Endocrine control of the immunosuppressive activity of the submandibular gland

BRAIN,

BEHAVIOR,

AND

Endocrine

6, 418-428 (1992)

IMMUNITY

Control EVA NAGY,

Department of Immunology,

of the Immunosuppressive Submandibular Gland ISTVAN

BERCZI,

Activity

of the

AND EDRIS SABBADINI

University of Manitoba,

Winnipeg. Manitoba

R3E OW3, Canada

Extracts of the submandibular gland (SMG) of rats contain fractions that stimulate the in vitro proliferation of Con A-treated lymphocytes. One of the stimulatory fractions was also shown to induce in vivo immunosuppression in rats and mice in several experimental models. Since many other biologically active factors of the SMG had been found to be hormone dependent, we investigated the effects on the immunosuppressive factor of hypophysectomy (Hx) and of hormonal reconstitution in male Fischer rats. Hx induced a marked atrophy of the SMG together with an almost complete disappearance of both the in vitro lymphocyte-stimulating activity and the in vivo immunosuppressive activity, the latter assayed with the contact sensitivity reaction in mice. The treatment of the Hx rats with pituitary hormones demonstrated that prolactin (PRL), thyroid stimulating hormone (TSH), and luteinizing hormone (LH) induced a significant reconstitution of these biological activities, growth hormone led to the recovery of the lymphocytestimulating activity but not of the immunosuppressive activity, while follicle-stimulating hormone, and adrenocorticotropic hormone did not induce any recovery of these biological activities. In view of the positive results obtained with TSH and LH further experiments were done to compare the effects of thyroid and sex hormones with those of PRL. The results demonstrated that testosterone and thyroid hormones induced significant recovery of the lymphocyte-stimulating and the immunosuppressive activity. The combination of these two hormones with PRL produced the most effective results. On the other hand, estrogens and progesterone had no significant effects. These results confirm the effectiveness of androgens and thyroid hormones in stimulating the production of biologically active factors by the SMG. Moreover, they demonstrate that PRL, a hormone not previously considered to increase the activity of the SMG. stimulates the production of immunoregulatory factors in Hx animals. ID 1992 Academic Press, Inc.

INTRODUCTION The submandibular gland (SMG) of rodents contains factors capable of suppressing immune and inflammatory reactions (Kongshavn & Bliss, 1970; Kongshavn & Lapp. 1972; Koch & Rowe, 1976; Roberts, Freston, & Reade, 1976; Hiramatsu, Hatakeyama, Hosoi, & Minami, 1979). In our studies with rats, in vivo immunosuppressive and/or antiinflammatory factors were shown to be associated with factors capable of stimulating in vitro lymphocyte proliferation (Kemp, Mellow, & Sabbadini, 1985). More recently, we have shown that semipurified fractions of the SMG suppress adjuvant-induced arthritis in rats and delayed-type hypersensitivity, allograft rejection, and plaque-forming cell production in mice. These results, especially with the plaque-forming cell assay, indicated a true immunosuppressive mechanism. An electrophoretically homogenous fraction was isolated and shown to be responsible for immunosuppression in the mouse CS model (Abdelhaleem & Sabbadini, 1992). The molecular weight of this immunosuppressive protein was estimated to be 40 kDa by gel filtration and SDS-polyacrylamide gel electrophoresis. Other fractions with different molecular sizes possessed in vitro comitogenic activity but were devoid of in vivo immunosuppressive properties. It has been shown that the size and the structure of the SMG as well as its 418 0889-1591/92 $5.00 Copyright 0 1992 by Academic Press, Inc. AU rights of reproduction in any form reserved

SUBMANDIBULAR

GLAND

AND

IMMUNITY

419

production of several biologically active factors in mice and rats are under the control of several hormones (Lacassagne, 1940; Raynaud, 1960; Gresik, 1980), especially androgens and thyroid hormones. Since the immunosuppressive factors we described in rats were easily detectable in males only, some form of endocrine control should be expected. In order to elucidate in detail the endocrine mechanisms that regulate the production of these factors. we studied in rats the effects of hypophysectomy (Hx) and of hormonal reconstitution on the immunosuppressive activity of SMG extracts. We report here that Hx reduced the immunosuppressive and the lymphocyte-stimulating activities of the SMG. Moreover, significant reconstitution of such activities was obtained with testosterone and thyroxine, two hormones that control the production of several other biologically active factors of the SMG and with prolactin which, to our knowledge, was never previously suggested to control the activity of the SMG. MATERIALS

