Cytokine responsiveness in germfree and conventional NMRI mice

CYTOKINE

RESPONSIVENESS IN GERMFREE CONVENTIONAL NMRI MICE

Tina Granholm,lJ

Berit Friiysa,l Cecilia Lundstriim,l Tore Midtvedt,s Olof Siiderl

AND

Aida Wahab,l

We have investigated the proliferative response of thymocytes from different mouse strains to cytokines in vitro. Interleukin 2 @L-2), IL-4 and IL-7 induced proliferation of thymocytes from NMRIKI (a locally bred NMRI mouse strain), NMRI/H (‘traditional’ NMRI mice), C3H/HeJ and C3WHeN mice. NMRI/KI thymocytes showed the most prominent proliferation in response to IL-lcx and IL-lp. IL-3, IL-5, IL-6, IL-S, IL-IO, tumour necrosis factor cy (TNF-a), interferon y (IFN-y), inhibin and lipopoiysaccharide (LPS) induced no thymocyte proliferation. Germfree NMRIKI mouse thymocytes showed a significantly lower proliferation in response to IL-la and IL-lp than conventional mice. Rat tissues, previously shown to contain lymphocyte activating factors (LAFs), were also tested. Skin, tongue, esophagus, proventricular stomach, testis and placenta were ail positive in the LAF assay utilizing NMRIKI thymocytes, whereas none of the tissue extracts could induce proliferation in NMRUH thymocytes. The higher cytokine responsiveness in conventional mice compared with germfree might suggest that exposure to microflora induces a higher state of activation of the immune system. The LAF assay, utilizing NMRYKI thymocytes, is a highly sensitive IL-1 bioassay with a detection level of 1 pg/mi for IL-lp and 2 pg/mI for IL-lo. The specificity of the assay is increased by utilizing NMRI/H mice to exclude the presence of IL-2, IL-4 and IL-7.

Animals of different species raised in a completely germfree milieu have previously been shown to differ in their immunological parameters and response to different stimuli in comparison to conventional animals of the same species. They have a lower count of peripheral blood leukocytes with a lower percentage of neutrophils and negligible amounts of serum gamma globulin fractions.1 Neutrophils of germfree rats seem to have less efficient killing activity as compared to conventional rats,2 and bronchoalveolar lavage performed on pigs shows half the normal number of T-cells in germfree animals, and a much smaller proportion of T-suppressor lymphocytes.3 Germfree mice have

From the IPediatric Endocrinology Unit, Karolinska Hospital, 2Department of Pediatric Surgery, S:t Giiran’s Children’s Hospital, and 3Department of Medical Microbial Ecology, Karolinska Institute, Stockholm, Sweden. Correspondence to: Olof Slider, M.D., Ph.D., Ass. Prof., Pediatric Endocrinology Unit (LS), Karolinska Hospital, S104 01 Stockholm, Sweden. Received 8 April 1992; revised and accepted for publication 15 July 1992 0 1992 Academic Press Limited 1043-4666/92/060545+06 $08.00/O KEY WORDS: thymocyte

CYTOKINE,

cytokine/germfree/interleukin

Vol.

4, No.

6 (November),

l/proliferation/

1992: pp 545-550

shown a higher susceptibility to, a slower recovery from, and a decreased immunological response to murine cytomegalovirus infections.4 In the present study, we have investigated the in-vitro proliferation of thymocytes from germfree, ex-germfree and conventional mice in response to different cytokines. The aim of the study was to assessthe importance of the immunological state of the thymocyte donors to the lymphocyte activating factor (LAF) responsiveness of the thymocytes. There is also previous evidence that thymocytes from different mouse strains respond differently when cultured in the presence of LAFs.~ Thus, we have compared the LAF activity induced in thymocytes from different mouse strains in response to lymphocyte activating cytokines.

RESULTS Cytokine Responsiveness

of Conventional

Mice

Thymocytes from the locally bred NMRI/KI mouse substrain showed LAF response with dosedependent proliferation to interleukin la (IL-la), IL-ll3, IL-2, IL-4, IL-7 and tIL-1, whereas IL-3, IL-5, IL-6 and IL-S were inactive (Figs lA, and 2). 54.5

546 ! Granholm

CYTOKINE,

et al.

