Advertisement

Streptozotocin-induced diabetes in rats diminishes the size of the osteoprogenitor pool in bone marrow

  • E. Weinberg
    Affiliations
    Department of Oral Biology, The Maurice and Gabriela Goldschleger School of Dental Medicine, Tel-Aviv University, Tel-Aviv, Israel
    Search for articles by this author
  • T. Maymon
    Affiliations
    Department of Oral Biology, The Maurice and Gabriela Goldschleger School of Dental Medicine, Tel-Aviv University, Tel-Aviv, Israel
    Search for articles by this author
  • O. Moses
    Affiliations
    Department of Periodontology, The Maurice and Gabriela Goldschleger School of Dental Medicine, Tel-Aviv University, Tel-Aviv, Israel
    Search for articles by this author
  • M. Weinreb
    Correspondence
    Corresponding author at: Department of Oral Biology, The Maurice and Gabriela Goldschleger School of Dental Medicine, Tel-Aviv University, Tel-Aviv 69978, Israel. Tel.: +972 3 6406430; fax: +972 3 6409250.
    Affiliations
    Department of Oral Biology, The Maurice and Gabriela Goldschleger School of Dental Medicine, Tel-Aviv University, Tel-Aviv, Israel
    Search for articles by this author
Published:December 05, 2013DOI:https://doi.org/10.1016/j.diabres.2013.11.015

      Abstract

      Aims

      Bone formation is reduced in animals and humans with type 1 diabetes, leading to lower bone mass and inferior osseous healing. Since bone formation greatly depends on the recruitment of osteoblasts from their bone marrow precursors, we tested whether experimental type 1 diabetes in rats diminishes the number of bone marrow osteoprogenitors.

      Methods

      Diabetes was induced by 65 mg/kg streptozotocin and after 4 weeks, femoral bone marrow cells were extracted and cultured. Tibia and femur were frozen for further analysis.

      Results

      The size of the osteoprogenitor pool in bone marrow of diabetic rats was significantly reduced, as evidenced by (1) lower (∼35%) fraction of adherent stromal cells (at 24 h of culture); (2) lower (20–25%) alkaline phosphatase activity at 10 days of culture; and (3) lower (∼40%) mineralized nodule formation at 21 days of culture. Administration of insulin to hyperglycemic rats normalized glycemia and abrogated most of the decline in ex vivo mineralized nodule formation. Apoptotic cells in tibial bone marrow were more numerous in hyperglycemic rats. Also, the levels of malondialdehyde (indicator of oxidative stress) were significantly elevated in bone marrow of diabetic animals.

      Conclusions

      Experimental type 1 diabetes diminishes the osteoprogenitor population in bone marrow, possibly due to increased apoptosis via Oxidative Stress. Reduced number of osteoprogenitors is likely to impair osteoblastogenesis, bone formation, and bone healing in diabetic animals.