AND METHODS

Animals. Inbred, pathogen-free male Fischer rats, weighing 175-200 g, were purchased from Canadian Breeding Farm Laboratories, Ltd. (Montreal, Canada). A/J female mice, 6-8 weeks old, were obtained from Jackson Laboratory (Bar Harbor, ME). The animals were maintained on Wayne’s laboratory blocks with 6% fat content (Chicago, IL) and on water ad libitum. Hormones. Adrenocorticotropic hormone (ACTH, purified cortropin, porcine, potency 1 IUimg, Sterivet Labs.), follicle-stimulating hormone (FSH, National Institute of Diabetes, Digestive and Kidney Diseases, NIAMDD-rat FSH B2, activity 3 x NIH-Sl), growth hormone (GH, NIH-GH B18 biopotency 0.81 IU/mg), luteinizing hormone (NIAMDD-bLH-4, bovine, biopotency 2.4 IUlmg), prolactin (NIAMDD-bPRL-6, bovine, biopotency 30 IUlmg), thyroid-stimulating hormone (TSH, thyrotron. human, potency 10 IU/2-5 ml (Nordic Pharmaceuticals, Ltd.), testosterone benzoate (T, ICN Pharmaceuticals), tri-iodothyronine (cytomel T,. liothyronine, SKF), progesterone (P, UPJ), and estradiol benzoate (E2, ICN Pharmaceuticals). Replacement hormone doses, established on the basis of previous experience (Berczi, Nagy, Kovacs, & Horvath. 1981; McKenzie & Berczi, 1987), were administered subcutaneously. ACTH, FSH, and LH at 20 +g daily, GH and PRL at 40 Fg, TSH at 0.66 IUiratiday, testosterone at 1.4 mgiweek, liothyronine at 0.5 kg/day, progesterone at 0.28 mg/week, and estradiol at 0.14 mglweek. Antiserum. Rabbits were immunized with rPRL as described by Vaitukaitis, Robbins, Nieschlag, & Ross (1971). Briefly, rPRL was dissolved in bicarbonatebuffered saline at 100 pglml and emulsified with an equal amount of Freund’s Complete Adjuvant. Female New Zealand White rabbits were injected with a total of 2 ml of emulsion intradermally at 30-50 sites on one side of the animal. Two weeks later, the rabbits were boosted with the same dose of the antigen injected at multiple sites intradermally but on the other side. The animals were bled 10 days later from the ear vein and the serum was tested by enzyme-linked immune assay as described by Signorella & Hymer (1984). Cross-neutralization tests in the Nb2 assay revealed that this serum did not affect bovine PRL or GH. Surgical procedure. Hx was done by parapharyngeal approach (Tarttelin & Gorshi, 1972). Ten days later, prolactin bioactivity was monitored in the serum of the Hx rats, using the Nb2 rat lymphoma proliferation assay (Tanaka, Shiu, Gout,

