. HIRSCH. 1973. Leukocyte locomotion and chemotaxis. New methods for evaluation and demonstration of cell-derived chemotactic factor. J. Exp. Med. 137: 387.
Dr. C. SORG, Department of Experimental Dermatology, University of Miinster, vonEsmarch-Str. 56, 4400 Miinster, Federal Republic of Germany
Abteilung fur Experimentelle Dermatologie, Universitats-Hautklinik, Munster, Federal Republic of Germany
Use of a Multiwell Assembly for Chemotaxis and Evaluation by Enzyme-Linked Immunosorbent Assay (ELISA) C. SORG, CHRISTIANE RAMB, B. OVERWIEN, and U. FEIGE Received February 12, 1982 . Accepted in Revised Form April 28, 1982
Multiwell Assembly for Chemotaxis
Abstract A multiwell chamber assembly for chemotaxis tests was designed, which integrates the established microtiter system. A microtiter plate is covered with a plastic plate containing up to 96 holes of the diameter of the microtiter wells. Between the plates, a Nucleopore filter sheet (5 lAm) and a silicon rubber gasket is placed. As a model system, human monocytes and lymphocyte-derived chemotactic factors were used. As it was observed that monocytes migrate through the membrane and settle on the bottom of the micro titer wells, an ELISA was adapted for quantitation of cells. After washing and incubation with a xenoantiserum against human monocytes, the bound antibody was quantitated using protein-A-conjugated alkaline phosphatase and p-nitrophenyl phosphate as detection system. The plates were read in a multichannel photometer. Cell numbers were determined directly from a calibration curve established before with varying numbers of monocytes. Current experience allows the following conclusions: The chemotaxis test in microtiter plates is simpler, faster and uses less material than conventional Boyden chambers. Evaluation by ELISA is much faster and more accurate than by microscopy.
Introduction Chemotaxis assays are usually performed in Boyden chambers, which require excessive handling and relatively large amounts of material (1). Evaluation of cells migrating into or through a micropore filter is done by staining and counting cells under a microscope, which is tedious and timeconsuming (2). The original Boyden chamber, which consists of an upper and a lower well, separated by a membrane filter, has been modified by many investigators (3, 4, 5, 6).One of the biggest drawbacks of the method is its capacity and speed of evaluation. An improvement towards increased capacity has been described by several authors who used multiwell chambers which can be handled as a single unit (7, 8,9). However, evaluation by counting cells in the microscope is still a bottle-neck prohibiting large-scale testing. More recently, very expensive automated image analyzers (9) for counting cells have come into use, whose price, however, is prohibitive for
Multiwell Assembly for Chemotaxis
193
most laboratories. In order to overcome the problem of limited capacity both for setting up chemotaxis chambers per working day and microscopic evaluation, we designed a multiwell chamber, which uses commercial 96well microtiter plates as bottom chamber and 5 micrometer Nucleopore filters to separate the bottom from the top chambers. The cells which migrate through the filter settle in the bottom chamber, where they are fixed and quantitated by an enzyme-linked immunosorbent assay.
