Immune parameters are affected differently after cyclosporine A exposure in Fischer 344 rats and B6C3F1 mice: implications for immunotoxicology

IOXICOIOGY ELSEVIER

Toxicology 94 (1994) 231-245

Immune parameters are affected differently after cyclosporine A exposure in Fischer 344 rats and B6C3F1 mice: implications for immunotoxicology'A" C. Blot a'b, H. Lebrec a, R. Roger a, R. Bohuon a'b, M. Pallardy *a'b aLaboratoire de Toxicologie, 1NSERM CJF 93-01, FacultO de Pharmacie Paris XI, rue Jean-Baptiste ClOment, F-92296 Ch6tenay-Malabry, France bDOpartement de Biologie Clinique (+14), lnstitut Gustave Roussy, rue Carnille Desmoulins, F-94805 Villejuif Cedex, France

Received 16 February 1994; accepted 15 May 1994

Abstract Knowledge of interspecies differences, commonly evaluated in other disciplines such as carcinogenesis, is a prerequisite for an appropriate assessment of immunotoxicological risks. The purpose of this study was to assess interspecies differences following exposure of Fischer 344 rats and B6C3F1 mice to cyclosporine A. These animals were exposed daily to cyclosporine A by oral gavage at 0, 5, 10, 25 mg/kg/day for 14 consecutive days. The results showed that splenocytes lymphoproliferation in response to concanavalin A or phytohemagglutinin, and lgM antibody-forming cells to sheep red blood cells, were affected in both species. Cytotoxic T-lymphocyte activity and mixed lymphocyte response were significantly inhibited in the rat following cyclosporine A exposure while they remained unaffected in the mouse. In contrast, natural killer cell activity was significantly depressed in the B6C3FI mouse but not in the Fischer 344 rat. The discrepancies between the two species in cytotoxic T-lymphocyte activity and mixed lymphocyte response assays could partially be explained by the constantly higher blood level of cyclosporine A in the rat than in the mouse. When these tests were performed using rat and mouse splenocytes exposed to cyclosporin A in vitro (10 -9 to 10-5 M) it was possible to correlate in vivo and in vitro data for concanavalin A- and phytohemagglutinininduced lymphoproliferation and for cytotoxic T-lymphocyte activity but not for mixed lymphocyte response. Natural killer activity was 10-fold more sensitive in mice than in rats in

* Corresponding author, Laboratoire de Toxicologie,INSERM CJF 93-01, Facult6 de Pharmacie Paris XI, rue Jean-Baptiste Cl&nent, F-92296 Chfitenay-Malabry, France. These results have been partially presented elsewhere: Toxicologist (1993) 13, 1353. 0300-483X/94/$07.00 © 1994 Elsevier Science Ireland Ltd. All rights reserved SSDI 0300-483X(94)02889-3

232

C Blot et al. /Toxicology 94 (1994) 231-245

vitro but these results did not clarify the in vivo difference. In conclusion, these results emphasize that the utilization of more than one species should be considered when assessing immunotoxicity.

Keywords: lmmunotoxicology; Immunosuppression; Cyclosporine A

1. Introduction Since the 1980s, considerable studies of methodology validations have been directed to develop animal models for the assessment of drug immunotoxicity and their relevance to humans (Van Loveren and Vos, 1989; Luster et al., 1988). Whereas methods for risk assessment of immunosuppressants are currently available, the choice of the reference species in such studies needs to be considered as for carcinogenicity studies which use at least two strains of rodents (Robens et al., 1989). With regard to the assessment of hazard to humans, comparison of data obtained between at least two species of research animals needs to be initiated as to draw systematic interspecies profiles. Commonly, the rat is the experimental animal most used in toxicology and pharmacology. In immunology, mice are more frequently used, but in some areas of immunological research, such as immunogenetics, transplantation, and cancer risk assessment, the rat is preferred (Gill et al., 1989). In order to identify the most predictive test among a number of immune tests, experiments have been conducted by the National Toxicology Program in USA using over 50 potential immunosuppressive compounds in B6C3FI mice (Luster et al., 1992). However, there are limited data comparing the results obtained in mice with those obtained in rats. Evidence of differences between Fischer 344 rats and B6C3F1 mice in the immunotoxic effect of some chemicals such as 2-methoxyethanoi, but not pharmaceutical drugs, has been reported (Smialowicz et al., 1992). Cyclosporin A (CsA) is a drug extensively used in prevention of transplantrejection. It can be used as a prototype in immunotoxicological assessment because of its selective action on lymphocyte function (Di Padova, 1989). Previous immunological studies have revealed that CsA blocked the calcium- and antigen-dependent proliferation of T cells mainly by inhibiting specific nuclear transcription factors that regulate the expression of early T cell activation genes such as interleukin 2 (IL-2) (Sigal and Dumont, 1992; Liu et al., 1991). It has been recently established that the formation of complexes between CsA and cyclophilin A, an ubiquitous cellular protein, blocks the calcium-activated phosphatase activity of calcineurin which plays a role in the dephosphorylation of a cytoplasmic form of a pre-existing T cell specific cytoplasmic component of the nuclear factor of activated T cells (NFATc) (Flanagan et al., 1991). The absence of dephosphorylation of NFATc results in the inhibition of its translocation into the nucleus and thus in the lack of IL-2 gene expression (Schreiber and Crabtree, 1992; Liu, 1993). Because of the specific action of CsA on CD4+ lymphocytes, the intensity of immunosuppression reflects the altered immune functions of immunocompetent cells dependent on interactions with CD4+ cells (Cockburn, 1987).