Thymocytes from NMRI/H mice also responded to IL-2, IL-4 and IL-7, but no significant LAF activity could be induced by IL-lcx, IL-lp or tIL-1 (Figs lB, 2A). The response induced by IL-2 and IL-7 in the NMRI/H mice was significantly lower in comparison to that in NMRI/KI mice (P < 0.001 at a concentration of 110 rig/ml for IL-2 and 22.5 rig/ml for IL-7). No LAF activity was demonstrated by IL-lo, tumour necrosis factor (Y (TNF-CY), rat or human interferon y (EN-y) inhibin or lipopolysaccharide (LPS) (data not shown). Under the present assay conditions, C3WHeJ and C3H/HeN mice showed slight response to IL-lo, IL-@ and tIL-1, and high LAF activity in response to IL-2, IL-4 and IL-7. This activity was, however, significantly lower compared to that in NMRIKI mice (P < 0.001 at a concentration of 20 r&ml for IL-2 and IL-7, and 35 rig/ml for IL-4; P < 0.05 at 1 rig/ml for IL-lo; P < 0.01 for IL-lp at the same concentration; P < 0.05 for tIL-1 at a protein concentration of 190 pg/ml; Fig. 2B). No significant difference in the cytokine response between C3IWHeJ and C3HHeN thymocytes could be detected.

Vol. 4, No. 6 (November 1992: 545-550)

IL-la (P < 0.01 at an IL-lo concentration of 15 pg/ml) and IL-ll3 (P < 0.05 at 15 pg/ml) in comparison to germfree animals, whereas the response to tIL-1 was slightly higher in the conventional mice (not statistically significant; Fig. 3). There was no significant difference in the response of thymocytes obtained from germfree and ex-germfree C3WHeJ and C3H/HeN mice to different cytokines (data not shown).

L.AF Activity Induced by Rat Tissues Crude aqueous extracts of normal adult rat tissues previously shown to contain LAF bioactivity, i.e. skin, tongue, esophagus, proventricular stomach, duodenum and placenta,627 were tested in the LAF assay, using conventional NMRI/KI and NMRI/H mice as thymocyte donors. LAF activity in the testis was tested with all mouse strains (tIL-1; see Materials and Methods). All tested tissue extracts produced LAF activity with NMRI/KI thymocytes, whereas very slight or no activity could be found with NMRIEI thymocytes (Fig. 4).

Cytokine Responsivenessof Germfree and Ex-germfree Mice

DISCUSSION

IL-la, IL-lp and tIL-1 were tested with germfree, ex-germfree and conventional NMRI/KI mice. The ex-germfree mice showed a slightly higher, though not statistically significant, LAF response to IL-lo, IL-lp and tIL-1 in comparison to germfree mice. Conventional mice showed 50-800/o higher LAF response to

In the present study we have compared the ability of different cytokines to induce thymocyte proliferation in vitro, using thymocytes from different mouse strains. IL-2, IL-4 and IL-7 could induce thymocyte proliferation in cell cultures from NMRI/KI, NMRI/H, C3WHeJ and C3H/HeN mice. whereas NMRI/KI

60

ca

/. /

n ‘0 -

50

; 8 -z 3 -1 B

40

5

I Relatlve

4

16 concentration

64

251

60 c

0 IL-la . IL-lp v IL-2 . IL-3 a IL-4 n IL-5 p IL-6 A IL-7 0 IL-B l ilL-I

B

30

q

rvkdl”rn Control

20 IO L

c I Relative

4

16

64

256

concentration

Figure 1. (A) Proliferative response of NMRFKI thymocytes to interleukins. Dose-dependent proliferation was induced by IL-lo (225 pgiml), IL-ll3 (225 pg/ml), IL-2 (450 rig/ml), IL-4 (140 ngiml), IL-7 (225 rig/ml) and tIL-1 (190 pg/ml final protein concentration), whereas no activity could be induced by IL-3 (450 rig/ml), IL-5 (450 U/ml), IL-6 (90 ngiml) or IL-8 (2.3 pg/ml). The maximal tested concentrations (relative concentration 256) are indicated within brackets. Thymocytes were cultured for 48 h, including a 2-h pulse with sH-thymidine prior to harvesting. (B) Proliferative response of NMRI/H thymocytes to interleukins. LAF activity was induced by IL-2, IL-4 and IL-7, whereas IL-la, IL-ll3 and tIL-1 produced no or slight activity. No activity was induced by IL-3, IL-5, IL-6 or IL-S. The cytokines were tested at the same concentrations as in A.

