key: cord-0252687-ay83obbp
authors: Ratkay, L G; Zhang, D; Tonzetich, J; Levy, J G; Waterfield, J D
title: Evaluation of a model for post-partum arthritis and the role of oestrogen in prevention of MRL-lpr associated rheumatic conditions.
date: 1994-10-03
journal: Clin Exp Immunol
DOI: nan
sha: a88bb74804613243fcb1d29a18e484afad1f6911
doc_id: 252687
cord_uid: ay83obbp

Sixty-eight percent of female MRL-lpr mice developed a post-partum exacerbation of their mild spontaneous arthritis within 30 days of parturition. The flare became evident between 5 and 15 days after delivery. Histologically it was characterized by a significant increase of subsynovial inflammation and synovial hyperplasia without changes in the level of cartilage and bone erosion. Immunohistologically, marked subsynovial and frequent synovial staining of MHC class II bearing cells was noted, along with the sporadic presence of CD3, CD4, and CD43 receptor-bearing cells in the subsynovium. Injection of physiological levels (0.08 mg/kg) of estradiol on days 2, 3, 9, 15 and 20 post-partum delayed and reduced the flare to 23% of the animals. Administration of pharmacological amounts (0.4 mg/kg per day for 2 weeks following Freund's complete adjuvant injection) prevented adjuvant-enhanced arthritis, reducing the incidence from 67% to the baseline 21% level. Deleterious changes in the underlying systemic lupus erythematosus (SLE), as demonstrated by proteinuria and mortality rate increases, were elicited only by the employed pharmacological amounts of estradiol. These results indicate that the MRL-lpr mice might serve as a model for post-partum flare of arthritis in SLE and rheumatoid arthritis by providing an approach to study the complexity of the effects of pregnancy on autoimmune diseases, and to obtain further evidence for the involvement of oestrogen in arthritis.

Systemic lupus erythematosus (SLE) is a progressive connective tissue disorder accompanied by a wide spectrum of autoantibody production. The tissues/organs commonly affected are the skin, kidney, brain, vasculature, heart, and diarthroid joints. Articular disease is the most frequent early manifestation of SLE, being present in more than 90% of cases. This articular pathology is often transient, and generally less severe than that seen in rheumatoid arthritis (RA). As articular changes often precede other features of SLE, they are frequently diagnosed as RA. Notwithstanding, RA also may coexist with SLE [1] .

SLE is a disease that most frequently afflicts women of childbearing age, and RA also occurs three times more frequently in women than men [2, 3] . Therefore the effect of pregnancy on these diseases is of particular importance. It is known that oestrogen levels increase during pregnancy, and decline immediately after delivery. Oestrogen has two known effects on the immune system: it suppresses antigen-specific T cell-dependent immune functions, characteristic of RA, and stimulates B cell responses, characteristic of SLE [4] . While the exacerbating effect of pregnancy in SLE is controversial [5, 6] , the ameliorating effect in RA is well described [7] . However, more than 90% of patients with RA remission during pregnancy develop a post-partum flare up of the disorder [8] . These findings suggest that oestrogen might influence these rheumatic conditions.

The MRL-lpr mouse strain has been successfully used as a model for studying both SLE and RA [9] [10] [11] [12] [13] [14] [15] . It develops a mild spontaneous form of arthritis which can be enhanced by an intradermal injection of Freund's complete adjuvant (FCA) [16] . Although the sex hormone involvement in SLE in MRLlpr is well characterized [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] , oestrogen has been studied less in respect to its effect on MRL-lpr arthritis [27] , and has not been described in relation to pregnancy in these mice. We noticed during routine breeding of MRL-lpr mice that a significant number of older female breeders developed postpartum erythema and swelling. Hence, this study examines the possibility of whether the MRL-lpr spontaneous arthritis could serve as a model for post-partum arthritis flare up in SLE or RA. We also investigated the effect of estradiol on the murine arthritis, using both the post-partum arthritis flare up and the adjuvant enhanced arthritis models in MRL-lpr mice.

MRL/MpJ-lpr/lpr (MRL-/pr) mice were obtained from a breeding colony maintained in the animal facilities at the Department of Oral Biology. This colony was established from stocks originally purchased from the Jackson Labs (Bar Harbor, ME) and bred for two to nine generations. Histological analysis of different generations of these mice demonstrated no-difference in the arthritic pathology of the spontaneous arthritis, in its post-partum flare up, or in the ability of FCA to enhance the disease (data not shown). The mice were kept on a standard diet with water ad libitum. Pregnant mice were housed one per cage. The colony was routinely screened for Mycoplasma pulmonis, and Myc. arthritidis, rodent coronaviruses (including hepatitis), and Sendai virus using the Murine ImmunoComb test (Charles River Labs, Wilmington, MA). Antibodies only against coronaviruses were detected.

Breeding protocol Ten-week-old virgin females were mated. In the first experiment the females were separated from the males after mating, but not during the second experiment. The litters were not weaned from the mothers until the termination of the studies. At that time the female parents were 17 weeks old.

Adjuvant injection was carried out as described previously [16] . Briefly, 13-week-old female mice were injected intradermally at two thoracic sites (selected for good lymph drainage) with 0 05 ml FCA, supplemented to 10 mg/ml with heat inactivated Mycobacterium tuberculosis H37 RA (Difco, Detroit, MI).

Estradiol administration Animals were randomly selected for estradiol injection. They were injected subcutaneously with 3-2 Ag (0-08 mg/kg) ,3estradiol-3-benzoate in olive oil (Sigma, St Louis, MO) on days 2, 3, 9, 15 and 21 after partition. This dosage of estradiol was shown previously to be adequate to provide physiological serum oestrogen levels characteristic of pregnancy [28] .

In a separate experiment a pharmacological dose of 0-4 mg/ kg per day in olive oil was administered subcutaneously, starting day 0 for 14 days, to non-mated adjuvant-injected animals. Controls were injected with olive oil only.

Clinical evaluation was carried out as described previously [16, 29, 30] . Mice were examined 'blind' every second day for 30 days after injection for visual appearance of arthritis and scored as positive if erythema and swelling of a fore or hind paw(s) was observed. Bimalleolar ankle width measurements were also recorded every 5 days using a micrometer.

On day 30 the experiment was terminated. Lymphadenopathy was evaluated using a subjective 0-4 non-invasive scoring scale based on palpation of the lymph nodes. The degree of lymph node involvement was rated as: 0, no palpable lymph nodes; 1, one palpable lymph node in one area; 2, seriously enlarged lymph node(s) in one area; 3, seriously enlarged lymph node(s) in both axillary and inguinal areas; and 4, massively enlarged lymph nodes in all areas.

Renal function was assessed on the basis of semiquantitative colorimetric determination of the urinary protein level, using Albustix sticks (Miles, Ontario, Canada). Protein was measured as g/l, with concentrations above 0 3 g/l considered abnormal.

After the animals were killed the joints and kidneys were dissected and fixed in 10% formalin. The joints were then decalcified for 48h in 10% formic acid. Both tissues were subsequently processed for paraffin embedding. Serial 3-5 pm thick sections were stained with haematoxylin and eosin (H&E).

Histopathological alterations of the joints were evaluated 'blind' for the presence of the following features: 1, subsynovial inflammation; 2, synovial hyperplasia; 3, cartilage erosion and pannus formation; and 4, bone destruction. Within each parameter scores were assigned from 0 to 2 (0, no; 1, mild; and 2, severe involvement).

Kidney sections were also examined 'blind' by a pathologist for mesangial cell proliferation, crescent formation, interstitial cellular infiltration, and vasculitis. The histological changes were evaluated by a grading system from 0 to 4, where 4 corresponded to severe, 3 moderate, 2 mild and 1 minimal presence of diseased lesions.

Immunohistological evaluation In order to identify subpopulations of cells in the affected areas and any changes during disease progression, the tarsometatarsal joints from mice of both control and experimental groups were fresh frozen and cryosectioned. The immunohistology was carried out applying commercially available biotinylated MoAbs directed against mouse cell surface antigens: MHC class II (lak) antigen and IL-2 receptor-(activation markers), CD3 (pan T), and CD4 (T helper) (Life Technologies, Gaithersburg, MD). Biotin-labelled anti-CD43 (activated T) antibody was kindly provided by Dr H. Ziltener (Biomedical Sciences, University of British Columbia).

The tissue blocks were immersed in Histo Prep (Fischer, Fair Lawn, NJ), snap frozen in liquid nitrogen-prechilled isopentane, and stored at -70°C for a maximum of 2 weeks before cryosectioning. Sagittal sections of 8-10,Im thickness were cut on a cryostat at -25°C. The sections were attached to adhesive tape (Scotch Brand Tape 600, 3M Canada) [31, 32] collected on glass slides, and stored at -20°C for a maximum of 1 week. Localization of the antibody binding was established by a biotin-streptavidin-alkaline phosphatase method. The nonspecific binding sites were saturated during 30 min incubation (0). One group of animals after delivery (0) received 0 08 mg/kg estradiol at days 2, 3, 9, 15, and 21 and showed an incidence comparable to, and a severity lower than the control group with spontaneous arthritis (A).

with rabbit serum (Dako, Carpinteria, CA). After washing in Tris-buffered saline pH 7 6 (TBS), biotin-labelled antibody was applied overnight at 4°C in 1: 50 dilution in TBS containing 1% bovine serum albumin (BSA). During incubation the slides were kept in a moist chamber to prevent them from drying. After rinsing with TBS, the alkaline phosphatase-conjugated streptavidin (Dako, Carpinteria, CA, and Life Technologies) was added for 30 min. After washing again in TBS the slides were treated with New Fuchsin containing TBS supplemented with 0 1% Levamisol (Sigma) according to the manufacturer's specifications. The slides were developed under the microscope for about 5 min until the desired intensity of red staining was achieved. The slides were then counterstained with Mayer's haematoxylin (Sigma), covered with Crystal Mount (Biomeda, Foster City, CA) overnight at room temperature, permanently mounted with Entellan (Merck, Darmstadt, Germany) and evaluated under a light microscope. For a negative control, TBS was substituted for the primary antibody. A lymph node from the same animal as well as the bone marrow within the test section served as a positive control. Alkaline phosphatase labelling was employed, as peroxidase staining has been reported as unreliable if the method includes the use of adhesive tape. Endogenous staining was reduced by 0 1 % Levamisol.

Statistical analysis Data are expressed as mean ± s.e.m. Statistical differences among groups were determined using the ANOVA and Bonferroni multiple comparison tests.

The incidence of swelling and erythema in post-partum animals during 30 days after delivery was observed in 76% (13/17) of the mice (Fig. la) . This was confirmed in a second experiment in which 67% of the animals demonstrated similar evidence of swelling and erythema (16/24) (data not shown). The clinical onset of post-partum flare was successfully prevented (21%, 3/14) and the time of the onset delayed by the administration of physiological amounts of estradiol. This frequency seen in estradiol-injected animals was comparable to the freqency of clinical signs of arthritis onset in non-mated, age-matched female MRL-lpr mice, which developed spontaneous arthritis (23%, 5/22). Bimaleolar ankle measurements, which quantify the severity of the swelling, showed similar patterns (Fig. lb) , although there was a remission of swelling in some of the cases in all three groups. Subsynovial inflammation, synovial hypertrophy, and the overall histological scores increased significantly post-partum (P < 0 05) (Fig. 2) . However, there was no significant increase in cartilage erosion, pannus formation, or bone destruction. Administration of physiological doses of estradiol significantly lowered the overall histological score (P = 0034), and prevented cartilage erosion, pannus formation, and bone destruction. It also decreased somewhat the level of inflammation and synovial hypertrophy. Figure 3 demonstrates the extent of subsynovial inflammation and synovial hypertrophy. Figure 3a depicts the sybsynovial infiltration of MHC class 11-positive cells. Cells also expressed the MHC II antigen at certain areas of the synovial lining (Fig. 3b) . There was a sporadic, but significant presence of CD3and CD43-expressing cells, as well as occasional cells staining for the CD4 marker (Fig. 3c) . Anti-IL-2R MoAbs were rarely detected in the test or the positive control sections.

The post-partum period, or the short-term administration of physiological amounts of estradiol, did not significantly alter any systemic disease manifestations, recorded at the termination of the experiment. These lymphadenopathy, significant proteinurea, glomerulonephritis, and autoantibody levels ( Table 1 ). The effect of pharmacological amounts of estradiol (0 4 mg/ kg) was also investigated on the development of adjuvantenhanced arthritis in female MRL-lpr mice. Thirteen-weekold mice received this dose daily for 2 weeks starting on the day of the FCA injection. This dose prevented the adjuvant enhancement of disease, as indicated by the similar frequency of clinical arthritis observed in treated and control animals ( Table 2 ). The severity of swelling was significantly lower than in adjuvant-injected mice (P = 0012). Histological analysis revealed a significant degree of prevention in all the pathological parameters examined, including the overall score (P < 0-05). This dose, however, appears to increase the mortality level to 34%, compared with 22% which is characteristic of this strain at 4 months old (Table 3 ). In the survivors this was accompanied by a significant decrease of lymphadenopathy (P < 0.02), and a marked increase of proteinurea in some animals, resulting in a non-significant (P > 0 183) increase in the mean proteinurea level.

The hormonal changes associated with pregnancy affect even the healthy mother, and can seriously alter the course of autoimmune diseases. Pregnancy causes flare up in active SLE. While it ameliorates RA during gestation, it exacerbates the disease after delivery [33] . Pregnancy and lactation also increase the dangers of the side effects that accompany the standard treatment in these diseases. Therefore, it is deemed imperative to develop animal models which allow the study of the mechanisms involved in these complex situations and test treatment modalities which could be used without endangering the health of both the mother and the fetus or breast-fed newborn.

The most frequently used model for studying the effect of pregnancy in arthritis is collagen-induced arthritis. The prepartum remission and post-partum exacerbation, as well as the acceleration of the onset of the disease by pregnancy, were described in a study that was conducted in a small number of DBA/1 mice with type II collagen-induced arthritis [34] . Recently it was reported that pregnancy prevented pristaneinduced arthritis [35] and proteoglycan-induced arthritis in mice [36] . As there are differences in both strain and in model in the reaction to pregnancy, it is important to investigate the effect of pregnancy in alternative model systems.

The MRL-lpr mouse strain is a well studied model of SLE. It carries a dysfunctional Fas gene, resulting in the lack of thymic apoptosis, and the proliferation of CD3+, CD4-, CD8lymphocytes [37] . This genetic defect results in an early onset of SLE-like symptoms, including immune complex glomerulonephritis, anti-nuclear and anti-extracellular matrix autoantibodies, rheumatoid factors, and vasculitis with a 50% mortality rate at 5-5 months of age [38] . It is the only described spontaneous SLE model which also develops an arthritis. SLE symptoms and some early histopathological signs of arthritis are already present in 2-month-old animals [39] .

The present study shows that this spontaneous form of arthritis in animals mated at 10 weeks of age is significantly exacerbated during the post-partum period. The reason this finding was not previously noticed is most probably that the MRL-lpr mice are generally bred at a younger age to avoid the severity of the SLE, thereby dampening the intensity of the flare up of arthritis. This age-dependence, reflecting the time required to reach a certain stage of underlying disease, was also noted in this strain during the development of the adjuvant-enhanced arthritis model (manuscript submitted). It is of interest that in both cases approximately 2/3 of the animals developed exacerbated arthritis. This indicates the presence of an underlying component of the disease at this age that can be accelerated by certain factors. The flare-up occurred as early as 5 days after delivery, with a mean onset of 13 days, and resulted in a massive subsynovial inflammatory infiltrate. There was, however, no considerable cartilage or bone destruction by day 30 post-partum. We are planning to examine the longer-term effect of the flare-up on the severity of arthritis histopathology in post-partum versus non-mated female mice.

The ameliorating effect of pregnancy on RA has been attributed to physiological factors, including oestrogen, progesterone, prolactin [40, 41] , pregnancy-associated proteins as immunosuppressors (a-2 glycoprotein), suppression of cellmediated immunity, and alterations in the glycosylation of IgG [8] acting as possible mediators [42] . In post-menopausal women oestrogen treatment of active RA increased bone mineral density [43] . The effect of oestrogen containing contraceptives has been described as well [44] . However, there is no clear evidence whether the use of oral contraceptives provides protection against RA. This inconclusiveness is probably due to differences in patient selection, as European hospital-based studies contradicted US population-based findings.

Oestrogen is a probable factor in the controversial effect of the pill. Although it has been previously reported that female sex hormones have only a minimal therapeutical effect on RA [45] , the latest controlled study showed that adjunct therapy with oestrogen, when applied to post-menopausal RA, can be effective if the resulting serum oestrogen levels are adequate [46] . It is noteworthy that most of the pills contain both oestrogen and progesterone [47] . Holmdahl et al. proposed that the remission of RA during pregnancy is primarily due to oestrogen, and that progesterone might potentiate this event. It is clear that progesterone alone had no effect in preventing type II collagen-induced arthritis in mice [48] . Mattsson et al. suggested that post-partum flare-up is due to both the fall in the elevated pregnancy steroid levels and the surge in prolactin release after delivery [49] . Holmdahl et al. described the role of oestrogen in RA prevention as model-and strain-dependent [50] . Schlaghecke et al. reported the preventive effect of oestrogen on adjuvant arthritis in rats [51] . The MRL-lpr arthritis models could serve as alternatives to the above models, since there is a need to study the described model and strain dependency of the oestrogen effect. Here we report the preventive effect physiological doses of estradiol have on the post-partum flare-up in spontaneous arthritis and the effect pharmacological doses have on adjuvant-enhanced arthritis in MRL-lpr mice.

These data extend and support the concept of oestrogen influence in RA, and provide a new spontaneous model for studying post-partum flare-up. Further investigations are needed to study the possible involvement of other factors, and to develop a strategy for safe therapeutic intervention. PF, ANOVA and Bonferroni multiple comparison test results at P significance level between spontaneous arthritis and post-partum arthritis groups; P2, ANOVA and Bonferroni multiple comparison test results at P significance level between post-partum arthritis with estradiol-injected and post-partum arthritis groups; NE, not examined. 

Systemic lupus erythematosus

Rheumatoid arthritis: cell and tissue pathology

The epidemiology of rheumatoid arthritis in Rochester, Minnesota: a study of incidence, prevalence, and mortality

Estrogen exaggerates lupus, but suppresses T-celldependent autoimmune disease

Frequency of lupus flare in pregnancy: the Hopkins Lupus Pregnancy Center experience

Lupus and pregnancy studies

A prospective clinical study of the effect of pregnancy on rheumatoid arthritis and ankylosing spondylitis

Changes in IgG glycoform levels are associated with remission of arthritis during pregnancy

Etiology and pathogenesis of a spontaneous lupus-like syndrome in mice

Spontaneous murine lupus-like syndromes. Clinical and immunopathological manifestations in several strains

Lymphoproliferation (lpr) and other single-locus models for murine lupus

Etiopathogenesis of murine SLE

A spontaneous rheumatoid arthritis-like disease in MRL/1 mice

Murine models of systemic lupus erythematosus

New inbred strains

Complete Freund's adjuvant induces an earlier and more severe arthritis in MRL-lpr mice

Effects of thymectomy or androgen administration upon the autoimmune disease of MRL/Mp-lpr/lpr mice

Effects of castration and sex hormones on immune clearance and autoimmune disease in MRL/Mp-lpr/lpr and MRL/Mp-+/+ mice

Effects of short-term administration of sex hormones on normal and autoimmune mice

Hormone control of autoantibodies to calf thymus nuclear extract (CTE) and DNA in MRL-lpr and MRL-+/+ mice

Sex hormones modulate the response of pulmonary perivascular inflammation to cyclophosphamide therapy in MRL/MpJ-lpr/lpr mice

Oestradioland testosterone-mediated effects on the immune system in normal and autoimmune mice are genetically linked and inherited as dominant traits

Oestrogen is a potent disease accelerator in SLE-prone MRL lpr/lpr mice

Differential effects of oestrogen in murine lupus: acceleration of glomerulonephritis and amelioration of T cell-mediated lesions

Pulmonary perviscular and interstitial inflammation in MRL/MpJ-lpr/lpr mice. III. Modulation by cyclophosphamide and sex hormones in 4-and 6-monthold animals

Estrogen accelerates immune complex glomerulonephritis but ameliorates T cell-mediated vasculitis and sialadenitis in autoimmune MRL lpr/lpr mice

Impact of estrogen on immune functions and progression of disease

Estrogen induced suppression of collagen arthritis. V: Physiological level of estrogen in DBA/1 mice is therapeutic on established arthritis, suppresses anti-type II collagen T-cell dependent immunity and stimulates polyclonal B-cell activity

Photodynamic therapy; a comparison with other immunomodulatory treatments of adjuvant-enhanced arthritis in MRL-lpr mice

Antibodies to extracellularmatrix proteins in the sera of MRL-lpr mice

Enzyme histochemical reactions in unfixed and undecalcified cryostat sections of mouse knee joints with special reference to arthritis lesions

A novel technique for immunohistoperoxidase staining of unfixed whole joints of small animals

Pregnancy and its effects on the risk of developing rheumatoid arthritis

Effect of pregnancy on collagen-induced arthritis in mice

Agalactosyl IgG in pristane-induced arthritis. Pregnancy affects the incidence and severity of arthritis and the glycosylation status of IgG

Effect of pregnancy on proteoglycan-induced progressive arthritis in BALB/c mice: remission of disease activity

Lymphoproliferation disorder in mice explained by defects in Fas antigen that mediates apoptosis

Induction of severe autoimmune disease in normal mice by simultaneous action of multiple immunostimulators

Etiopathogenesis of the rheumatoid arthritis-like disease in MRL/l mice. I. The histomorphologic basis of joint destruction

A possible role of prolactin in adjuvant arthritis

Prolactin deficiency in rheumatoid arthritis

Adjuvant oestrogen treatment increases bone mineral density in postmenopausal women with rheumatoid arthritis

The pill, parity and rheumatoid arthritis

Estrogens and rheumatoid arthritis

Estrogen replacement (ERT) as an adjunct therapy in postmenopausal rheumatoid arthritis (RA); a randomized study of 200 patients

Oral contraception and its possible protection against rheumatoid arthritis

Oestrogen induced suppression of collagen arthritis. IV. Progesterone alone does not affect the course of arthritis but enhances the oestrogen-mediated therapeutic effect

Maintained pregnancy levels of oestrogen afford complete protection from postpartum exacerbation of collagen-induced arthritis

Oestrogen is a potent immunomodulator of murine exerimental rheumatoid disease

Preventive and therapeutic effect of estrogen administration in adjuvant arthritis