      Keywords

      1. Introduction

      Osteoporosis is a common complication of diabetes mellitus (DM) and patients with either type 1 or type 2 DM experience a higher incidence of fractures [
      • Schwartz A.V.
      Diabetes mellitus does it affect bone?.
      ,
      • Yamagishi S.
      • Nakamura K.
      • Inoue H.
      Possible participation of advanced glycation end products in the pathogenesis of osteoporosis in diabetic patients.
      ]. In addition, type 1 DM, induced in rats or mice, results in reduced bone mass [
      • Verhaeghe J.
      • van Herck E.
      • Visser W.J.
      • Suiker A.M.
      • Thomasset M.
      • Einhorn T.A.
      • et al.
      Bone and mineral metabolism in BB rats with long-term diabetes. Decreased bone turnover and osteoporosis.
      ,
      • Hamada Y.
      • Kitazawa S.
      • Kitazawa R.
      • Fujii H.
      • Kasuga M.
      • Fukagawa M.
      Histomorphometric analysis of diabetic osteopenia in streptozotocin-induced diabetic mice: a possible role of oxidative stress.
      ]. Common to all these studies is that DM is associated with a reduction in bone formation rate [
      • Hamada Y.
      • Kitazawa S.
      • Kitazawa R.
      • Fujii H.
      • Kasuga M.
      • Fukagawa M.
      Histomorphometric analysis of diabetic osteopenia in streptozotocin-induced diabetic mice: a possible role of oxidative stress.
      ,
      • Krakauer J.C.
      • McKenna M.J.
      • Buderer N.F.
      • Rao D.S.
      • Whitehouse F.W.
      • Parfitt A.M.
      Bone loss and bone turnover in diabetes.
      ]. Systemically, reduced bone formation leads to bone fragility, and locally, it hampers bone healing, such as in femoral fractures [
      • Funk J.R.
      • Hale J.E.
      • Carmines D.
      • Gooch H.L.
      • Hurwitz S.R.
      Biomechanical evaluation of early fracture healing in normal and diabetic rats.
      ], bone marrow ablation [
      • Lu H.
      • Kraut D.
      • Gerstenfeld L.C.
      • Graves D.T.
      Diabetes interferes with the bone formation by affecting the expression of transcription factors that regulate osteoblast differentiation.
      ], and insertion of titanium implants [
      • Takeshita F.
      • Murai K.
      • Iyama S.
      • Ayukawa Y.
      • Suetsugu T.
      Uncontrolled diabetes hinders bone formation around titanium implants in rat tibiae. A light and fluorescence microscopy, and image processing study.
      ]. Thus, reduced osteoblastogenesis and bone formation are prominent features of DM.
      The actual mechanism of DM-associated osteopenia is yet unclear. Lack of insulin action and the presence of hyperglycemia per se are some obvious possibilities. Recently, attention has been drawn to AGEs (advanced glycation end products) and oxidative stress (OxS, exaggerated production of reactive oxygen species (ROS)) as major mechanisms of many diabetic complications [
      • Figueroa-Romero C.
      • Sadidi M.
      • Feldman E.L.
      Mechanisms of disease: the oxidative stress theory of diabetic neuropathy.
      ,
      • Prasad A.
      • Bekker P.
      • Tsimikas S.
      Advanced glycation end products and diabetic cardiovascular disease.
      ].
      Accelerated AGE formation due to hyperglycemia in DM [
      • Kim W.
      • Hudson B.I.
      • Moser B.
      • Guo J.
      • Rong L.L.
      • Lu Y.
      • et al.
      Receptor for advanced glycation end products and its ligands: a journey from the complications of diabetes to its pathogenesis.
      ] leads to their accumulation in bone tissue, which may participate in the pathogenesis of DM-related osteoporosis [
      • Yamagishi S.
      • Nakamura K.
      • Inoue H.
      Possible participation of advanced glycation end products in the pathogenesis of osteoporosis in diabetic patients.
      ].
      Hyperglycemia in DM causes OxS either directly (via glucose overloading of the mitochondria) or indirectly (secondary to signaling pathways such as AGEs and Polyol)) [
      • Figueroa-Romero C.
      • Sadidi M.
      • Feldman E.L.
      Mechanisms of disease: the oxidative stress theory of diabetic neuropathy.
      ]. The increased ROS levels result in cell death, tissue damage and impaired healing. As in many other DM-associated pathologies, OxS has been implicated in the pathogenesis of DM-related loss of bone mass [
      • Hamada Y.
      • Kitazawa S.
      • Kitazawa R.
      • Fujii H.
      • Kasuga M.
      • Fukagawa M.
      Histomorphometric analysis of diabetic osteopenia in streptozotocin-induced diabetic mice: a possible role of oxidative stress.
      ,
      • Manolagas S.C.
      • Almeida M.
      Gone with the Wnts: beta-catenin, T-cell factor, forkhead box O, and oxidative stress in age-dependent diseases of bone, lipid, and glucose metabolism.
      ], possibly via osteoblast apoptosis.
      It is well known that bone marrow includes hemopoietic and stromal compartments and that osteoblast precursors reside within the latter [
      • Owen M.
      Lineage of osteogenic cells and their relationship to the stromal system.
      ]. In vitro, explanted bone marrow cells form fibroblastic colonies initiated by cells called CFUf (colony forming units-fibroblastic) [
      • Friedenstein A.J.
      • Chailakhjan R.K.
      • Lalykina K.S.
      The development of fibroblast colonies in monolayer cultures of guinea-pig bone marrow and spleen cells.
      ,
      • Friedenstein A.J.
      • Deriglasova U.F.
      • Kulagina N.N.
      • Panasuk A.F.
      • Rudakowa S.F.
      • Luriá E.A.
      • et al.
      Precursors for fibroblasts in different populations of hematopoietic cells as detected by the in vitro colony assay method.
      ]. Given appropriate culture conditions (such as dexamethasone, organic phosphate, and ascorbate), these cells express bone-associated markers and form mineralized bone-like nodules [
      • Maniatopoulos C.
      • Sodek J.
      • Melcher A.H.
      Bone formation in vitro by stromal cells obtained from bone marrow of young adult rats.
      ,
      • Herbertson A.
      • Aubin J.E.
      Dexamethasone alters the subpopulation make-up of rat bone marrow stromal cell cultures.
      ]. The number of osteogenic CFUs (CFU-O) declines in conditions characterized by reduced bone formation rate, such as aged rats [
      • Keila S.
      • Kelner A.
      • Weinreb M.
      Systemic prostaglandin E2 increases cancellous bone formation and mass in aging rats and stimulates their bone marrow osteogenic capacity in vivo and in vitro.
      ], ovariectomized osteopenic rats [
      • Tabuchi C.
      • Simmons D.J.
      • Fausto A.
      • Russell J.E.
      • Binderman I.
      • Avioli L.V.
      Bone deficit in ovariectomized rats. Functional contribution of the marrow stromal cell population and the effect of oral dihydrotachysterol treatment.
      ], IL-10 knockout osteopenic mice [
      • Dresner-Pollak R.
      • Gelb N.
      • Rachmilewitz D.
      • Karmeli F.
      • Weinreb M.
      Interleukin 10-deficient mice develop osteopenia: decreased bone formation, and mechanical fragility of long bones.
      ], and unloaded, osteopenic rat bones [
      • Keila S.
      • Pitaru S.
      • Grosskopf A.
      • Weinreb M.
      Bone marrow from mechanically unloaded rat bones expresses reduced osteogenic capacity in vitro.
      ]. On the other hand, we reported that administration of anabolic doses of PGE2 into young and old rats increases bone formation and, in parallel, the number of bone marrow osteoprogenitors [
      • Keila S.
      • Kelner A.
      • Weinreb M.
      Systemic prostaglandin E2 increases cancellous bone formation and mass in aging rats and stimulates their bone marrow osteogenic capacity in vivo and in vitro.
      ,
      • Weinreb M.
      • Suponitzky I.
      • Keila S.
      Systemic administration of an anabolic dose of PGE2 in young rats increases the osteogenic capacity of bone marrow.
      ]. Taken together, these studies indicate that the rate of bone formation greatly depends on the recruitment of osteoblasts from their marrow progenitors [
      • Marie P.J.
      Human osteoblastic cells: a potential tool to assess the etiology of pathologic bone formation.
      ] and a lower number of osteoprogenitors within the marrow stroma may impede osteoblast generation and bone formation and will compromise bone mass and healing.
      The objective of this study was to test whether type 1 DM in rats, which is known to cause diminished bone formation and lower bone mass, results in a reduced number of osteoprogenitors in bone marrow.