C. Blot et al. / Toxicology 94 (1994) 231-245

233

The present study was designed to compare the in vivo and in vitro immunosuppressive potential of CsA on splenocytes of inbred Fischer 344 female rats and B6C3FI female mice. For the in vivo protocols, rats and mice received a similar oral dose of CsA daily for 14 consecutive days, corresponding to 5, 10 and 25 mg/kg/day. No overlapping toxicity was found for these doses, as mainly revealed by the absence of renal toxicity and modification in body weight and body weight gain: thus risks of non-specific immunotoxic effects were avoided. For in vitro studies, splenocytes were exposed to CsA at concentrations ranging from 10 -9 M to 10 -5 M. Our approach has implied the choice of several immunological tests previously validated in our laboratory (Lebrec et al., 1994; also Lebrec et al., unpublished data) and in others (Van Loveren and Vos, 1989; Luster et al., 1988; Smialowicz, 1987). The immune parameters chosen for this evaluation were lymphoproliferation in response to T cell mitogens (concanavalin A, (?on A; phytohemagglutinin, PHA), mixed lymphocyte reaction (MLR), cytotoxic T lymphocyte activity (CTL), lgM antibodyforming cells (AFC) to sheep erythrocytes (SRBC) and natural killer cell activity (NK).

2. Materials and methods

2.1. Experimental animals Female strain Fischer 344 rats and B6C3F1 mice were used at 8-12 weeks of age. They were supplied by Iffa Credo laboratory (L'Arbresle, France) and by Janvier laboratory (Le Genest, France), respectively. Animals were acclimatized to laboratory conditions for 7 days before starting the study. Mice and rats were housed in polycarbonate cages containing heat treated pine shaving (UAR, Villemoisson, France) and maintained under non-pathogen-free conditions. Food (A0410, UAR, France) and water were given ad libitum. The ambiant temperature was 20 ± I°C, humidity 60 ± 5%, and a 12-h light/12-h dark cycle was provided. Female Brown Norway rats (Janvier) and DBA/2 mice (Iffa Credo) were used as a source of allogeneic spleen cells for the mixed lymphocyte reaction. 2.2. In vitro and in vivo exposure to cTclosporin A In vivo: CsA (100 mg/ml; Sandimmune ®) was purchased from Sandoz laboratory (Rueil-Malmaison, France) and diluted in olive oil. The doses of CsA used for rats and mice were 5, 10, and 25 mg/kg/day and olive oil was used as the control. Groups of six rats and groups of seven mice were dosed each day by oral gavage for 14 consecutive days. At day 7, animals were weighed and the concentration of CsA adjusted according to weight gain. In vitro: CsA (a kind gift from Sandoz) was dissolved in ethanol at 10 -2 M and diluted in RPMI 1640 for studies of the in vitro action of CsA on naive splenocytes. Splenocytes were always pretreated with CsA at concentrations ranging from 10 -9 M to 10 -5 M for 1 h before the activation signal was added. Ethanol (0.1%) was used as the control. The final concentration of the diluent (ethanol) was always equal to 0.1%, and CsA remained in the culture media throughout the assay.

234

c. Blot et al. / Toxicology 94 (1994) 231-245

2.3. Preparation o f cells suspension Non-CsA-exposed (in vitro study) or CsA-exposed (ex vivo study) rats and mice were scarified by CO2 anoxia and spleens were removed by sterile dissection. A splenocyte (SC) suspension was then prepared by pressing spleens with a syringe valve and washed once in complete medium (CM), i.e. RPMI 1640 medium containing 22 mM HEPES buffer (Eurobio, Les Ulis, France), 2 mM L-glutamine (Eurobio), 0.1 mg/ml streptomycin (Eurobio), 100 U/ml penicillin (Eurobio), 5.10 -5 M 2-mercaptoethanol (Sigma, St. Louis, MO) and 5% heat-inactivated fetal bovine serum (FBS) (J. Bio, France). Cells were then suspended in CM and counted using a Coulter counter (Coulter S+, Coulter, France). For in vivo protocols, weights of the thymus and spleens were recorded for each animal and a single cell suspension of thymocytes and splenocytes was prepared for counting. Cell viability was always found around 90% as assessed by trypan blue exclusion. Following in vitro exposure to CsA, cell viability was never statistically different from control cells. 2.4. Determination of CsA concentrations, creatinine blood levels" and peripheral blood leukocyte counts On day 13, and 24 h after the last gavage, blood samples with EDTA as an anticoagulant were collected from the retro-orbital venous plexus of rats and mice, slightly anesthetized with ether. CsA concentrations were assessed in rats and mice in whole blood by a fluorescence polarization immunoassay (FPIA; Abott Laboratory, France) that involves one selective monoclonal antibody and specifically measures the parent drug (Yatscoff et al., 1990). Creatinine concentration, as a biological marker of CsA-induced nephrotoxicity, was measured in plasma using a CX7000 apparatus (Beckman, France). Peripheral blood leukocyte counts were performed with a Coulter S+ apparatus (Coulter). 2.5. Immune parameters Lymphoproliferative responses to mitogens. Proliferation of naive or CsA-exposed splenocytes, in response to T-cell mitogens, concanavalin A (Con A; Sigma) and purified phytohemagglutinin (PHAp; Sigma), was evaluated using a microculture assay. Cells were cultured in CM (4 x 105 cells per well) using mitogen at optimal concentration. After 48 h, the culture was pulsed for 6 h with 0.5 t~Ci of [3H]thymidine (specific activity, 5 Ci/mmol, Amersham, Les Ulis, France) per well and harvested with a semi-automatic cell harvester (Skatron, OSI, France). Radioactivity was then measured in a scintillation counter (Beckman, France). Mixed lymphocyte reaction ( M L R ) . An unidirectional MLR was performed using irradiated (2000 Rad-X irradiation) DBA/2 and Brown Norway (BN) splenocytes as stimulatory cells for B6C3F1 mice and F344 rats respectively. F344 and B6C3F1 splenocytes were employed at 2 x l 0 6 cells/ml and cocultured in CM with 2 x l0 b BN or 5 x l 0 6 DBA/2 splenocytes respectively. After 4 days of culture for rats and 5 days for mice, cultures were pulsed for 6 h with 0.5 t~Ci of [3H]thymidine per well and harvested, as described above for mitogenic proliferations. In this experiment the following controls were used: inactivated stimulatory cells alone and with Con A. Cytotoxic T lymphocyte activity (CTL). CTL function was assessed by Brunner's

