The role of suppressor factors in the regulation of immune responses by ultraviolet radiation-induced suppressor T lymphocytes

CELLULAR

IMMUNOLOGY

126,255-267

(1990)

The Role of Suppressor Factors in the Regulation of Immune Responses by Ultraviolet Radiation-Induced Suppressor T Lymphocytes 111.Isolation of a Suppressor Factor with the B16G Monoclonal Antibody’ GENE

K. Y EE,~“.* JULIAG.LEVY,~ MARGARETL. KRIPKE,* ANDSTEPHENE. ULLRICH*

*Department ofImmunology, The University of Texas, M. D. Anderson Cancer Center, Houston, TX: and the tDepurtment yfMicrobiology. The University of British Columbia. Vancouver, British Columbia, Canadu Received July 27, 1989; accepted October 7, 1989 The purpose of this study was to determine whether the ultraviolet (UV) radiation-induced systemic suppression of the immune response results from the release of soluble suppressor factors (TsF) by W-induced suppressor T cells (UV Ts). Injecting a TsF-specific monoclonal antibody (B16G) significantly reduced the UV radiation-induced suppression of contact hypersensitivity (CHS). The transfer of spleen cells from the UV-irradiated, Bl6G-treated mice into normal recipients suppressed CHS in the recipients, indicating that while the suppression of CHS was reversed in the UV-irradiated, B16G-treated mice, suppressor cells were still present. Supernatants from cultures containing UV Ts were incubated on BI6G-immunoadsorbent columns. The antibody-bound fraction (45 to 60-kDa, non-disulfide-linked proteins) suppressed CHS when injected into normal recipients. These results demonstrate that the B t6G antibody reacts with TsF from UV Ts and suggestthat B I6G acts in vivo by inhibiting the activity of TsF. Thus, suppressor factors appear to play an essential role in the regulation of immune responses by UV Ts. 0 1990 Academic press. lnc.

INTRODUCTION The generation of suppressor T lymphocytes (Ts)~ has been reported by many investigators using very different experimental systems (1). A common feature of Ts ’ This work was supported by Grant RR55 It-22 from the National Institutes of Health, Grant 83-191 from the Sid Richardson Foundation, and funds from the M. D. Anderson Annual Campaign, a project of The University of Texas Cancer Board of Directors. * A Rosalie B. Hite Predoctoral Fellow of The llniversity of Texas M. D. Anderson Cancer Center. 3 Present address: Department of Pathology, Boston University School of Medicine, 80 East Concord Street, Boston, MA 02 118-2394. 4 Abbreviations used: CHS, contact hypersensitivity; CTL, cytotoxic T lymphocytes; DNFB, 2,4-dinitrofluorobenzene; HSV-2, herpes simplex virus type-2; DNP-NSC, dinitrophenyl-modified normal spleen cells: NR, nonirradiated; SDS-PAGE, sodium dodecylsulfate polyacrylamide gel electrophoresis; TCSA/ UVA radiation, tetrachlorosalicylanilide plus UVA radiation; TNBS, 2,4,6-trinitrobenzenesulfonic acid; TNCB, 2,4,6-trinitrochlorobenzene; Ts, suppressorT cells; TsF, T lymphocyte-derived suppressorfactors; UV Ts, ultraviolet radiation-induced suppressorT lymphocytes. 255 000%8749/90 $3.00 Copyngbt 0 1990 by Academic Press, Inc. All rights of reproduction in any farm reserved.

256

YEE ET AL.

induced by these protocols is the sequential interaction between Ts and suppressor ceJJfactors (TsF), resulting in the induction of a suppressor cell cascade.The mode of action and specificity of the TsF generated at various steps of the cascadereflect those of the Ts associatedwith the respective factors. The suppression ofthe immune response is believed to result from the interaction between the ultimate TsF and the effector cell (2). Exposing mice to ultraviolet (UV) radiation has been shown to suppressthe induction ofcontact hypersensitivity (CHS), in part through the induction ofantigen-specinc Ts. Jt appears that at least two different mechanisms are involved. The first has been called local or low-dose suppression of immunity, and the second is referred to as systemic suppression. Toews et al. (3) demonstrated that when haptens were applied directly to skin previously exposed to relatively low doses of UV radiation ( 1.6 kJ/m2), the induction ofCHS was suppressed.The mechanism appears to be a modulation of Langerhans cell function at the site of sensitization, so the hapten cannot be presented to the immune system in a way that activates a response In addition, antigen-specific suppressor T cells were found in the spleensof the UV-irradiated animals (4). A recent report by Cruz et al. (5) demonstrated that injecting mice with Langerhans cells that were irradiated in vitro with UV radiation resulted in the induction of splenic Ts. Thus, it appears that exposure to UV radiation is modulating the function of Langerhans cells in a way that results in the activation of the suppressor arm of the immune response. Systemic suppression of CHS occurs when mice are first exposed to large (30 to 40 kJ/m2) dosesof UV radiation and then sensitized with hapten at a distant nonirradiated site (6). In this casealso, antigen-specific Ts are found in the spleens of the UVirradiated animals (7). Although the function of Langerhans cells at the site of sensitization is normal (8) the ability of spleen cells from these mice to act as antigenpresenting cells is depressed(6). Recent studies have suggestedthat soluble mediators, released from UV-irradiated epidermal cells, may be involved in the induction of systemic suppression (9, 10). The involvement of TsF in UV radiation-induced local suppression of immune responseshas recently been demonstrated. Aurelian and co-workers ( 11, 12) cultured spleen cells from UV-irradiated herpes simplex virus (HSV-2)-inoculated mice with viral antigens and found that supernatants from these cultures suppressed the immune responseagainst HSV-2. Both antigen-specific and nonspecific TsF were found in the culture supernatants. Similarly, Tokura and collaborators (13) reported that lysatesobtained by freezing and thawing spleen cells from UV-irradiated mice treated with tetrachlorosalicylanilide plus UVA radiation (TCSA/UVA radiation) contained antigen-specific TsF. Using the model of UV-induced systemic suppression of CHS (14, 15), we reported recently that supernatants obtained from the cultures of UV radiation-induced Ts (UV Ts), normal responder lymphocytes, and hapten-modified stimulator cells suppressedCHS and the in vitro generation of cytotoxic T lymphocytes (CTL). The supernatants suppressed the induction but not the elicitation of CHS, and did so in a hapten-specific manner. To investigate further the role of TsF in the UV-induced systemic suppression of CHS, the TsF-specific B16G monoclonal antibody was employed. This antibody was originally raised against tumor-specific TsF releasedby Ts isolated from animals injected with the P8 J5 murine mastocytoma. The Ts and their associated TsF sup-

