Explore The Role Of Osteocytes In Myeloma Metastasis

Background and Objectives:Multiple myeloma(MM)arises and resides in the bone marrow(BM)microenvironment where it is supported by various stromal populations.The BM microenvironment is a complex system comprised of many cell types,including hematopoietic cells,stromal cells,adipocytes,and endothelial cells,along with cells within the surrounding bone itself.Of these cell types,bone marrow stromal cells,osteoblasts,and osteoclasts have been extensively studied in the context of MM progression and osteolytic disease.However osteocytes,the most abundant cell type in the bone accounting for over 90%of the total cell population,has been largely ignored.Osteocytes were historically thought to be largely inactive cells,but newer studies have revealed that osteocytes carry out many important biological functions largely via the secretion of important regulatory molecules such as the WNT antagonist sclerostin and receptor activator of nuclear factor kappa b ligand(RANKL).Interestingly,when osteocytes undergo apoptosis,they and/or the surrounding live osteocytes secrete soluble molecules,such as RANKL and vascular endothelial growth factor(VEGF)that activate bone resorption and angiogenesis.In addition,osteocyte apoptosis occurs during a number of pathological conditions,including microfracture and skeletal unloading.Osteocyte apoptosis was also shown in MM patients,where it contributes to bone destruction characteristic of MM.However,whether loss of osteocyte viability contributes to MM cell growth and progression/metastasis in bone remains largely unknown.Therefore,in this study the role of osteocyte apoptosis on both MM growth and metastasis,as well the underlying mechanisms regarding regulation of the BM microenvironment by osteocytes,was determined.Methods:1.Cell lines and cell cultureMLO-Y4 or MLO-A5 osteocyte-like cells were kindly provided by Dr.Linda Bonewald(Indiana University).These cells were maintained at 70%to 80%confluence in Minimum Essential Medium Eagle-Alpha Modification(?MEM)with nucleosides(Corning,Corning,NY)supplemented with 5%calf serum,5%fetal bovine serum(FBS),100 U/mL penicillin,and 100 ?g/mL streptomycin in a 37?incubation chamber at 5%CO2.Mouse 5TGM1 or 5TGM1 expressing firefly luciferase(5TGM1-luc)MM cells were a gift from Dr.Fenghuang Zhan(Iowa University),and human CAG MM cells were provided by Dr.Ralph Sanderson(University of Alabama,Birmingham).5TGM1 and CAG cells were grown in RPMI 1640 medium(Corening)supplemented with 10%FBS,100 U/mL penicillin,and 100?g/mL streptomycin in a 37? incubation chamber at 5%CO2.2.Generating a MM mouse model with selective osteocyte apoptosisDiphtheria toxin receptor(DTR)transgenic mice were obtained from Riken BioResource Center,Japan.These mice express the human DTR selectively in osteocytes,and a single injection of diphtheria toxin(DT)results in specific osteocyte apoptosis in vivo.These DTR founder mice were crossed with C57B1/KaLwRij mice(Harlan Laboratories Inc.),an immunocompetent syngeneic model of mouse MM that faithfully replicates many aspects of human MM,to generate "syngeneic DTR"mice.The C57Bl/KaLwRij phenotype was verified in syngeneic DTR mice by PCR for a characteristic deletion of the Samsn1 gene.Wild type and DTR positive progeny were injected with a single dose of either PBS of DT(12.5 ?g/kg)and mice were sacrificed after two days to confirm osteocyte apoptosis.TUNEL staining of mouse hind limbs demonstrated significantly increased osteocyte apoptosis in DTR mice injected with DT compared to DTR mice injected with PBS or WT mice injected with DT.In addition,osteoblasts on the bone surface and cells in the bone marrow remained unstained,indicating the specificity for apoptosis in osteocytes.The percent of dead osteocytes remained mostly unchanged by 7 days post DT injection(?25%),however by 40 days this number had increased slightly(?35%).3.SCID-hu animal modelSCID-hu animals were generated as previously described.Briefly,human fetal long bones were halved and subcutaneously implanted into each side of the dorsum of a SCID mouse.After 6 to 8 weeks of engraftment,105 of either parental CAG MM cells(mild MM)or CAG cells engineered to be more aggressive(aggressive MM)were injected directly into the marrow cavity of one human bone implant.Mice were sacrificed after 7 weeks and both human bone grafts were isolated for further analysis.4.5TGM1 i.v.and intra-tibial injection animal modelMale and female 8 to 10 week old syngeneic DTR mice were initiall injected i.p.with 6.25 or 12.5 ?g/kg DT or PBS as control.After 7 days,2×106 5TGM1 cells were injected i.v.via lateral tail veins,or 1 ×105 5TGM1 cells were directly injected into the right tibia of each mouse(primary leg)while the other leg remained uninjected(contralateral leg).In a separate experiment,2×106 5TGM1 cells were injected i.v.into mice 40 days after the initial dose of DT or PBS.Serum was collected biweekly,and IgG2b?