c. Blot et aL /Toxicology 94 (1994) 231-245

235

modified technique (Brunner et al., 1976). CTL was generated in 20-ml cocultures of 30 x 10 6 splenocytes with 3 × 105 irradiated stimulatory cells (2000 Rad-X irradiation) in Minimal Eagle's Medium (MEM, Gibco, France) supplemented with 2 mM L-glutamine, 0.1 mg/ml streptomycin, 100 U/ml penicillin, 2 × 10 -5 M 2mercaptoethanol and 10% heat-inactivated FBS. Stimulatory cells were the WFu/G1 rat lymphoma cell line (kindly provided by Dr R. Smialowicz) and the P815 mouse mastocytoma cell line (kindly provided by Dr G. Burleson) for rats and mice, respectively. Four days later for rats and five for mice, lytic activity was tested with a standard 4-h 51Cr (Amersham) release assay using 51Cr-labeled WFu/GI and 51Crlabeled P815 cells for rats and mice, respectively. Splenocytes and target cells (2 x 10 5 cells/ml, 100/~l/well) were cocuitured for 4 h at a 100:1 effector-to-target ratio in round-bottomed 96-well plates (Nunc, ATGC, France). Supernatants were then collected and radioactivity quantitated in a 3,-counter (Beckman). The percentage of specific cytolysis of target cells was calculated as follows: percentage cytotoxicity = experimental release - spontaneous release maximum release - spontaneous release The maximal release and the spontaneous release were obtained by incubating target cells in 0.1% Triton X-100 (Sigma) and in MEM 10% FBS, respectively. Natural killer cell activity ( N K ) . Fresh naive or CsA-exposed splenocytes from rats and mice were tested for NK activity using the YAC-1 mouse lymphoma as a target cell (a kind gift from Dr R Smialowicz) in a standard 4-h 51Cr release assay (Kiessling et al., 1975). Briefly, YAC-I cells were 51Cr-labeled in 1 ml FBS for 1 h and washed three times in PBS 2% FBS, and resuspended in CM. Splenocytes and target cells (105 cells/ml, 100 p.l/well) were cocultured at 100:1, 50:1 and 25:1 effector-to-target ratios in round-bottomed 96-well plates (Nunc). Plates were incubated for 4 h at 37°C in a humidified 5% CO2 incubator. Supernatants were then collected and quantitated in a ,),-counter (Beckman). The percentage of specific cytolysis was determined using the formula as described above for CTL activity and lyric units calculated as described by Grabstein (1980). Briefly, the lytic units were calculated after establishing a linear relationship between the percentage 5~Cr release and the E:T ratio. Lytic units (LU) were calculated with the following formula:

LU =

10 7 ETI0%

×

10 4

where 10 7 represents an arbitrary number of effector cells, 104 represents the number of targets per well, and ETi0% represents the E:T ratio necessary to obtain 10% lysis. Antibody response to sheep red blood cells. The primary response of rats in vivo to the T-cell dependent antigen sheep red blood cell (SRBC, Biom6rieux, Lyon,

(2 Blot et al. / Toxicology 94 (1994) 231-245

236

France) was determined using a modification of the standard plaque forming cells assay (PFC) (Cunningham, 1965). Animals were intravenously immunized on day 25 with 1 ml of SRBC suspension (2 x l0 s cells), lgM antibody plaque forming cells were assessed 4 days later using a direct plaque method: splenocytes were introduced into tubes placed in a 47°C water bath containing 350/zl of a 0.5% bacto-agar solution (Sigma) (2 x 105 cells in 100 /A), 0.05% DEAE dextran (Sigma) in Earles's balanced salt solution (EBSS; Sigma) with SRBC and guinea pig complement (Biom~rieux, France). This suspension (250 tA) was poured into a Petri dish, quickly covered with a coverslip, and incubated at 37°C, 5% CO2 for 3 h. The number of plaques was determined and the number of PFC per 10 6 splenocytes calculated. Calculation of the concentration inducing 50% of inhibition (IC5o). IC50 was calculated after establishing a linear relationship between the concentration used and the effect observed expressed as percentage of control. The following formula was used to calculate the IC50: 50% of effect = a + b(ICs0) where a = ~y. .-. . bI2~ . n

b-

n£xy - £ x . Ey ngx 2 - (2x) 2

n represents the number of couples x,y (x, concentration; y, effect expressed as percentage of control). 2.6. Statistical analysis Comparisons between treatments were carried out using one way ANOVA. If statistical significance was reached (P < 0.05), Dunnett's multicomparison of Student's t-test was used. P < 0.05 was considered statistically significant. The same procedure was utilized for in vitro and in vivo studies. 3. Results

