antigen presentation by human monocytes: effects of modifying major histocompatibility complex class...

370 J. Rhodes, J. Ivanyi and P. Cozens Eur. J. Immunol. 1986.16: 370-375

John Rhodes, Juraj Ivanyi and Peter Cozens+

Antigen presentation by human monocytes: effects of modifying major histocompatibility complex class I1 antigen expression and interleukin 1 production by using recombinant interferons and corticosteroids

Department of Experimental Immunobiology and Department of

v

Molecular Biology', Wellcome Research Laboratories, Beckenham, Kent Lymphocyte proliferation in response to monocytes pulsed with an antigenic extract

of Candida albicans was measured in vitro and the effects of modifying major histocompatibility complex (MHC) class I1 antigen expression at the surface of the antigen-presenting cells was investigated. The study shows that no simple correlation exists between changes in MHC class I1 antigen expression and changes in the effec- tiveness of antigen presentation. Recombinant interferon-al (rIFN-al), rIFN-y and hydrocortisone were found to increase the expression of monocyte class I1 MHC antigens. In contrast, rIFN-aZ did not increase class I1 antigen expression although it did increase MHC class I expression. Treatment of monocytes with rIFN-al, rIFN-a2 or corticosteroids during antigen pulsing resulted in a reduction in the subsequent proliferative lymphocyte response. In all cases this inhibitory effect was restricted to antigen-specific proliferative responses since the polyclonal lymphocyte response to pokeweed mitogen-pulsed monocytes remained unaffected. Only rIFN-y treatment of antigen-pulsed monocytes resulted in enhancement of the subsequent specific lymphocyte proliferative response. The suppressive effects of hydrocortisone could not be attributed to its well documented inhibitory effects on arachidonic acid metabolism. The effect of C. albicans antigen, IFN and corticosteroids on interleukin 1 (IL 1) production by monocytes was also investigated. C. albicans antigen alone induced IL 1 production. So too did IFN-a1 and IFN-y. IFN-a2 did not induce IL 1 production. Addition of interferons together with C. albicans, however, resulted in the same level of IL1 productions as with C. albicans antigen alone. Neither antigen nor IFN had any effect on IL 1 action in the thymocyte assay. Corticosteroids did not affect IL 1 production by monocytes but were potent antagonists of IL 1 in the thymocyte proliferation assay. Mitogen-induced thymocyte proliferation was also inhibited by corticosteroids. Pretreatment of monocytes with hydrocortisone followed by washing did not markedly affect their subsequent ability to produce IL 1 neither was it possible to reverse the inhibitory effects of hydrocortisone on antigen presentation by addition of exogenous IL 1. Thus, signals which alter class I1 MHC antigen expression influence the antigen-presenting capacity of monocytes by a mechanism independent of IL 1. No simple correlation exists between class I1 expression and antigen-presenting capacity.

1 Introduction

In the induction of immune responses two signals are required to initiate the proliferative response of specific T helper cells. One is antigen associated with self-major histocompatibility complex (MHC) class I1 molecules at the surface of antigen- presenting cells (APC) [l]. The other is the maturational signal interleukin 1 (IL 1) [ 2 ] . The expression of MHC antigens is a dynamic property of macrophages and other accessory cells and it seems reasonable to ask whether quantitative changes in the expression of class I1 MHC molecules at the surface of APC can influence the induction of an immune response. Here we investigate the effects of signals which alter the expression of monocyte class I1 MHC antigens on the ability of these cells to present antigen in vitro. We show that no simple correlation

[I 50781

Correspondence: John Rhodes, Department of Experimental Immunobiology, Wellcome Research Laboratories, Beckenham, Kent BR33BS, GB

Abbreviations: APC: Antigen-presenting cells IFN: Interferon IL 1: Interleukin 1 MHC: Major histocompatibility complex PWM: Pokeweed mitogen PHA: Phytohemagglutinin

exists between monocyte class I1 product expression and the presentation of a class I1 region-restricted antigen.