Materials and Methods Sera reagents
Tissue culture media and sera were purchased from Seromed, Munich. All chemicals, if not indicated otherwise, were purchased from Serva, Heidelberg. A rabbit anti-human-leukocyte serum was prepared by injection of mononuclear leukocytes in complete Freund's adjuvant and a booster shot after 4 weeks. The serum was used without absorption. Cells
Peripheral blood leukocytes were gained by leukapheresis (10) either from individual donors or from pooled buffy coats. Mononuclear cells were purified by centrifugation on FicollPaque (Pharmacia) (11). Monocytes were prepared by further centrifugation on Percoll (Pharmacia) following the manufacturer's instructions. Briefly, Percoll stock solutions were made isotonic by dilution with 10-fold concentrated minimum essential medium (MEM) and adjusted with phosphate-buffered saline (PBS) to a density of 1.07 g/mL 38 ml of this solution were filled into a 50-ml centrifuge tube and spun at 18,000 rpm at 4°C for 16 min in a Sorvall centrifuge (Sorvall RC-SB, Du Pont Instruments, Newtown/Conn., USA), using the rotor SS 34. 3 ml of cell concentrate (3 X 108 cells) were then placed on top of the gradient and centrifuged for 15 min at 4°C at 400 X g. A band which was highly enriched for monocytes at a density of 1.055 g/ml was recovered. The monocytes were thoroughly washed, resuspended in McCoy's medium supplemented with 20 % autologous serum at a density of 2 X 106 cells per ml and placed into a Teflon membrane bag, which then was sealed (12). When monocytes from pooled buffy coats were used, 10 % fetal calf serum (FCS) was added to the medium. After 3-5 days of cultivation, the cells were used in the chemotaxis assay. Chemotactic factor
Mononuclear leukocytes were purified by Picoll centrifugation as described above. The cells were suspended in RPMI 1640 which was supplemented with 1 % Hepes and S X 10- 5 M 2-mercaptoethanol at a concentration of 1-3 X 106 cells per ml with (10 !!g/ml) or without phytohemagglutinin (PHA-P, Difco). Soluble Concanavalin A (Con A, Serva, Heidelberg) was used at a concentration of 15 !!g/ml and insolubilized Con A-Sepharose (Pharmacia) at a concentration of 100 !!g/mL Both concentrations had been determined before as optimal for chemotactic factor production. Two hours after exposure to mitogens, the cells were washed three times with complete RPM I 1640 and then incubated for another 42--48 hrs at 37°C. The supernatants were spun at 48,000 X g for 40 min, aliquoted and stored at - 80°C. Description of chemota.xis chambers
The chemotaxis chambers consist of a bottom plate and a top plate (Fig. 1). A standard 96well microtiter plate is placed between the two plates, covered with a filter sheet (either Nucleopore or Millipore) and a silicone gasket with the same number of holes as the top plate. The top plate has 24, 48, or 96 holes of the size of the microtiter wells. Six threaded nods fitted with wing nuts are spaced along the edges of the apparatus to supply an evenly distributed pressure.
194 . C. SORG, CHRISTIANE RAMB, B. OVERWIEN, and V. FEIGE
Fig. 1. Multiwell assembly for chemotaxis. Right: Bottom plate with standard 96-well microtiter plate, covered with nitrocellulose filter sheet, silicone gasket and top plate (48 wells). Left: Completely assembled unit. Chemotaxis assay
The chemotaxis assay was carried out with two modifications. After filling the microtiter wells with chemotaxins, using the first modification, an 8 ILm cellulose nitrate filter (SMl1301, Sartorius, Giittingen) was placed above the microtiter plate. A 5 /!ffi polycarbonate filter sheet (Nucleopore Cooperation, Pleasanton, Ca.) was laid on tOp. In the second modification, only a 5 [.1m polycarbonate filter was placed between the chamber plates. Before covering the micro titer plate, the filter sheets were wetted with culture medium. Air bubble inclusion in the lower chambers was carefully avoided. After assembly of the chambers, the upper wells were filled with 3 X 105 cells/well, suspended in MEM-E plus 10 % FCS and incubated for 2 hrs at 37°C. The cells in the upper wells were then aspirated, and the chambers were opened. When using the first modification, the cellulose nitrate filter was fixed with isopropanol and stained with Mayer's hemalaun. Filter areas over chambers were cut out and mounted on microscopic slides. Results were expressed as cells per oil immersion field. A minimum of 5 fields was counted. With the second modification, the microtiter plate (V-shaped; Costar No. 3796) was centrifuged at 1200 rpm (PR-6000; International Equipment Company, Nathan Heights, Mass.) for 5 min. The cells were resuspended, washed three times serum-free, and transferred to a Poly-L-Lysine-coated flat-bottomed microtiter plate (flat-bottomed; Costar No. 3596). The microtiter plates had been prepared by incubation of 100 ILl of a Poly-L-Lysine solution (25 [.Ig/ml, No. 1886, Sigma) for 30 min in each well. Immediately before use, the plates were washed three times with PBS. After fixation for 3 min with 0.25 % glutaraldehyde, the cells were stable and could be stored in PBS containing 0.2 % gelatin (E. Merck, Darmstadt) and 0.05 % tween 20 at 4°C for several days. For quantitation of cells by ELISA, 50 [.II of an optimal dilution of a rabbit anti-human leukocyte serum was added to the cells, incubated at 37°C for 30 min, washed and deVeloped with protein-A alkaline phosphatase conjugates as described before (13).