3, l. Effects ~[" in vivo exposure to CsA on thymus and spleen weight and cellularity Exposure of Fischer 344 rats and B6C3FI mice to CsA up to 25 mg/kg for 14 consecutive days did not induce nephrotoxicity, as assessed by creatinine blood levels. At0, 5, 10, 25 mg/kg/day ofCsA, creatinine levels were, respectively 46 4- 3, 43 4- 3, 44 ± 3 and 48 4- 9/~mol/l for F344 rats and 9 4- 2, 14 4- 1, 13 ± 0.7, and 6 4- 2 t~mol/1 for B6C3FI mice. In addition, no modification in body weight and bodyweight gain have been found in the two species (data not shown). Peripheral blood leukocyte counts were significantly decreased in Fischer 344 rats at l0 mg/kg (-27% compared with controls) and 25 mg/kg (-32% compared with controls) but only at

237

C. Blot et al. / Toxicology 94 [1994) 231-245 Table 1 Effects of cyclosporin A on immune parameters (immunopathology) following in vivo exposure Cyclosporin A (mg/kg/day)

Fischer 344 rats Spleen weights (mg) Spleen cellularities ( × 106 splenocytes) Thymus weights (mg) Thymus cellularities B6C3FI mice Spleen weights (mg) Spleen cellularities ( × 106 splenocytes) Thymus weights (mg) Thymus cellularities

0

5

10

25

665 + 15 267 4- 9

664 ± 16 276 4- 16

681 ± 15 276 4- 7

627 ± 13 236 ± 10

294 ± 17 404 4- 25

310 + 11 478 + 23

331 ± 11" 556 + 43*

270 4- 8 517 ± 30

143 + 5 127 4- 7

139 4- 7 141 4- 11

164 4- 8 148 ± 5

202 • 16"** 215 4- 23***

53 4- 4 89 + 3

51 + 3 71 4- 7

56 ~ 3 84 ± 11

52 ± 4 65 4- 7

*P < 0.05;

***P < 0.001.

25 mg/kg in B6C3F1 mice (-30% compared with controls; data not shown). Spleen weight and cellularity in B6C3F1 mice were significantly increased at 25 mg/kg (Table 1). 3.2. Lymphoproliferative responses to Con A and P H A following in vivo or in vitro exposure to CsA

In

vivo exposure

to

CsA

induced

140 -]

A

a dose-dependent

decrease

in

lyre-

140

[

120

120

I(X)

100

80

80

60

60

"

40

40

r~

20

20

q +1 eo t)

~D

0

0 5

10

25

5

10 CsA

25

(mg/kg/d)

Fig. I. Lymphoproliferative response to Con A (A) and PHA (B) following in vivo exposure to CsA ([Bold], Fischer 344 rats; [right hatched], B6C3FI mice, *P < 0.05: **P < 0.01; ***P < 0.001).

C. Blot et al. / Toxicology 94 (1994) 231-245

238

phoproliferative responses to Con A and PHA in both Fischer 344 rats and B6C3F1 mice (Fig. 1). Lymphoproliferation in response to Con A was significantly decreased at 10 and 25 mg/kg in B6C3F1 mice (respectively -35% and -39% compared with controls) and at 25 mg/kg in Fischer 344 rats (-28% compared with controls; Fig. 1A). Lymphoproliferation of lymphocytes in response to PHA was decreased significantly in the two species, only at 25 mg/kg (-48% in Fischer 344 rats and -37"/,, in B6C3F1 mice compared with controls; Fig. 1B). After in vitro exposure of naive splenocytes, we found strikingly comparable results between IC50 values obtained in rats and mice. The calculated ICs0 values were for Fischer 344 rats and B6C3FI mice respectively, 1.0 x 10 -7 M and 2.0 x 10 -7 M for Con A-induced lymphoproliferation and 6.2 x 10 -8 M and 6.3 x 10 -8 M for PHA-induced lymphoproliferation (data not shown). 3.3. Antibody response to S R B C in Fischer 344 rats following in vivo exposure to CsA

PFC response following in vivo exposure of F344 rats to CsA was significantly reduced at 10 mg/kg and dramatically suppressed at the highest dose (Fig. 2). The 5 mg/kg dose did not affect the PFC response. 3.4. M i x e d lymphocyte response Jollowing in vivo or in vitro exposure to CsA

In vivo exposure of F344 rats to CsA resulted in a dose-dependent inhibition of T-cell proliferation (up to 88% compared with controls at 25 mg/kg), whereas no alteration of lymphoproliferation was observed in B6C3FI mice at any dosage group tested (Table 2). Whatever the daily dose of CsA, CsA blood levels were always higher in Fischer 344 rats than in B6C3FI mice (Table 2). However, unlike in vivo results described above, we observed a dose-dependent inhibition of lymphocyte proliferation in both species after in vitro exposure of naive splenocytes to CsA. The

1200 '

800 o

% 400

E 0

5

10

25

CsA ( m g / k g / d ) Fig. 2. Effect of in vivo exposure to CsA on antibody response to SRBC in the Fischer 344 rats (**P < 0.001 compared with control).