      2. Materials and methods

      Four-month-old Wistar rats (8–10 per group) were used in all experiments in this study, except where noted otherwise. All animal procedures were approved by the animal use ethics committee of the Faculty of Medicine, Tel-Aviv University.

      2.1 Materials

      Chemicals for tissue culture were from Biological Industries (Beit Haemek, Israel), unless otherwise stated. Dexamethasone (DEX), streptozotocin (STZ), ascorbic acid, naphthol AS-MX phosphate, fast red violet B, phosphatase substrate, alkaline buffer solution, silver nitrate, sodium carbonate and formalin were from Sigma–Aldrich (Rehovot, Israel). Beta-glycerophosphate (β-GP) was from Calbiochem (La Jolla, CA, USA). Tissue culture dishes were from Nunc (Rosekilde, Denmark). HbA1c levels were measured with the Glyco-tek column kit from Helena Laboratories (Beaumont, TX, USA). Ketamine chlorhydrate was from Rhone-Mérieux (Lyon, France) and Xylazine from Vitamed (Bat-Yam, Israel). Sustained-release insulin implants were from LinShin (Toronto, ON, Canada). OXI-TEK TBARS assay kit was from Enzo Life Sciences (Lausen, Switzerland) and BCA protein determination kit was from Pierce (Rockford, IL, USA).

      2.2 Induction of DM and insulin repletion

      Diabetes in rats was induced with a single intra-peritoneal (IP) administration of streptozotocin (65 mg/kg of body weight) diluted in citrate buffer (0.01 M, pH = 4.3). Control animals receive the buffer alone. Animals were given food and water ad libitum and body weight was continuously monitored. Blood glucose level was evaluated at regular intervals using a glucometer (Accu-Check, Roche Diagnostics, Basel, Switzerland) and rats with blood glucose level >250 mg/dL were considered diabetic.
      In one of the experiments, STZ-injected animals were treated with insulin via sustained-release implants or with identical implants without insulin (sham). Success of insulin repletion was monitored via blood glucose and HbA1c levels.

      2.3 Isolation and enumeration of BMSCs

      Four weeks after injection of STZ, rats were sacrificed with an overdose of ketamine chlorhydrate (90 mg/kg) and xylazine (10 mg/kg), followed by asphyxiation with carbon dioxide. Blood was collected from the tail vein for final glucose and HbA1C measurements. Femurs were retrieved and bone marrow was expelled and pooled from both femurs of each animal. Cells were counted with a hemocytometer and seeded in 6-well plates at 2 × 107 cells/well in a medium composed of Minimum Essential Medium-Alpha containing 5.5 mmol/L d-glucose in triplicates for each assay. This medium was supplemented with 13% fetal Calf Serum (FCS) + 2 mM glutamine + 100 μ/mL penicillin + 100 mg/mL streptomycin + 12.5 U/mL Nystatin (basic medium). After 24 h, cultures were washed with PBS to remove non adherent cells and attached cells were collected with 0.25 w/v trypsin/0.02 w/v EDTA, counted and their number calculated as percent of the cells seeded.

      2.4 Osteoblastic differentiation

      Explanted cells were seeded in 6-well plates as described, in basic culture medium supplemented with 10 mM β-GP + 50 μg/mL ascorbic acid + 10 nM DEX (osteogenic medium). Cultures were washed with PBS after 24 h to remove non-adherent cells and were cultured for 10 or 21 days at 5% CO2 and 37 °C in the same medium, which was changed twice a week. The number of osteoprogenitor cells was evaluated by measuring alkaline phosphatase (ALP) activity and mineralized nodule formation as described below.