ISOLATION OF A UV-INDUCED

SUPPRESSOR FACTOR

257

pressedtumor rejection in vivo and the generation of P8 15-specific CTL in vitro. The B 16G antibody recognizes 40- to 50-kDa and 80- to 90-kDa determinants on a number of murine ( 16-23) and human (22) TsF. B 16G has been reported to react against a monoclonal first-order suppressor/inducer TsF (TsF,) but not with second- or thirdorder TsF (TsF2 and TsF3) (23). These findings suggestthat B 16G recognizes a common determinant present on a number of TsF, . In the studies reported here, the B 16G antibody was injected iv into UV-irradiated mice to determine its effect on the UV-induced systemic suppression of CHS. To determine the target of B16G, spleen cells from UV-irradiated, B16G-treated mice were transferred to normal recipients to determine whether the antibody depleted either the UV Ts themselves or the associated TsF. Finally, the B16G monoclonal antibody was used to isolate and partially characterize the TsF from UV Ts culture supernatants that possessedsuppressive activity. MATERIALS AND METHODS Mice. Specific-pathogen-free female C3H/HeN Cr (MTV-) mice were supplied by the Animal Production Area of the Frederick Cancer Research Facility (Frederick, MD). The mice were cared for according to the guidelines set forth in the Guide for the Care and Use of Laboratory Animals (D.H.H.S. Publication No [NIH] 78-23) in an AAALAC-accredited facility, and all protocols were approved by the Institutional Animal Care and Use Committee. The animals were between 10 and 12 weeks old at the beginning of each experiment. Induction and elicitation of CHS. Either 50 ~1of a 0.3% solution (v/v) of 2,4-dinitrofluorobenzene (DNFB; Sigma, St. Louis, MO) in acetone or 100 ~1of a 3% solution (w/v) of 2,4,6-trinitrochlorobenzene (TNCB, Pfaltz and Bauer, Inc., Waterbury, CT) in acetone was used as the sensitizing antigen. The contact allergen was applied to the shaved abdominal skin of the mice. Six days later, the ear thickness was measured with an engineer’s micrometer (Swiss Precision Instruments, Los Angeles, CA), and the mice were challenged by painting each ear surface with 5 ~1 of either a 0.2% solution of DNFB or a 1% solution of TNCB. One day later, the ear thickness was remeasured; the specific ear swelling was determined by subtracting the swelling produced by painting the hapten on the ears of unsensitized mice from the swelling produced in sensitized animals. Generation of suppressor T lymphocytes. The shaved dorsal skin of mice was exposed to a single 3-hr dose (40 W/m’) of UVB (280-320 nm) radiation from a bank of 6 Westinghouse FS-40 sunlamps, as described previously (24). During the irradiation, the ears of the mice were shielded with electrical tape. Five days after irradiation, the mice were sensitized by epicutaneous application of a contact allergen onto their shaved, unirradiated abdominal skin, and they were challenged, as described above. Control mice were similarly sensitized and challenged, but they were not irradiated (NR). The spleensof mice that exhibited a suppressedCHS response (UV-irradiated, hapten-sensitized mice) were removed, single-cell suspensionswere prepared, and the red blood cells were subjected to osmotic lysis in water. To enrich for T lymphocytes, the cells were incubated on nylon wool columns (25). Spleen cells from the NR mice were treated similarly and used as a source of control cells. The suppression of CHS was determined by comparing the responsesof UV-irradiated, hapten-sensitized mice with that of unirradiated, hapten-sensitized mice according to the following formula:

258 Percentageof suppression = 1 -

YEE ET AL.

Specific ear swelling of UV-irradiated mice X 100 Specific ear swelling of nonirradiated mice .

