Role Of Glucocorticoid-Induced Leucine Zipper(GILZ) In Bone Formation

Glucocorticoids (GCs) have both anabolic and catabolic effects on bone. However, no GC anabolic effect mediator has been identified to date. Here we show that targeted expression of glucocorticoid-induced leucine zipper (GILZ), a GC anti-inflammatory effect mediator, in mice enhances bone acquisition.Three independent GILZ transgenic (Tg) lines, each carrying a GILZ-HA fusion construct downstream of Col3.6promoter, were obtained on a C57BL/6background. Genotyping results indicated that each line contains varying copies of GILZ transgene. The expression of mRNA and protein from GILZ transgene was confirmed by qRT-PCR and Western blot analyses, respectively, using RNA and protein samples extracted from bone tissues, and by immunohistochemical staining of bone sections. The optimal lines expressed GILZ transgene at a level comparable to that normally seen in cultured bone marrow MSCs induced by dexamethasone (100nM for6hrs), and showed no signs of abnormalities in body size, body weight, liter size, sex ratio, or gross anatomy.As an initial assessment of GILZ function in bone, we performed densitometiy analysis (GE Lunar PIXImus system) of femurs from3-month-old mice. Results show that BMD and BMC were increased significantly in both male (BMD:12%; BMC:24%, p<0.05) and female (BMD:12%; BMC:16%, p<0.05) Tg mice compared with their littermate controls. Micro-CT analysis of femoral samples of male mice showed that the integrity of trabecular bone is remarkably higher in Tg mice than that in the WT mice, especially at older age. Quantitative results showed that at3months of age, the femurs of Tg mice had significantly increased bone volume (BV/TV)(+28%), trabecular numbers (Tb.N)(+15%), and trabecular thickness (Tb.Th)(+13%) as compared with WT controls. The trabecular separation (Tb.Sp), connectivity density (Conn. D) and structure model index (SMI) were not significantly different in Tg vs. WT bones at3months of age. The magnitude of increase in these parameters seems to become more evident in Tg mice as the animals aged. At6months, the difference between Tg and control mice increased to73%in BV/TV,53%in Tb.N, and15%in Tb.Th. Also at6months, the parameters that showed no changes at3months became significantly different; Conn. D was150%higher and Tb.Sp and SMI41%and22% lower than the WT mice, respectively.To determine whether bone mass increase in GILZ Tg mice was due to an increase in bone formation, we performed calcein double-labeling experiments using femoral samples of3-month-old male mice. Results showed that bone formation (2mm below the proximal end of femurs) was dramatically increased in Tg mice; with a2.0-fold increase in mineral apposition rate (MAR),26%increase in mineralizing surface (MS/BS), and a2.5-fold increase in bone formation rate (BFR). The cortical thickness showed a13%increase.Histology and histomorphometry studies of decalcified bone sections showed that GILZ Tg mice had significantly increased trabecular bones, remarkably increased numbers of osteoblasts (N.Ob/B.Pm)(+93.0%) and osteoblast surface (Ob.S/BS)(+46.7%). A significant reduction in the numbers of marrow adipocytes in Tg mice was also evident. Osteoblasts and marrow adipocytes share a common progenitor cell and these two pathways have a reciprocal relationship.To determine whether increased osteoblasts and bone acquisition in Tg mice was due to a shift of marrow MSC lineage commitment, we performed colony forming unit (CFU) assays. Aliquots of single-cell bone marrow aspirates were seeded in six-well plates and cultured either in regular growth media (DMEM) or, induced with osteogenic or adipogenic induction media. At the indicated time points, cells were fixed and stained with Giemsa (CFU-F), Oil-red O (CFU-Ad), or incubated with alkaline phosphatase (ALP) substrates (CFU-Ob). No significant difference was detected in the numbers of CFU-F colonies between Tg and WT mice, indicating that normal stem/progenitor cell function was not affected. However, significant increase in the numbers of CFU-Ob and decrease in the numbers of CFU-Ad colonies were detected in Tg marrow cell cultures. In line with the CFU-Ob data, separate osteogenic induction experiments with a higher initial cell seeding density confirmed a pro-osteogenic activity of GILZ.To confirm the lineage preferential effect of GILZ at gene expression level, we performed real-time qRT-PCR analysis and found that the basal levels of key osteogenic regulators Runx2and Osx mRNA