      2.5 ALP activity

      For in situ histochemical assessment, cultures were fixed on day 10 in a 1:1:1.5 solution of 10% formalin/methanol/water for 2 h. Dishes were stained with naphthol AS-MX phosphate and fast red violet B solution and the area of positively stained (purple) nodules (representing CFU-ALP = ALP-positive colony forming units) was determined per dish using an image analysis system (NOVA, BIOQUANT Corporation, Nashville).
      For biochemical measurement of ALP activity, cells were washed in PBS on day 10, scraped in 10 mmol/L Tris-HCl buffer (pH 7.6) containing 10 mmol/L MgCl2 and 0.1% Triton X-100, and stored at −20 °C until use. Protein content was measured using the BCA protein determination kit and lysates were incubated with phosphatase substrate and alkaline buffer solution for 15 min at 37 °C and then quenched by addition of 0.2 N NaOH. The resulting optical density was measured at 405 nm using a SpectraMax 190 Microplate reader (Molecular Devices, Sunnyvale, CA). ALP activity was expressed as International Units per milligram protein (IU/mg protein).

      2.6 Mineralized nodule formation

      On day 21, cultures were washed in PBS, fixed as indicated above and stained with the Von Kossa method [
      • Keila S.
      • Pitaru S.
      • Grosskopf A.
      • Weinreb M.
      Bone marrow from mechanically unloaded rat bones expresses reduced osteogenic capacity in vitro.
      ]. Briefly, cells were washed with distilled water, treated with 5% silver nitrate for 15 min, washed again with distilled water, treated with 5% sodium carbonate in 25% formalin for 5 min and washed with tap water. The area of mineralized nodules (stained black, representing CFU-O = colony forming unit, osteoblastic) was determined per dish using the same image analysis system. Due to the vast number of dishes for BMSC cultures generated from each animal, the experiment described hitherto was divided into 2 similar parts. In each part, values belonging to control (normoglycemic) animals were transformed to 100%, and data from the 2 parts were thus merged.

      2.7 In situ measurement of bone marrow cell apoptosis

      Tibiae from 4 animals from each group were decalcified with 10% EDTA, embedded in paraffin and sectioned longitudinally at 5 μm. Apoptotic cells were detected using the DeadEnd Colorimetric TUNEL System (Promega, CA) as per the manufacturer's instructions. TUNEL positive cells in bone marrow were counted in the epiphyseal bone marrow in 6–8 adjacent fields under a 10× objective of a Zeiss Axioplan 2 microscope.

      2.8 Measurement of lipid peroxidation in bone

      Tibiae (5–6 per group) were cleaned from the adhering muscles; the proximal metaphysis was frozen at −70 °C until use (4–6 weeks). Then, thawed bones were homogenized with a polytron® tissue disperser (Kinematica, Luzern, Switzerland) in PBS, and centrifuged at 8000 × g for 8 min at 4 °C to remove cortical bone and epiphyseal cartilage. The levels of malondialdehyde (MDA), an indicator of lipid peroxidation and hence tissue oxidative stress, were determined in supernatants by the thiobarbituric acid-reacting substances (TBARS) assay (OXI-TEK kit) according to the manufacturer's instructions. The intensity of fluorescence was read using a SpectraMax M5 Microplate reader (Molecular Devices) at 530 nm excitation and 550 nm emission. Protein content was measured using BCA protein determination kit. Values of TBARS fluorescence were expressed as nmol MDA per mg of protein.

      2.9 Statistical analysis

      Means and standard deviations were calculated for all parameters and analyzed with non-paired t tests using the SPSS (IBM) software.