Production of suppressorfactors in UV Ts culture supernatants. Supernatants from cultures containing dinitrophenyl (DNP)-specific UV Ts, normal responder lymphocytes, and DNP-modified syngeneic stimulator cells were used as a source of TsF ( 14). The UV Ts (1 X 10’ cells) were added to cultures containing 5 X lo6 responder and 5 X lo6 stimulator cells. The cultures were incubated at 37°C for 5 days, and the supematants were harvested from the cultures by centrifugation at 25Og for 8 min. The supernatants were then spun for 12 min at 4OOg,passedthrough a 0.45~pm filter (Millipore, Bedford, MA), and used immediately as a source of TsF. Cultures that contained responder and stimulator cells plus NR control cells were treated similarly and used as the source of control supernatants. Monocional BZ6G antibody. The TsF-specific B 16G monoclonal antibody ( 16) was tested for its ability to inhibit UV-induced suppression of CHS. The affinity-purified antibody ( 100 pg in 0.125 ml) was injected iv into UV-irradiated mice 2 days before sensitization with the hapten. The mice received an additional 100 pg 2 days following sensitization, and they were challenged 4 days after the last injection. The ear swelling response was determined as described above. As a control, similar amounts of an irrelevant isotype-matched monoclonal antibody (IgG2a) were injected. Zmmunoadsorbents. The B16G and control antibodies were linked to cyanogen bromide-activated Sepharose CL-4B (Pharmacia, Piscataway, NJ) according to the manufacturer’s instructions. Coupling of antibody was carried out at a concentration of 5 mg antibody per ml of reconstituted Sepharose beads. The immunoadsorbant columns were stored at 4°C in PBS plus 0.02% sodium azide and washed with PBS prior to use. Isolation of TsF. Seven milliliters of supernatant derived from the cultures of DNPspecific UV Ts, normal responder cells, and DNP-modified stimulator cells, was loaded on lo-ml Econo-Chromatography columns (Pharmacia) containing 1 ml of B 16G-Sepharose,according to the procedure described by Steele et al. (23). This volume of supernatant contained 9 units of suppressive activity, with 1 unit defined as the amount that causesa 50% inhibition of CHS. The supernatant was cycled through the column three times, and the columns were incubated at 4°C for 30 min. The effluent (approximately 6 ml) was collected, PBS was added to the columns, and the unbound material was collected at a constant flow rate of approximately 0.2 ml/min. When the absorbance of the fractions reached the baseline value, the bound material was eluted with 1 M glycine-HCl (pH 2.0). One-milliliter fractions were collected, the absorbanceswere measured at 280 nm (Beckman Model DU-65 spectrophotometer), and the fractions were pooled and dialyzed against PBS (pH 7.2). For biochemical analyses, the fractions were dialyzed first in PBS, followed by dialysis against distilled water to remove the PBS. Alternatively, the protein was acetone-precipitated and resuspended in the SDS sample buffer. The fractions from the control antibody columns were treated identically. In addition, control supematants derived from cultures with NR cells in lieu of UV Ts were subjected to the identical isolation procedure. Assay of TsF biological activity. The PBS-dialyzed fractions were injected iv into normal recipients immediately before sensitization with 0.3% DNFB. The mice were challenged 6 days later with 0.2% DNFB, and ear swelling was measured as described

ISOLATION OF A UV-INDUCED Group Treatment ----

Antibody

Hapten

1

NW?

2

NW

NCW

DNFB

3

w

NC#2

DNFB

4

w

816G

DNFB

5

w

COfltrOl

DNFB

6

NR

NOW!

DNFB

7

NR

B16G

DNFB

NR

COlltrOl

DNFE

6

I 0

259

SUPPRESSOR FACTOR

5

Ear Swelhg

I

I

I

10

15

20

+ SD (cm x 16 3,

FIG. I. Effect of B16G on UV radiation-induced suppression of CHS to DNFB. Mice were exposed to UV radiation on Day 0. Five days after exposure the animals were sensitized with DNFB. On Days 3 and 7 (2 days before and 2 days after sensitization), 100 yg of B16G monoclonal antibody or 100 rg of an isotype-matched control monoclonal antibody was injected into each mouse. Six days after sensitization (Day 1I), the mice were challenged with DNFB, and CHS was assessedon the next day. Control mice (NR) were treated similarly but not irradiated.

above. The unfractionated culture supernatants and the effluents from the immunoadsorbant columns were injected in volumes of 0.75 ml, with protein concentrations ranging from 1 X lo-’ to 4 X 1O-2mg. S&S-PAGE. The bound material eluted from the B16G columns (100 pg in 1.O ml) was subjected to analysis with SDS-PAGE. Forty microliters of each sample was electrophoresed on a Bio-Rad Mini-Protean two-gel apparatus (Bio-Rad, Richmond, CA) at 4°C for 45 min using a 10% SDS gel. A constant current of 22 mA and a voltage of 150 V (with a final limiting voltage of 300 V) was maintained. Since 100 pg of TsF was generated from 35 X lo6 UV Ts, we estimate that the amount of TsF loaded onto each well (4 pg) is equivalent to the amount of TsF generated by 1.4 X lo6 UV Ts. For analysis in reducing conditions, the samples were suspended in a sample buffer containing dithiothreitol at a final concentration of 10 n&? In some experiments, N-ethylmaleimide was added following reduction in a lo- to 20-fold excess to prevent oxidation of reduced components. The gels were fixed, silverstained with the Bio-Rad silver staining kit, and preserved for analysis. A determination of the molecular weight was accomplished by comparing the relative mobility of the B16G-bound material to that of the low molecular weight standards. These included soybean trypsin inhibitor (21.5 kDa), carbonic anhydrase (31 kDa), ovalbumin (45 kDa), bovine serum albumin (66.2 kDa), and phosphorylase (92.5 kDa). Statistical analyses. The two-tailed Student t test was used to assessthe significance of differences between ear swelling responses.Each experiment was repeated at least

260

YEE ET AL. Group Treatment ----

Antibody

Hapten

1

NOW

NW?

2

NWP

None

3

W

None

TNCB

6

NFI

NOW

TNCB

7

Nu

B16G

TNCB

6

wi

C0ntr0l

t---j--

TNCB

0

I

I

5

10

Ear Swelling

f SD (cm x 10

-3

I

I

15

20

)

FIG. 2. Effect of B 16G on UV radiation-inducedsuppressionof CHS to TNCB. Mice were exposed to UV radiation on Day 0. Five daysafter exposurethe animalsweresensitizedwith TNCB. On Days3 and

7 (2 daysbefore,and 2 daysafter sensitization),100Ggof B16G monoclonalantibody or 100pg of an isotype-matchedcontrol monoclonalantibody wasinjectedinto eachmouse.Six daysafter sensitization (Day I 1).the mice were challengedwith TNCB, and CHSwasassessed on the nextday.Control mice(NR) weretreatedsimilarly but not irradiated.

twice; representative examples are shown. Five to 10 mice per group were used in the experiments. RESULTS

Efect qfthe B16G monoclonal antibody on the UV-induced suppressionof CHS. The B16G monoclonal antibody was injected into UV-irradiated animals to determine its effect on the suppression of CHS. Three days after exposure to UV radiation and 2 days before sensitization with DNFB, mice were injected iv with 100 fig of antibody. An additional 100 pg was injected 2 days after sensitization. As shown in Fig. 1, mice exposed to UV radiation and sensitized with DNFB (group 3) exhibited a decreased CHS response compared with the positive control (group 2). When B16G was administered to UV-irradiated, hapten-sensitized mice (group 4), the suppression of CHS was significantly reduced (group 3 versus group 4, P < 0.001). Suppression was not reduced in animals exposed to UV radiation and injected with the isotypematched control antibody (group 5). Injecting either B16G or the control antibody into unirradiated mice before and after sensitization with DNFB had no effect (groups 7 and 8). These results indicated that administration of the TsF-specific B 16G monoclonal antibody into UV-irradiated animals reduced the UV radiation-induced systemic suppression of CHS. Similar results were obtained when TNCB was used as the sensitizing hapten (Fig. 2). Partial restoration of the CHS response was observed in UV-irradiated mice injected with B 16G (group 4) but not with the control antibody (group 5). As observed

ISOLATION OF A UV-INDUCED Group -

Cells

transferred

261

SUPPRESSOR FACTOR

Hapten

1

NW

NOCK?

2

NCW

DNFE

3

UV + DNFB

DNFB

4

UV + B16G + DNFB

DNFB

5

UV + Control Ab + DNFB

DNFB

6

NR + DNFB

DNFB

7

NR + B16G + DNFB

DNFB

6

NR + Control Ab + DNFB

DNFE

0

I

I

I

I

I

5

10

15

20

25

Ear Swelhng

+ SD (cm x 16 3,

FIG. 3. Ability of spleen cells from B I6G-treated mice to suppressCHS. Spleen cells ( I X 10’) from UVirradiated, BI6G-treated, DNFB-sensitized mice (UV + B16G + DNFB) were transferred into normal recipients before contact sensitization with DNFB. The recipients were challenged 6 days later with DNFB, and CHS was measured 24 hr later. Controls included groups of mice injected with cells from nonirradiated animals (NR + DNFB); nonirradiated mice injected with B16G (NR + B16G + DNFB) or the isotypematched control antibody (NR + Control Ab + DNFB); and UV-irradiated mice injected with the isotypematched control antibody (UV + Control Ab + DNFB).