increased2.4-and4.7-fold, respectively, and the master adipogenic regulator PPARy2decreased by67.5%in Tg cells. Following6days of osteogenic induction, the mRNA levels of Osx increased2.7-fold and PPARy2decreased by92%in Tg cells. Interestingly, no significant increase of Runx2mRNA was detected between Tg and WT cells under this condition, possibly due to the maximal expression of wild type cells in osteogenic conditions even in the absence of GILZ overexpression. The mRNA levels of GILZ in WT and Tg cells before and after osteogenic induction were also examined and, as expected, the levels of GILZ were significantly higher in Tg cells at baseand osteogenic induction conditions. Together, these results demonstrated that transgenic expression of GILZ in bone altered bone marrow mesenchymal stem/progenitor cell lineage commitment favoring osteogenic, rather than adipogenic, differentiation pathway.To determine whether GILZ-mediated increase in bone acquisition was accompanied by an increase in bone turnover, we examined osteoclastogenesis both in vivo, using decalcified tibia samples. TRAP stain experiment showed that neither the numbers of osteoclasts (N.Oc/B.Pm) nor the osteoclast surfaces (N.Oc/BS) were different between GILZ Tg and WT bone samples. These results indicated that overexpression of GILZ in MSC/osteoprogenitor cells did not have an effect on osteoclastic activity. This conclusion is in line with the fact that the GILZ transgene is not expressed in BMMs of Tg mice.PPARy is a key factor controlling bone marrow MSC osteogenic and adipogenic lineage commitment. We showed previously that GILZ binds specifically to a tandem repeat C/EBP-binding site in the promoter region of the PPARy gene and represses its transcription (16). To investigate the molecular mechanism by which GILZ inhibits PPARy gene transcription, we examined protein-protein and protein-DNA interactions between GILZ and C/EBPs in the PPARy promoter region using co-immunoprecipitation (co-IP) and electrophoretic mobility shift assays (EMSA). HA-tagged GILZ and Flag-tagged C/EBP expression vectors were cotransfected into293T cells.48hr after transfection, cells were harvested and whole cell lysates were immunoprecipitated with anti-Flag monoclonal antibody, separated by SDS-gel, and then detected with anti-HA antibody in Western blot. Results showed that HA-tagged GILZ co-precipitates with Flag-tagged C/EBPα,-β,and-δ. Same results were obtained when the experiment was performed in a reverse order, i.e., immunoprecipitate with anti-HA antibody and detect with anti-Flag antibody (not shown). The expression of Flag-C/EBPs and HA-GILZ from the transfected plasmids is shown by Western blot using anti-Flag and anti-HA antibodies, respectively. Because C/EBPs transcriptionally activate, and GILZ inhibits, PPARy transcription via binding to the same DNA element in the PPARy promoter, we hypothesized that GILZ represses PPARy gene transcription by inhibiting the binding of C/EBPs to the PPARy promoter, or by disrupting C/EBP transactivation functions. To test these hypotheses, we performed EMSA assay using affinity-purified GST-C/EBPβ, His-GILZ, and a30bp DNA probe containing the tandem repeat C/EBP-binding site corresponding to nucleotides-316to-346within the PPARy promoter. Results showed that GILZ did not disrupt C/EBPβ DNA-binding activity; instead, by forming a GILZ-C/EBP-DNA tertiary complex, it enhanced or stabilized C/EBPβ DNA-binding activity and this protein complex can be super-shifted by anti-C/EBPβ antibody. These interactions were confirmed by DNA pull-down assay using biotin-labeled DNA probe and cell lysates from GILZ (or GFP)-expressing MSCs in the presence or absence of60x excess of unlabeled DNA probe as competitors, and by GST pull-down assay using purified GST-C/EBP and His-tagged GILZ proteins. Based on these, as well as our previous studies showing that GILZ inhibits C/EBP-mediated PPARy gene transcription, we propose that binding of GILZ-C/EBP complex to the PPARy promoter interferes the transactivation function of C/EBPs, possibly, by blocking the physical interactions between C/EBPs and co-activators assembled in the PPARy gene promoter region or altering the conformation of the general transcriptional machinery.In conclusion, our results showed that GILZ is capable of increasing bone acquisition in vivo and this action is mediated via a mechanism involving the inhibition of PPARy gene transcription and shifting of bone marrow MSC/progenitor cell lineage commitment in favor of osteoblast pathway.