Eur. J. Immunol. 1984.14: 47-52 Marginal zone B cells express CR1 and CR2 receptors 47

4 Askenase, P. W., Bursztajn, S. , Gershon, M. D. and Gershon, R. K., J . Exp. Med. 1980. 152: 1358.

5 Askenase, P. W., Metzler, C. M. and Gershon, R. K., Immunol- ogy 1982. 47: 239.

6 Van Loveren, H., Meade, R. and Askenase, P. W., J . Exp. Med. 1983.157: 1604.

7 Ptak, W., Askenase, P. W., Rosenstein, R. W. and Gershon, R. K., Proc. Natl. Acad. Sci. USA 1982. 79: 1969.

8 Askenase, P. W., Rosenstein, R. W. and Ptak, W., J . Exp. Med. 1983. 157: 862.

9 Askenase, P. W., Van Loveren, H., Rosenstein, R. W. and Ptak, W., Monogr. Allergy 1983. 18: 249.

10 Liu, F. T., Bohn, J. W., Ferry, E. L., Yanamoto, H., Molivaro, C. A, , Sherman, L. A, , Klinman, N. R. and Katz, D. H., J . Immunol. 1980. 127: 2728.

11 Britz, J. S., Askenase, P. W., Ptak, W., Steinman, R . M. and Gershon, R. K., J . Exp. Med. 1982. 155: 1344.

12 Kulczycki, Jr . , A,, Isersky, C. and Metzger, H., J . Exp. Med. 1979. 139: 600.

13 Linthicum, D. S . , Munoz, J. J . and Blashett, A., Cell. Immunol. 1982. 73: 299.

14 Kops, S. K., Van Loveren, H., Rosenstein, R. W., Ptak, W. and Askenase, P. W., Lab. Invest. 1984, in press.

15 Karnovsky, M., J. Cell Biol. 1965. 27: 137A.

16 Richardson, K. L., Larrett, L. and Finke, E. H., Stain Technol.

17 Shore, P. A , , Pharmacol. Rev. 1962. 14: 531. 18 Carlsson, A, , Handbook of Experimental Pharmacology, Springer

19 Van Orden, L. S . , Vugman, I., Bensch, K. G. and Giarman, N. J.,

20 Hallenbeck, R. A., Shore, P. A, , Naunyn-Schmiedebergs Arch.

21 Theoharides, T. C., Bondy, P. K., Tsakalos, N. D. and Askenase,

22 Lewis, R. A. and Austen, K. F., Nature 1981. 293: 103. 23 Bach, M. K., Ann. Rev. Microbiof. 1982. 36: 371. 24 Tamir, H., Theoharides, T. C., Gershon, M. D. and Askenase, P.

25 Tada, T. and Okumura, K., Adv. Immunol. 1979. 28: 1. 26 Taussig, M. J., Immunology 1980. 41: 759. 27 Germain, R. N. and Benacerraf, B., Springer Semin.

Immunopathol. 1980. 3: 93. 28 Askenase, P. W . , Schwartz, A., Siegel, J. N. and Gershon, R. K.,

Int. Arch. Allergy Appl. Immunol. 1981. 66 (Suppl. I): 125. 29 Rocklin, R. E., in Sirois, P. and Rola-Pleszczynski, M. (Eds.),

Immunopharmacology, Elsevier, Amsterdam 1982, p. 49.

1960. 35: 313.

Verlag, New York 1966, p. 529.

J . Pharmacol. Exp. 1967. 158: 195.

Exp. Pathol. Pharmakol. 1971. 203: 263.

P . W., Nature 1982. 297: 229.

W., J . Cell Biol. 1982. 93: 638.