(a soluble marker of 5TGM1 cells)levels were measured by ELISA.Weekly bioluminescent imaging also noninvasively tracked the extent of bone-homing and tumor growth in vivo.All animal studies were performed in accordance with NIH guidelines and approved by the University of Alabama at Birmingham(UAB)Institutional Animal Care and Use Committee(IACUC).5.Statistical analysisStatistical comparisons between two experimental groups were analyzed by t-test.For comparisons among multiple groups,analysis of variance followed by a post hoc Bonferroni correction was used.Data were considered significantly different when P<0.05 and are reported as such.Results:1.Osteocytes apoptosis occurs in distant bone sites prior to arrival of MM cells in the SCID-hu modelWe initially evaluated osteocytes in bone biopsies from 14 healthy donors,10 MGUS,and 35 untreated MM patients.In agreement with the findings of Giuliani et al.,an increase in empty lacunae(dead osteocytes)was observed in MM patients compared to both MGUS or healthy bone.To study this important observation,the SCID-hu MM model,in which human fetal long bones are implanted and engraft in each side of the dorsum of a SCID mouse,was utilized.One bone was injected directly with either "mild" or "aggressive" human CAG MM cells,while the other remained uninjected.Significant osteocyte death was noted in CAG MM cell-injected primary human bone grafts(data not shown)compared with uninjected grafts.More interestingly,however,was the significant increase in osteocyte death(indicated by both empty lacunae and TUNEL positive osteocytes)in the contralateral bone of mice bearing aggressive MM tumors compared to mild tumors.Importantly,tumor cells were not detected in the contralateral bones of either group.These data suggest that MM cells secrete soluble factor(s)that induce osteocyte apoptosis in distant bone sites,even prior to MM colonization of the new bone sites.2.Osteocyte apoptosis results in increased dissemination and growth of MM in boneTo investigate whether the increase in apoptotic osteocytes in distant bones contributes to MM cell homing and progression,a mouse model with inducible osteocyte apoptosis(syngeneic DTR)was developed as described in the "Materials and methods".Syngeneic DTR mice were initially injected i.p.with DT(6.25 or 12.5?g/kg)and separated into two groups:"Pre-MM" or "Tumor bearing".Pre-MM mice were sacrificed 7 days following DT injection to evaluate the status of the bone and BM immediately prior to MM tumor injection.The short term(2×106 5TGM1-luc i.v.injection 7 days post DT injection)and long term(2×106 5TGM1-luc i.v.injection 40 days post DT injection)effects of osteocyte apoptosis on MM bone homing and growth in bone were also investigated.For short term effects,two separate experiments were performed using either 6.25 or 12.5 ?g/kg DT or PBS as a control for each.Osteocyte-ablated(DTR+DT)mice showed increased tumor burden via bioluminescent imaging compared to DTR+PBS which was confirmed by a significant increase in serum IgG2b?(a biomarker of tumor burden)in DTR+DT mice.Intriguingly,a dose dependent effect of osteocyte apoptosis on MM growth was apparent,as the high dose of DT yielded an increased ratio of tumor burden compared to the low dose of DT.The long term effects of osteocyte apoptosis were similar as DTR+DT mice showed significantly larger tumors as determined by bioluminescence imaging and serum IgG2b? levels.Together,these data demonstrate that osteocyte apoptosis promotes MM progression in bone and that these effects persist long after osteocyte death.Importantly,no difference in tumor burden was observed among WT+PBS and WT+DT mice(Supplemental figure 3),confirming the conclusion that the increased tumor growth in DTR+DT mice was the result of DT-induced osteocyte apoptosis.Intra-tibial injections MM cells was next used to determine the effects of osteocyte apoptosis on MM bone-to-bone metastasis.Syngeneic DTR mice were injected with DT(12.5 ?g/kg)or PBS,and 7 days later were injected with 1×1055TGM1-luc MM cells directly into the BM cavity of the right tibia.After 4 weeks,mice showed no overall difference in primary tumor size via bioluminescent imaging in the primary tumor-injected bones.To identify distant metastases,BM was harvested from primary(injected)and contralateral(uninjected)legs from DTR+DT and DTR+PBS mice with equivalently sized primary tumors and BM cells were stained with mouse anti-CD 138 antibodies(another marker of MM cells.Flow cytometry confirmed a similar tumor burden in primary