2 Materials and methods

2.1 Preparation of cells

Thirty ml of normal human blood was defibrinated with a U- shaped rod in a siliconized tube and centrifuged to obtain 15 ml serum. The original volume was then restored, and the blood further diluted 1 : 2 with phosphate-buffered saline. The mononuclear leukocyte fraction was obtained by Metrizoate/ Ficoll gradient separation and the cells were separated into monocyte and lymphocyte fractions by differential adherence to glass or plastic in the presence of 10% autologous serum. Two cycles of adherence of at least 2 h each at 37 "C were used to deplete monocytes from lymphocyte preparations. All man- ipulations were performed in RPMI 1640 medium (Flow Laboratories, Ayshire, Scotland) containing antibiotics and supplemented with 10% autologous serum. In a proportion of experiments B lymphocytes were depleted by adherence to nylon fiber columns (Fenwal Laboratories, Deerfield, IL). Lymphocytes, at the concentration of 5 x 107-10 x lo7 cells/

0014-2980/86/0404-0370$02.50/0 0 VCH Verlagsgesellschaft mbH, D-6940 Weinheim, 1986

Eur. J. Immunol. 1986.16: 370-375 Monocyte antigen presentation, class I1 MHC antigen expression and IL 1 production 371

L

ml, were added to the column and incubated at 37°C for 45 min after which the nonadherent cells were gently eluted with pre-warmed medium, washed and resuspended at a concentration of 2 x lo7 cells/ml.

:

2.2 Assays for monocyte MHC antigen expression

Monocyte class I MHC molecules were defined here by the anti-monomorphic monoclonal antibody (mAb) W6.32 [3]. Class I1 MHC molecules were defined by the anti-monomorphic monoclonal reagent YE2.36 [4, 51. mAb were purified from ascites fluid by protein A-Sephadex column chromatography. MHC antigen expression was determined by a titration rosette assay in which increasing amounts of mAb were cou- pled to sheep red blood cells by means of chromic chloride [6]. Rosette formation by adherent monocytes in Lab-tek chamber slides (Flow Labs.) was determined as a function of the dose of mAb. The resultant dose-response curves provide a sensitive reflection of changes in MHC antigen expression. Changes in MHC antigen expression in response to interferon were also demonstrated by means of flow cytometry. Monocytes were stained with monoclonal anti-MHC reagents as adherent monolayers. The cells were then washed and stained with fluorescein-conjugated anti-IgG secondary reagents and then removed from the glass by means of 24 mM xylocaine (1 h treatment followed by vigorous pipetting). The monocytes were then fixed with 1% formaldehyde in the presence of 1% fetal calf serum and subjected to flow cytometric analysis using an Ortho H50 cytofluorograph interfaced with a 2150 computer system. Operation of the instrument and computer gating were as previously described [7].

s - t L: 30 ; B

$ 2 0 -

E z 1 0 -

0 -

2.3 Lymphoproliferative assay for monocyte antigen presentation

A soluble antigenic extract of Candida albicans was obtained from Bencard Laboratories (Brentford, GB). The solution was dialyzed extensively against phosphate-buffered saline prior to use. Adherent monocytes in microtiter plates (Titer- tek, U-shaped wells, Flow Labs.) were pulsed with antigen at the concentration specified below for 20 h and then washed 3 times with medium containing 10% autologous serum. Auto- logous lymphocytes were then added and the cells cultured for a further 6 days in humidified air containing 5% COz. Triti- ated thymidine (2 Ci/mmol = 74 GBq/mmol) was present for the last 18 h of culture at a concentration of 10 yCi/ml (1 yCi/ well). Cells were harvested onto glass microfiber paper by means of a Titertek cell harvester (Flow Labs.) and subjected to liquid scintillation spectrometry. The polyclonal lymphoproliferative response to pokeweed mitogen (PWM; Gibco Labs., Grand Island, NY) was assayed in the same manner. Monocytes were pulsed with PWM for 20 h, washed 3 times and exposed to autologous lymphocytes for a further 6 days.

-

2.4 Assay for monocyte IL 1 production

Adherent monocytes in Linbro 24-well plates (Flow Labs.) were cultured for 24 h in the presence of the reagents specified. Cell-free supernatants were assayed for IL 1 content at 1/10 dilutions using thymocytes from C3H/He mice. Thymo- cytes at lo7 cellshl in microtiter plates were stimulated with a suboptimal dose of phytohemagglutinin (PHA; Wellcome

Labs., Beckenham, GB; 1 pg/ml) for 3 days. Proliferation was determined using [3H]thymidine and liquid scintillation spectrometry as described in Sect. 2.3.