      3. Results

      Administration of STZ successfully induced diabetes in all animals, as evidenced by increased mean blood glucose at sacrifice (588.2 ± 19.3 mg/dL vs. 128.3 ± 29.7 mg/dL in control rats (P < 0.001)). HbA1c levels were elevated accordingly (11.54 ± 1.33% (102 mmol/mol) vs. 5.25 ± 0.96% (34 mmol/mol), respectively, P < 0.001).
      The size of the osteoprogenitor pool in bone marrow was estimated by 3 parameters: the relative number (%) of adherent (stromal) cells (CFU-F), the extent of CFU-ALP (ALP-positive colonies) and the extent of CFU-O (mineralized tissue forming colonies). Fig. 1 shows that the fraction of stromal bone marrow cells is significantly reduced (∼35% lower) in hyperglycemic rats compared with normoglycemic animals. When these cells were induced to differentiate into osteoblasts for 10 days, the area of CFU-ALP was significantly (>20%) lower in hyperglycemic, compared with normoglycemic animals (Fig. 2A ). In support of this finding, we used an alternative biochemical method to measure ALP activity in BMSC cultures and found it to be significantly (>20%) lower in hyperglycemic rats (Fig. 2B). When osteoblastic differentiation was induced for 21 days, and mineralized bone-like tissue was formed in culture, the extent of CFU-O (Von-Kossa positive colonies) was significantly (∼40%) smaller in hyperglycemic animals (Fig. 3A ). These data support the notion that STZ-induced hyperglycemia is associated with a reduction in the osteoprogenitor population in bone marrow.
      Figure thumbnail gr1
      Fig. 1Number (mean and standard deviation (SD)) of adherent (stromal) bone marrow cells in normoglycemic (control) and diabetic (STZ) rats expressed as % of the number found in control animals. Difference is statistically significant by non-paired t-test at P < 0.001.
      Figure thumbnail gr2
      Fig. 2(A) Dish area (mean and SD) occupied by ALP-positive colonies (CFU-ALP) in BMSC cultures of normoglycemic (control) and diabetic (STZ) rats expressed as % of the value found in control animals. Difference is statistically significant by non-paired t-test at P < 0.05. Under the bar of each group a representative ALP-stained dish is depicted. (B) ALP activity (determined biochemically, mean and SD) of BMSC cultures of normoglycemic (control) and diabetic (STZ) rats expressed as % of the values found in control animals. Difference is statistically significant by non-paired t-test at P < 0.04.
      Figure thumbnail gr3
      Fig. 3(A) Dish area (mean and SD) occupied by mineralized, Von Kossa-positive, colonies (CFU-OB) in BMSC cultures of normoglycemic (control) and diabetic (STZ) rats, expressed as % of the value found in control animals. Difference is statistically significant by non-paired t-test at P < 0.001. Under the bar of each group a representative Von Kossa-stained dish is depicted. (B) Dish area (mean and SD) occupied by mineralized, Von Kossa-positive, colonies (CFU-OB) in BMSC cultures of normoglycemic (control), diabetic (STZ) and insulin-treated diabetic (Ins) rats, expressed as % of the value found in control animals. Difference between the diabetic and the other 2 groups is statistically significant by non-paired t-test at P < 0.01 or P < 0.05, as indicated.
      Administration of insulin to STZ-injected animals via sustained-release implants normalized glycemia (i.e. returned blood glucose and HbA1C levels to normal values (data not shown) and abrogated most of the decline in CFU-O colonies (Fig. 3B).
      In a follow-up experiment, we found that apoptotic cells were more numerous (>2-fold) in the tibial bone marrow of hyperglycemic, compared with normoglycemic rats (Fig. 4).
      Figure thumbnail gr4
      Fig. 4Representative histologic micrographs of tibial epiphysis showing in situ detected apoptotic (TUNEL-stained) cells (marked by black arrows). B = bone trabeculae; M = bone marrow. Hematoxylin counter-stain. Scale bar = 100 μm. Mean and SD of cell counts are given under each photomicrograph. Difference is statistically significant by non-paired t-test at P < 0.001.
      Finally, to test the possible involvement of OxS in the STZ-induced reduction in the bone marrow osteoprogenitor population, the levels of malondialdehyde (MDA), an indicator of lipid peroxidation and hence tissue oxidative stress, were found to be significantly (∼35%) elevated in the proximal tibial metaphysis (which comprises mainly of bone marrow) of diabetic, compared with normoglycemic, animals (Fig. 5).
      Figure thumbnail gr5
      Fig. 5Concentrations (mean and SD) of malondialdehyde (MDA, expressed as nmol/mg protein) in the tibial metaphysis of normoglycemic (control) and hyperglycemic (STZ) rats. Difference is statistically significant by non-paired t-test at P < 0.01.
      Thus, bone marrow of hyperglycemic rats contains fewer osteoprogenitors, a higher number of apoptotic cells and increased levels of an oxidative stress biomarker.