David Gray', Ian McConnell', Dinakantha S. Kumararatne', Ian C. M. MacLennan', John H. Humphrey' and HervC BazinA

Department of Immunology, University of Birmingham Medical School, Birmingham', Department of Veterinary Pathology, University of Edinburgh, Edinburgh', Department of Immunology, Royal Postgraduate Medical School, Hammersmith Hospital, London' and Experimental Immunology Unit, University of Louvain. BrusselsA

Marginal zone B cells express CRl and CR2 receptors

Opsonized yeast is known to bind strongly to the marginal zones in frozen sections of rat spleen (Kumararatne, D. S . et al., Eur. J. Immunol. 1981. 11: 858). This study reports an analysis of the cells involved in this binding. Sheep red cells coated respec- tively with C3b, C3bi or C3d were used as indicator cells. These showed homoge- neous binding of both C3b and C3bi to marginal zones and germinal centers. C3d- coated red cells bound in a uniform speckled pattern to marginal zones. They also bound to germinal centers and the small lymphocyte zones of the follicles. Selective depletion experiments were undertaken to show that binding to marginal zones was a property of the IgM' and IgD- B cells characteristic of this area. Binding to germinal centers was attributable to follicular dendritic cells. The C3d receptors in follicles were shown to be on IgM' and IgD' small B lymphocytes

1 Introduction

The marginal zone (MZ) is the outermost compartment of the white pulp of mammalian spleens. Most cells found in this area are B lymphocytes, which differ in several respects from the follicle-seeking B cell subset. MZ B cells are predominantly surface membrane IgM positive (IgM') but lack IgD (IgD-); in contrast, most follicular cells are IgM' and IgD' [l]. Most

[I 41881

Correspondence: David Gray, Department of Immunology, Univer- sity of Birmingham, Medical School, Vincent Drive, Edgbaston, Bir- mingham B15 2TJ, GB

Abbreviations: EA: Antibody-coated erythrocytes EAC: Comple- ment-coated EA HGG: Human gammaglobulin FITC: Fluorescein isothiocyanate HES: Hydroxyethyl starch TI-2: Thyrnus-indepen- dent type 2 antigens MZ: Marginal zone(s) CY: Cyclophos- phamide PALS: Periarteriolar lymphocytic sheath

0 Verlag Chemie GmbH, D-6940 Weinheim, 1984

B cells of the follicles are small lymphocytes, while B lympho- cytes of the MZ are medium-sized cells. Furthermore, MZ cells do not recirculate, unlike the small B lymphocytes of the follicles [2]. The origin and function of MZ B cells have been discussed in details elsewhere [l, 3-51, Complement-coated yeast and IgG-coated red cells (EA) bind with exceptional avidity to the MZ when layered on frozen sections of the spleen [ 2 ] . However, it has not been clear whether these indi- cator cells adhere to the MZ B cells or to macrophages, which constitute less than 5% of MZ cells [6] .

Treatment of rats with a single dose of cyclophosphamide (CY), at a dose of 500 mg/m2, selectively destroys the lympho- cytes of the MZ [3, 61. In the present study, we show that this CY-induced depletion is associated with selective loss of IgM', IgD- B cells from the system. The macrophages of the splenic MZ are not removed by this CY treatment. This selec- tive effect of CY is used to study which cells in the MZ carry receptors €or C3 breakdown products.

0014-2980/84/0101-0047$02.50/0

48 D. Gray, I. McConneIl, D. S. Kumararatne et al. Eur. J. Immunol. 1984.14: 47-52

Complement receptors on tissue cells have three distinct specificities for different parts of activated C3. The CR1 recep- tor binds C3b and C4b and has a wide tissue distribution, being found on primate red cells, neutrophils, macrophages and lym- phocytes [7, 81. This receptor has recently been shown to bind to C3bi also [9]. The CR2 receptor has been detected on lym- phocytes only and it binds C3d [8], although lymphocytes have also been shown to bind C3bi via this receptor [8, 101. The CR3 receptor has a simlar tissue distribution to CR1 [lo] and was thought to bind to C3bi, although recently it has been shown to react with the g region of C3d-g and not C3bi [9]. Red cells, respectively coated with C3b, C3bi and C3d, are used in the present study to analyze which of C3 breakdown products will bind to rat spleen sections.

2 Materials and methods

2.1 Animals

(Lou x DA)F1 hybrid rats derived from highly inbred parental strains bred and maintained under standard laboratory condi- tions at the Department of Immunology, University of Bir- mingham, were used in this study. Animals were age and sex matched within any one experiment.

2.2 Antisera

Rabbit anti-rat IgM and rabbit anti-rat IgD antibodies were produced in the Medical School, University of Louvain, Brus- sels, as described by Bazin et al. [ll].