above, the injection of neither the B 16G nor the control antibody into nonirradiated animals significantly affected the CHS responseto TNCB (groups 7 and 8). The target of B16G monoclonal antibody. To determine whether B 16G treatment depleted UV Ts in vivo, spleen cells from UV-irradiated, antibody-treated donors were transferred iv to normal recipients just before sensitization with DNFB (Fig. 3). The CHS response of mice injected with spleen cells from UV-irradiated, haptensensitized donors (group 3); UV-irradiated, B 16G-treated, hapten-sensitized donors (group 4); or UV-irradiated, control antibody-treated, hapten-sensitized donors (group 5) was significantly suppressedcompared with the positive control (group 2, P < 0.00 1). In contrast, CHS in animals receiving control spleen cells from nonirradiated, hapten-sensitized donors (group 6); nonirradiated, B 16G-treated, hapten-sensitized donors (group 7); and nonirradiated, control antibody-treated hapten-sensitized donors (group 8) was not significantly different from that ofthe positive control group (group 2). These results indicated that although B16G treatment inhibited the suppression of CHS in the donor animals, it did not achieve this effect by depleting UV Ts. Isolation and in vivo assay of TsF. To test whether the inhibition of suppression after B 16G injection was directed against a TsF associated with the UV Ts, TsF was incubated on B 16G-immunoadsorbent columns, and the B 16G-bound material was eluted and tested for suppressive activity. Supernatants from cultures containing DNP-specific UV Ts, normal responder cells, and DNP-modified stimulator cells contain a TsF whose activity mimics that of the UV Ts ( 14). The supernatants were incubated on B 16G columns, and the effluent, unbound, and B 16G-bound fractions

YEE ET AL.

262

d

‘OA Ah20 Fraction

30

40

number

FIG. 4. Elution profile of supematants from immunoadsorbant columns. The B 16G or control monoclonal antibody was linked to Sepharose4B. Culture supematants (7 ml) were cycled through the column 3 times and then incubated for 30 mitt, and the effluents were collected. The unbound material was eluted with PBS (pH 7.2) in l-ml fractions until absorbance at 280 nm approached the baseline value. GlycineHCl buffer (pH 2.0) was added to elute the bound material (represented by the arrowheads).

were collected and assayedfor their ability to suppressCHS in vivo. The elution profiles for the UV Ts and NR control supernatants from both B 16G and control IgG columns are shown in Fig. 4. Only one major peak was observed after the antibodybound material was eluted with glycine-HCl. The elution profile of control supernatants derived from cultures containing NR control cells in lieu of UV Ts did not contain significant amounts of B 16G-bound material. In addition, neither of the supernatants incubated on the control IgG columns exhibited a major peak of absorbance following addition of the eluting buffer. To determine whether the antibody-bound material had suppressive activity, it was injected iv into normal recipients just prior to sensitization with DNFB (Table I). The unfractionated culture supernatants, effluent, and antibody-unbound fractions were tested similarly. Injecting the unfractionated UV TsF significantly suppressed CHS (group 3, P < 0.00 1) when compared to the positive control (Group 2). Neither the effluent (group 5) nor the B16G-unbound fraction (group 6) suppressed CHS. However, the B 16G-bound material suppressedCHS at all three concentrations injected (1 X lo-* mg, 2 X lo-* mg, and 4 X lo-* mg of protein; groups 7-9). In contrast, the corresponding fractions eluted from the control IgG column (groups 12- 14) did not contain any suppressive activity. Moreover, passing the TsF over the control column did not deplete the suppressive activity. These results indicated that a factor capable of suppressing CHS was recognized by the B 16G antibody. To determine the minimal amount of TsF required to suppress CHS, dilutions of the B16G-bound material were injected into mice (Table 2). The TsF significantly suppressedCHS when administered at dilutions containing as little as 1 X lop4 mg of protein (groups 3-5 vs group 2; P < 0.001). In contrast, suppression was not observed in animals injected with fractions from the control IgG column (groups 8-l 1). On the basis of a comparison between the specific activity of Bl6G-bound TsF (9 X 1O3U/mg) and unfractionated starting material ( 10 U/mg), the B 16G-bound TsF contained 900 times more activity than the starting material. Biochemical analyses of the TsF. To determine the molecular weight of the TsF and the presence of disulfide linkages, the B 1BG-bound TsF (as shown in Fig. 4) was subjected to SDS-PAGE under reducing and nonreducing conditions (Fig. 5). Under

ISOLATION OF A UV-INDUCED

263

SUPPRESSOR FACTOR

TABLE 1 Effect of B 16G-Bound TsF on CHS

croupa

Supernatanth None None UVTs NR UVTs UVTs UVTs

8

UVTs

9

UVTs

10 I1 12

UVTs UVTs UVTs

13

UVTs

14

UVTs

Column fraction’

DNFB

None None Unfractionated Unfractionated B 16G effluent B I6G unbound B 16G eluate (I X lo-* mg) B I6G eluate (2 X IOe2mg) B 16G eluate (4 X IOe2mg) IgG effluent IgG unbound IgG eluate (3 X 1O-3mg) IgG eluate (6 X IO-‘mg) IgG eluate ( 1 X 10e2mg)

None + + + + + +

i + + +

A

% suppressiorrj

19.3 k 2.3 18.5 f 2.0 8.2 + 1.2

0 13.0 4.5 14.2 13.7 14.5 3.4

0 65* 0 0 0 74*

9.5 f 1.0

4.1

64*

f 2.0

3.7

72*

9.0 k2.3 9.2 f 1.7 18.9 f 1.9

4.2 4.4 14.1

68* 0 0

18.8

+ 2.3

14.0

0

18.5

+ 2.5

13.7

0

Ear swelling f SD (cm X 10m3) 4.8 20.8 17.8 F 1.6 9.3 I? 1.0

19.0 -+ 2. I

8.5

’ Normal recipients were injected iv with various fractions and sensitized immediately with 50 pl of a 0.3% solution (v/v) of DNFB in acetone. They were challenged 6 days later with 5 11of a 0.2% solution of DNFB on each ear surface, and CHS was measured 1 day later. ’ The UV Ts supernatants are derived from cultures that contained UV Ts, normal responder cells, and mitomycin C-treated, DNP-modified syngeneic lymphocytes. The control supernatants are those with NR cells substituted for UV Ts. ’ The column effluent is that fraction remaining following three cycles of passingthe supernatant through the immunoadsorbent columns. The unbound fraction was eluted with PBS (pH 7.2) and the eluate is that fraction eluted with glycine-HCI buffer (pH 2.0). d The significance of differences between test groups and group 2 was determined by the Student t test. The asterisk indicates P i 0.00 1.

reducing conditions, the B16G-reactive TsF showed two strong bands migrating at 60 and 50 kDa, with a number of faint bands migrating at 40 to 45 kDa (lane 1). Under nonreducing conditions, the TsF yielded three strong bands migrating at 45, 50, and 60 kDa (lane 2). DISCUSSION In this study, we provide direct evidence that TsF releasedby UV Ts play a role in the systemic suppression of CHS. Injecting the TsF-specific monoclonal antibody B 16G inhibited the immunosuppression causedby exposing mice to UV radiation. In addition, the TsF-specific monoclonal antibody bound a substance with suppressive activity, and we were able to isolate and partially characterize that substance. In these studies, hapten-specific immunosuppression in vivo was achieved by irradiating mice with relatively high doses (30 to 40 kJ/m2) of UV radiation at one site

264

YEE ET AL. TABLE 2 Dose Responseof B 16G-Bound TsF in the Suppression of CHS

Group”

Fraction injected

Protein cont. b-u4

DNFB

Ear swelling f SD (cm X IO-‘)

A

% suppressionb

1 2 3 4 5 6 7 8 9 10 11 12 13 14

None None TsF TsF TsF TsF TsF TsF Control Control Control Control Control Control

None None 1 x lo-* 1 x 10-3 1 x 1o-4 1 x loms 1 x 10-h 1 x lo-’ I x 10-l I x 1om7 I x 1om4 1 x 10-S 1 x tom6 1 x 10~’

None + + + + + + + + + + + + +

2.9 +- 1.1 18.5 _+2.6 7.7 2 1.0 10.3 f 1.4 10.0 -I-2.2 14.0 k 2.2 17.5 + 1.2 19.0 + 1.8 18.9 k 2.7 17.7 + 3.2 20.0 A 2.3 18.4 +- 1.3 17.4 +- 2.6 17.6 + 2.8

15.6 4.8 7.1 7.1 I I.1 14.6 16.1 16.0 14.8 17.1 15.5 14.5 14.7

69** 51** 54** 29* 6 0 0 1 0 0 7 6

’ Normal recipients were injected with either the B 16G-bound, UV-induced TsF or control eluates from NR supernatants incubated on B16G columns. The animals were sensitized immediately with 50 ~1of a 0.3% solution (v/v) of DNFB in acetone. They were challenged 6 days later with 5 pl of a 0.2% solution of DNFB on each ear surface, and CHS was measured I day later. b Significance of difference as determined by Student’s t test when the test groups are compared with Group2.*=P~0.001;**=P~0.001.

and applying the hapten to distant, unit-radiated skin. A different model of suppression (low dose or local) involves inhibiting the immune response with low doses(1.3 to 3.4 kJ/m2) of UV radiation and applying the antigen (or hapten) directly to the irradiated skin. Aurelian and co-workers (11, 12) have studied the Ts and TsF involved in the low-dose model of suppression. Both antigen-specific and antigen nonspecific suppressive factors were found. The antigen-specific TsF is a 115-kDa heterodimer composed of disulfide-linked 70- and 52-kDa components (12). The lower molecular weight of our antigen-specific TsF and the absence of disulfide linkages suggestthat their factor and ours are not the same.Becausethe protocol for generating Ts to HSV-2 is different from that used to generate UV Ts in our studies, these differencesare not surprising. Tokura and colleagues (13) also examined the role of TsF in the UV radiationinduced local suppression of photoallergy to TCSA. Spleen cells from the UV-irradiated mice were lysed and the cell lysateswere analyzed by molecular sieve chromatography. Antigen-specific TsF were found in the 60- to SO-kDa and lOO- to 200-kDa fractions. These single chain TsF suppressedphotoallergy against TCSA/UVA radiation in vivo, and the activity of the TsF was not altered by reduction and alkylation. In this system, the dose ofUVB radiation used was markedly lower (0.96 kJ/m2) than that used in our studies and the TCSA was applied to the UV-irradiated site. Our TsF appears to be a monomer; thus, it may be similar to the TsF described by Tokura and co-workers ( 13). Since molecular sieve chromatography does not permit a precise identification of molecular weight, perhaps our TsF and their smaller molec-

ISOLATION OF A UV-INDUCED

SUPPRESSOR FACTOR

265

66.2

31.0

21.5 FIG. 5. SDS-PAGE of B I6G-bound TsF. Lane 1, B 1hG-bound TsF from UV Ts supernatants (reducing conditions): lane 2, B16Gbound TsF from UV Ts supernatants (nonreducing conditions). The columnbound materials were electrophoresed at 150 V, with a final limiting voltage of 300 V on a 10% SDSPAGE minigel. For the reducing conditions, the running buffer contained a final concentration of 10 mM dithiothreitol. Protein bands were visualized by silver staining.

ular weight TsF (60 to 80 kDa) are similar. On the other hand, our TsF (45 to 60 kDa) can be easily distinguished from the higher molecular weight (100 to 200 kDa) TsF described by Tokura and colleagues (13). The differences in size and structure (disulfide linked vs nondisulfide linked) of the three UV-induced TsF point out the diversity of TsF associatedwith UV Ts. Apparently, the different protocols for inducing systemic and local suppression by UV radiation result in the generation of several different TsF. Others have also shown that TsF are involved in the suppression of CHS. Using DNP, Claman and co-workers (26) reported a 35 to 60-kDa TsF in the culture supernatants of Ts, as determined by hapten-linked immunoadsorbent columns and molecular sieve chromatography. Asherson and colleagues (27) reported the presence of disulfide-linked, two-chain 90-kDa heterodimers composed of 50-kDa components in the culture supernatants of lymphoid cells from mice injected with the hapten TNBS and sensitized epicutaneously with TNCB. They also detected components with estimated molecular weights of approximately 25- and 35-kDa in the supernatants, and they postulated that these represented separate hapten-binding and I-J+bearing chains. In addition, using molecular sieve chromatography, Zembala and collaborators (28) reported nonspecific TsF with estimated molecular weights of 30to 50-kDa in a similar system using TNBS and TNCB. Ptak et al. (29) also identified in the supernatants of cultures containing lymph node and spleen cells from TNBS/ TNCB-sensitized mice. The molecules of interest were postulated to be 70-kDa pro-

266

YEE ET AL.

teins in the acidic range from pH 5.6 to 6.8. Preliminary studies indicate that our B 16G-bound TsF has an isoelectric point of approximately pH 5.1 (data not shown). Although precise comparisons of the TsF described above with our TsF are difficult becauseof the various techniques used to generate and analyze the respective factors, our TsF appears to differ in size from the factors described by others. Another major difference is the lack of disulfide-linked components in the biologically active, B 16Greactive material. It must be recalled that the procedure for inducing Ts in the DNP and TNP systems mentioned above involves the iv injection of soluble haptens, followed by epicutaneous sensitization. Although the haptens used are similar, exposure to large doses of UV radiation followed by epicutaneous sensitization with hapten clearly represents a different means of inducing Ts and may explain the fact that different TsF are produced. Because the B16G antibody was utilized to isolate the TsF and because it clearly possessesspecificity for TsF, (23), other TsF of differing molecular weight, structure, and specificity may be present in the culture supernatant that was the source of TsF for this study. However, the B 1GG-bound material isolated from UV Ts supernatants suppressedCHS and contained 900 times more suppressive activity compared with the starting material. Previous reports from groups using this antibody to isolate TsF from murine splenocytes and hybridomas have found that B 16G binds to an 80- to 90-kDa dimer from P8 15-immune spleen cells (17), an SO-to 90-kDa heterodimer composed of 43- and 45-kDa components with an associated 25-kDa component from the P8 15-specific A 10 hybridoma (18) a 70-kDa molecule from the P8 15-specific A29 hybridoma (20) and 80- and 35-kDa molecules from the ferredoxin-specific Fd 11 hybridoma (2 1). In addition, the specificity of B 16G for TsF is further demonstrated by its ability to recognize an 80- to 90-kDa heterodimer composed of 45and 50-kDa components from human tonsil cells (22). Our results using the B16G antibody are not consistent with those mentioned above, becausethe major components contained in the B 16G-reactive UV-induced TsF appear in native form as nondisulfide-linked 45-, 50-, and 60-kDa molecules, rather than similar disulfide-linked heterodimers. This difference may be due to the source of TsF used in the previous studies. For example, when spent media from the hybridoma cell cultures were used as the source of the TsF, 25-35 kDa proteins were isolated. If, however, cell lysates were used, the antibody-bound material was in the range of 40-50 kDa (Levy, unpublished observation). Thus, this finding suggeststhat differences in the size of the B16G-reactive material may be the result of degradation of the TsF in culture. Despite the apparent differences in structure and reactivity of our TsF and those reported in previous studies, the ability of the monoclonal antibody to react with our TsF and the other human and mouse TsF indicates a serological relationship between the various TsF. B16G reacted against TsF, , but not against either TsF, or TsF3 from hybridomas in the NP hapten system (23). Evidence from recent studies in our laboratory indicate that TsF isolated from UV Ts can induce suppressor cells when injected into normal mice, which is reminiscent of the ability of TsF, to induce Tsz (data not shown). Moreover, previous reports (14, 15) demonstrated a number of similarities between the activity of TsF releasedby UV Ts and those reported for TsF, (I). Thus, these data suggestthat the UV Ts may use a mechanism similar to that described by others (1) to inhibit the immune response. The known specificity of B16G for TsF and the ability of spleen cells from UVirradiated, hapten-sensitized, B 1GG-treated mice to suppress CHS when transferred