Results
Calibration of ELISA Human monocytes were seeded at different densities into a 96-well microtiter plate precoated with Poly-L-Lysine, spun and fixed with
Multiwell Assembly for Chemotaxis . 195
E
1.2 0.8 0.4
105 cells/well Fig. 2. Calibration curve for ELISA. Cell numbers were plotted against extinction at 405 nm after reaction of cells with antibody, protein A-alkaline phosphatase conjugate and chromogenic substrate.
glutaraldehyde. The cells were quantified using a xenoantiserum against human monocytes and protein-A-conjugated alkaline phosphatase with pnitrophenyl phosphate-bis-(2-amino-2-ethyl-l.3-propanediol) salt as substrate. Results are shown in Figure 2, where the extinction at 405 nm was plotted against the cell number. The assay is linear over a wide range of cell numbers and is sensitive enough to detect 500-1000 cells per well. Comparison of chemotactic activity after evaluation by microscopy and
ELISA
The chemotactic activity of a standard preparation of Con-A-stimulated human lymphocyte-culture supernatants was assayed in the multiwell
1:2
1:4 1:8
1:20
1:50 1:100 1:200
1:~
Dilutions
Fig. 3. Comparison of chemotactic activity against Con-A-stimulated lymphocyte supernatant after evaluation by microscopy (e--e) and ELISA (T--T).
196 . C. SORG, CHRISTIANE RAMB, B. OVERWIEN, and U. FEIGE
assembly, and the microscopic evaluation of a cellulose nitrate filter was compared with the quantitation of cells migrated through a nucleopore filter by ELISA. As can be seen in Figure 3, the microscopic evaluation shows a maximal titer of 1 :4, while the enumeration by ELISA yields a titer of 1 :400. It also shows a linear relation between migrated cells and chemotactic activity. In order to make the two methods comparable, the total area of one microtiter well has been reduced to the area of 1 microscopic field, which is approximately 1/ !ODD of a microtiter well.
Effect of serum proteins on chemotactic response of monocytes It has been shown by others (6) that the presence of protein is essential for the chemotactic response of cells. Therefore, as a routine, proteins were added to all media used in chemotactic assays. Since it was intended to use this assay for monitoring chemotactic activity in fractions containing varying amounts of proteins, the influence of human serum albumin or FCS on the chemotactic response was investigated. Figure 4 shows the chemotactic activity of PHA-stimulated lymphocyte supernatants in the presence of 2 % human serum albumin or 10 % FCS. It can be seen that the addition of 10% FCS leads to a more pronounced chemotactic response. In contrast to Con-A-stimulated supernatants, a prozone is regularly observed with PHA-stimulated supernatants.
E
1.4
1.0
0.6 Controls:
0.2
~
HSA
o
FCS
1:3 1:9 1:27 1: 81 1:2431:729 Dilutions Fig. 4. Effect of serum proteins on chemotactic response of monocytes to PHA-stimulated lymphocyte supernatants in presence of 2 % human serum albumin ( . - - . ) or 10 % FCS
(e--e).
Multiwell Assembly for Chemotaxis . 197
Discussion
The application of the Boyden chamber technique to large-scale testing of chemotactic activity has been limited by the capacity of this technique. This has been overcome in part by the design of various multiwell chemotaxis chamber assemblies (7, 8, 9). Here we describe another version of a multiwell chemotaxis chamber, which uses a commercial 96-well microtiter plate as bottom wells. The assembly can be used as a 24, 48, or 96-well device. The time needed for handling two plates until the point where the cells are glutaraldehyde-fixed and may be stored for several days is 4 hrs. This would mean that eight plates (48 wells per plate) or approximately 400 samples may be handled per person per working day. Quantitation of cells by ELISA can easily keep up with the capacity of the test. Using multichannel micropipettes and multichannel photometers for reading the ELISA, 400 samples can be assayed within 3 hrs. The sensitivity of enumeration by ELISA is approximately IOO-fold higher than that by microscopy The evaluation by ELISA takes advantage of the fact that monocytes migrate through the Nucleopore membrane, detach and settle to the bottom of the lower chamber. The question was whether these cells were representative in number and type, or whether the bulk of cells was adherent to the lower site of the filter. When the filter was stained and the lower site was examined microscopically, only few cells were adherent to the filter. Thus it seems reasonably safe to assume that the amount of cells passing and settling to the bottom chamber is directly proportional to the chemotactic activity. Similar observations have been described by KELLER et al. (1972) (2) of polymorphonuclear leukocytes. The quantitation of cells that migrate through the filter poses less problems than the quantitation of cells migrating into a cellulose nitrate filter. It has been suggested by others to determine only the cells of the leading front, neglecting all the other cells that were chemotactically responsive (14). The adaptation of a commercial 96-well microtiter plate into the chemotactic assembly opens some novel experimental perspectives. It has been described that the response to a certain chemotactic stimulus is not uniformly expressed in all types of macrophages but is a phenotypic trait (NEUMANN and SORG, 1980). With the microtiter plate those cells that have responded to a chemotactic factor can be collected and characterized further. Under the described conditions, 4-6 X 104 cells/chamber may migrate into the bottom well within 2 hrs. Further cultivation of these cells is possible. They might be used in further studies to elucidate the phenotypic changes of monocytes or macrophages that have undergone chemotactic activation. Acknowledgement We thank Ms. MARION SCHULZ for skilful technical assistance and Ms. BRUNHILDE SCHEIBEL for typing the manuscript.