2.3 Drug administration

Rats were given a single i.p. injection of CY (Montedison Pharmaceuticals, Barnet, GB) at a dose of 500 mg/m2 body surface, as described elsewhere [12].

2.4 Fluorescent conjugates

Fluorescein-labeled heat-aggregated human gamma globulin (HGG) and rhodamine-labeled hydroxyethyl starch (HES) were prepared as indicated in [13]. Localization of fluorescein isothiocyanate-coupled (FITC) heat-aggregated HGG was assessed in rats killed 48 h after an i.v. injection at 1 mg/rat. Rhodamine-labeled HES was injected i.p. at a dose of 200 pg/ rat, 48 h before killing.

2.5 Double rosette assay

The preparation of spleen cell suspensions, the coating of rhodamine- or fluorescein-labeled ox red cells with anti-p or an t i4 antibodies, respectively, and the technique of double rosetting to simultaneously detect surface membrane p and 6 are described in detail by Gray et al. [l].

2.6 Preparation of C3-coated sheep red cells

Indicator erythrocytes carrying activated human C3 in various stages of breakdown were used. Rabbit-IgM-coated sheep

erythrocytes (EA) were incubated in a guinea pig R3 reagent to prepare EAC142gp [14]. These were then incubated with purified human C3 (6 mg/ml) for 20 min at 37°C to produce EAC42gp3hv (EAC3b). Some of these indicators were then treated with purified C3b inactivators (Factor I + Factor H) for 30 min at 37 "C to produce EAC3bi. Aliquots of the latter were then treated with trypsin (2 yglrnl) for 5 min at 37°C. Finally, a 2-fold molar excess of soybean trypsin inhibitor was added for 30 s and the cells then washed 3 times to produce EAC3d. The state of the activated C3 on these indicators was checked with: (a) radiolabeled "'I conglutinin used to dis- criminate between C3bi and the other preparations. The counts for EAC3bi rosettes were 5280 cpm, while counts for EAC3b, EAC3d and EAIgM were 414, 360 and 300 cpm, respectively. (b) A monoclonal antibody C9, which reacts with C3bi and C3d but not C3b was also used. Counts for EAC3b with 12'I-labeled C9 were 210 cpm, those for EAC3bi and EAC3d were 27000 cpm and 25000 cpm, respectively. (c) Polyclonal anti-C3 was used to confirm the sensitization of the EAC3b. This reagent agglutinated EAC3b, EAC3bi and EAC3d but not EAIgM. These counts are for the first set of cells used in the experiments reported in this study, but are also representative of the other two sets used.

2.7 Histology and red cells overlays

Portions of spleen were snap frozen in liquid nitrogen. Frozen sections cut at 5 pm were mounted on formol-gelatinized slides and then dried thoroughly for 1 h at room temperature. C3- coated red cells were used at a 0.5% or 1% suspension, the lower dilution giving satisfactory results. Sections were over- laid with 2-3 drops of this suspension and incubated at room temperature for 1 h. The red cells which had not bound to the section were rinsed off the slide by gentle dipping into phos- phate-buffered saline (PBS) + 2% fetal calf serum (FCS). The preparations were then fixed for 5 min in methanol before staining in hematoxylin and eosin.

Fluorescent conjugate localization was observed in frozen sec- tions which were immediately fixed in cold methanol to pre- vent leaching of the rhodamine-labeled HES.

2.8 Microscopy

Fluorescent preparations were observed using a Leitz Dialux microscope fitted with a Ploemopak 2.3 fluorescence illuminator with two interchangeable filter systems; filter sys- tem 12 for fluorescein and filter system N2 for rhodamine.

3 Results

3.1 The binding patterns of sheep red cells coated with different C3 breakdown products to frozen sections of rat spleen

These are demonstrated in Fig. 1 a-d. Both C3b and C3bi- coated cells form a continuous mat over the MZ (Fig. 1 a-b). Binding to germinal centers is also strong. Conversely, there is minimal binding to: (a) the small lymphocyte areas of the follicles, which contain recirculating B cells; (b) the periar- teriolar lymphocytic sheath, which is populated mainly by recirculating T cells and (c) the red pulp.