C. Blot et al. / Toxicology 94 (1994) 231-245

239

Table 2 Mixed lymphocyte response in B 6 C 3 F I mice and Fischer 344 rats following in vivo exposure to CSA CsA (mg/kg/day)

5 10 25

Fischer 344 rats

B6C3FI mice

CsA blood concentration (M x 10 -8 )

Cell proliferation a (n = 6)

CsA blood concentration (M x 10 -8 )

Cell proliferation a (n = 7)

6 4- 0.7 23 4- 3 83 4- 7

73 ± 9* 47 4- 8*** 12 + 2***

0.6 4- 0.1 ! 4- 0.1 12 + 1.6

106 4- 10 92 4- 6 102 4- 7

aValues are expressed as a percentage of control 4- S.E.M. *P < 0.05; ***P < 0.001.

IC50 values found were 1.06 x 10 -~° for Fischer 344 rats and 1.08 x 10 -8 M for B6C3F1 mice (Table 3). These results showed that the sensitivity of the [F344 x BN] model to CsA was 100-fold superior to that of the [B6C3FI x DBA/2]. 3.5. Cytotoxic T-lymphocyte activity after in vivo or in vitro exposure to CsA In vivo CsA treatment for 14 days significantly reduced ex vivo CTL activity of rats but not that of mice (Table 4). In rats, the decrease in CTL activity against the allogeneic WFu/G1 target cell line occurred at 25 mg/kg corresponding to a CsA blood level of 83 x 10 -8 M. In contrast, and whatever the dose group considered, no alteration of ex vivo CTL activity in B6C3FI mice against the allogeneic P815 murine mastocytoma target cell line was observed (Table 4). CsA blood levels of mice were always below that of rats at all doses assessed (Table 4). Moreover, in vitro exposure of naive splenocytes to CsA showed identical sensitivity for the CTL test, when the two species were compared. The ICs0 values were 1.27 x 10 -6 M for Fischer 344 rats and 0.96 x 10 -6 M for B6C3FI mice respectively (Table 5). 3.6. Natural killer activity after in vivo and in vitro exposure to CsA N K activity directed to YAC lymphoma cells was significantly affected in B6C3FI

Table 3 Mixed lymphocyte response in B6C3FI mice and Fischer 344 rats following in vitro exposure to CSA Cyclosporin A (M)

Fischer 344 rats B6C3FI mice

IC50 (M)

10 -9 M

10 -8 M

10 -7 M

41 4- 15 ***~'b 70 ± 11 **a,b

19 + 3*** 55 ± 17"**

11 4- 6*** 31 + 11"**

aValues are expressed as a percentage of control + S.E.M. bMean 4- S.E.M. of three independent experiments. *P < 0.05; **P < 0.01; ***P < 0.001.

1.06 x 10 -1° 1.08 x 10 -8

240

C. Blot et al. / Toxicology 94 (1994) 231-245

Table 4 C y t o t o x i c T - l y m p h o c y t e activity in B6C3F1 mice a n d F i s c h e r 344 r a t s f o l l o w i n g in vivo e x p o s u r e to C S A CsA (mg/kg/day)

5 10 25

Fischer 344 r a t s

B6C3F1 mice

CsA blood concentration (M x 10 -8 )

C T L activity a (n = 6)

CsA blood concentration (M X 10 -8 )

C T L activity a (n = 7)

6 • 0.7 23 :~ 3 83 4- 7

97 + 11 112 4- 5 66 4- 8**

0.6 4- 0.1 1 :~ 0.1 12 4- 1.6

109 + 6 96 4- 9 107 + 4

aValues are expressed as a p e r c e n t a g e o f c o n t r o l + S.E.M. **P < 0.01.

Table 5 C y t o t o x i c T - l y m p h o c y t e activity in B 6 C 3 F I mice a n d Fischer 344 rats f o l l o w i n g in vitro e x p o s u r e to C S A C y c l o s p o r i n A (M)

Fischer 344 rats B 6 C 3 F I mice

IC50 (M)

10 -7 M

10 -6 M

10 -5 M

96 4- 18 a.b 87 + 3 a'b

55 + 8** 57 + 2**

12 ~ 3*** 4 + 2***

1.27 x 10 -6 0.96 x 10 6

aValues are expressed as a p e r c e n t a g e o f c o n t r o l 4- S.E.M. CMean + S.E.M. o f three i n d e p e n d e n t experiments. **P < 0.01; ***P < 0.001.

Table 6 N a t u r a l killer activity in B6C3F1 mice a n d Fischer 344 rats following in vivo e x p o s u r e to C S A CsA (mg/kg/day)

Fischer 344 rats CsA blood concentration (M x 10 ~s)

0 5

0 2 4. 0.2

10

16 + 2

25

50 + 11

B 6 C 3 F I mice LUI0 a (n = 6)

CsA blood concentration (M x 10 -s)

LUI() ~ (n = 7)

26 4. 3 21 4- 1 (83%) b

0 0.6 + 0.1

28 + 3 19 + 2* (68%)

24 + 4 (93%) 26 4. 3

I + 0.1

17 4- 4* (61%) 19 + 1"

(102%) aValues are expressed in lytic units 10% (LUI0) 4. S.E.M. b p e r c e n t a g e o f c o n t r o l ± S.E.M. *P < 0.05.