198 . C. SORG, CHRISTIANE RAMB, B. OVERWIEN, and U. FEIGE
References 1. BOYDEN, S. V. 1962. The chemotactic effect of mixtures of antibody and antigen on polymorphonuclear leucocytes. J. Exp. Med. 115: 453. 2. KELLER, H. U., J. F. BOREL, B. C. WILKINSON, H. N. HESS, and H. COTTIER. 1972. Reassessment of Boyden's technique. for measuring chemotaxis. J. Immunol. Methods 1: 165. 3. KELLER, H. U., and E. SORKIN. 1968. Chemotaxis of leukocytes. Experientia 24: 641. 4. TEMPEL, T. R., R. SNYDERMAN, H. V. JORDAN, and S. E. MERGENHAGEN. 1970. Factors from Salliva and other bacteria chemotactic for polymorphonuclear leukocytes: Their possible role in gingival inflammation. J. Periodont. 41: 71. 5. BOUMSELL, L., and M. S. MELTZER. 1975. Mouse mononuclear cell chemotaxis. I. Differential response of monocytes and macrophages. J. Immunol. Methods 115: 1746. 6. WILKINSON, P. c., and R. B. ALLAN. 1978. Assay systems for measuring leukocyte locomotion: an overview. In: Leukocyte Chemotaxis, eds. J. I. Gallin and P. G. Quie (Raven Press, New York), p. 1. 7. SWANSON, M. J. 1977. A simple multiple chamber apparatus for measuring chemotaxis of polymorphonuclear leukocytes utilizing centrifugation of the chambers before incubation. J. Immunol. Methods 16: 387. 8. ARVILOMMI, H., and A. LAATIKAINEN. 1977. A simple technique for measuring leucocyte chemotaxis in reversible Boyden chambers. Z. Immun. Forsch. 153: 179. 9. FALK, W., R. H. GOODWIN, Jr., and E. J. LEONARD. 1980. A 48-well micro chemotaxis assembly for rapid and accurate measurement of leucocyte migration. J. Immunol. Methods 33: 239. 10. FEIGE, U., J. KNOP, and C. SORGo 1980. Large scale production of human monocytes by leukapheresis. Immunobiology 157: 217. 11. BOYUM, A. 1968. Isolation of mononuclear cells and granulocytes from human blood. Scand. J. Clin. Lab. Invest. Vol. 21, Supp!. 97, 77: 89. 12. NEUMANN, c., and C. SORGo 1980. Sequential expression of functions during macrophage differentiation in murine bone marrow liquid culture. Eur. J. Immunol. 10: 834. 13. SUTER, L., J. BROGGEN, and C. SORGo 1980. Use of an enzyme-linked immunosorbent assay (ELISA) for screening of hybridoma antibodies against cell surface antigens. J. Immunol. Methods 39: 407. 14. ZIGMOND, S. H., and J. G. HIRSCH. 1973. Leukocyte locomotion and chemotaxis. New methods for evaluation and demonstration of cell-derived chemotactic factor. J. Exp. Med. 137: 387.
Dr. C. SORG, Department of Experimental Dermatology, University of Miinster, vonEsmarch-Str. 56, 4400 Miinster, Federal Republic of Germany