Eur. J . Immunol. 1984.14: 47-52 Marginal zone B cells express CR1 and CR2 receptors 49

Figure I. (a) Normal adult rat spleen X 110. Section has been overlaid with EAC3b. Nuclei are stained with hematoxylin, while the red cells stain with eosin. These indicator cells bind to the MZ (M) and to germinal centers (G). There is negligible binding to the PALS (P), the small lymphocyte zones of the follicle (F) or the red pulp (R). (b) Normal adult rat spleen X 135. Section overlaid with EAC3bi. The binding pattern of these indicator cells is essentially the same as that of EAC3b. (c) Normal adult rat spleen X 55. Section overlaid with EAC3d. These indicator cells bind to the MZ (M), germinal centers (G) and to the small lymphocyte zones of follicles (F). No indicator cells adhere to the PALS (P) or red pulp (R). The binding of EAC3d to follicles and MZ gives a characteristic speckled appearance. (d) Normal adult rat spleen x 135. Section overlaid with EA, which adhere only to cells in the marginal sinus (arrowed). The MZ (M), follicle (F), PALS (P) and red pulp are marked.

The C3d-coated cells bind to the MZ in a less uniform manner than C3b or C3bi-coated cells, giving a speckled pattern (Fig. lc). Binding to germinal centers is also seen by the C3d-sen- sitized cells and there is definite binding to the small B lym- phocytes of primary follicles and the corona of secondary folli- cles. However, binding to the small lymphocyte corona of the follicles is less marked than to the MZ. This pattern of binding indicates that follicular B cells express CR2 but not CR1 receptors. Fig. Id shows the binding of sheep cells coated with rabbit anti-sheep red cells IgM. There is some adherence to the marginal sinus with these indicator cells, but none to ger- minal centers or the MZ.

3.2 Experiments to assess the nature of the cells or other structures binding C3 breakdown products in rat spleen sections

3.2.1 Sensitivity of MZ B cells to CY treatment

As discussed in Sect. 1, lymphocytes of the splenic MZ are depleted by CY at a dose of 500 mg/m2. Also MZ B cells can be

identified as being surface membrane v' but 6-. Conversely, follicles are rich in p', 6' B cells. Groups of rats were treated with CY 500 mg/m2 and 2 days later their spleens were taken. A portion of the spleen was snap frozen to assess C3 binding on sections. A lymphocyte suspension was made of the remainder and the proportion of B cells expressing surface membrane IgM and/or IgD was assessed. The results of these assays on cell suspensions are shown in Fig. 2. It will be seen that p', 6- cells are reduced by twelvefold. Overlays of indi- cator cells of all types, on frozen sections of these spleens, showed almost total loss of binding to MZ (Fig. 3a). Binding of all C3 indicators to germinal centers was retained, as was binding of C3d-coated cells to the follicles (Fig. 3b) and the EA to the marginal sinuses. These results are summarized in Table 1.

3.2.2 The effect of CY on MZ macrophages and follicular dendritic cells

These cells can be identified in frozen sections, for MZ mac- rophages take up HES [ 131, while follicular dendritic cells

D. Gray, I. McConnell, D. S. Kumararatne et al. Eur. J . Immunol. 1984.14: 47-52

TOTAL TOTAL LYMPHOCYTES ROSETTING p+6- p+ 6+ p- 6 +

N C N C N C N ’ C N C

1 . 0 ’

Figure 2. Lymphocyte subpopulations in dissociated spleen from rats treated with 500 mgim’ CY. The results are expressed as percent devi- ation from log means of rats of a control group receiving no treatment. Standard errors of the groups of 6 animals are shown. N = control rats and C = rats treated with CY.

acquire immune complex on their surface by an indirect pro- cess which is only poorly understood [El. Groups of rats were injected with rhodamine-labeled HES and fluorescein-labeled heat-aggregated HGG. Half the animals were treated 2 days later with CY 500 mg/m2 and were killed after a further 3 days. The amount of HES retained in MZ macrophages (Fig. 4a) and heat-aggregated HGG localized on follicular dendritic cells (Fig. 4b) was assessed by fluorescence microscopy in fro- zen sections. The retention of both these products was not appreciably diminished by CY treatment.