12 ± 1.6

(68%)

241

C Blot et al. / Toxicology 94 (1994) 231-245 Table 7 Natural killer activity in B6C3FI mice and Fischer 344 rats following in vitro exposure to CSA Cyclosporin A (M)

IC50 (M)

Vehicle

10 -7 M

10 -6 M

10 -5 M

Fischer 344 rats

37 ± 8 a'¢

25 ± 4 (68%)b

19 ± 2* (51%)

15 4- 5*** (40%)

1.7 × I0-6M

B6C3FI mice

24 ± 8 a'c

12 ± 4* (50%)

8 ± 1"* (33%)

7 + 1"** (29%)

1.0 x 10 7 M

aValues are expressed in lytic units 10% (LU10) ± S.E.M. bPercentage of control ± S.E.M. CMean ± S.E.M. of three independent experiments. *P < 0.05; **P < 0.01; ***P < 0.001.

mice in all dosage groups (Tables 6 and 7). However, NK activity directed to the same target cell was not affected in Fischer 344 rats even though CsA blood levels were higher in Fischer 344 rats than in B6C3F1 mice (Table 6). When splenocytes were exposed to CsA in vitro, we found that NK activity in B6C3F1 mice was 10 times more sensitive to CsA than that measured in the Fischer 344 rat. The calculated IC50 values were 1.7 × 10 -6 M and 1.0 x 10 -7 M for Fischer 344 rats and B6C3FI mice respectively (Table 7).

4. Discussion The aim of this study was to assess species differences between two inbred strains of rodents widely employed in immunotoxicology testing, namely the Fischer 344 rat and the B6C3F1 mouse. CsA was employed here as a reference immunosuppressive drug because it is specific for the immune system and not cytotoxic (for review, see Sigal and Dumont, 1992). The choice of these two species of rodents was motivated by their previous use in numerous interlaboratory validations of iminunotoxicological methods mainly directed and developed by the NTP (Luster et al., 1988) and in our laboratory (Pallardy et al., 1991; Lebrec et al., 1994; also Lebrec et al., unpublished data). To investigate the effects of CsA on immunity, we used a panel of tests: lymphoproliferation in response to T-cell specific mitogens (Con A, PHA), mixed lymphocyte response, IgM antibody-forming cells to SRBC, cytotoxic Tlymphocyte activity, and natural killer cell activity. In the in vivo approach, mice and rats received the same daily regimen of CsA for 14 days using the following doses (0, 5, 10, and 25 mg/kg/day). For in vitro assays spleen cells were exposed to concentrations ranging from 10 -9 to 10 -5 M during experiments. Results found in the in vivo study showed three distinct situations. (1) Immune parameters were affected in both species; we observed a significant reduction in lym-

242

C. Blot et al. / Toxicology 94 (1994) 231-245

phoproliferative responses to both Con A and PHA as well as a decrease in IgM AFC response to SRBC. (2) Immune parameters were affected in rats but not in mice; the MLR assay was inhibited at the three doses tested in Fischer 344 rats while it remained unaffected in B6C3F1 mice; CTL activity was only modified in Fischer 344 rats and at the highest dose tested. (3) Immune parameters were affected in mice but not in rats; N K cell function was altered in mice at 5, 10, and 25 mg/kg but not in rats. Reductions in ex vivo lymphoproliferative responses to PHA and Con A were observed in the two strains of rodents, as extensively described by Borel and coworkers (Borel et al., 1977; Borel, 1981, 1989). Con A-induced lymphoproliferation was more affected in B6C3F1 mice following in vivo exposure to CsA than in Fischer 344 rats although the CsA blood level in mice was always below that of rats. After in vitro exposure to CsA, splenocyte proliferation in response to Con A was equally affected in both species thus ruling out a possible difference in cellular sensitivity to CsA in this case. Earlier studies with female B6C3F1 mice in our laboratory demonstrated that in vivo CsA exposure resulted in an alteration of IgM AFC response to SRBC, as we had observed for Fischer 344 rats (Lebrec et al., unpublished data). However, compared with these data, rat PFC response to CsA at 25 mg/kg was more affected than that of mice, indicating that the sensitivity of the assay was highest in the rat. The difference between the two species found for this test could be directly imputable to the constantly higher CsA blood concentrations in rats. After in vivo treatment, a decrease in lymphoproliferation in the MLR assay was only observed in Fischer 344 rats. This ex vivo interspecies difference could be attributable to the notable finding that after in vitro exposure to CsA, the sensitivity of the rat MLR assay is 100-fold more elevated than that of mice, as depicted by the calculated IC50 values. For example, at 5 mg/kg corresponding to a CsA blood concentration of 6.0 × 10 -8 M, we noted a decrease of 27% in ex vivo lymphoproliferation in rats but no inhibition of murine T lymphoproliferation at 12 × 10-8 M of CsA concentration in the blood, corresponding to the 25-mg/kg dose. Moreover, differences in assay conditions (i.e. stimulatory ceils) must be taken into consideration in our study since Ryffel et al. (1988) suggested that the considerable variability of 1C50 values, described in the literature for CsA and its metabolites, is certainly due to distinct experimental procedures. The inhibitory effect of CsA on CTL function has been widely described in the literature (Borel, 1989), and could stem from a functional decrease in CD4+ helper T cells, especially in the production of cytokines which promote the proliferation (IL-2) and differentiation (1FN-3,, IL-4, IL-6) of resting CTL precursors into antigen-specific and mature CTL (Sigal and Dumont, 1992; Bubeck et al., 1989). After in vivo exposure, we noted a difference between species, namely a 34% decrease in rat CTL activity at 25 mg/kg whereas no corresponding decrease was observed in murine CTL activity at the same dose. Interestingly, at this dosage, the effect of CsA on ex vivo rat CTL activity was similar to that of calculated after in vitro exposure (-42% at 83 x 10 -8 M, this value being extrapolated from the dose-response curve). Furthermore, the CsA blood level (12 × 10 -8 M) found in