      4. Discussion

      The animal model we used, i.e. 4 weeks post-diabetes induction with STZ in rats, has been repeatedly reported to result in significant bone loss [
      • Bain S.
      • Ramamurthy N.S.
      • Impeduglia T.
      • Scolman S.
      • Golub L.M.
      • Rubin C.
      Tetracycline prevents cancellous bone loss and maintains near-normal rates of bone formation in streptozotocin diabetic rats.
      ,
      • Shyng Y.C.
      • Devlin H.
      • Sloan P.
      The effect of streptozotocin-induced experimental diabetes mellitus on calvarial defect healing and bone turnover in the rat.
      ] and is therefore most valid for the exploration of diabetes-induced osteopenia.
      The actual mechanism of DM-associated osteopenia is yet unclear. Since several studies demonstrated unaltered [
      • Kemink S.A.
      • Hermus A.R.
      • Swinkels L.M.
      • Lutterman J.A.
      • Smals A.G.
      Osteopenia in insulin-dependent diabetes mellitus; prevalence and aspects of pathophysiology.
      ] or decreased [
      • Gunczler P.
      • Lanes R.
      • Paz-Martinez V.
      • Martins R.
      • Esaa S.
      • Colmenares V.
      • et al.
      Decreased lumbar spine bone mass and low bone turnover in children and adolescents with insulin dependent diabetes mellitus followed longitudinally.
      ] bone resorption in type 1 DM patients, and in diabetic rats [
      • Locatto M.E.
      • Abranzon H.
      • Caferra D.
      • Fernandez M.C.
      • Alloatti R.
      • Puche R.C.
      Growth and development of bone mass in untreated alloxan diabetic rats. Effects of collagen glycosylation and parathyroid activity on bone turnover.
      ] and mice [
      • Botolin S.
      • Faugere M.C.
      • Malluche H.
      • Orth M.
      • Meyer R.
      • McCabe L.R.
      Increased bone adiposity and peroxisomal proliferator-activated receptor-gamma2 expression in type I diabetic mice.
      ], reduced bone formation, as documented in type 1 DM patients [
      • Kemink S.A.
      • Hermus A.R.
      • Swinkels L.M.
      • Lutterman J.A.
      • Smals A.G.
      Osteopenia in insulin-dependent diabetes mellitus; prevalence and aspects of pathophysiology.
      ,
      • Bouillon R.
      • Bex M.
      • Van Herck E.
      • Laureys J.
      • Dooms L.
      • Lesaffre E.
      • et al.
      Influence of age, sex, and insulin on osteoblast function: osteoblast dysfunction in diabetes mellitus.
      ] and in in vivo models of experimental diabetes [
      • Bain S.
      • Ramamurthy N.S.
      • Impeduglia T.
      • Scolman S.
      • Golub L.M.
      • Rubin C.
      Tetracycline prevents cancellous bone loss and maintains near-normal rates of bone formation in streptozotocin diabetic rats.
      ,
      • Shyng Y.C.
      • Devlin H.
      • Sloan P.
      The effect of streptozotocin-induced experimental diabetes mellitus on calvarial defect healing and bone turnover in the rat.
      ,
      • Shires R.
      • Teitelbaum S.L.
      • Bergfeld M.A.
      • Fallon M.D.
      • Slatopolsky E.
      • Avioli L.V.
      The effect of streptozotocin-induced chronic diabetes mellitus on bone and mineral homeostasis in the rat.
      ] is likely to be the cause of osteopenia. This suggests that a decrease in osteoblast number and/or activity is the major mechanism for the DM-related reduction in bone formation. While systemic bone remodeling requires constant recruitment of new osteoblasts, trabecular bone formation in response to “acute” healing models is even more dependent on rapid recruitment of osteoblasts from their stromal precursors. Thus, according to our working hypothesis, it is not surprising that bone formation in response to femoral fractures [
      • Funk J.R.
      • Hale J.E.
      • Carmines D.
      • Gooch H.L.
      • Hurwitz S.R.
      Biomechanical evaluation of early fracture healing in normal and diabetic rats.
      ], bone marrow ablation [
      • Lu H.
      • Kraut D.
      • Gerstenfeld L.C.
      • Graves D.T.
      Diabetes interferes with the bone formation by affecting the expression of transcription factors that regulate osteoblast differentiation.
      ], and dental implants [
      • Takeshita F.
      • Murai K.
      • Iyama S.
      • Ayukawa Y.
      • Suetsugu T.
      Uncontrolled diabetes hinders bone formation around titanium implants in rat tibiae. A light and fluorescence microscopy, and image processing study.
      ] is diminished in diabetic animals.
      Taken together, these studies imply that the decreased bone formation in type 1 DM may stem from deficit in the recruitment of osteoblasts from their marrow osteoprogenitors. Our results in this study, showing reduced numbers of osteoprogenitor cells in bone marrow of hyperglycemic rats (also seen in diabetic mice [
      • Lozano D.
      • de Castro L.F.
      • Dapía S.
      • Andrade-Zapata I.
      • Manzarbeitia F.
      • Alvarez-Arroyo M.V.
      • et al.
      Role of parathyroid hormone-related protein in the decreased osteoblast function in diabetes-related osteopenia.
      ]) could point to this direction. Repletion of insulin in diabetic animals in this study normalized glycemia and abrogated most of the decline in ex vivo mineralized nodule formation, suggesting causal relationship between the diabetic condition and decreased OP number within the bone marrow. Our finding of reduced size of the OP pool four weeks after induction of DM agrees well with many reports documenting reduced bone formation as early as 4 weeks after injection of STZ [
      • Bain S.
      • Ramamurthy N.S.
      • Impeduglia T.
      • Scolman S.
      • Golub L.M.
      • Rubin C.
      Tetracycline prevents cancellous bone loss and maintains near-normal rates of bone formation in streptozotocin diabetic rats.
      ,
      • Shyng Y.C.
      • Devlin H.
      • Sloan P.
      The effect of streptozotocin-induced experimental diabetes mellitus on calvarial defect healing and bone turnover in the rat.
      ].
      In addition, marrow adiposity is known to increase in diabetic humans and rodents, in whom bone formation is suppressed [
      • Botolin S.
      • McCabe L.R.
      Bone loss and increased bone adiposity in spontaneous and pharmacologically induced diabetic mice.
      ,
      • Fowlkes J.L.
      • Bunn R.C.
      • Liu L.
      • Wahl E.C.
      • Coleman H.N.
      • Cockrell G.E.
      • et al.
      Runt-related transcription factor 2 (RUNX2) and RUNX2-related osteogenic genes are down-regulated throughout osteogenesis in type 1 diabetes mellitus.