2.5 Recombinant interferons (rIFN)

The IFN-al cDNA was obtained from Cetus Inc., Emmery- ville, CA. The cDNA designated IFN-a2 was cloned as described [9] and has the same sequence as IFN-aA of Goed- del and colleagues [lo]. The IFN-y cDNA was obtained as a nonexpressing plasmid from Dr. Jan Vilcek of New York Uni- versity. The genes were expressed in E. coli on plasmids bear- ing the tryptophan promoter and the trp L ribosome binding site. The construction of the expression plasmids will be described elsewhere (Cozens, Johnston and MacKenzie, manuscript in preparation). IFN production was induced with 3-0-indoleacrylic acid in 1.5 liters of M9 salts medium supplemented with casamino acids. Cells were concentrated 25- fold and treated with 10 mg/ml lysozyme for 30 min at 0°C before lysing by freezing and thawing. Cell debris was removed by low speed centrifugation and SlOO lysates were prepared by centrifugation in a Beckman 50Ti rotor to an 02t of 7.57 X lo1’. The recombinant IFN subspecies used in the present study were further purified by affinity chromatography using polyclonal antibodies and mAb. Antiviral activity was determined by means of cytophathic stain-uptake inhibition in V3 cells infected with Semliki Forest Virus. Titers for each IFN in the range of 3 x lo8 to 6 x lo8 units/A650/liter of bacte- rial culture were obtained routinely. The specific activity of IFN-a, was 3.5 x lo7 unitslmg, of IFN-a2 2 x lo8 unitslmg, and of IFN-y 5 x 10’ unitsimg protein.

3 Results

3.1 Effects of recombinant IFN and hydrocortisone on monocyte MHC antigen expression

The data in Fig. 1 shows the effects of IFN and hydrocortisone on the ‘expression of class I1 and class I MHC antigens by normal monocytes cultured for 20 h. IFN-al at a concentration of 2 X lo3 unitdm1 markedly increased the expression of class

& A B C

T

2 A E C = Anti-MHC Clarr I 1

1 5 0 7 8 [ Y E 2 361 IW6 321 A m - M H C Clarr

Figure I. Changes in class I1 and class I MHC antigen expression induced by rIFN and corticosteroids. (A) rIFN-a, at 2 X lo3 unitsiml. (B) rIFN- u, at 2 x 10’ unitsirni. (C) rIFN-y at 10’ unitslml. (D) Hydrocortisone at M (0.5 ygiml).

372 J. Rhodes, J. Ivanyi and P. Cozens Eur. J. Immunol. 1986.16: 370-375

I1 antigens whereas IFN-a2 had no significant effect. IFN-a2 markedly increased the expression of class I MHC whereas IFN-al had no significant effect. The preferential enhancement of class I1 by IFN-al and class I by IFN-a2 was main- tained over a dose range of lo3 to lo4 antiviral unitdm1 (data not shown). IFN-y was found to increase monocyte class I1 MHC antigen expression at doses one tenth (in terms of both weight and antiviral activity) of that required to produce con- sistent increases by IFl\T-al. Hydrocortisone hemisuccinate at a concentration of M was also found to increase the expression of both class I and class I1 MHC antigens after a 20-h period of culture. Increases in monocyte class I1 MHC antigen expression in response to IFN-al were also demonstrated by flow cytofluorometry. The data in Fig. 2 may be compared with the data for IFN-al shown in Fig. 1 where the same dose, time course and conditions were employed.

- Untreated

Fluorescence intensity

Figure 2. Changes in class I1 MHC antigen expression in response to IFN-al at 2 X lo3 unitslml demonstrated by means of flow cytometry. Fluorescence intensity is on a linear scale.