c. Blot et al. / Toxicology 94 (1994) 231-245

243

the mouse at 25 mg/kg, corresponded to no effect level in in vitro experiments. Fukuzawa and Shearer (1989) previously noted, however, that the ex vivo generation of CTL in response to alloantigens could be detectably affected from 75 mg/kg upward in B6C3F1 mice. Thus, these differences observed between species in the CTL assay following in vivo exposure to CsA could likely be attributed to different pharmacokinetic profiles. Under these in vivo conditions of exposure to CsA, we noted a persistent suppression of NK cell activity in B6C3F1 mice which did not occur in Fischer 344 rats. We have found a similar effect on rat NK using the Sprague-Dawley rat strain and identical doses (unpublished data). However, our results were in agreement with those obtained by Yanagihara and Adler (1982) with respect to the mouse but in contradiction with those described by others [30,31]. In one study, murine NK activity after 14 days of exposure to CsA administered i.p. at doses ranging from 1.0 to 100 mg/kg was not altered (Murray et al., 1991). In another study using C57BL/6 mice, in which CsA was added in the basal diet for 4 weeks, NK activity was not affected although a very high CsA blood level was noted (Yabu et al., 1991). In our report, this ex vivo difference between species was not expected since the CsA blood levels of rats at 5, 10, and 25 mg/kg were always higher than those of mice. In addition, in vitro NK cell activity was significantly affected in the two species despite increased sensitivity of murine cells, as attested by the IC50 values. Indeed, in Fischer 344 rats, the CsA blood level measured at 25 mg/kg (50 x 10 -8 M) corresponded to an effective in vitro concentration. With regard to the wide interspecies differences in physiological and pharmacokinetic parameters which have been extensively described in the literature for CsA, it appears that this discrepancy could likely be explained by intrinsic parameters such as metabolisation (Maurer, 1985; Sangalli et al., 1988). Further work is in progress to investigate this possibility. In conclusion, this report indicates that the interspecies differences observed following in vivo CsA exposure using a panel of currently validated tests, could depend on pharmacokinetic profiles (as revealed in CTL assays), cellular sensitivity (as depicted in MLR assays) or metabolisation (as hypothetized for NK cell assays). We believe that this study could have an impact on immunotoxicological testing and may present exploitable opportunities for insights into species differences regarding the toxicity of pharmaceutical drugs with selective immunosuppressive potential like CsA.

Acknowledgments The authors thank Patrice Ardouin, Gilberte Bretou and Mallory Perrin for their constant help and Drs Gary Burleson, Robert House and Ralph Smialowicz for providing us the P815, YAC and WFu-G1 cell lines.

References Borel, J.F. (1981) CyclosporinA -- present experimental status. Transplant. Proc. 13, 344-350. Borel, J.F. (1989) Pharmacologyof cyclosporine(Sandimmune).IV. Pharmacologicalproperties in vivo. Pharmacol. Rev. 41, 339-351.