ISOLATION OF A UV-INDUCED

SUPPRESSOR FACTOR

267

to normal recipients suggeststhat B16G does not act by depleting suppressor cells. B 16G may act in viva by neutralizing the TsF generated by the UV Ts. This activity would explain why B 16G inhibits the suppression of CHS but fails to deplete the UV Ts, and it is consistent with the failure of this antibody to bind to Ts in indirect immunofluorescence assays( 17). The inability to remove UV Ts following incubation on B 16G-coated culture dishes (data not shown) is also consistent with this hypothesis. In summary, the ability of the TsF-specific monoclonal antibody to inhibit the suppression of CHS that results from exposing mice to UV radiation provides evidence for the role of TsF in the UV-induced systemic suppression of CHS. In agreement with previous reports, our data suggestthat the target of the B 16G is TsF and not Ts. We suggestthat the B16G monoclonal antibody inhibits the suppression of CHS by inactivating a 4% to 60-kDa TsF releasedby the UV Ts. ACKNOWLEDGMENTS We thank Drs. Stephen J. LeGrue, Zvi Keren, Keith A. Knisely, and Scott L. Rodkey for their advice and assistancein the biochemical studies.

REFERENCES I. 2. 3. 4.

Germain, R. N., and Benacerraf, B., Stand. J. Immunol. 13, 1, 198 1. Dorf, M. E., and Benacerraf, B., Annu. Rev. Immunol. 2, 121, 1984. Toews, G. B., Bergstresser,P. R., and Streilein, J. W., J. Immunol. 124,445, 1980. Elmets, C. A., Bergstresser,P. R., Tigelaar, R. E., Wood, P. J.. and Streilein, J. W., J. Exp. Med. 158, 781, 1983. 5. Cruz, P. D., Nixon-F&on, J., Tigelaar, R. E., and Bergstresser,P. R., J. Invest. Dermafol. 92, 160, 1989. 6. Noonan, F. P., Kripke, M. L., Pedersen,G. M., and Greene, M. I., Immunology43,527, 1980. 7. Noonan, F. P., DeFabo, E. C., and Kripke, M. L., Photochem. Photobiol. 34,683, 1981. 8. Noonan, F. P., Bucana, C., Sauder, D., and DeFabo, E. C., J. Immunol. 132,2408, 1984. 9. Ross,J. A., Howie, S. E. M., Norval, M., and Maingay, B. A., Photodermatology 5,9, 1988. 10. Noonan, F. P., Morrison, H.. and DeFabo, E. C., J. Invest. Dermatol. 90,92, 1988. 1I. Yasumoto, S., Hayashi, Y., and Aurelian, L., J. Immunol. 139,2788, 1987. 12. Aurelian, L., Yasumoto, S., and Smith, C. C., J. Viral. 62,2520, 1988. 13. Tokura, Y., Miyachi, Y., Tokigawa, M., and Yamada, M.. Cell. Immunol. 110,305, 1987. 14. Yee, G. K., Ulhich, S. E., and Kripke, M. L., Cell. Immunol. 121,74, 1989. 15. Yee, G. K., Ulhich, S. E., and Kripke, M. L., Cell. Immunol. 121,88, 1989. 16. Maier, T., Tenth Stammers. A., and Levy, J. G., J. Immunol. 131, 1843, 1983. 17. Steele, J. K., Tenth Stammers, A., Chart, A., Maier, T., and Levy, J. G., Cc/l. Immunol. 90,303, 1985. 18. Steele, J. K., Tenth Stammers, A., and Levy, J. G., J. Immunol. 134,2767, 1985. 19. Steele,J. K., Singhai, R., Tenth Stammers, A., and Levy, J. G., J. Immunol. 137,3025, 1986. 20. Steele, J. K., Tenth Stammers, A., Chan, A., and Levy, J. G., J. Immunol. 137,3550, 1986. 21. Steele, J. K., Chu, R. N., Chart, A., North, J., and Levy, J. G., J. Immunol. 139,469, 1987. 22. Steele,J. K., Tenth Stammers, A., and Levy, J. G., J. Immunol. 135, 1201, 1985. 23. Steele, J. K., Kawasaki, H., Kuchroo, V. K., Minami, M., Levy, J. G., Dorf, M. E., J. Immunol. 139, 2629.1987. 24. Ulhich, S. E., Yee, G. K., and Kripke, M. L., Immunology 58, 185, 1986. 25. Julius, M. H., Simpson, E., and Herzenberg, L. A., Eur. J. Immunol. 3,645, 1973. 26. Claman, H. N., Miller, S. D., Conlon, P. J., and Moorhead, J. W., Adv. Immunol. 30, 121, 1980. 27. Asherson, G. L., Watkins, M. C., M. A., Zembala, and Colizzi, V., Cell. Immunol. 86,448, 1984. 28. Zembala, M. A., Asherson, G. L., James, B. M., Stein, V. E., and Watkins, M. C., J. Immunol. 129,

1823, 1982. 29. Ptak, W., Gershon. R. K., Rothstein, R. W., Murray, J. H.. and Cone, R. E., J. Immunol. 131, 2859, 1983.