      ], suggesting a reciprocal relationship between adiposity and bone formation. Therefore, our hypothesis that diabetes reduces the number of osteoprogenitors in bone marrow, certainly fits within this paradigm. Interestingly, metformin, a well-known anti-diabetic drug, exerts concomitant pro-osteogenic as well as anti-adipogenic effects on rat mesenchymal stem cells [
      • Gao Y.
      • Xue J.
      • Li X.
      • Jia Y.
      • Hu J.
      Metformin regulates osteoblast and adipocyte differentiation of rat mesenchymal stem cells.
      ,
      • Molinuevo M.S.
      • Schurman L.
      • McCarthy A.D.
      • Cortizo A.M.
      • Tolosa M.J.
      • Gangoiti M.V.
      • et al.
      Effect of metformin on bone marrow progenitor cell differentiation: in vivo and in vitro studies.
      ], in line with this notion.
      Regarding the mechanism for fewer osteoprogenitors in bone marrow, our data of increased number of apoptotic cells in bone marrow of diabetic rats suggests (although it does not prove definitively) that apoptosis of BMSCs/OPs is a likely explanation for the reduction in their number, induced by STZ. There are many studies showing that diabetic conditions have detrimental effects (including apoptosis) on osteoblastic cells in vivo (e.g. [
      • Motyl K.J.
      • McCauley L.K.
      • McCabe L.R.
      Amelioration of type I diabetes-induced osteoporosis by parathyroid hormone is associated with improved osteoblast survival.
      ,
      • Liang J.
      • Chen H.
      • Pan W.
      • Xu C.
      Puerarin inhibits caspase-3 expression in osteoblasts of diabetic rats.
      ]) or that AGEs (whose levels increase in diabetic bone) cause in vitro apoptosis of various osteoblastic cells [
      • Alikhani M.
      • Alikhani Z.
      • Boyd C.
      • MacLellan C.M.
      • Raptis M.
      • Liu R.
      • et al.
      Advanced glycation end products stimulate osteoblast apoptosis via the MAP kinase and cytosolic apoptotic pathways.
      ,
      • Mercer N.
      • Ahmed H.
      • Etcheverry S.B.
      • Vasta G.R.
      • Cortizo A.M.
      Regulation of advanced glycation end product (AGE) receptors and apoptosis by AGEs in osteoblast-like cells.
      ]. Although very little is known about the possible apoptogenic effect of the diabetic conditions on osteoprogenitors within the bone marrow, this possibility seems likely since AGEs and high concentrations of glucose in vitro inhibit the proliferation and/or osteogenic differentiation of mesenchymal stem cells [
      • Gopalakrishnan V.
      • Vignesh R.C.
      • Arunakaran J.
      • Aruldhas M.M.
      • Srinivasan N.
      Effects of glucose and its modulation by insulin and estradiol on BMSC differentiation into osteoblastic lineages.
      ,
      • Stolzing A.
      • Coleman N.
      • Scutt A.
      Glucose-induced replicative senescence in mesenchymal stem cells.
      ]. In clear support for our hypothesis that diabetic conditions induce apoptosis of osteoprogenitors, we recently demonstrated directly that AGE-BSA increases apoptosis of rat BMSCs in vitro 2- to 3-fold in a caspase-dependent manner (Weinberg et al., [
      • Weinberg E.
      • Maymon T.
      • Weinreb M.
      AGEs induce caspase-mediated apoptosis of rat BMSCs via TNFα production and oxidative stress.
      ]).
      The most prominent mechanisms proposed for the impairment of bone formation in diabetes include hyperglycemia per se, lack of insulin signaling, accumulation of AGEs and oxidative stress (OxS). Recently, mounting evidence focuses on the latter two as plausible mechanisms to many of the DM-associated pathologies such as nephropathy, retinopathy, neuropathy, and impaired dermal healing [
      • Ahmed N.
      Advanced glycation endproducts – role in pathology of diabetic complications.
      ], including osteoporosis [
      • Yamagishi S.
      Role of advanced glycation end products (AGEs) in osteoporosis in diabetes.
      ].
      Specifically, the search for the intracellular pathways in which diabetic conditions induce apoptotic cell loss in various tissues, has highlighted the participation of oxidative stress (OxS), which may be associated with the pathogenesis of diabetic bone disorder. In vitro data confirm that OxS can cause death of many cell types [
      • Maiese K.
      • Chong Z.Z.
      • Shang Y.C.
      Mechanistic insights into diabetes mellitus and oxidative stress.
      ], including osteoblasts [
      • Chen R.M.
      • Wu G.J.
      • Chang H.C.
      • Chen J.T.
      • Chen T.F.
      • Lin Y.L.
      • et al.
      2,6-Di-isopropylphenol protects osteoblasts from oxidative stress-induced apoptosis through suppression of caspase-3 activation.
      ]. Studies that examined the relationship between OxS and diabetic bone disease in vivo have suggested that OxS may be involved in the pathogenesis of diabetic osteopenia [
      • Hamada Y.
      • Kitazawa S.
      • Kitazawa R.
      • Fujii H.
      • Kasuga M.
      • Fukagawa M.
      Histomorphometric analysis of diabetic osteopenia in streptozotocin-induced diabetic mice: a possible role of oxidative stress.
      ,
      • Hamada Y.
      • Fujii H.
      • Kitazawa R.
      • Yodoi J.
      • Kitazawa S.
      • Fukagawa M.
      Thioredoxin-1 overexpression in transgenic mice attenuates streptozotocin-induced diabetic osteopenia: a novel role of oxidative stress and therapeutic implications.
      ]. We found that the levels of MDA, an indicator of tissue OxS, were significantly elevated in bone marrow of diabetic, compared with normoglycemic animals, suggesting that the local involvement of OxS is a strong possible explanation for osteoprogenitor apoptosis and reduced bone formation in DM. Our finding is in agreement with those of Mordwinkin et al. [
      • Mordwinkin N.M.
      • Meeks C.J.
      • Jadhav S.S.
      • Espinoza T.
      • Roda N.
      • diZerega G.S.
      • et al.
      Angiotensin-(1-7) administration reduces oxidative stress in diabetic bone marrow.
      ], who documented increased ROS (reactive oxygen species) levels in bone marrow of diabetic mice. In clear support for our hypothesis that OxS is involved in apoptosis of osteoprogenitors in diabetes, we recently demonstrated directly that rat BMSC apoptosis induced by AGE-BSA involves excessive ROS production and is inhibited by the antioxidant NAC (N-acetylcysteine) (Weinberg et al., [
      • Weinberg E.
      • Maymon T.
      • Weinreb M.
      AGEs induce caspase-mediated apoptosis of rat BMSCs via TNFα production and oxidative stress.
      ]).