legs.However,a significant increase in CD138+MM cells was observed in DTR+DT contralateral legs compared to the DTR+PBS group,suggesting the increased metastasis of MM cells to the contralateral limb.No difference in CD138+ cells was seen in the BM between groups prior to MM cell injection.To confirm this result,another group of uninjected legs was harvested from mice with comparable primary tumors.Histological analysis of H&E sections revealed that 3 out of 4 DTR+DT mice contained defined tumors in uninjected legs compared to only 1 out of 3 DTR+PBS mice.Combined,these data support the idea that osteocyte apoptosis facilitates MM progression from a primary bone site to new bone sites.3.Osteocyte apoptosis results in dynamic alterations in the BM,including an immuno-suppressive bone microenvironment,low immune response to MM cells and ostocyte signaling.1)To elucidate how osteocyte apoptosis promotes MM growth and metastasis,the levels of cytokines/chemokines in Pre-MM BM was examined 7 days after osteocyte apoptosis was induced.Cytokine arrays revealed many molecules important for MM progression were increased in DTR+DT mice compared to control.Increased IL-6,VEGF,MIP-1?,and VCAM-1 levels were observed.All of these cytokines have be shown to support MM growth and bone homing.Additionally,increased expression of many other molecules that regulate different aspects of the BM microenvironment in MM was detected.These factors include EGF and VEGF(known regulators of angiogenesis)as well as M-CSF,RANKL,IL-11 and MMP-9 which are important mediators of bone resorptio.Many of these factors have only recently been appreciated to aid in myeloid derived suppressor cell(MDSC)activation and immune suppression,such as GM-CSF,IL-6,VEGF,and IL-1.Intriguingly,increased IL-10 expression was observed in MDSCs from the BM of DTR+DT mice,implicating these cells as a direct source of the IL-10 elevated in the BM supernatant.These data demonstrate that osteocyte apoptosis regulates BM cytokines that directly promote MM progression through attracting MM cells and supporting MM growth,and indirectly via alterations in the BM microenvironment.2)Osteocyte apoptosis results in dynamic alterations in the BM cellular compartment and osteocyte signaling.Osteocytes have been reported to be the primary source of RANKL to initiate of bone resorption and bone turnover.Because increased RANKL was observed in BM supernatant after induction of osteocyte apoptosis,whether osteocytes were the direct source was investigated via IHC.Indeed,a significantly greater number of RANKL positive osteocytes was observed in DTR mice injected with DT compared to control(7 days post DT or PBS injection;Pre-MM).In addition,at this early time point a significant increase in the number of blood vessels in the BM was also apparent.These changes persisted in tumor bearing DTR+DT mice 4 weeks after 5TGM1-luc MM cell injection,as those with increased osteocyte apoptosis also had more RANKL positive osteocytes compared to tumor bearing DTR+PBS mice(Fig.5E).Interestingly,tumor bearing DTR+DT mice had significantly more osteoclasts on the bone surface,suggesting that osteocyte apoptosis,likely through upregulation of RANKL,also promotes osteoclastogenesis in the setting of tumor.Hif-1a was reported to be related with angiogenesis,indeed we observed increased experssion of Hif-1a in osteocyte from opoptotic osteocyte mice.And also TRAP+osteoclast were increased.Blood vessels also remained significantly increased in the BM in DTR+DT mice compared to DTR+PBS after 4 weeks of tumor growth.Osteoclastogenesis,bone destruction,and angiogenesis are critical physiologic functions for MM progression.therefore these data provide direct evidence that osteocyte apoptosis promotes a BM microenvironment that supports MM progression by independently regulating osteoclasts and endothelial cells,and potentially other cells in the bone and BM microenvironment.3)Osteocyte apoptosis induces an immuno?suppressive bone microenvironment Defects in osteocyte signaling or osteocyte apoptosis can disrupt normal hematopoiesis.Considering our findings regarding cytokines in the BM,the effects of osteocyte apoptosis on a broad range of immune cells in the BM was investigated.At 7 days following induction of apoptosis by DT,a significant increase in a number of immune-suppressive cells,including myeloid derived suppressor cells(MDSCs),T regulatory cells(Tregs),and a loosely defined population of B cells with regulatory functions(Bregs)was observed,whereas the number of dendritic cells(antigen presenting cells)decreased.The difference in immune cell numbers was specific to the BM,as no changes were observed in the circulation or spleen(data not shown).Both male and female mice showed