Studies on the effect of cortisol on the primary immune response to sheep erythrocytes in vivo by mouse spleen cells.


HELLIG, HERMIONE and WALDEK, J. F., 1974. Studies on the effect of cortisol on the primary immune response to sheep erythrocytes in vivo by mouse spleen cells. Onderstepoort J. vet. Res. 41, 2938 (1974). Female mice were injected with a suspension of 0,2 ml of 10 per cent v/v sheep red blood cells (SRBC) and sacrificed on the 5th day for recording of spleen mass and assay of plaque-forming cells (PFCs) in the spleens. All PFCs from a pool of five spleens and the individual spleen masses were subjected to rigid statistical analysis to verify whether differences from the control groups were significant (P < 0, 05). Three doses of 4 mg cortisol administered at 24-hourly intervals were given at various times relative to the time of immunization to determine the optimum stage at which to inhibit the formation of PFCs. The greatest inhibitory effect was obtained when the first dose was given at -6 h, with a lesser though still significant depression of PFCs with the -30 h dose regimen. Earlier and later injections had relatively little effect. An attempt was made to determine the response to different dosage levels by injecting doubling doses of steroid starting at the -6 h period. A maximal effect was attained using either 1 to 2 mg cortisol suspension or 0, 5 mg of the soluble pharmaceutical preparations, Efcortolan* and Betsolan*. Inhibition of PFCs appeared to be a more sensitive indicator of steroid action than loss of mass. In addition, spleen mass was determined after injection of either 4 mg cortisol three times or SRBC. During steroid treatment the spleen mass decreased, but the masses started to increase again almost immediately this treatment was discontinued; the maximum antigenic response, as evinced by attainment of maximal mass, occurred at +4 days, when the PFC production reached its peak. INTRODUCTION Since the classical experiment of Selye (1936), it has been known that g1ucocorticoids secreted by the adrenal exert a marked effect on the reduction of lymphatic tissue mass. This effect has been measured quantitatively by numerous workers such as Ringertz, Fagreus & Berglund (1952), Frank, Kumagai & Dougherty (1953), Santisteban & Dougherty (1954), and Elliot & Sinclair (1968), and cortisol has been found to have the most potent action (Santisteban & Dougherty, 1954). However, conflicting results were obtained when attempts to correlate antibody levels with hormone treatment were made (Dougherty, 1952; White, 1958, 1963). Fagreus (1952) using rabbits as experimental animals and Salmonella typhimurium Ag H as antigen, showed conclusively that glucocorticoids also produced a marked reduction in humoral antibody response. Berglund (1952) confirmed this effect using rats. Berglund (1956a) measured the effect of cortisol on the primary haemolysin response (fifth day) in rats and showed the vital importance of the time of hormone administration relative to the time of administration of antigen. Suppressive effects on haemolysin titres were observed only when cortisol was given from 2 to 0 days. He also showed (Berglund, 1956b) that the amount of cortisol needed for suppression was proportional to the dose of sheep red blood cells (SRBC) injected. Later he showed that the period of maximum sensitivity to corticoid in his system was at about -10 hours (Berglund, 1962), and that the effect of steroid could be reversed by injecting spleen or thymus cells shortly after SRBC (Berglund & Fagreus, 1956, 1961). The effects of a single large dose of steroid varying from 2, 5 to 15 mg per 20 mg mass on the haemolysin response to SRBC in mice was studied by Dietrich (1966), Elliott & Sinclair (1968) and Petranyi, Benczur & Alfody (1971); the latter authors were the first to (1) Veterinary Research Institute, Onderstepoort (2) Department of Agricultural Technical Services, Pretoria * Glaxo Laboratories Ltd. Received on 8 February 1974.-Editor 29 demonstrate a drop in plaque-forming cells (PFCs) in the spleens of cortisol-treated mice. Haemolysin levels, at Day + 5 at least, are mainly accounted for by splenic production (Rowley, 1950; Adler, 1965). Following the description by Jerne & Nordin (1963) of the simple assay method for splenic PFCs, which form lgM haemolysin, the more direct assay procedure can now be applied. PFC formation and the depression thereof caused by cortisol, have not previously been thoroughly investigated. Accordingly we have studied time and dose-response relationships using the dual criteria of PFCs/1 06 spleen cells and reduction of spleen mass to assess glucocorticoid effect. The mice used were of the strain maintained in the Onderstepoort colony. MATERIALS AND METHODS Animals Eight-week old female white mice of the strain maintained in the Onderstepoort closed colony were used throughout. In some of the