81

4c A l

i 3.2 Effects of rIFN and corticosteroids on antigen presentation

by monocytes

Treatment of monocytes with IFN-al at doses which increased class I1 MHC antigen expression during a 20-h pulse with Can- dida albicans antigen (0.1 %) consistently inhibited the subsequent proliferative response of autologous lymphocytes in a dose-dependent manner (Fig. 3A). The same dose-dependent inhibition of antigen presentation was observed when IFN-a2 was present during antigen pulsing, when no detectable increase in class I1 MHC had been induced (Fig. 3B). Hyd- rocortisone and dexamethasone likewise produced a dose- dependent inhibition of antigen presentation (Fig. 3C). Only IFN-y treatment of monocytes during the 20-h pulse with antigen enhanced the subsequent specific proliferative response of lymphocytes (Fig. 3D). The same results were observed whether the autologous responding lymphocyte population was complete or depleted of B cells. Because variation occurs in the responsiveness of normal individuals to C. albicans, each observation reported here was repeated in at least 5 independent experiments.

2 31

B I T 6 ’= 24

0 8 ’; 16

q

0

c

5 ‘ i e I

I *? C

fl I-L 1

- 2

L

3 -L 4

- 5 2 3 4 5

I 5 6 7

1

T T D I

h i 1 2

c

r. 3.3 Effects of rIFN and hydrocortisone on mitogen presentation by monocytes

Purified peripheral blood T lymphocytes mounted a substan- tial proliferative response when exposed to monocytes pulsed with PWM (1/200 dilution of Gibco commercial grade). Unlike the presentation of antigen by monocytes, the “presentation” of mitogen was unaffected by treatment with IFN-al (Fig. 4A) and hydrocortisone (Fig. 4B). These observations on the polyclonal lymphocyte response to mitogen-pulsed monocytes also argue against any possible trivial effects of residual IFN in pulsed and washed monocyte cultures directly inhibiting the proliferation of lymphocytes responding to specific antigen.

Figure 3. Lymphocyte proliferation in response to autologous monocytes pulsed for 20 h with C. albicans: effects of IFN and corticosteroids. (A) Effects of IFN-a, treatment of monocytes was as follows. (1) No antigen; (2) C. albicans at 0.1%; ( 3 ) C. albicans at 0.1% plus IFN-al at 5 x lo3 unitslml. (4) C. albicans at 0.1% plus 1FN-q at 5 X 102units/ml. (5) C. albicans at 0.1% plus IFN-a, at 50 unitslml. (B) Effects of IFN-a,. Treatment of rnonocytes was as follows. (1) No antigen; (2) C. albicans at 0.1%; (3) C. albicans at 0.1% plus IFN-a2 at 2 X lo3 unitslml. (4) C. albicans at 0.1% plus IFN-a’ at 2 X 10’ units/ ml; ( 5 ) C. albicans at 0.1% plus IFN-y, at 20 unitdml. (C) Effect of corticosteroids. Treatment of monocytes was as follows. (1) No antigen; (2) C. albicans at 0.1%; (3) C. albicans at 0.1% plus hydrocortisone at 10 pg/ml; (4) C. albicans at 0.1% plus hydrocortisone at 1 pg/ ml; (5) C. albicans at 0.1% plus hydrocortisone at 0.1 yglml; (6) C. albicans at 0.1% plus dexamethasone at 10 yg/ml; (7) C. albicans at 0.1% plus dexamethasone 0.1 pg/ml. (D) Effects of IFN-y. Treatment of monocytes was as follows. (1) No antigen; (2) C. albicans at 0.1%; (3) C. albicans at 0.1% plus IFN-y at 1 X lo3 unitslml; (4) C. albicans at 0.1% plus IFN-y at 1 X lo2 unitshl.

3.4 The effect of C. albicans antigen, IFN and corticosteroids on monocyte IL 1 production and action

Monocytes pulsed with C. albicans antigen for 24 h produced significant amounts of IL 1, compared with control cultures, in a dose-dependent manner (Fig. 5A). IFN-a1 and IFN-y in the absence of C. albicans antigen also induced IL1 production whereas IFN-a2 had no significant effect (Fig. 5B). Neither antigen nor IFN exerted any significant effect on the action of

Eur. J. Immunol. 1986.16: 370-375 Monocyte antigen presentation, class I1 MHC antigen expression and IL 1 ,production 373

m 1 2 3 4

'i. r I 3 4

Figure 4. Lymphocyte proliferation in response to autologous monocytes pulsed for 20 h with PWM: effects of IFN-a and hydrocortisone. (A) IFN-a2 treatment of monocytes was as follows. (1) No mitogen; ( 2 ) PWM 11100 dilution; (3) PWM 1/200; (4) PWM M O O plus IFN-a, 2 X lo3 units/ml. (B) Effects of hydrocortisone. Treatment of monocytes was as follows. (1) No mitogen; ( 2 ) PWM 1/100; (3) PWM 11200; (4) PWM 1/100 plus hydrocortisone at M.