Cytokine response in mice I 547

)-

I-

)-

I-

I?

NMRI/KI

NMRI/H

NMRI/KI

C3H/HEJ

C3H/HEN

Figure 2. (A) LAF activity in NMRIiKI and NMRIiH mice from a similar experiment as in Fig. 1. Dose-response curves were established for all interleukins, and the curves were compared at a level where they were linear and parallel. LAF activity was induced by IL-la (55 pgiml), IL-ll3 (55 pgiml), IL-2 (110 ngiml), IL-4 (35 rig/ml), IL-7 (225 rig/ml) and tIL-1 (50 pg/ml of tissue protein) in NMRI/KI thymocytes, and by IL-2, IL-4 and IL-7 in NMRIiH thymocytes. Tested concentrations are indicated within brackets. IL-2 and IL-7 induced a higher response in the NMRI/KI thymocytes (P < O.OOl), whereas IL-4 induced an equal resoonse in both substrains. (B) LAF activity in C3H/HeJ and C3WHeN thymocytes compared to the NMRIIKi substrain. C3WHeJ and HeN produced siight proliferation in response to iL-l& or IL-#under these conditions. LAF activitv was induced bv IL-2. IL-4 and IL-7 in both C3H/HeJ and C3HHeN mice. but this was significantly lower thanin the NMRIIKI substrain (P < 0.001). Dose-response curves were established for all cytokines, and the response compared where the curves were linear and parallel. The concentrations inducing the present LAF response were 1 rig/ml for IL-la and IL-lp, 20 rig/ml for IL-2, IL-3 and IL-7, 35 @ml for IL-4,20 U/ml for IL-5,5 rig/ml for IL-6 and 100 rig/ml for IL-& tIL-1 was tested at a final protein concentration of 190 pgiml

showed

the highest and most reproducible LAF to IL-lo and IL-P. IL-3, IL-5, IL-6, IL-8, IL-lo, TNF-CY, EN-y, inhibin and LPS showed no LAF activity in the present assay. A significant difference in response between conventional and germfree NMRUKI mice was demonstrated, where thymocytes from conventional mice showed a higher in-vitro proliferation in response to IL-lo and IL-@. Adult rat tissues, previously shown to contain LAFs,6,7 were tested with NMRVKI and NMRUH mice, and LAF activity was demonstrated in all tested tissues, i.e. skin, tongue, esophagus, proventricular stomach, duodenum, testis and placenta, using thymocytes from the NMRUKI substrain. None of the tissues could induce any LAF activity using thymocytes from the NMRUH substrain. Gery et al.5 have reported that thymocytes from different mouse strains showed a difference in LAF activity in response to soluble macrophage-derived LAFs in vitro. The underlying cause for this difference is, as yet, unexplained. The NMRI/KI and NMRI/H mouse substrains share a common ancestry in 1960, and yet NMRUH thymocytes are unable to proliferate in response to IL-l. Both substrains are healthy normal mice and the NMRUH mice show no signs of immune dysfunction. The demonstrated difference might thus be an in-vitro effect, and the thymocytes might activity

in response

respond in a normal way to IL-1 induction in vivo. Alternatively, they may have developed a different mode of T-cell activation in response to antigens or mitogens. The C3WHeJ and C3WHeN thymocytes do respond to IL-lo, IL-lp, IL-2, IL-4 and IL-7 by proliferation in vitro during the present conditions, but the response is significantly lower than in the NMRI/KI thymocytes. The C3IYHeJ and C3IWHeN mice respond equally well, however, in spite of their difference in susceptibility to LPS. The demonstrated difference in cytokine responsiveness in germfree and conventional NMRI/KI mice might suggest that the exposure of environmental microflora induces a higher state of activation of the immune system. The germfree animals in our studies are exposed to bacterial antigens from dead bacteria in the food, but the absence of live bacteria and other micro-organisms in the incubators is carefully and continuously monitored. The ex-germfree animals were colonized with normal intestinal microflora, and after three weeks the LAF assay indicated a slightly higher responsiveness, though not statistically significant. These findings might have clinical implications in patients receiving bone marrow transplants, since some studies indicate that the graft-versus-host reaction is less vigorous and more easily treated in patients who are subject to strict protective isolation

548 I Granholm

CYTOKINE,

et al.

and gastro-intestinal bacterial decontamination prior to the transplantation.*,9 IL-l was first described as a lymphocyte activating factor (LAF). s~O,ll Other cytokines have also been shown to exhibit LAF activity, such as IL-4,12 IL-6,13 IL-7,i4 IL-9,is IL-lo,16 and TNF-a,17 as well as other substances, e.g. inhibini* and tachykinins (neurokinin A and physalaemin). 19 IL-l has an important function in the induction of the immune system in response to infections and other antigen stimuli. In the present study we have shown that IL-2, IL-4 and IL-7 can induce thymocyte proliferation in the LAF bioassay under the conditions used in our laboratory, whereas other T-lymphocyte inducers, such as IL-6, IL-lo, TNF-a, inhibin and LPS are unable to produce any LAF activity. However, since the NMRI/KI substrain shows LAF activity in response to IL-lo and IL-lp, whereas the NMRI/H substrain does not, we are able to exclude the presence of IL-l, IL-4 and IL-7 in tissue samples that are positive in the NMRUKI assay and negative in the NMRUH assay. So far, the only characterized lymphocyte activating factors that have shown this pattern are IL-la and IL-lp.

Vol. 4, No. 6 (November 1992: 545-550)

I

di

I T A

B

C

D

E

F

G

IL-2

Figure 4. LAF response induced by aqueous extracts of adult rat tissues in thymocytes from conventional NMRI/KI and NMRI/H mice. Dose-response curves were established, and the tissue extracts compared at the dose inducing the maximal response. A, skin (tested at a final protein concentration of 1.3 &l/ml); B, tongue (40 pg/ml); C, esophagus (25 l.rg/ml); D, proventricular stomach (65 pg/ml); E, duodenum (25 pg/ml); F, testis (240 pg/ml); and G, placenta (300 pg/ml). The tested tissues induced no or slight LAF activity in the NMRI/H mice, whereas high levels of LAF activity could be demonstrated with all tested tissues with responder thymocytes from NMRI/KI mice. For comparison, the responsiveness to IL-2 (110 rig/ml), which is active with both NMRIiKI and NMRI/H mouse thymocytes, is shown.

**

)-

I-

IL- la NMRI I7 kXS E3

/KI Germfree Ex-germfree Conventional

m

NMRI/H

IL- Ip

IL-6

tlL-I

Figure 3. LAF activity in response to IL-la (4 pg/ml), IL-1B (4 pg/ml), IL-6 (20 rig/ml) and tIL-1 (50 pg/ml of tissue protein) in germfree NMRIiKI mice in comparison to ex-germfree and conventional NMRI/KI mice and conventional NMRI/H mice. The depicted concentrations are indicated within brackets. The LAF activity induced by IL-la and IL-1B in thymocytes from ex-germfree NMRUKI mice was slightly higher as compared to germfree mice (not statistically significant). Thvmocvtes from conventional NMRI/KI mice showed a significantly higher proliferation in resoonse to IL-la (**P < 0.01) and IL-1B (*P < 0.05) in comparison with germfree mice. Proliferation induced by tIL-i was also higher, though not statistically significant, in conventional compared to germfree mice. Only slight thymocyte proliferation could be induced in NMRIiH mice, and IL-6 induced no activity in any of the tested animals.

The present assays were used on tissues previously shown to contain LAFs, and we can thus conclude that the LAF(s) present in the testis, skin, proximal gastrointestinal tract and placenta either consists of IL-l, or some other hitherto unknown functionally related polypeptide factor(s). In conclusion, we have shown a distinct difference in responsiveness to different cytokines in germfree and conventional NMRUKI mice. This difference is in accordance with previous results that germfree animals show a difference in immunological parameters and response to antigen stimulation,1-4 and might reflect a difference in the state of activation of the immune system in germfree and conventional animals. The present LAP assay using NMRUKI mouse thymocytes has been shown to be highly sensitive to IL-la, with a detection level of 2 pg/ml (< 120 fM), and to IL-lp, with a detection level of 1 pg/ml (< 60 fM). By utilizing NMRUH thymocytes, and comparing the results between the two assays, any presence of IL-2, IL-4 and IL-7 in the sample can be excluded. Furthermore, we have been unable to detect any direct or additive stimulatory effect of IL-6 under the present conditions. The present LAF assay is thus a highly sensitive IL-l assay, where other presently known cytokines can be excluded as the source of LAF activity.

Cytokine responsein mice i 549 MATERIALS

AND METHODS

Experimental Animals Conventional (non-germfree) male 4-8-week-old NMRI mice, originally bred at Institut fur Versuchstierzucht, Hannover, Germany (here designated NMRIIH), were obtained from ALAB, Sollentuna, Sweden. Male mice from a NMRI substrain locally bread since 1960 at the Department of Medical Microbial Ecology at the Karolinska Institute (NMRFKI), and originating from the same strain as NMRI/H the same year, were tested as germfree, ex-germfree and conventional mice at ages 4-S weeks. Germfree and ex-germfree male C3HHeJ and C3HHeN mice, also raised at the Department of Medical Microbial Ecology, were used at ages 4-6 weeks. The germfree animals were reared as described by Gustafsson.20 To obtain ex-germfree mice, the animals were taken out of their germfree environment, colonized with normal intestinal microflora as previously described21 and kept in the same cage as a conventional mouse; they were sacrificed 2-3 weeks later. All germfree mice were removed from the germfree environment immediately before being sacrificed. Adult male and female Sprague-Dawley rats (aged 60-90 days) were obtained from ALAB. All animals were sacrificed by CO, inhalation.

Cytokines Recombinant human IL-la and IL-ll3 and recombinant murine IL-3, IL-4 and IL-5 were obtained from Genzyme, Boston, MA, USA. Recombinant human IL-2, IL-6 and interferon-y @N-y) were purchased from BoehringerMannheim Scandinavia AB, Bromma, Sweden, and recombinant human IL-7 and IL-8 from Biosource International, Westlake Village, CA, USA. Recombinant murine IL-10 was obtained from Pharmingen, San Diego, CA, USA and recombinant human tumour necrosis factor-o (TNF-a) from Genentech, San Francisco, CA, USA. Tissue-derived IL-6 from a human bladder carcinoma was a gift from Dr B. Andersson, Stockholm, and recombinant rat IFN-y from Dr A. Scheynius, Stockholm, Sweden. Inhibin sources were ovine rete testis fluid obtained from Dr F. J. de Jong (Erasmus University, Rotterdam, the Netherlands) and purified bovine follicular fluid (M, 31 000) inhibin obtained from Dr D. M. Robertson (Monash University, Victoria, Australia). Lipopolysaccharide (LPS) from Escherichiu coli was purchased from Difco Laboratories, Detroit, MI, USA. Interleukins 1 to 8 were tested for lymphocyte activating factor (LAF) activity at concentrations stated in the figure legends. IL-10 was tested at a maximal concentration of 45 U/ml, bovine inhibin at 4000 rig/ml, ovine inhibin at 2 U/ml, rat IFN-y at 2000 U/ml and TNF-cx at 45 pg/ml. LPS was tested at 1:4 intervals ranging from a concentration of 20 pg/ml down to 70 pg/ml

15 min. The supernatants were frozen, thawed and again centrifuged at 10 000 x g in order to remove cryoprecipitated material. Supernatants were filtered (0.22 pm) and kept at -20°C until analyzed in the LAF bioassay. The protein concentration was determined using the method of Bradford22 with bovine serum albumin (BSA) as a standard. Rat testicular extract as a source of testicular IL-l-like factor (tIL-123) was used as a control in the LAF assay, and was tested at a maximal protein concentration of 190 @ml.

LymphocyteActivating Factor (LAF) Assay The murine thymocyte proliferation assay was performed as previously described.5,24 The mouse thymus was removed aseptically immediately after sacrificing the animal and placed in Hank’s medium (Flow Laboratories, Irvine, Scotland). Cells were released from the thymic tissue by careful disruption with a Pasteur pipette and passed through a plastic mesh filter (500 pm pore size). Cell aggregates were removed by passage through two additional mesh filters (80 and 25 pm). After washing the filters with Hank’s medium, the cell suspension was centrifuged at 250 X g for 7 min. The supernatant was removed, and the cells resuspended in a-modified Eagle’s minimum essential medium (a-MEM; Flow Laboratories)? and again centrifuged at 250 x g for 7 min. The supernatant was again discarded, and the cells resuspended in a-MEM supplemented with L-glutamine (2 mM), penicillin (100 IU/ml), streptomycin (100 yg/ml), fetal calf serum (l%), c-r-thioglycerol (75 pM) and 5 pg/ml of phytohemagglutinin-P (Difco Laboratories, Detroit, MI, USA). Each cell suspension consisted of thymocytes from only one animal. The cell concentration was adjusted to 4 X 106 cells per milliliter (2.0 - 2.6 x 106 in the C3H/HeJ and C3WHeN experiments), and the cells were transferred in 100 pl volumes to flat-bottomed 96-well microtiter plates (Falcon; Becton Dickinson Labware, Lincoln Park, NJ, USA). The individual cytokines were added from the start in 10 pl volumes at six different concentrations (1:4 dilutions) in order to establish dose-response curves. The cells were incubated for 48 h at 37°C. Tritiated thymidine (0.5 pCi/well; Amersham Sweden AB, Solna, Sweden) was added for the last 2 h of incubation. The cells were harvested onto glass fiber filters using a Titertek Cell Harvester 500 (Flow Laboratories), and the incorporated radioactivity was measured in a liquid scintillation counter. Thymocyte proliferation (LAF activity) was assessedas incorporation of sH-thymidine and expressed as mean counts per minute (cpm) of duplicate cultures. Dose-response curves were established for all samples. Statistical significance was evaluated by using Student’s t-test.

Tissue Extraction

Acknowledgements

Sprague-Dawley rats were dissected immediately after sacrificing, and tissues were frozen at -20°C. After thawing, the tissues were homogenized in 0.15 M sodium chloride (saline; 1:4 or 1:lO w/v) and centrifuged at 7.50 x g for

Supported by the Swedish Medical Research Council, the King Gustaf V Jubilee Foundation, H.R.H. Crown Princess Lovisa’s Society of Pediatric Health Care, the Swedish Society of Medicine, the

550 / Granholm

et al.

Children Cancer Foundation, Magn. Bergvall Foundation, the Sven Jerring Foundation, and the Society for Child Care.