420

NAGY,

BERCZI,

AND

SABBADINI

Beer, Noble & Friesen, 1980). At the end of the experiments, the completeness of Hx was determined macroscopically in each animal during autopsy. SMG extract. The method for the preparation of the crude SMG extract was modified from Takeda and Grollman (1968). Rat glands were homogenized in phosphate-buffered saline (PBS) in an ice bath using a Brinkmann homogenizer (Brinkmann Instruments, Rexdale, Ontario, Canada). The insoluble material was removed by centrifugation at 18,OOOg for 20 min at 4°C. Ammonium sulfate was added to 36% saturation and the extract was left at 4°C for 1 h and then centrifuged at 18,000g for 15 min at 4°C. The resulting supernatant was made up to 64% saturation with ammonium sulfate, left at 4°C for 1 h, and centrifuged, The precipitate was redissolved in a small volume of PBS and dialyzed against PBS. Gelfiltration. Sephacryl S200 superfine (Pharmacia Canada Ltd., Dorval, Canada) was packed in a 2.6-cm diameter column to a bed height of 60 cm. The column was calibrated with a Pharmacia gel filtration calibration kit. Separations were carried out at room temperature using PBS as the buffer at a flow rate of 22 ml/h/cm’. The protein concentrations of the fractions were measured spectrophotometrically at 280 nm using bovine serum albumin in PBS as the standard. Tissue culture medium. The medium was RPMI-1640 supplemented with 5 x lo-’ M 2-mercaptoethanol, 100 l-&ml penicillin, 100 kg/ml streptomycin, and 10% fetal calf serum (FCS). Cultures were done in 96-well U-bottomed plates (Flow Labs., Rockville, MD). Con A reaction. Samples of the SMG extracts were dialyzed against medium without FCS, sterilized by filtration, and then supplemented with FCS. Tenmicroliter samples were added to each well in quadruplicate. Normal A/J lymph node cells (IO’ in 100 pl) and 0.2 )*g of Con A (Sigma Chemical Co., St. Louis, MO) in 100 ~1 of medium were added to each well. The plates were incubated for 3 days. On the third day, 0.5 pCi of tritiated thymidine (New England Nuclear, Lachine, Canada) in 10 ~JJof medium was added. The cultures were incubated for further 6 h and then the cells were harvested on glass fiber discs with a cell harvester (Skatron A.S., Lienbyen, Norway). The amount of tritiated thymidine incorporated by the cells was measured by liquid scintillation. Contact sensitivity (CS). The CS variety of the delayed-type hypersensitivity reaction was carried out according to the method of Asherson and Ptak (1968). A/J mice were anaesthetized with 54 kg/gm body weight of Nembutal. The skin of the abdomen was shaved. Then, 100 ~1 of 5% picryl chloride-ethanol solution was used to paint a confined area of the abdominal skin. The day of sensitization was taken as Day 0. On Day 5, the thickness of the left ear was measured with calipers. Treatment with SMG fractions was carried out as two daily injections on Days 5 and 6. Each injection contained the material corresponding to one SMG. On Day 6, the animals were challenged by painting both sides of the left ear with 1% picryl chloride in olive oil. The thickness of the ear was measured again 24 h after the challenge. Experimental protocol. Hx was performed by the parapharyngeal approach. Ten days later the PRL level was measured in the serum of both Hx and control rats. Hx animals showing ~20% of control serum lactogenic activity were used for the experiments. Two weeks after Hx, treatments of groups of five animals were initiated with various hormones and were continued for 7 days. All hormones were given subcutaneously at the doses indicated above. At the same time, one group of Hx rats was treated with rabbit-anti-rat PRL serum (50 pi/day, SC) for 7

SUBMANDIBULAR

GLAND

421

AND IMMUNITY

days. Last injections were given 3 h before the termination of the experiments. Submandibular glands were removed under overdose of ether anesthesia. The glands were immediately frozen with liquid nitrogen and kept at -70°C until the time of extraction. Statistical methods. In all experiments involving multiple groups, one-way analysis of variance was used. Comparisons between individual groups were used to test the experimental hypotheses (Dixon & Massey, 1977). In experiments to study the comitogenic activity of multiple chromatographic fractions, the standard deviation (SD) of the control group with PBS was calculated. Confidence limits for the null hypothesis were assumed to be the mean of the control group + 2 SD. RESULTS

The data reported below represent the typical results of two types of experiments, one with Hx rats reconstituted with pituitary hormones and the other with Hx animals reconstituted with nonpituitary hormones. The first type of experiment was repeated twice with similar results, the second type of experiment was repeated once, also with similar results. Effects of Hx. After Hx, serum prolactin levels were measured in all animals. Such levels were well below the maximum level considered acceptable (20% of control) in all Hx groups (data not shown), thus demonstrating that Hx had been complete. Completeness of Hx was confirmed at autopsy at the end of the experiments. In all experiments, Hx groups were compared with intact and with shamoperated controls. In none of the experiments were significant differences between these two control groups observed. For reasons of space, in some of the figures only the results of one control group are shown. Hx induced some growth retardation in our young experimental animals (data not shown). At the same time, there was appreciable atrophy of the SMG as shown by decreased weight of the gland (see control groups of Figs. IA and 64. Such a reduction of gland weight was roughly proportional to the reduction of

Sham

Hx

l-lx Hx+uPRL Hx+ACTH l-lx+ FSH HX+LH Hx+TSH tix+GH Hx+ PRL

Weight

of SMG

igm)

Protecn

content

of SMG

(mg)

1. Effects of Hx and of the treatment of Hx rats with various pituitary hormones on the weight of the SMG (A) and on the amount of protein recovered from each gland in the crude extract (B). Rats were subjected to Hx and treated with pituitary hormones for 7 days at the doses described under Materials and Methods. ACTH, adrenocorticotropic hormone; FSH, follicle-stimulating hormone; LH, luteinizing hormone; TSH, thyroid-stimulating hormone; GH, growth hormone; PRL, prolactin; a-PRL, anti-PRL serum. FIG.