s- = 0 0.002 0.01 o. 1

Concentratton B c. Alblcans 4g (%) 4

4 5l

0 0.01 0.5 1.0

Concentratlon Cortlcosterold (pg/ml)

115078.51

Concentration IFN (units/ml)

0

0- 0 100 500 1000

Concentratton IFN (un l tsh l )

Figure 5. (A) Monocyte I L l production in response to C. albicans antigen. (B) Monocyte IL 1 production in response to IFN, IFN-y (U), IFN-al (e), IFN-a2 (A). (C) Effect of corticosteroid on monocyte IL 1 production in response to C. albicans antigen. No addition (0), C. albicans 0.1% (O), C. albicans hydrocortisone ( k O ) , C. albicans plus dexamethasone (m), C. albicans plus hydrocortisone, dialyzed supernatants (0). (D) Monocyte IL 1 production in response to C. albicans antigen plus IFN. No addition (0), C. albicans 0.1% respec- tive controls (0 0 A). IFN-y (U) IFN-al (O), IFN-a? (A).

12

10

a

6

m 4

: x 2 ; C

.- 8 0 L

E

E ._

0.03 0.1 0.3 1.0 +/ '

0.5 2.5 10

Concentration Concentration hydrocortisone (Mg/ml) corticosteroid (pg/ml)

1.5 6 25 100 0 0.03 0.1 0.3 1.0

Concentration PHA (pg/ml) Concentration rn hydrocortisone (pg/ml)

Figure 6. (A) Effect of pre-treating monocytes with hydrocortisone on subsequent IL 1 production in response to C. albicans. Monocyte cultures were treated with hydrocortisone for 24 h then washed 3 times. C. albicans was then added at 0.1%. (B) Direct addition of corticosteroids to the thymocyte assay for monocyte IL 1. Hydrocor- tisone (0), dexamethasone (0). (C) Dose-dependency of thymocyte response to PHA. (D) Effect of hydrocortisone on the thymocyte response to an optimal dose of PHA (6 pg1ml).

IL 1 in the murine thymocyte assay. Corticosteroids dramatically inhibited the detection of IL 1 in monocyte supernatants stimulated with C. albicans antigen (Fig. 5C). When such supernatants were dialyzed to remove corticosteroid, however, inhibition was abolished, showing that IL 1 production was not affected by the presence of corticosteroids. When IFN-y, IFN-al, or IFN-a2 was added together with C. albicans antigen at 0.1% no additive or synergistic effect on IL 1 production occurred (Fig. 5D). Instead, IL 1 production was the same level as with antigen alone.

Pretreatment of monocytes with hydrocortisone for 24 h followed by washing only slightly reduced the subsequent detection of IL1 produced in response to C. albicans (Fig. 6A). Although corticosteroids did not affect IL 1 production, they were found to be potent antagonists of IL1 action. Steroids dramatically inhibited the thymocyte response to human IL 1 (Fig. 6B). This inhibition of thymocyte proliferation by steroids was not specific for IL 1. The data in Fig. 6C show the dose-dependence of the thymocyte response to PHA. When an optimal dose of PHA was used to stimulate thymocytes the proliferative response was inhibited by corticosteroids in a dose-dependent manner (Fig. 6D). Since monocyte cultures were thoroughly washed after antigen pulsing, it is unlikely that the inhibitory effect of corticosteroid on antigen presenta-

374 J. Rhodes. J. Ivanyi and P. Cozens Eur. J. Immunol. 1986.16: 370-375

T Figure 7. Lymphocyte prolifera-

T : 6 7

tion in response to autologous monocytes pulsed with C. albicans: failure to reverse the effects of hydrocortisone with exogenous IL1, and failure to reproduce the effects of hydrocortisone with nonsteroidal inhibitors of eicosanoid production. (1) No addition; (2) C. albi-

hydrocortisone 10 pgiml; (4) C. albicans plus hydrocortisone plus IL1 at 20 unitsiml; ( 5 ) C. albicans plus BW755C at 10 pg/ml; (6) C. albicans plus PW755C at 50 pg/ml; (7) C. albicans plus in- domethacin 5 pgiml.