REFERENCES 1. Travnicek J, Mandel L, Trebichavsky I, Talafantova M (1989) Immunological state of adult germfree miniature Minnesota pigs. Folia Microbial (Praha) 34:157-164. 2. Tsuda M, Ohkubo T, Tamura M, Ogoshi K, Ohnishi N. Gzawa A. Yamamura M (1990) Perinheral blood neutronhil function in germfree and conventional rats. Tokai J Exp Clin Med 1.5:107-109. 3. Gehrke I, Pabst R (1990) Cell composition and lymphocyte subsets in the bronchoalveolar lavage of normal pigs of different ages in comparison with germfree and pneumonic pigs. Lung 168:79-92. 4. Tazume S, Umehara K, Leung WC, Ando K, Yoshida T, Hashimoto K (1990) Immunological responses in germfree mice infected with murine cytomegalovirus. Tokai J Exp Clin Med ,15:13-20. 5. Gery I, Gershon RK, Waksman BH (1972) Potentiation of the T-lymphocyte response to mitogens. I. The responding cell. J Exp Med 136:128-142. 6. Granholm T, Siider 0 (1991) Constitutive production of lymphocyte activating factors by normal tissues in the adult rat. J Cell Biochem 46:143-151. 7. Granholm T, Froysa B, Midtvedt T, Seder 0 (1992) Ontogeny of lymphocyte activating factors in conventional and ge:mfree rats. Regional Immunol (to appear in vol. 4). /’ 8. Vossen JM, Heidt PJ, van den Berg H, Gerritsen EJA, Hermans J, Dooren LJ (1990) Prevention of infection and graft-versus-host disease by suppression of intestinal microflora in children treated with allogeneic bone marrow transplantation. Eur J Clin Microbial Infect Dis 9:14-23. 9. Vossen JM, de Be1 CE, van den Berg H, Gerritsen BJA, Heidt PJ (1990) Successful complete suppression of the gut microflora prevents acute graft-versus-host disease in children grafted with a-matched non-T-cell depleted bone marrow transnlant. In Heidt PT. Vossen JM . Rusch VC (eds) Mikrookolozie ;nd Therapie - Microecology and Therapy, Vol. 20, pp. 515-519. Institut fiir Mikrookologie, Herborn-Dill, Germany. 10. Gery I, Waksman BH (1972) Potentiation of the Tlymphocyte response to mitogens. II. The cellular source of potentiating mediator(s). J Exp Med 136:143-155.

CYTOKINE,

Vol. 4, No. 6 (November 1992: 545-550)

11. di Giovine FS, Duff GW (1990) Interleukin 1: The first interleukin. Immunol Today 11:13-20. 12. Palacios R, Sideras P, von Boehmer H (1987) Recombinant interleukin 4/BSF-1 promotes growth and differentiation of intrathymic T cell precursors from fetal mice in vitro. EMBO J 6:91-95. 13. Helle M, Brakenhoff JP, de Groot ER, Aarden LA (1988) Interleukin 6 is involved in interleukin l-induced activities. Eur J Immunol18:957-959. 14. Chantry D, Turner M, Feldmann M (1989) Interleukin 7 (murine pre-B cell growth factor/lymphopoietin 1) stimulates thymocyte growth: Regulation by transforming growth factor beta. Eur J Immunol19:783-786. 15. Uyttenhove C, Simpson RJ, van Snick J (1988) Functional and structural characterization of P40, a mouse glycoprotein with Tcell growth factor activity. Proc Nat1 Acad Sci USA 85:6934-6938. 16. Chen WF, Zlotnik A (1991) IL-lo: A novel cytotoxic T cell differentiation factor. J Immunol147:528-534. 17. Hurme M (1988) Both interleukin 1 and tumor necrosis factor enhance thymocyte proliferation. Eur J Immunol 18:1303-1306. 18. Hedger MP, Drummond AE, Robertson DM, Risbridger GP, de Kretser DM (1989) Inhibin and activin regulate sHthymidine uptake by rat thymocytes and 3T3 cells in vitro. Mol Cell Endocrinol61:133-138. 19. Seder 0, Hellstrom PM (1989) The tachykinins neurokinin A and physalaemin stimulate murine thymocyte proliferation. Int Arch All Appl Immunol90:91-96. 20. Gustafsson BE (1959) Lightweight stainless steel systems for rearing germfree animals. Ann NY Acad Sci 78:17-28. 21. Midtvedt T, Carlstedt-Duke B, Hoverstad T, Midtvedt AC, Norin KE, Saxerholt H (1987) Establishment of a biochemically active intestinal ecosystem in ex-germfree rats. Appl Environ Microbial 53:286&2871. 22. Bradford MM (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72:248-254. 23. Khan SA, Seder 0, Syed V, Gustafsson K, Lindh M, Ritz&n EM (1987) The rat testis produces large amounts of an interleukin-l-like factor. Int J Androl10:459-503. 24. Seder 0, Madsen K (1988) Stimulation of chondrocyte DNA synthesis by interleukin-1. Br J Rheumatol27:21-26.