244

C Blot et al. / Toxicology 94 (19943 231-245

Borel, J.F., Feurer, C., Magnee, C. and St6helin, H. (19773 Effects of the new anti-lymphocyte peptide cyclosporin A in animals. Immunology 32, 1017-1025. Brunner, K.T., Engers, H.D. and Cerrottini, J.C. (1976) The 5tCr-release assay as used for the quantitative measurement of cell mediated cytolysis in vitro. In: B.R. Bloom and J.R. Davis (Eds), In Vitro Methods of Cell-Mediated and Tumor Immunity, Academic Press, New York, pp. 423-428. Bubeck, R., Miethke, T., Heeg, K. and Wagner, H. (19893 Synergy between interleukin 4 and interleukin 2 conveys resistance to cyclosporin A during primary in vitro activation of murine CD8 cytotoxic T cell precursors. Eur. J. lmmunol. 19, 625-631. Cockburn, I. (19873 Assessment of the risks of malignancy and lymphomas developing in patients using Sandimmune. Tansplant. Proc. 19, 885-890. Cunningham, A.J. (1965) A method of increased sensitivity for detecting single antibody-forming cells. Nature 207, 1106-1107. Di Padova, F.E. (1989) Pharmacology of cyclosporine (Sandimmune). V Pharmacological effects on immune function: in vitro studies. Pharmacol. Rev. 41, 373-405. Flanagan, W.M., Corthesy, B., Bram, R.J. and Crabtree, G.R. ( 19913 Nuclear association of a T-cell transcription factor blocked by FK-506 and cyclosporin A. Nature 352, 803-806. Fukuzawa M. and Shearer, G.N. (19893 Effect of cyclosporin A on T cell immunity. 1. Dose-dependent suppression of different murine T helper cell pathway. Eur. J. lmmunol. 19, 49-56. Gill, T.J. III, Smith, G.G., Wissler, R.W. and Kunz, H.W. (19893 The rat as an experimental animal. Science 245, 269-276. Grabstein, K. (1980) Cell-mediated cytolytic response. In: B.B. Mishell and S.M. Shiigi (Eds), Selected Methods in Immunology, Freeman, New York, pp. 124-135. Kiessling, R., Klein, E. and Wigzell, H. (19753 'Natural' killer cells in the mouse. I. Cytotoxic ceils with specificity for mouse Moloney leukemia cells. Specificity and distribution according to genotype. Eur. J. Immunol. 5, 112-118. Lebrec, H., Blot, C., Pequet, S., Roger, R., Bohuon, C. and Pallardy, M. (19943 lmmunotoxicologica[ investigation using pharmaceutical drugs. In vivo evaluation of immune effects. Fundam. Appl. Toxicol. 23, 159-168. Liu, J., Farmer, I.D., Lane, W.S., Friedman, J., Weissman, I. and Schreiber, S. (19913 Calcineurin is a common target of cyclophilin-cyclosporin A and FKBP-FK506 complexes. Cell 66, 807-815. Liu, J. (19933 FK506 and cyclosporin, molecular probes for studying intracellular signal transduction. Immunol. Today 14, 290-295. Luster, M.I., Munson, A.E., Thomas, P.T., Holsapple, M.P., Fenters, J.D., White K.L. Jr, Lauer, L.D., Germolec, D.R., Rosenthal, G.J. and Dean, J.H. (19883 Development of a testing battery to assess chemical-induced immunotoxicity: National Toxicology Program's guidelines for immunotoxicity evaluation in mice. Fundam. Appl. Toxicol. 10, 2-19. Luster, M.I., Portier, C., Pait, D.G., White, K.L. Jr, Gennings, G., Munson, A.E. and Rosenthal, G.J. (1992) Risk assessment in lmmunotoxicology. I. Sensitivity and predictability of immune tests. Fundam. Appl. Toxicol. 18, 200-210. Maurer, G. (19853 Metabolism of cyclosporine. Transplant. Proc. 17, 19-26. Murray, M.J., Horn, P.A. and Thomas, P.T. (19913 Specificity of host resistance assay for identifying immunomodulatory compounds. Toxicologist 11, 208. Pallardy, M., Lebrec, H., Blot, C., Burleson, GR. and Bohuon, C. (t991) In vitro evaluation of druginduced toxic effects on the immune system as assessed by proliferative assays and cytokine production. Eur. Cytokine Net. 2, 201-206. Robens, J.F., Piergorsch, W.W. and Schueler, R.L. (19893 Methods of testing for carcinogenicity. In: A.W. Hales (Ed), Principles and Methods of Toxicology, Raven Press. New York, pp. 251-273. Ryffel, B., Foxwell, B.M.J., Mihatsch, M.J., Donatsch, P. and Maurer, G. (19883 Biologic significance of cyclosporine metabolites. Transplant. Proc. 20, 575-584. Sangalli, L., Bortolotti, A., Jiritano, L. and Bonati, M. (1988) Cyclosporine pharmacokinetics in rats and interspecies comparison in dogs, rabbits, rats and humans. Drug Metab. Dispos. 16, 749-755. Schreiber, S.L. and Crabtree, G.R. (19923 The mechanism of action of cyclosporin A and FK506. lmmunol. Today. 13, 136-142.

C. Blot et al. / Toxicology 94 (1994) 231-245

245

Sigal, N.H. and Dumont, F.J. (1992) Cyclosporin A, FK-506, and Rapamycin: pharmacologic probes of lymphocytes signal transduction. Annu. Rev. Immunol. 10, 519-549. Smialowicz, R.J., Riddle, M.M., Williams, W.C., Copeland, C.B., Luebke, R.W. and Andrews, D.L. (1992) Differences between rats and mice in the immunosuppressive activity of 2-methoxyethanol and 2-methoxyacetic acid. Toxicology 74, 57-65. Smialowicz, R.J., Rogers, R.R., Rowe, D.G., Riddle, M.M. and Luebke, R.W. (1987)The effects of nickel on immune function in the rat. Toxicology 44, 271-278. Van Loveren, H. and Vos, J.G. (1989) Immunotoxicological considerations: a practical approach to immunotoxicity testing in the rat. In: A.D. Dayan and A.S. Paine (Eds), Advances in Applied Toxicology, Taylor & Francis, London, pp. 143-153. Yabu, K., Warty, V.S., Gorelik, E. and Shinozuka, H. (1991) Cyclosporine promotes the induction of thymic lymphomas in C57BL/6 mice initiated by a single dose of "),-radiation. Carcinogenesis 12, 43-49. Yanagihara, R.H. and Adler, W.H. (1982) Inhibition of mouse natural cell killer activity by cyclosporin A. Immunology 45, 325-33L Yatscoff, R.W., Copeland, K.R. and Faraci, C.J. (1990) Abott TDx monoclonal antibody assay evaluating for measuring cyclosporine in whole blood. Clin. Chem. 36, 1969-1973.