dosage experiments only selected mass-measured mice, 20,5 ± 1, 0 g, were used. In the other experiments their mass ranged from 15,5 to 25,5 g. Treatment of animals The mice were fed a balanced pelleted ration and tapwater ad lib. They were subjected to normal seasonal fluctuations in temperature throughout the 18-month period during which the experiments were conducted, as the animal room is not air-conditioned. No stringent light and dark cycles were maintained. They were housed in batches of from five to seven to a box, the same numbers being used for both controls and for each of one or two test groups per experiment. The only exception was the case of the measurement of spleen mass variation with time after either steroid treatment or immunization, when 10 animals were used for each experimental measurement. Both controls and test groups received SRBC. Test groups were treated with steroids as well. Steroids were injected intraperitoneally (i.p.) at the designated dosage and time relative to the time of STUDIES ON EFFECT OF CORTISOL ON IMMUNE RESPONSE TO SHEEP ERYTHROCYTES BY MOUSE CELLS immunization. Immunization was with 0,2 ml 10 per cent (vfv) SRBC in saline, by intravenous (i.v.)* injection, at 15h00 on day 0. The animals were sacrificed by cervical dislocation at 08h00 on day 5, i.e. four days and 17 h after immunization, unless otherwise specified. Spleen mass Immediately after sacrificing the animals, the spleens were excised and put into 5 ml cold balanced salt solution (BSS) (Mishell & Dutton, 1967). They were then mass-measured to 0,1 mg on wax paper and placed in a fresh solution of BSS. The spleens usually measured from 100 to 200 mg, though chronological "bunching" occurred giving occasional high mass group spleens, i.e. measuring from 200 to 300 mg. About 5 per cent measured more than 300 mg, a few as much as 600 to 1 500 mg. All the spleens that measured more than 300 mg were rejected as they probably indicated a pathological condition and moreover they lie outside the range of the experimentally determined normal population distribution. It should be noted that the mass accepted as normal was considerably higher than the mean of 70 mg reported for specific pathogen-free animals (Gisler, Bussard, Mazie & Hess, 1971). Preparation of cell suspensions For each experimental measurement of either control or test mice, the total cells from five to seven spleens were combined. The maximum PFC response of these mice is at Day 4 (Oellermann, Veterinary Research Institute, Onderstepoort, personal communication, 1973). However, for convenience we adopted the regimen given above, resulting in PFC measurements a day after the maximum. The cell suspensions were obtained by teasing two to three spleens in 5 ml BSS with a forceps and a scalpel. The suspensions so obtained were then aspirated J 0 times using a J 0 ml syringe fitted with a 20 gauge needie. The 5 ml of suspension was then mixed with 5 ml BSS in a conical centrifuge tube and the debris allowed to settle for three minutes. The suspension was then decanted into a graduated conical centrifuge tube and centrifuged at 500 g for 10 minutes (1 000 rpm on a Christ centrifuge, model No. IV KS). The supernatant fluid was decanted, the packed cell volume (PCV) measured, and the precipitated cells taken into suspension in 80 volumes ofBSS. All these operations were carried out at 4 oc. A differential count utilizing nigrosin dye exclusion for viability was then made in a haemocytometer and generally > 99,0 per cent were found to be viable. An aliquot of the cell suspension was then made to contain 1 >< 107 total spleen cells per mi. C omp/emen t Adult guinea-pigs were bled from the heart. After standing at room temperature for 4 h, the clotting blood was left in a refrigerator at 4 oc overnight and then centrifuged. The serum was frozen in 1 ml aliquots and stored at -20 cc for up to three weeks before use. SRBC Forty ml fresh sheep blood was mixed with 5 ml I 0 per cent LiCI and stored at 4 cc for not longer than three days before use. The solution was centrifuged at 4 cc at 750 g for I 0 min, the cells washed with 8, 5 per cent NaCl, and subsequently centrifuged again. After three washes. the packed cell volume was *Mice placed in a 37 ' Coven for 10 minutes-humidity and Yentilation controlled-to expose tail vein 30 measured and either a 7 per cent vfv solution made in BSS for PFC assay, or a JO per cent vfv solution made in isotonic NaCl. For immunization 0,2 ml of the latter solution was injected i. v. This was the standard immunizing dose of SRBC used in our experiments. It has been reported to be maximally effective for IgM haemolysin production (Dietrich, 1966). Assay ofPFC Five dilutions of 1 x 107 ; 7 , 5 x 10 6 ; 5,0 x 106 ; 2, 5 X I 0 6 and I , 0 X 106 cells/ml were made and 0, 1 ml aliquots taken for PFC assay (in triplicate for each dilution) performed with 0, 55 m1 0, 