3.3 The effect of CY on the localization of aggregated HGG on to follicular dendritic cells

Groups of animals were treated with 500 m g h 2 CY and 8 days later, were injected with fluorescein-labeled heat-aggregated HGG. Upon killing, 48 h after the latter procedure, localiza- tion into germinal centers in the spleen was shown to be almost completely abolished.

Figure 3. (a) Spleen from CY-treated rat ( x 110). Section overlaid with EAC3b. Binding of these indicators to such spleens is markedly reduced. Only a small band of red cells bind to cells in the MZ. Hematoxylin staining shows that the MZ is depleted of lymphoid cells. A similar picture is obtained when such spleen sections are overlaid with EAC3bi. (b) Spleen from CY-treated rat ( X 55). Binding of EAC3d to the MZ is abolished, while adherence to follicles (F) and germinal centers (not shown) remains. Other symbols as in Fig. 1 (d).

Figure 4. Frozen section of spleen from normal adult rat ( x 300) injected 48 h previously with rhodamine-labeled HES and fluorescein-labeled aggregated HGG. (a) Localization of rhodamine-HES in macrophages in the MZ. (b) Localization of fluorescein-HGG in germinal center in the same section.

Eur. J. Immunol. 1984.14: 47-52 Marginal zone B cells express CR1 and CR2 receptors 51

Table 1. Binding of EAC to spleen histological compartments

Spleen compartment - EAC binding") EAC MZ GCb' F PALS RP

Normal b +++ +++ - - -

d ++ ++ ++ - -

bi + i- +++ - - - d + l- ++ ++ - -

- - - bi +++ +++ a)

b) - - - CY- treated b + i- +++

4 Discussion

CY treatment causes the depletion of most MZ cells. How- ever, MZ macrophages, which have localized rhodamine- labeled HES (Fig. 4a), remain following CY treatment. We have shown previously that the MZ B cells possess surface membrane p but not 6 [l]. The double rosetting of spleen cell suspensions from CY-treated rats indicate that the p' 6- B cell population is selectively destroyed (Fig. 2). Hence the loss of binding of C3b-, C3bi- and C3d-coated red cells to spleens from treated animals provides good evidence that the p' 6- B cells of the MZ possess receptors for the C3 breakdown prod- ucts, i.e. they carry CR1 and CR2 receptors on their surface. We can reliably make a statement about CR1 and CR2 recep- tors only, because both CRl and CR2 receptors also bind C3bi 191. To detect the CR3 receptor, with confidence, would require the CRl and CR2 receptors to be blocked with anti- sera. However, this has only been carried out, to date, in man [8, 101. Binding of coated red cells to red pulp macrophages was minimal. It seems unlikely that this reflects an absence of complement receptors on these cells, but rather that they dif- fer quantitatively and/or qualitatively from those of MZ B cells.

Binding of C3b-, C3bi- and C3d-coated red cells to germinal centers after CY treatment is probably by adhesion to recep- tors on follicular dendritic cells [16], since the dose of CY used causes the loss of lymphoid cells from germinal centers [3], while not affecting prelocalized immune complexes. This result suggests that follicular dendritic cells possess CR1 and CR2 receptors and is interesting in that CR2 has, until now, been demonstrated only on lymphocytes. The preservation of C3d binding by the splenic follicles after CY treatment may reflect the relative preservation of a subpopulation of B cells. The small lymphocyte areas of splenic follicles do contain both a major recirculating and minor static B cell components [ l , 41. Further studies are required to determine which of these is responsible for the observed binding of C3d-coated red cells. The speckled nature of binding of C3d-coated red cells sug- gests that receptors for this molecule are less uniformly ex- pressed than those for C3b. This may be due to subpopulations of cells which are respectively CR2' and CR2-. Whether this is due to heterogeneity of CR2 expression within the same devel- opmental population or reflects two distinct subsets is unclear.