422

NAGY,

BERCZI,

AND

SABBADINI

body weight. At the same time, there was a parallel, but more marked, reduction of the protein content of the SMG extracts (control groups of Figs. IB and 6B). There was a severe reduction of the lymphocyte stimulatory activity in the Con A bioassay in the various gel filtration fractions of the Hx groups compared to the controls (Fig. 2, p < .Ol). Since most of the comitogenic activity of the control group was concentrated in fractions 45-54 (mol wt 14-50 kDa) and since these fractions contain also the in viva immunosuppressive activity (see below), in subsequent experiments only the activity of these fractions was compared in the various experimental groups. To establish whether the loss of lymphocyte stimulatory activity was merely a reflection of the decreased protein content of the fractions, serial dilutions of the pooled 14- to 50-kDa fractions from controls and Hx animals were tested. As shown in Fig. 3, the control fractions demonstrated some lymphocyte-stimulating activity even when diluted I:8 (12 p,g protein per culture), while the two Hx groups did not show any activity even if higher amounts of protein per culture were used. Finally, the pooled 14- to 50-kDa gel filtration fractions of the controls induced significant suppression of the CS reaction in A/J mice (p < .Ol), while such a

7

i

c

LT16 P

45

35

Fraction

sl E

.c

a z

18

55

65

Number

t 9

Cl

2

5 4

25

1

232 ”

..,

t 27

30

35

40

Fraction

45

50

55

60

1

65

Number

2. Proliferative activity of Con A-stimulated murine lymphocytes in the presence of the various gel filtration fractions from the SMG of sham-Hx (top) and Hx rats (bottom). The incorporation of tritiated thymidine (0) and the protein content of each fraction (0) are shown. The two dotted horizontal lines across each profile indicate the expected range for the PBS controls (means t 2 SD). FIG.

SUBMANDIBULAR

GLAND

423

AND IMMUNITY O-0

cont.

e-e 0-D .-.

Sham Hx Hx l-lx+ UPRL

8?’ x 5 B

64--g=g

2 t I

01

,

,

1.5

I

1.0

.

,

I

1

,

,

I*

0

0.5 Protein (mg/ml)

FIG. 3. Proliferative activity of Con A-stimulated murine lymphocytes in the presence of serial dilutions of the pooled 14- to 50-kDa gel filtration fractions of SMG extracts from untreated controls (O), sham-Hx (0). Hx (U), and Hx rats treated with anti-PRL antiserum (m).

suppressive activity was not seen in the fractions from Hx animals (control groups of Figs. 5 and 8). The 50- to IOO-kDa gel filtration fractions, used for control purposes in some experiments, had no immunosuppressive effects in any of the experimental groups (data not shown). Reconstitution of Hx animals with pituitary hormones. Hx rats were treated with various pituitary hormones. The weight of the SMG (Fig. IA) was partially restored by treatment with FSH, LH, TSH, GH, and PRL, while ACTH had no such effect. On the other hand, partial recovery of the protein content of the extracts was seen only in the groups treated with GH or PRL (Fig. 1B). In the Con A bioassay, no restoration of the lymphocyte-stimulatory activity of the pooled 14- to 50-kDa gel filtration fractions was induced by ACTH or FSH, while partial restoration was obtained with LH, TSH, PRL. and GH (Fig. 4).

Sham

HX

HX Hx+nPFiL Hx+ACTH Hx+FSH Hx+LH Hx+TSH Hx+GH Hx+ PRL

I

a

I

L

5

I

I

10 CPM

(x

I1

/

15

I

II

/

20

10-31+SE

FIG. 4. Proliferative activity of Con A-stimulated murine lymphocytes in the presence of the 14- to 50-kDa fractions from the SMG of Hx rats and of Hx rats treated for 7 days with various pituitary hormones at the doses described under Materials and Methods. For abbreviations, see legend to Fig. 1.

424

NAGY,

BERCZI,

AND

SABBADINI

In the in viva CS assay (Fig. 5), significant restoration of the immunosuppressive activity was seen only in the groups treated with LH (JJ < .Ol), and TSH (p < .Ol). In this experiment, treatment with PRL induced partial, but statistically nonsignificant, recovery of the immunosuppressive activity. However, in subsequent experiments the restoration of immunosuppressive activity induced by PRL was consistently sigmficant (e.g., see the experiment of Fig. 8. below). Reconstitution of Hx animals with pituitary-controlled hormones. Since the experiments ‘with pituitary hormones had shown that the immunosuppressive activity of SMG extracts could be restored with LH and TSH, experiments were carried out with sex and thyroid hormones. The interactions of these hormones among themselves and with PRL were also studied. The weight of the SMG of Hx rats and the protein content of the extracts was restored by the treatment with testosterone, PRL, and all combinations containing these hormones. On the other hand, T,, E,, and P had no appreciable effects on these parameters (Fig. 6). In the Con A bioassay (Fig. 7), the treatment of Hx mice with testosterone, T,, or PRL resulted in the reappearance of significant levels of lymphocyte-stimulating activity. On the other hand. so significant reconstitution of the lymphocyte-stimulating activity of various gel filtration fractions was seen with either E, or progesterone. Of the various combinations, those in which PRL was included appeared to induce a more definite level of reconstitution. In the CS assay, testosterone, T,, PRL, and the various combinations of these hormones all induced significant restoration of immunosuppressive activity @ < .Ol for all these treatments), while progesterone and E, had no such effect (Fig. 8). DISCUSSION

We had previously shown that fractions in the molecular range of about 15 to 50 kDa obtained by gel filtration from the rat SMG stimulated the in vitro proliferation of Con A-treated lymphocytes and suppressed in vivo immune reactions

PBS Control Sham

Hx

Hx Hx +crPRL Hx+ACTH Hx + FSH Hx+LH Hx +TSH Hx+GH Hx+

PRL 10 Ear

30

20 thickness

(mm

x lo?

40

50

& SE

FIG. 5. CS reactions of A/J mice challenged with picryl chloride 6 days after immunization. The mice received two injections of phosphate-buffered saline (PBS) or of the 14- to 50-kDa gel filtration fractions from the SMG of unoperated controls (Control), of sham-operated controls (Sham Hx), of Hx rats (Hx), or of Hx rats treated with various pituitary hormones (see legend to Fig. 1 for abbreviations). Ear thickness was measured 24 h after challenge.

SUBMANDIBULAR

AND

GLAND

425

IMMUNITY

Control Hx Hx+T Hx+T, Hx+PRL

f3

Hx+P Hx+E* Hx+T+T3 Hx+T+PRL Hx+PRL+T3

0.05 Werght

5

0.14

0.10 of SMG

cgm)

Protem

14

10 content

per

SMG

tmgl

FIG. 6. Effects of Hx and of the treatment of Hx rats with PRL or nonpituitary hormones on the weight of the SMG (A) and on the amount of protein recovered from each gland in the crude extract (B). T, testosterone; T,, tri-iodothyronine: PRL, prolactin: P, progesterone; E2. estradiol benzoate.

(Kemp et al., 1985). Further fractionation of the 15 to 50-kDa fractions demonstrated the presence of several fractions with in vitro lymphocyte-stimulating activity. Only one of these lymphocyte-stimulating fractions, a 40-kDa protein, also suppressed in vivo immune reactions (Abdelhaleem & Sabbadini, 1992). Thus, while some SMG proteins stimulate in vitro lymphocyte proliferation and are devoid of in vivo immunosuppressive activity, a 40-kDa protein seems to induce both in vivo immunosuppressive effects and in vitro stimulation of lymphocyte proliferation. This clearly indicates that, while the CS assay for in vivo immunosuppression may be expected to measure the activity of the immunosuppressive factor, the in vitro mitogen assay may have been affected by other components of the semipurified extracts used for these investigations. The concomitance of an immunosuppressive activity with the stimulation of

PBS Control Hx Hx+T Hx+T, Hx+PRL Hx+P Hx+E2 Hx+T+T3 Hx+T+PRL Hx+PRL+T3 Hx+T+T3+PRL I 5 CPM

I 10

1 ) 15

(x 10-3)kSE

FIG. 7. Proliferative activity of Con A-stimulated murine lymphocytes in the presence of the 14 to 50-kDa gel filtration fractions from the SMG of Hx rats and Hx rats treated with PRL or various nonpituitary hormones. For abbreviations, see the legend to Fig. 6.

426

NAGY, BERCZI,

AND SABBADINI

PBS

Control Hx Hx+T Hx+T3 Hx+ PRL Hx+P Hx+E2 Hx+T+T3 Hx+T+PAL Hx+PRL+Ts Hx+T+Ts+PRL

10 Ear

1 30

20 thickness

(mm

x lo-‘)

I 40

* 50

?SE

Frc. 8. CS reactions of A/J mice challenged with picryl chloride 6 days after immunization. The mice received two injections of phosphate-buffered saline (PBS), or of low mol wt (14-50 kDa) fractions from the SMG of unoperated controls (Control), of Hx rats (Hx), or of Hx rats treated with various hormones (see legend to Fig. 6 for abbreviations). Ear thickness was measured 24 h after challenge.

lymphocyte proliferation in the same apparently pure fraction is somewhat puzzling. One might speculate that the cells which respond with proliferation in the presence of the immunosuppressive factor may have suppressor effects on certain immune reactions. Otherwise, it is possible that several cell types may be sensitive to the effects of the 40-kDa protein. Some of these cells would respond with increased proliferation, while others may be inhibited in their immune responsiveness. Since several biologically active factors had been shown to be hormone dependent, we decided to investigate the effects of Hx and of the reconstitution with various hormones on the immunosuppressive activity of SMG extracts and on the lymphocyte-stimulating one. The results demonstrated that Hx induced a marked drop of both the in vitro lymphocyte stimulating activity and the in vivo immunosuppressive activity. Since the SMG fractions from Hx animals failed to stimulate in vitro lymphocyte proliferation even at levels of protein concentration that allowed the detection of stimulating effects of fractions from normal animals, it would appear that the reduction of this activity after Hx was not a mere consequence of the atrophy of the SMG. Rather, these results indicate the existence of specific regulatory endocrine mechanisms for the immunologically active factors. The reconstitution experiments suggested that such regulatory mechanisms are different for different factors of the SMG. Thus, GH induced a significant recovery of the weight of the SMG, of the protein concentration of the extracts, and of the in vivo lymphocyte stimulation. However, GH did not induce the recovery of the in vivo immunosuppressive activity. As noted above, the lymphocyte proliferation assay and the CS assay for immunosuppression measure the activity of different but partially overlapping factors. On the other hand, PRL and testosterone restored all these parameters, while T, induced a marked restoration of the in

SUBMANDIBULAR

GLAND

AND

IMMUNITY

421

vivo and in vitro immunoregulatory activities without measurable increase of the total weight of the gland or of the protein content of the extracts. As noted above, not all the fractions that stimulate in vitro lymphocyte proliferation possess in vivo immunosuppressive activity. Therefore, a dissociation of the effects of hormones indicates differential effects on different SMG proteins. The reconstitution experiments also indicated that the immunosuppressive and the lymphocyte-stimulating activities are under the control of endocrine mechanisms to some extent similar to those which regulate other biologically active factors of the SMG. Thus, testosterone and thyroxine have been shown to stimulate the production of several SMG factors produced by the cells of the convoluted granular tubules (Gresik, 1980), while corticosteroids are required for a full restoration of the trophism of the SMG induced by thyroid hormones and testosterone (Chretien, 1977). Our results with TSH and T,, on the one hand, and with LH and testosterone, on the other, show the existence of similar mechanisms for immunologically active factors. On the other hand, to our knowledge, there have been no reports indicating that PRL stimulates the production of biologically active factors in the SMG. Therefore, our finding that PRL stimulates the production of factors with in vivo immunosuppressive and in vitro comitogenic activities may indicate that these factors have a somewhat different endocrine regulation. Alternatively, the silence of the literature about PRL may merely indicate a certain lack of interest of previous investigators in the effects of this hormone. By itself, the presence of an immunosuppressive factor in the SMG does not necessarily imply that this gland has a significant role in the regulation of immune responses. On the other hand, findings in the recent (Mathison, 1991) and less recent literature (Hiramatsu et al., 1979) indicate that the SMG may indeed have such a regulatory role. If this is the case, the 40-kDa factor with in viva immunosuppressive properties we described may be one of the mechanisms mediating such a regulatory activity. REFERENCES Abdelhaleem, M., & Sabbadini, E. (1992). Identification of immunosuppressive fractions from the rat submandibular gland. Zmmuno/ogy 76, 331-337. Asherson, G. L., & Ptak, W. (1968) Contact and delayed hypersensitivity in the mouse. Immttnolog~ 15, 405416. Berczi, I., Nagy, E., Kovacs, K., & Horvath. E. (1981) Regulation of humoral immunity in rats by pituitary hormones. Acta EndocrinoL. 98, 5065 13, Chretien, M. (1977). Action of testosterone on the differentiation and secretory activity of a target organ: The submaxillary gland of the mouse. Int. Rev. Cytol. 50, 333-395. Dixon, W. .I., & Massey, F. J. (1977). Introducrion fo staristicnl ana/ysis. McGraw-Hill, New York. Gresik, E. W. (1980). Postnatal developmental changes in submandibular glands of rats and mice. J. Histochem.