cans 0.1%; (3) c. albicans plus

tion is due to residual steroid inhibiting IL 1 action. The observation that pretreatment of monocytes with corticosteroid did not markedly affect the production and action of IL 1 supports this view, as does the observation that corticosteroid inhibition of antigen presentation could not be reversed by the addition of exogenous IL 1 (Fig. 7). In addition, the inhibitory effects of corticosteroids on antigen presentation were found not to be due to inhibition of monocyte arachidonic acid metabolism [8, 111 since nonsteroidal inhibitors of arachidonic acid metabolism failed to reproduce the suppressive effects of steroids (Fig. 7).

4 Discussion

Although it is clear that IFN and other mediators can alter the expression of class I1 MHC molecules at the surface of APC it is by no means clear what effects these changes have on antigen presentation. In the present report we investigate the ability of IFN to increase the expression of monocyte class I1 MHC antigens and we examine their effects on the subsequent antigen-presenting capacity of these cells. It is generally agreed that IFN-y enhances the expression of class I1 MHC molecules [12-141 and the present report confirms this view. A number of studies on IFN-a, however, have concluded that this IFN does not affect class I1 MHC [12, 131. The present report clarifies this point by investigating two recombinant subspecies of IFN-a. We show here that IFN-a2 (IFN-aA) does not increase the expression of monocyte class I1 antigens although it does increase class I antigen expression. Con- versely, IFN-al (IFN-aD) increases monocyte class I1 expression but does not affect class I. Previous negative findings with IFN-a have dependent on the use of IFN-a2 [12, 131. We have previously shown that lymphoblastoid IFN-a (Wellcome Biotech, Beckenham, GB), which contains at lease ten sub- types, increases the expression of human monocyte class I1 antigens in vivo [15]. Others have used lymphoblastoid and melanoma cell lines to show that all three types of IFN (a, fl and y) increase mRNA for class I and class I1 MHC molecules although this increase is not necessarily reflected in increased surface expression of the products [16].

The effects of corticosteroids on MHC antigen expression have similarly presented a complex picture. In murine systems corticosteroids have been shown to diminish the expression of class I1 MHC antigens [17]. Steroids can also reduce the

expression of class I MHC antigens by human lymphocytes [18]. We show here, however, that hydrocortisone increases the expression of class I and class I1 MHC antigens on human monocytes. This is in agreement with the findings of Gerrard et al. who used a similar system [19]. Clearly there are significant differences between murine peritoneal macrophages and human peripheral blood monocytes as evidenced by the fact that the latter constitutively express class I1 MHC whereas most of the former do not [20].

In the present study we find that no simple correlation exists between MHC class I1 expression and the capacity of human monocytes to present C. albicans antigen even though the presentation of this antigen has been shown to be class I1 region restricted [21, 221. A similar dissociation between antigen presentation and class I1 expression has been found for keyhole limpet hemocyanin [19]. In the present study two signals which increased class I1 MHC (IFN-al and hydrocortisone) nevertheless inhibited or suppressed the subsequent response of lymphocytes to antigen-pulsed monocytes. A third signal (IFN-a*) which increased class I but not class I1 expression also inhibited antigen presentation. Only IFN-y both increased class I1 expression and enhanced the subsequent lymphoproliferative response to antigen-pulsed, IFN-treated monocytes. The suppressive effects of IFN-a and hydrocortisone appear to be at the level of antigen processing and/or associa- tion with class I1 antigens because the polyclonal lymphocyte response induced by monocytes pulsed in the same way with PWM remained unaffected. In some circumstances an increase in the ratio between class I1 MHC determinants and antigenic determinants may inhibit the specific response of T lymphocytes and we are currently investigating this possibility.

The second obligatory signal in the induction of immune responses is IL 1 produced by the APC. We show here that C. albicans antigen alone is a sufficient signal for monocyte IL 1 production. This is in agreement with the findings of Lombardi and colleagues who examined the response to a polysaccharide extract of C. albicans [23]. We also found that rIFN-y and rIFN-al induced IL 1 production whereas rIFN-a2 failed to induce IL 1 production. Thus, IFN-al differs from IFN-a2 in its effects on IL 1 production as well as on MHC antigen expression. Others found that leukocyte IFN-6 and fibroblast IFN did not induce IL1 production in monocytes [24]. When monocytes were pulsed with C. albicans antigen together with IFN-al, IFN-a2 or IFN-y, IL 1 production was comparable to that obtained with antigen alone. It therefore seems that the regulatory effects of IFN on the presentation of this antigen are not due to effects of IL 1 production.

The same conclusion can be drawn for corticosteroids. Steroids did not induce IL1 production, neither did they inhibit the production of IL1 in response to C. albicans antigen. However, corticosteroids were found to be potent antagonists of IL 1 at the level of thymocyte proliferation. This inhibitory effect was not specific for IL 1 responses since PHA- induced thymocyte proliferation was inhibited to a comparable extent. Because monocyte cultures were washed extensively after antigen pulsing and treatment with IFN or corticosteroid, it is unlikely that residual steroid could affect IL 1 action in the lymphocyte proliferative response to antigen. The observation that pre-treatment of monocytes by steroid followed by washing did not markedly affect the subsequent detection of monocyte IL1 supports this view, as does the observation that exogenous IL 1 did not reverse the inhibitory effects of steroid on antigen presentation.

Eur. J. Irnmunol. 1986.16: 375-380 Suppressed IFN-y production in carrageenan-treated mice 375

In summary the present study shows that no simple correlation exists between class I1 MHC antigen expression and antigen- presenting capacity, and that the differences observed are not due to effects on IL 1. It might be argued that this would not be found in a situation where class I1 antigen expression is constitutively very low as in the case of murine peritoneal macrophages and murine cell lines. However, recent work on the antigen-presenting capacity of the murine line P388Dl has shown that signals which down-regulate Ia expression have little effect on antigen presentation by the same cells [25]. It seems that T lymphocytes may require only low levels of class I1 MHC antigen expression on APC.

We are grateful to Dr. T. Prospero, Department of Pathology, Uitiver- sity of Cambridge, who performed the flow cytofluorometric analysis. We thank Barbara Pearce for technical assistance.

Received May 9, 1985; in revised form November 5, 1985.

5 References

1 Matis, L. A., Glimcher, L. H., Paul, W. E. and Schwartz, R. H.,

2 Mizel, S. B., Immunol. Rev. 1982. 63: 51. 3 Barnstable, C. J., Bodmer, W. F., Brown, G., GalfrC, G., Mil-

stein, C., Williams, A. F. and Ziegler, A., Cell 1978. 14: 9. 4 Brickell, P. M., McConnell, I., Milstein, C. and Wright, B.,

Immunology 1981. 43: 493. 5 Brickell, P. M., Richardson, N. E., McConnell, I. and Feinstein,

A., Clin. Exp. Immunol. 1983. 54: 117. 6 Coombs, R. R. A , , Wilson, A. B., Eremin, D., Gurner, B. W.,

Haegert, D. , Lawson, Y., Bright. S. and Munro, A., J . Immunol. Methods 1977. 18: 45.

Proc. Natl. Acad. Sci. U S A 1983. 80: 6019.

Fujio Suzuki", Hiroshi Maeda' and Richard B. Pollard'

Department of Microbiology, Kumamoto University Medical School, Kumamoto' and Division of Infectious Diseases, Department of Internal Medicine and Department of Microbiology, University of Texas Medical Branch and Virology Division, Shriners Burns Institute, Galveston'

7 Zamoyska, R., Waldmann, H. , Prospero, T. and Lennox, E.,

8 Salmon, J., Prostaglandins 1978. 15: 383. 9 Slocombe, P., Easton, A,, Boseley, P. and Burke, D. C., Proc.

Natl. Acad. Sci. USA 1982. 79: 5455. 10 Goeddel, D. V., Leung, D. W., Dull, T. J., Gross, M., Lawn, R.

M., McCandliss, R., Seeburg, P. A , , Ullrich, A., Yelverton, E. and Gray, P. W., Nature 1981. 290: 20.

11 Rhodes, J., Salmon, J. and Wood, J., Eur. J. Immunol. 1985. 15: 222.

12 Basham, T. Y. and Merigan, T. C., J . Im unol. 1983. 130: 1492.

Mol. Biol. Med. 1983. I : 137.

13 Kelley, V. E., Fiers, W. and Strom, T. B., 7 J . Immunol. 1984.132: 240.

14 Wong, G. W. H., Clark-Lewis, I., McKimrn-Breschkin, J. L., Harris, A. W. and Schrader, J. W., J . Immunol. 1983. 131: 788.

15 Rhodes, J., Jones, D. H. and Bleehen, N. M., Clin. Exp. Immunol. 1983. 53: 739.

16 Rosa, F., Hatat, D., Abadie, A., Wallach, D., Revel, M. and Fellous, M., EMBO J. 1983. 2: 1583.

17 Snyder, D. S. and Unanue, E. R., J . Immunol. 1982.129: 1803. 18 Hokland, M., Larsen, B., Heron, I . and Plesner, T., Clin. Exp.

19 Gerrard, T. L., Cupps, T. R., Jurgensen, C. H. and Fauci, A. S . ,

20 Beller, D. I., Kiely, J.-M. and Unanue, E. R., J. Immunol. 1980.

21 Raff, H., Picker, L. J. and Stobo, J. D., J. Exp. Med. 1980. 152:

22 Gonwa, T. A., Picker, L. J., Raff, H. V., Goyert, S. M., Silver, J.

23 Lombardi, G., Piccolella, E., Vismara, D., Colizzi, V. and Asher-

24 Arenzana-Siesdos, F. and Vierelizier, J. L., Eur. J . Immunol.

25 Zlotinik, A,, Shirnonkervitz, R., Kappler, J. and Marrack, P.,

Immunol. 1981. 44: 239.

Cell. Immunol. 1984. 85: 330.

124: 1426.

581.

and Stobo, J. D., J. Immunol. 1983. 130: 706.

son, G. L., Clin. Exp. Immunol. 1984. 58: 581.

1983. 13: 437.

Cell. Immunol. 1985. 90: 154.

Suppression of interferon gamma production in mice treated with carrageenan Effects of carrageenan (CAR) treatment on the response of interferon (IFN) production in vivo and in vitro after stimulation with an IFN-y inducer, staphylococcal enterotoxin A (SEA), was investigated. The IFN-y production in mice stimulated with SEA was impaired after i.v. administration of a 20 mglkg dose of CAR. Spleen cells (SC) from CAR-treated mice had decreased ability to produce IFN in vitro after stimulation with the same inducer. SC obtained from mice during the suppressive state inhibited IFN-y production when they were co-cultured with mononuclear cells prepared from spleens of untreated control mice. This suppressor cell activity could be removed from SC by an adherence technique to plastic surface. The SC with suppressor activity were not inactivated by treatments with monoclonal anti-Thy-1.2 antibody, anti-asialo GM1 antisera and anti-mouse immunoglobulin antisera followed by complement. The suppressive activity was detected in cell-free culture fluids of macrophage fractions containing suppressor cell activity. These results suggest that the decrease in IF"-y production in mice pretreated with CAR may associate with the presence of suppressor cells characterized to the monocytelmacrophage lineage.

[I 51171 1 Introduction ~

Correspondence: Fujio Suzuki, Department of Microbiology, Kumamoto University Medical School, Kumamoto 860, Japan

Abbreviations: IFN-y: Interferon gamma SEA: Staphylococcal enterotoxin A CAR: Carrageenan McP: Macrophages NK: Natural killer MNC: Mononuclear cells C: Complement PAC: Plastic-adherent cells CAR-mice: Mice treated with CAR N-mice: Normal control mice PACF Cell-free fluids from cultures of PAC SC: Spleen cells mAb: Monoclonal antibody(ies)

Carrageenans (CAR), high molecular weight substances extracted from seaweed [ 1, 21, interfere with humoral and cell- mediated immune responses in mice [3] and rats [4] through the modifications of macrophage (MQ) functions. It was reported [5] that a single injection of relatively small doses of CAR abrogated or weakened natural resistance of irradiated mice to foreign bone marrow transplants. I n vivo administra-

0 VCH Verlagsgesellschaft mbH, D-6940 Weinheim, 1986 0014-298Ol8610404-0375$02.50/0

antigen presentation by human monocytes: effects of modifying major histocompatibility complex class...

Documents