5 per cent agarose gel containing 0 , 05 ml 7 per cent SRBC in BSS on microscope slides (Mishell & Dutton, 1967). Each slide was treated with 0,2 ml 5 per cent guinea-pig serum in BSS as the source of complement and incubated for 3 h in a humidified incubator at 37 oc before plaque counts were made. The elapsed time from sacrifice to incubation was about 1, 5 h. The slide was scored into squares with a diamond pencil and plaque counts from 50 to 450 recorded with a strong background source of illumination. Background counts from unimmunized animals never exceeded 8 per I 0 6 spleen cells and were therefore not routinely determined. Cortisol preparations A suspension of 20 mg/ml cortisol(l) was made by agitating the crystalline steroid powder in {volume ethanol on a rotary vibrator, and then adding ! volume almond oil(2) and vibrating for two min. The oil was stored in a dark bottle at 4 °C. This preparation was used after toxicity testing had shown most commercial preparations to be unsuitable. Routinely a standard dose of0 ,2 ml (4 mg) of this oil suspension of steroid was injected i.p. at 08h00 using a 26 gauge needle. The suspension was stable for at least 30 min. Later aqueous suspensions of the same and lower concentrations were prepared, but these were very unstable and reproducibility could only be attained by continuously vibrating the 1 ml syringe between injections. Betsolan Soluble is supplied as an aqueous solution of 2 mg/ml and was used as such (0 ,25 to 0, 5 ml injected i.p.). Efcortolan is supplied as a powder of monosodium cortisol succinate soluble in 2 ml H 20. This was further diluted with saline to give the necessary concentrations for i.p. injection. Statistical methods Spleen mass and PFCs/ I 0 6 spleen cells for control and test groups were analyzed by analysis of variance (Steel & Torrie, 1960). The assumptions underlying the analysis of variance were then checked by analysing the residuals (Draper & Smith, 1966). If the assumptions seemed to hold, the means were compared by (a) Tukey's method if the number of observations per treatment was the same and (b) Bonferroni's method if the numbers differed (Miller, 1966). If, however, the assumptions seemed unrealistic, a Kruskal-Wallis nonparametric analysis of variance was performed (Reinach, 1966). The different types of analysis were necessary because there were such large variations in the mass data. This was reflected by the high coefficient of variance (C.V.) found in most of these experiments. If a significant (P < 0 , 05) result was obtained tf.e treatments were compared using the method of Dunn (1964). (1) Merck (2) Sweet almond oil manufactured by Harmon & Reinar, S.A. (Pty) Ltd, received as a gift from Dr W. A. Parsons of the firm To plot the spleen mass after immunization or steroid treatment, the curves were fitted to the data by means of polynomials (Graybill, 1961). The adequacy of the model to describe the observed data was tested by means of a "lack of fit" test (Miller, 1966 In some cases some observations were discarded by the standard residual technique (Draper & Smith, 1966). As stated previously, anomalous spleen mass data were rejected on both statistical and biological grounds. The statistical method consists essentially of checking the probability that a certain observation belongs to its group (Draper & Smith, 1966). It is worth noting that the discards obtained for the controls (SRBC-injected) by the statistical method of calculating residuals agreed perfectly with an empirical distribution study, made at the conclusion of the experiments, on all the control SRBC data. RESULTS A chronological study of variation in spleen mass after HERMIONE HELLIG & J. F. WALDEK experiment 10 animals were sacrificed every 24 h. The results are depicted in Fig. 2. The curve obtained by using all but four very high observations shows a maximum at Day 4, a result in accord with the maximum of PFC production. However, there was reason to doubt the mathematical validity of this curve as the "lack of fit" test was significant, indicating that the curve does not fit the data well. After abstracting data by the standardized residual technique a rather different curve was obtained. Although the ''lack of fit" " test for this curve was indeed nonsignificant, its shape is rather difficult to explain on biological grounds. Since we encountered such large variations in individual spleen masses and the rise is only of the order of 25 per cent it is possible that the time curve can only be validated by the use of a much larger number of mice.

11 Figures and Tables

Cite this paper

@article{Hellig1974StudiesOT, title={Studies on the effect of cortisol on the primary immune response to sheep erythrocytes in vivo by mouse spleen cells.}, author={H R Hellig and J F Waldek}, journal={The Onderstepoort journal of veterinary research}, year={1974}, volume={41 1}, pages={29-37} }