Cytochem.

28, 860470.

Hiramatsu, M., Hatakeyama. K., Hosoi, K., & Minami. N. (1979). Regulation of delayedhypersensitivity response by the submandibular gland of male mice. Immunology 37, 869-872. Kemp, A., Mellow, L., & Sabbadini, E. (1985). Suppression and enhancement of in vitro lymphocyte reactivity by factors in rat submandibular gland extracts. Immunology 56, 261-267. Koch, J. H., & Rowe, J. (1976). Immunosuppressive activity of submaxillary gland extracts of the mouse. I. Effect on antibody formation in response to sheep red blood cells, Eur. .I. Zmmunol. 6, 583-588. Kongshavn, P. A. L., & Bliss, I Q. (1970). Effect of mouse submandibular gland extracts on survival of H-2 incompatible skin allografts. Immunology 19, 363-367. Kongshavn, P. A. L., & Lapp. W. S. (1972). Immunosuppressive effect of male mouse submandibular gland extracts on plaque-forming cells in mice: Abolition by orchiectomy. Immunology 22, 227230.

428

NAGY, BERCZI,

AND SABBADINI

Lacassagne, A. (1940). Dimorphisme sexuel de la glande sous-maxillaire chez la souris. C. R. Sot. Biol. 133, 180. Mathison, R., Hogan, A., Helmer, D., Bauce, L., Woolner, J., Davison, J. S., Schultz, G., & Befus, D. (1992). Role for the submandibular gland in modulating pulmonary inflammation following induction of systemic anaphylaxis. Bruin Behav. Immun. 6, 117-129. McKenzie, C. E., & Berczi. I. (1987). Oestrogen dependent suppression of mitogen response in rat lymphocytes. Acta Endocrinol. 116, 200-204. Raynaud, J. (1960). Control hormonal de la glande sous-maxillaire de la souris. Bull. Biol. 94, 400. Roberts, L., Freston, J. A., & Reade, C. (1976). Suppression of immune responsiveness by submandibular salivary gland factor. Immunology 30, 81 I-814. Signorella, A. P., & Hymer, W. C. (1984). An enzyme-linked immunosorbent assay for rat prolactin. Anal. Biochem. 136, 372-381. Takeda, T.. & Grollman, A. (1968). Inhibitory action of submaxillary gland on thymus and lymphoid tissues of the mouse. Am. J. Physiol. 215, 1337-1342. Tanaka, T., Shiu, R. P. C., Gout, P. W.. Beer, C. T., Noble, R. L.. & Friesen, H. (1980). A new sensitive and specific bioassay for lactogenic hormones: Measurement of prolactin and growth hormone in human serum. J. C/in. Endocrinol. Metab. 51, 1058-1063. Tarttelin, M. F., & Gorshi, R. A. (1972). A ventral surgical approach to the anterior tuberal hypothalamus of the rat with specific emphasis on recovery. Endocrinology 90, 1128-1132. Vaitukaitis, J., Robbins, J. B., Nieschlag, E., & Ross, G. T. (1971). A method for producing antisera with small doses of immunogen. .I. C/in. Endocrinol. 33, 988-991. Received July 28, 1992