      5. Conclusions

      This study reports that streptozotocin-induced type 1 diabetes in rats results in a diminution in the size of the osteoprogenitor pool in bone marrow, possibly due to oxidative stress-mediated apoptosis. This detrimental effect of hyperglycemia on bone marrow osteoprogenitors could help explain the reduction in bone formation and bone mass and healing, which are constant features of this disease in humans and animals.

      Conflict of interest

      The authors declare that they have no conflict of interest.

      Acknowledgments

      This work was performed in partial fulfillment of the requirements for a Ph.D. degree of E. Weinberg, Sackler Faculty of Medicine, Tel Aviv University, Israel. This study was supported by grant no. 4824 from the Chief Scientist Office of the Ministry of Health, Government of Israel .

      References

        • Schwartz A.V.
        Diabetes mellitus does it affect bone?.
        Calcif Tissue Int. 2003; 73: 515-519
        • Yamagishi S.
        • Nakamura K.
        • Inoue H.
        Possible participation of advanced glycation end products in the pathogenesis of osteoporosis in diabetic patients.
        Med Hypotheses. 2005; 65: 1013-1015
        • Verhaeghe J.
        • van Herck E.
        • Visser W.J.
        • Suiker A.M.
        • Thomasset M.
        • Einhorn T.A.
        • et al.
        Bone and mineral metabolism in BB rats with long-term diabetes. Decreased bone turnover and osteoporosis.
        Diabetes. 1990; 39: 477-482
        • Hamada Y.
        • Kitazawa S.
        • Kitazawa R.
        • Fujii H.
        • Kasuga M.
        • Fukagawa M.
        Histomorphometric analysis of diabetic osteopenia in streptozotocin-induced diabetic mice: a possible role of oxidative stress.
        Bone. 2007; 40: 1408-1414
        • Krakauer J.C.
        • McKenna M.J.
        • Buderer N.F.
        • Rao D.S.
        • Whitehouse F.W.
        • Parfitt A.M.
        Bone loss and bone turnover in diabetes.
        Diabetes. 1995; 44: 775-782
        • Funk J.R.
        • Hale J.E.
        • Carmines D.
        • Gooch H.L.
        • Hurwitz S.R.
        Biomechanical evaluation of early fracture healing in normal and diabetic rats.
        J Orthop Res. 2000; 18: 126-132
        • Lu H.
        • Kraut D.
        • Gerstenfeld L.C.
        • Graves D.T.
        Diabetes interferes with the bone formation by affecting the expression of transcription factors that regulate osteoblast differentiation.
        Endocrinology. 2003; 144: 346-352
        • Takeshita F.
        • Murai K.
        • Iyama S.
        • Ayukawa Y.
        • Suetsugu T.
        Uncontrolled diabetes hinders bone formation around titanium implants in rat tibiae. A light and fluorescence microscopy, and image processing study.
        J Periodontol. 1998; 69: 314-320
        • Figueroa-Romero C.
        • Sadidi M.
        • Feldman E.L.
        Mechanisms of disease: the oxidative stress theory of diabetic neuropathy.
        Rev Endocr Metab Disord. 2008; 9: 301-314
        • Prasad A.
        • Bekker P.
        • Tsimikas S.
        Advanced glycation end products and diabetic cardiovascular disease.
        Cardiol Rev. 2012; 29: 177-183
        • Kim W.
        • Hudson B.I.
        • