Osteopontin regulates hindlimb-unloading-induced lymphoid organ atrophy and weight loss by modulating corticosteroid production

Abstract

Osteopontin (OPN), a multifunctional secreted phosphoglycoprotein, plays diverse roles in bone biology, immune regulation, cell survival, inflammation, and cancer metastasis. Here we show its role in determining lymphocyte homeostasis and body mass in response to hindlimb unloading (HU), a model for evaluating effects of weightlessness on the musculoskeletal and other physiological systems. Using this stress model, we compared OPN^−/− mice with OPN^+/+ mice subjected to HU for 3 days. Whereas OPN^+/+ mice suffered a marked reduction of body weight and significant spleen and thymus atrophy, OPN^−/− mice exhibited minor weight loss and much less spleen and thymus atrophy. The HU-induced lymphoid organ atrophy was the result of dramatically diminished numbers, respectively, of T and B cells in the spleen and CD4⁺CD8⁺ double-positive cells in the thymus of OPN^+/+ mice. Increased levels of corticosterone, which modulates lymphocyte activation responses and apoptosis during stress, were found only in OPN^+/+ mice. Apoptotic cell death was evident in the spleen and thymus of OPN^+/+ mice subjected to HU but not in OPN^−/−mice. Importantly, lymphocytes from both OPN^+/+ and OPN^−/− mice were equally sensitive to corticosteroid-induced apoptosis. These results reveal that OPN is required for enhanced corticosterone production, immune organ atrophy, and weight loss in mice subjected to HU.

Osteopontin (OPN) is a multifunctional protein involved in both cell adhesion and cell signaling by its interactions with certain integrins and CD44 variants. It is an active player in many physiological and pathological processes, including tissue remodeling and cell survival (1, 2). OPN has been identified as a major factor regulating immune responses and inflammation; its expression is up-regulated in activated T cells, macrophages, and natural killer cells (3, 4). It is classified as a T helper-1 (Th1) cytokine because it enhances production of the Th1 cytokines IFN-γ and IL-12 and inhibits production of the T helper-2 (Th2) cytokine IL-10 (5, 6). Although it is a proinflammatory cytokine associated with autoimmune diseases, the Th1 inflammatory response, B cell function, and natural killer T cell function, it also has antiinflammatory functions, such as inhibiting the induction of inducible nitric oxide synthase and inhibiting the production of Th2 cytokines (7–14). Its up-regulation is also a hallmark for many major pathological disorders, such as cardiovascular disease, cancer, and lung disease (12, 15).

OPN expression is increased in response to various stressors, including physical and chemical stress (16, 17). A connection between stress and immune function has been demonstrated in various experimental paradigms (18). Stress can either enhance or impair immune system function depending on its type and duration. Stress hormones such as cortisol play a pivotal role in modulating proinflammatory/antiinflammatory responses and the TH1/Th2 cytokine balance, thereby influencing the susceptibility to and outcome of various immune-related diseases (19). Whereas chronic stress is immunosuppressive and leads to increased vulnerability to infections and cancer, acute stress promotes immunoprotection through cell-mediated immunity (20–22). To our knowledge, the role of OPN in the response of the immune organs to stress has not been examined.

Hindlimb unloading (HU) is a widely used ground-based model for simulating stresses experienced by astronauts during space flight, for example, to study the influence of limb disuse on the immune, musculoskeletal, cardiovascular, renal, neural, metabolic, and reproductive systems. Compared with physical-restraint stress, mice under HU have free access to food and water and thus are not transiently deprived of either. Exposure to HU causes alterations in the immune system, including immune organ atrophy, similar to those observed after space flight (23). Wei et al. (24) demonstrated that HU caused a dramatic reduction in the number of mouse splenocytes and thymocytes; the reduction in lymphocyte number resulting from stress was corticosteroid-dependent and apparently caused by the induction of apoptosis. OPN has been found to inhibit apoptosis in various situations (1, 13, 14). It has previously been reported that OPN is necessary for the HU-induced remodeling of bone (1, 25). Whether OPN plays a role in immune system dysfunction under HU stress is unknown. We have used OPN^−/− and OPN^+/+ mice to study the consequence of OPN deficiency to the response of the animals to HU. We report here the finding that OPN is required for immune organ atrophy and weight loss in mice subjected to HU by modulating stress-induced production of corticosteroid.

Results

OPN Deficiency Protects Mice from Weight Loss and Lymphoid Organ Atrophy.

Mice subjected to certain stressors, HU for example, suffer significant reductions in both body weight and in the cellularity of the spleen and thymus (24, 26). To examine the role of OPN in this stress response, we exposed OPN^−/− and OPN^+/+ mice to HU for 3 days. This treatment caused a greater reduction in the body weight of the OPN^+/+ mice compared with the OPN^−/− mice (Fig. 1). The spleen and thymus were considerably reduced in size in the suspended OPN^+/+ mice compared both with control OPN^−/− mice and with the suspended OPN^−/− mice (Fig. 2). The spleens and thymuses of OPN^−/− mice were comparable in size with or without HU. Analysis of lymphoid tissues revealed a substantial reduction in the number of splenocytes and thymocytes in the OPN^+/+ mice. The suspended OPN^−/− mice exhibited a much smaller, and statistically insignificant, reduction in the number of both thymocytes and splenocytes (Fig. 3). To determine whether the loss of cellularity in the immune organs of the OPN^+/+ mice subjected to HU was caused by apoptosis, we assessed the extent of apoptosis in histological sections of spleen and thymus by using the TUNEL in situ staining technique. We observed a larger number of apoptotic splenocytes and thymocytes in HU OPN^+/+ but not in HU OPN^−/− mice or control mice (Fig. 4). Thus, it appears that OPN is necessary for immune cell apoptosis in response to the HU stress challenge. Because stress-induced elevation of glucocorticoids leads to cell apoptosis (19), it is possible that OPN may act by increasing the sensitivity of lymphocytes to corticosterone-induced apoptosis or by up-regulating corticosterone levels during HU.

Fig. 1.

HU causes significant body weight loss. Mice were weighed before and after 3-day HU. Data are expressed as percent weight loss (mean ± SEM) relative to the body weight before HU. *, Statistically significant difference between OPN^+/+ and OPN^−/− mice at P < 0.05; two-tailed t test.

Fig. 2.

HU causes significant lymphoid organ atrophy. The spleen and thymus were dissected from control and suspended mice. Surgically extracted whole organs were put into a cell culture dish filled with medium and scanned on a digital scanner.

Fig. 3.

HU leads to significant reduction in the cellularity of both thymus and spleen in OPN^+/+ but not OPN^−/− mice. Single-cell suspensions were prepared from spleens (A) and thymuses (B) of suspended and control OPN^+/+ and OPN^−/− mice. The total cell numbers were enumerated with a hemocytometer. Data represent means ± SEM of 6–10 animals at each data point. ***, Statistical significance at P < 0.001 with two-tailed t test.

Fig. 4.

OPN^−/− mice are resistant to HU-induced apoptotic cell death in both thymus and spleen. Wild-type and OPN^−/− mice were subjected to HU for 20 h. TUNEL-positive apoptotic cells in paraffin sections of spleen and thymus tissues are shown in dark brown on a methyl green counterstained background.

OPN Promotes Stress-Induced Loss of Splenocytes and Immature T Cells in Thymus.

To ascertain which lymphocyte cell types were affected by HU, we examined immune cell subpopulations with flow cytometry. Analysis of untreated OPN^+/+ and OPN^−/− mice revealed a normal distribution and equivalent numbers of B and T cells as well as T cell subpopulations in both strains (Table 1). These data suggest that OPN deficiency is not causing a skewed distribution of immune cell subpopulations in mice. Splenocytes and thymocytes were harvested from suspended OPN^+/+ and OPN^−/− mice and stained with fluorescence-conjugated antibodies specific for cell surface markers CD4, CD8, and B220 cell populations. In the thymus, a dramatic reduction of CD4⁺CD8⁺ double-positive cells (up to 80%) was observed in suspended OPN^+/+ mice, whereas a much smaller reduction of this population was observed in suspended OPN^−/− mice or untreated controls (Fig. 5A). Interestingly, the total number of single-positive CD4⁺ and CD8⁺ cells in the thymus was not significantly affected by HU in either OPN^+/+ or OPN^−/− mice. In the spleen, HU caused a >50% reduction in every subpopulation examined, including CD4⁺ T cells, CD8⁺ T cells, and B cells of OPN^+/+ mice but only a slight reduction of these lymphocyte types in OPN^−/− mice (Fig. 5B). These results revealed that OPN is necessary for the stress-induced reduction of immature T cells in thymus and both T cells and B cells in spleen.

Table 1.

Lymphocyte subpopulation distribution in the spleen and thymus

Organ	Total cell no.	% CD4+	% CD8+	% CD4+CD8+	% B220
Thymus
OPN+/+	104.2 ± 8.9	11.2 ± 1.2	3.8 ± 0.4	83.5 ± 1.0	—
OPN−/−	85.6 ± 13.9	13.2 ± 3.1	4.4 ± 1.0	77.4 ± 3.6	—
Spleen
OPN+/+	84.1 ± 9.1	23.1 ± 1.4	12 ± 0.8	—	42.8 ± 5.8
OPN−/−	56.6 ± 13.1	26.3 ± 1.84	14.1 ± 1.1	—	38.1 ± 4.7

Single-cell suspensions were generated from the spleens and thymuses of OPN^+/+ and OPN^−/− mice. Cells were stained with fluorochrome-conjugated cell surface marker antibodies against B220 (B cells), CD4 (T helper cells), and CD8 (cytotoxic T cells) subpopulations. Data represent the mean ± SEM of five to seven mice. The percentages of positively stained cells are recorded. —, subpopulation does not exist.

Fig. 5.

Analysis of lymphocyte subpopulations by flow cytometry. (A) Thymocytes isolated from suspended and control OPN^+/+ and OPN^−/− mice were stained with fluorochrome-conjugated anti-CD8 and anti-CD4 antibodies and analyzed by flow cytometry. Bar graphs indicate the number of cells in different subpopulations in OPN^+/+ and OPN^−/− mice (mean ± SEM, n = 4–7 animals). (B) Splenocytes isolated from suspended and control OPN^+/+ and OPN^−/− mice were stained with fluorochrome-conjugated anti-B220, anti-CD8, and anti-CD4 antibodies and analyzed by flow cytometry. Data represent means ± SEM of four to seven animals in each data point. ***, Statistical significance at P < 0.001 with two-tailed t test.

Endogenous Corticosterone Levels Are Not Elevated in OPN^−/− Mice Subjected to HU.

Corticosterone, the major stress hormone in rodents, is quickly up-regulated in response to stress, leading to the induction of substantial lymphocyte apoptosis (19). Interestingly, it has been shown that blocking steroid receptors in mice can suppress the HU-induced lymphocyte reduction (24). To determine whether OPN plays a role in regulating the production of glucocorticoid after HU, we measured both corticosterone and OPN levels in serum of the mice after HU. We found that although the OPN level was not changed in OPN^+/+ mice (data not shown), HU caused an ≈5-fold increase in corticosteroid levels in the serum of OPN^+/+ mice but not in OPN^−/− mice (Fig. 6). This finding suggests that OPN is required for up-regulation of the steroid hormone in response to HU-induced stress. When exogenous dexamethasone was supplied to cells isolated from OPN^+/+ and OPN^−/− mice, there was no difference in the rate of induced cell death (Fig. 7), indicating that steroid-induced lymphoid cell death is not impaired in OPN^−/− cells. The inference is that endogenous OPN acts upstream of the corticosterone hormone, possibly contributing to its increased production in response to the HU-induced stress and thereby indirectly contributing to lymphoid cell death.

Fig. 6.

Serum corticosterone levels are elevated in OPN^+/+ mice but not in OPN^−/− mice after HU. Serum samples from suspended and control of OPN^+/+ or OPN^−/− mice were diluted in corticosterone-free serum matrix and assayed with the corticosterone ELISA kit. Data shown represent the mean ± SEM of 5–11 animals from several independent experiments for each data point. **, Statistical significance at P < 0.01 with two-tailed t test; ns, not significant.

Fig. 7.

Dexamethasone-induced cell death was not impaired in OPN^−/− mice. OPN^+/+ and OPN^−/− cells were treated with dexamethasone for 16 h at 37°C, and the percentage of remaining live cells was assessed by flow cytometry with propidium iodide staining. WT, wild type; KO, knockout; spl, spleen; thy, thymus.

OPN Regulates Immune Cell Proliferation and Cytokine Production in Response to HU.

The reduction in spleen and thymus cellularity during HU of OPN^+/+ mice could be the result of decreased proliferation/maturation or increased emigration/apoptosis of the lymphocytes. Because OPN is known to promote cell proliferation in certain contexts, we investigated whether lymphocyte proliferation ([³H]thymidine incorporation) was regulated by OPN in mice subjected to HU. We quantified splenic and thymic T cell proliferation by stimulating equal numbers of these OPN^+/+ and OPN^−/− cells with anti-CD3/CD28 for 48 h and then labeling with [³H]thymidine for 16 h. Although there was no difference in [³H]thymidine incorporation in thymocytes from untreated OPN^+/+ or OPN^−/− mice, a significant increase in splenocyte DNA synthesis was observed from suspended OPN^−/− mice compared with suspended OPN^+/+ mice (Fig. 8A). The increased proliferative ability of the splenocytes derived from the suspended OPN^−/− mice may be caused by a higher percentage of activated cells in the OPN^−/− mice. In contrast, the thymocytes from the suspended OPN^+/+ mice exhibited a dramatic increase in [³H]thymidine incorporation, which may be caused by the substantial loss of immature CD4⁺CD8⁺ double-positive cells in the OPN^+/+ mice after HU, resulting in a higher percentage, relative to OPN^−/− mice, of single-positive T cells (CD4⁺ or CD8⁺) that exhibit a strong proliferative response when stimulated by anti-CD3/anti-CD28. Parallel to the cell proliferation study, we assessed the levels of cytokine production in the supernatants of the cells activated by anti-CD3/anti-CD28 (Fig. 8B). The activated OPN^+/+ thymocytes from the HU mice secreted substantially higher amounts of IL-2 than did unstressed controls at 48 h, whereas very low IL-2 production in OPN^−/− cells was observed at the same time point. In OPN^−/− splenocytes, we observed an increased production of the Th2 cytokine IL-10, which is consistent with the notion that OPN promotes Th1 cytokine expression and inhibits Th2 cytokine expression. Secretion of the chemokine monocyte chemotactic protein-1 (MCP-1, a macrophage chemoattractant) was significantly less in activated OPN^−/− splenocytes compared with OPN^+/+ splenocytes.

Fig. 8.

Lymphocyte proliferation and cytokine production induced by anti-CD3/anti-CD28 stimulation. Total lymphocytes prepared from the spleens and thymuses of OPN^+/+ or OPN^−/− mice were cultured in 96-well plates at 2 × 10⁵ cells per well in complete RPMI with and without (control) anti-CD3 and anti-CD28. (A) After 48 h, the cells were pulsed with [³H]thymidine; incorporation of ³H was quantified by scintillation counting. Open bars, OPN^+/+; filled bars, OPN^−/−. (B) After 48 h, the supernatants were analyzed for cytokine production with a DuoSet ELISA kit. IL-2 was measured in medium conditioned by thymocytes; IL-10 and MCP-1 were measured in media conditioned by splenocytes. The results are expressed as the mean ± SEM of triplicate wells. ***, Statistical significance at P < 0.001 with two-tailed t test. Open bars, control; filled bars, HU.

Discussion

It is well documented that various forms of physical stress (e.g., restraint, HU, stroke, and microgravity) can cause the involution of the spleen and/or thymus, thus impairing the organism's ability to mount an immune response (23, 24, 26, 27). The loss of spleen and thymus cellularity could result from one or a combination of factors, including increased cell death or reduced proliferation and/or redistribution of the lymphocytes in the body (28). Depending on the specifics, stress can be immunoenhancing or immunosuppressing. Acute stress can enhance the immune response, stimulating redistribution of leukocytes and enabling their response to the injury (20, 22). Chronic stress tends to suppress the immune response and leads to lymphocyte apoptosis (26) and increased susceptibility to diseases (19). These contrary responses are believed to be mediated in part by stress-induced adrenal glucocorticoids (20). There is evidence that the loss of splenocytes and thymocytes in response to HU is the result of increased glucocorticoid-dependent apoptosis (24).

In our studies, the mice were suspended for 3 days, a chronic stress model. We found that after 3 days the OPN^+/+ mice had lost ≈20% of their original weight compared with ≈10% for the OPN^−/− mice; during this same 3-day period, control mice gained ≈5% in weight (Fig. 1). Parallel to the weight loss, the reduction in the cellularity of the spleen and thymus was >70% in OPN^+/+ mice. However, the composition of the immune cell population in the circulation was not changed significantly (data not shown), indicating that migration/redistribution of immune cells was not a major contributing factor to the loss of cellularity in spleen and thymus. In contrast, mice lacking OPN were protected from significant spleen and thymus atrophy after HU. At 3 days, corticosterone levels in the plasma of the HU OPN^+/+ mice were increased by 5-fold. Evidently, OPN is involved in the response to this stress and is necessary for augmentation of the corticosterone in response to HU. Mice subjected to physical restraint stress (immobilized in a ventilated 50-ml conical tube for two 12-h periods separated by 12 h of freedom) revealed that OPN was also required for the immune organ atrophy similar to that induced by HU (data not shown).

In the absence of OPN, the depletion of the major lymphocyte populations (B, CD4⁺, and CD8⁺ cells) in the spleen and CD4⁺CD8⁺ double-positive cells in the thymus was much less dramatic (Fig. 5). We observed a significant increase in the proliferation of thymocytes from stressed OPN^+/+ mice because of diminished double-positive cells. This double-positive subpopulation was previously defined as steroid-sensitive thymocytes (29) that cannot be stimulated to proliferate with phorbol myristate acetate/ionomycin or anti-CD3/anti-CD28. Instead, they undergo activation-induced cell death (30). Accompanying the dramatic increase in thymocyte proliferation was an ≈3-fold increase in IL-2 secretion by the thymocytes (Fig. 8B). Increased splenocyte proliferation in cells from stressed OPN^−/− mice indicated that these cells are protected from stress-induced apoptosis, therefore they were more responsive to stimulation. Our results correlate well with atrophy of the lymphoid organs, indirectly demonstrating that OPN stimulates the loss of certain lymphocyte subpopulations as reflected in the proliferative ability of the total lymphocyte population. The effect of OPN on cytokine production provides further evidence that OPN may mediate the stress response in a broad area in addition to regulating corticosteroid levels.

OPN is implicated in innate and adaptive immune responses, for example recruiting macrophages and lymphocytes to sites of injury and regulating their activities. Its expression is elevated in response to mechanical stress in bone and the vasculature (12, 25, 31). OPN levels are also elevated in response to the oxidative stresses existing in tumors, after reperfusion injury, and in inflammatory diseases such as inflammatory bowl disease, chronic obstructive pulmonary disease, and Parkinson's disease (17, 32, 33). Our finding that mice lacking OPN are less subject to stress-induced lymphoid organ atrophy is intriguing because it implies that OPN functions as a stress mediator.

Microarray analysis has revealed that OPN is required for enhanced p53 gene expression in HU mice (34). Elevated p53 levels drive cells to commit to apoptosis or to cease dividing in response to the harmful effects of stress; however, the role of p53 in HU-induced apoptosis in lymphocytes has not been investigated. Our results showed less cell loss in OPN^−/− mice after 3 days of HU, consistent with a diminished p53 response in OPN^−/− mice. These data reveal that OPN acts to mediate stress signaling upstream from genes regulating apoptosis and corticosterone biosynthesis or release. It is of interest to note that Marsh et al. (39) have reported that the OPN^−/− mice have increased mechanosensory thresholds, perhaps contributing to their increased tolerance of the stresses imposed in this research.

Steroid hormones have long been associated with lymphocyte apoptosis. Glucocorticoids, released through activation of the hypothalamus–pituitary–adrenal axis, are believed to mediate chronic stress-induced immune modulation. It has been reported that acute stress can enhance the delayed-type hypersensitivity response (35) and wound healing (36). Serum corticosterone levels are elevated in OPN^+/+ but not in OPN^−/− mice after HU, and that appears to be the proximate cause of lymphoid organ atrophy. Stress hormones regulate cytokine production and influence the balance of pro- and antiinflammatory and Th1/Th2 cytokines, for example suppressing IL-12 production from antigen-presenting cells and enhancing IL-4 production by T cells (19). This effect in the acute response may protect the organism from excessive proinflammatory Th1 cytokines. In the case of acute stress, glucocorticoids result in suppression of Th2 cytokines and inflammatory mediators, which facilitate inflammation, one of the reasons that corticosteroids are potent antiinflammatory antiallergic drugs. The role of OPN in regulating the Th1/Th2 cytokine balance has been well documented (4, 5). Our data show that OPN may act by enhancing corticosterone production, thereby contributing to the immunosuppressive effect of stress. This work reveals the importance of OPN in stress responses and opens an area for exploring manifold functions of OPN.

Materials and Methods

Mice.

OPN^−/− mice were generated (37) and maintained in a 129 background, along with isogenic wild-type controls, in the Rutgers Nelson Animal Facility (Piscataway, NJ), which is accredited by the Association for Assessment and Accreditation of Laboratory Animal Care and is under the care of a board-certified veterinarian. All animals used in the experiments were age- and sex-matched.

Hindlimb Unloading.

Mice were caged individually and suspended for 3 days by the tail by using a strip of adhesive surgical tape attached to a chain hanging from a pulley. Mice were suspended at a 25° angle to the floor with only the forelimbs touching the floor (24, 38). Food and water were supplied ad libitum. Control mice were housed under similar conditions except that they were not suspended. This protocol was approved by the Rutgers Institutional Animal Care and Use Committee (no. 97:031). Mice were weighed before and after being subjected to HU.

Cell Counts and Surface Marker Analysis.

After 3 days of HU, the mice were killed by CO₂ inhalation. The thymus and spleen were excised, and single-cell suspensions were prepared. Total cell counts were determined with a hemocytometer. Specific lymphocyte subpopulations were identified on the basis of cell surface markers by flow cytometry; fluorescence-conjugated monoclonal antibodies, including rat anti-mouse CD4, CD8, and CD45R/B220 (all from BD Biosciences–PharMingen, San Diego, CA), were used. Cells were incubated with BD Fc Block (BD Biosciences–PharMingen) to block nonspecific binding before staining. Staining was determined by FACScalibur (Becton Dickinson, San Jose, CA). Data were acquired and analyzed with CellQuest software.

Determination of Corticosterone in Serum.

Blood was collected from experimental and control OPN^+/+ and OPN^−/− mice immediately after HU. Serum was collected and stored at −80°C. The levels of corticosterone in serum samples were assessed with a corticosterone ELISA kit from IBL America (Minneapolis, MN) according to the manufacturer's instructions.

Induction of Apoptosis by Dexamethasone.

Single-cell suspensions of splenocytes and thymocytes were cultured in 96-well plates at 10⁶ cells per well in the presence of dexamethasone at various concentrations. After a 16-h incubation at 37°C, the cells were washed in PBS and stained with propidium iodide in the presence of 100 μg/ml RNase/0.1% saponin. The percent of apoptotic cells was determined by DNA content analysis with flow cytometry.

In Situ Apoptosis Detection (TUNEL Assay).

Nuclear DNA fragmentation in apoptotic cells was detected on paraffin sections by using a TACS.XL DAB in situ apoptosis detection kit from Trevigen (Gaithersburg, MD). BrdU incorporation by terminal deoxynucleotidyl transferase at the site of DNA fragmentation was detected by a highly specific and sensitive biotinylated anti-BrdU antibody and visualized by a streptavidin–horseradish peroxidase conjugate. The tissue sections were counterstained with methyl green.

T Cell Proliferation Assay.

Single-cell suspensions of thymocytes and splenocytes were plated in 96-well plates that were precoated with anti-CD3 at 5 μg/ml in PBS for 24 h at 4°C. Cells were seeded at 5 × 10⁵ cells per well in complete medium (RPMI medium 1640/10% heat-inactivated FBS/55 μM 2-mercaptoethanol/2 mM l-glutamine/50 units/ml penicillin/50 μg/ml streptomycin) supplemented with 1 μg/ml anti-CD28. The cells were cultured for 3 days and then labeled with [³H]thymidine for 16 h. The incorporation of [³H]thymidine was measured by β-scintillation counting.

Determination of Cytokine Production.

The supernatants of anti-CD3/anti-CD28-stimulated thymocytes or splenocytes cultures were harvested after 48 h. Cytokines IL-2, IL-10, and MCP-1 were measured with DuoSet ELISA kits (R&D Systems, Minneapolis, MN), according to the manufacturer's instructions.

Statistical Analysis.

Each experiment was performed at least twice with three or more mice in each group. Data are expressed as mean ± SEM. Differences between groups were analyzed by using Student's t test associated with Excel software.

Abbreviations

HU

hindlimb unloading

MCP-1

monocyte chemotactic protein-1

OPN

osteopontin

Th1

T helper cell-1

Th2

T helper cell-2.