Several lines of evidence have led us to believe that MZ B cells may be involved in responses to thymus-independent type 2 (TI 2) antigens [ 171, including pneumococcal antigens. The predominant and most effective antibody against pneumococ-

These results represent the observations from 8 normal and 8 CY-treated rats. +/- = occasional indicator cells bound. GC: germinal centers, F: follicles, PALS: periarteriolar lymphocytic sheath, RP: red Pulp.

cal infection in mice is y3 [HI, which is equivalent to yzc in rats [19]. In the order of 20% of MZ B cells express yZc on their surface [4]. In addition, yzc production is selectively retained in rats depleted of 6 expressing B cells [4]. The importance of spleen in resistance to pneumococcal infections is well documented [20, 211. Recent work by Van Wyck et al. [20] and Brown et al. [21] has shown that the spleen is of particular importance in the clearance of bacteria when complement levels are lowered or when organisms are opsonized with com- plement components alone. MZ cells are positioned in the primary blood sinusoids of the spleen [22], in the proximity of carbohydrate-localizing macrophages [13] and also possess receptors for C3b, C3bi and C3d. Hence these lymphocytes may well be of importance in the trapping of and the initiation of antibody response to pneumococcal antigens.

Secondly, MZ B cells may help to transport immune com- plexes into germinal centers. Antigen-antibody complexes are necessary for germinal center formation; antigen alone cannot initiate germinal center formation [23]. Complement is also required for the formation of these structures [23-251. It is known that complexes become trapped in the MZ prior to localization in the germinal center [26]. The possession of Fc receptors and avid C3b, C3bi and C3d receptors make MZ B cells ideally equipped to bind complexes. Thus, if MZ B cells play a role in immune complex transport to germinal centers, this may, at least in part, explain the complement dependence of this process. The experiments of Brown et al. [15], Van Rooijen [27,28] and Veerman and Van Rooijen [29] show that procedures which deplete the MZ of cells abolish immune complex localization into germinal centers. We find that the selective depletion of MZ B cells with CY inhibits localization of aggregated HGG within follicle centers but does not affect follicular dendritic cells. Systemic administration of endotoxin [30] produced appearances which suggest that MZ cells, which do not normally recirculate [2], may be induced to migrate into follicles by this agent. The hypothesis that MZ cells help to transport antigen-antibody complexes into the follicle center is consistent with these data.

We are grateful to Professor P. J . Lachmann for providing the purified conglutinin and the monoclonal antibody to C3c used for assessing C3b, C3bi and C3d-coating of indicator erythrocytes and D. Grennan in assistance with preparation of fluoresceinated HGG and rhodamine- labeled HES.

Received June 24, 1983; in revised form September 13, 1983.

52 G. H. W. Wong, I. Clark-Lewis, A. W. Harris and J. W. Schrader Eur. J. Immunol. 1984.14: 52-56

5 References

1 Gray, D., MacLennan, I. C. M., Bazin, H. and Khan, M., Eur. J .

2 Kumararatne, D. S. , Bazin, H. and MacLennan, I. C. M., Eur. J .

3 Kumararatne, D. S. and MacLennan, I. C. M., Eur. J . Immunol.

4 Bazin, H., Gray, D., Platteau, B. and MacLennan, I. C. M., Ann.

5 MacLennan, I. C. M., Gray, D., Kumararatne, D. S. and Bazin,

6 Kumararatne, D. S. , Gagnon, R. F. and Smart, Y., Immunology

7 Ross, G. D., Winchester, R. J., Rabellino, E. M. and Hoffman,

8 Dobson, N. J., Lambris, J . D. and Ross, G. D., J . Immunol. 1981.

9 Ross, G. D., Fed. Proc. 1982. 41: 3089.

Immunol. 1982. 12: 564.

Immunol. 1981. 11: 858.

1981. 11: 865.

N. Y. Acad. Sci. 1982. 399: 157.

H., Immunol. Today 1982. 3: 305.

1980. 40: 123.

T., J. Clin. Invest. 1978. 62: 1086.

126: 693.

10 Ross, G. D. and Lambris, J. D., J. Exp. Med. 1982. 155: 96. 11 Bazin, H., Platteau, B., Beckers, A. and Pauwels, R., J . Immunol.

1978. 121: 2083. 12 Gagnon, R. F. and MacLennan, I. C. M., Clin. Exp. Immunol.

1979. 37: 89. 13 Humphrey, J. H. and Grennan, D., Euv. J . Immunol. 1981. 11:

221. 14 Lachman, P. J . and Hobart, M. J., in Weir, D. M. (Ed.), Hand-

book of Experimental Immunology, Blackwell Scientific Publica- tions, Oxford 1978, vol. 1, p. 5A1.

15 Brown, J. C., Harris, G., Papamichail, M., Sljivic, V. S. and Hol-

16 Humphrey, J. H. and Grennan, D., Adv. Exp. Med. Biol. 1982.

17 Mosier, D. E., Zitron, 1. M., Mond, I. J., Ahmed, A., Scher, I.

18 Briles, D. E., Claflin, J. L., Schroer, K. and Forman, C., Nature

19 Dt r Balian, G. P., Slack, J., Clevinger, B. L., Bazin, H. and

20 Van Wyck, D. B., Witte, M. H. and Witte, C. L., J . Infect. Dis.

21 Brown, E. J., Hosea, S. W. and Frank, M. M., J . Clin. Invest.

22 Herman, P. G., Monogr. Allergy 1980. 16: 126. 23 White, R. G., Henderson, D. C., Eslami, M. B. and Nielson, K.,

24 Klaus, G. G. B. and Humphrey, J. H., Immunology 1977.33: 31. 25 Papamichail, M., Gutierrez, C., Embling, P., Holborow, E. J. and

26 Nossal, G. J. V., Austin, C. M., Pye, J. and Mitchell, J . , Int. Arch.

27 Van Rooijen, N., Immunology 1974. 27: 617. 28 Van Rooijen, N., Immunology 1975. 28: 1155. 29 Veerman, A. J. P. and Van Rooijen, N., Cell Tissue Res. 1975.

30 Veerman, A. J. P. and Vries, H. D., Z. Immunitarsforsch. 1976.

borow, E. J., Immunology 1973. 24: 995.

149: 823.

and Paul, W. E., Immunol. Rev. 1977. 38: 89.

1981. 294: 88.

Davie, J. M., J . Exp. Med. 1980. 152: 209.

1982. 145: 514.

1981. 67: 975.

Immunology 1975. 28: 1.

Pepys, M. B., Scand. J. Immunol. 1975. 4: 343,

Allergy Appl. Immunol. 1966. 29: 368.

161: 211.

151: 202.

Grace Ha W- W o w , Ian Clark-Lewis, Effect of cloned interferon-y on expression of H-2 and Ia antigens on cell lines of hemopoietic, lymphoid,

Alan W. Harris and John W. Schrader - -

The Walter and Hall Institute Of epithelial, fibroblastic and neuronal origin* Medical Research, Royal Melbourne Hospital, Victoria

Interferon-y (IFN-y), tested in the form of the product of a cloned murine IFN-y gene, was found to increase the expression of H-2 antigens on cultured cell lines of a wide variety of cell types including factor-dependent hemopoietic cells, pre-B and B cells, macrophages, T cells, mast cells and cell lines of epithelial, fibroblastic and neuronal origin. However, IFNy induced Ia antigens on only B cells, macrophages and T-dependent mast cells or persisting cells. On the basis of these results, we suggest that a major function of IFN-y is to potentiate immune responses by enhanc- ing the expression of H-2 and Ia antigens on a variety of cell types.

[I 43111

* This work was supported by the National Health and Medical Research Council, Canberra, Australia, the Bushell Trust, the Drakensberg Trust and the Windermere Hospital Foundation (Australia) and National Institutes of Health (USA) Grant A103958 to G.J.V. Nossal.

Correspondence: Grace H. W. Wong, The Walter and Eliza Hall Insti- tute of Medical Research, Post Office, Royal Melbourne Hospital, Victoria 3050, Australia

Abbreviations: MHC: Major histocompatibility complex IFN-y: Interferon-y P cell: Persisting cell

0014-2980/84/0101-0052$02.50/0

1 Introduction Class I and class I1 major histocompatibility complex (MHC) antigens are important for initiation, expression and regula- tion of immune responses [l]. We and others have observed that a soluble factor produced by activated T cells increases the expression of Ia antigens on mast cells (21, B cells [>5] and macrophages [5-lo]. The cellular source and biochemical properties of this factor indicated that it might be interferon-y (IFN-y) [2, 4, 5, 111 and recent experiments with material derived from a cloned gene have established that the MHC- inducing factor is indeed IFN-y [S, 12-16]. However, the effect of IFN-y on the expression of MHC antigens on a wide range

0 Verlag Chemie GmbH, D-6940 Weinheim, 1984