similar immune cells profiles in the BM after osteocyte apoptosis.These data suggest that osteocyte apoptosis leads to an immunosuppressive BM microenvironment that has the potential to suppress an anti-tumor response and allows tumor proliferation and disease progression.4)Tumor bearing mice with apoptotic osteocytes have compromised immune reactivityDue to the marked alterations in the immune profile following induction of osteocyte apoptosis,the effect of osteocyte apoptosis on immune activity in tumor bearing mice was examined next.Because a physical interaction between tumor cells and T cells is required for cell killing,total T cells(CD3+)and cytotoxic T cells(CD8+)in tumor bearing legs were identified to determine T cell localization within the BM.Interestingly,the number of tumor-infiltrated CD3+and CD8+T cells was significantly decreased in DTR+DT mice with osteocyte apoptosis compared to DTR+PBS control.The activation of BM-derived T cells was also examined by flow cytometry,7 and 28 days after tumor inoculation.Activated cytotoxic T cells destroy tumor cells by producing and releasing the cytotoxins perforin and granzyme.Importantly,after 7 days of tumor cell injection,DTR+DT mice showed a significantly decreased response to the tumor,indicated by granzyme B and perforin expression,compared to control mice.By week 4 both groups showed heavy immune suppression;nevertheless,DTR+DT mice maintained a significantly lower number of cytotoxic T cells than DTR+PBS mice.Taken together,these data indicate osteocyte apoptosis suppresses the immune interaction and response to MM cells in bone.5)Osteocyte-derived soluble molecules have direct anti-MM effects in vitroLittle is known about the potential direct effect(s)that osteocytes have on MM cells,therefore to investigate potential mechanisms a series of in vitro experiments examining the interaction between osteocytes and MM cells were performed.Because osteocytes secrete both RANKL and VEGF,known MM chemoattractants,the ability of osteocytes to directly attract MM cells was elucidated.Surprisingly,compared to control fresh medium(FM),MM cells migrated significantly less towards healthy osteocyte conditioned medium(CM).However,this effect was partially mitigated by CM from osteocytes that were previously serum starved to induce apoptosis(as described in the supplemental methods),which was significantly more chemoattractive to MM cells than healthy osteocyte CM.Based on this data,the effect of osteocyte-derived soluble molecules on MM proliferation was determined.As observed with tumor cell migration,osteocyte CM had a significant inhibitory effect on MM cell proliferation which was reduced in apoptotic osteocyte CM.MM cells were next cultured with healthy and apoptotic MLO-Y4 osteocytes separated by a membrane that only allowed crossing of soluble molecules.The results revealed that in the absence of cell-cell contact,healthy osteocytes significantly inhibited MMI cell growth,but this effect was partially diminished when MM cells were cultured with apoptotic osteocytes.Finally,to determine the molecular underpinnings of this decreased proliferation,western blots were performed on 5TGM1 MM cells after 24 hours of culture in CM of healthy or apoptotic MLO-Y4 osteocytes.Interestingly,both a decrease in Ki-67,a marker of proliferation,and an increase in cleaved caspase-3,a marker of apoptosis was observed,in MM cells cultured in healthy osteocyte CM compared to FM.In agreement with the functional assays,MM cells in apoptotic osteocyte CM had increased Ki-67 and decreased cleaved caspase-3 compared to healthy osteocyte CM.Collectively,these data reveal a surprising,direct inhibitory effect of healthy osteocyte-derived soluble molecules on MM cell processes important for progression.The molecules secreted by healthy osteocytes significantly decreased the proliferation of MM cells as well induced MM cell apoptosis in vitro.Conversely,when MM cells were cultured in the CM of apoptotic osteocytes or in transwell dishes separated from apoptotic osteocytes,MM cell proliferation increased.Although the mechanism(s)responsible for this effect remain to be elucidated,it is clear that both in vivo and in vitro apoptotic osteocytes support MM progression in the BM microenvironment.Conclusions:1.Osteocytes apoptosis occurs frequently in MM patients and animal model further validate osteocyte apoptosis occures in distant bone sites prior to arrival of MM cells in the SCID-hu model2.Osteocyte apoptosis results in increased dissemination and growth of MM in bone3.Osteocyte apoptosis results in dynamic alterations in the BM microenvironment including cytokine profile favorable for MM progression,an immuno-suppressive bone microenvironment,low immune-response to MM.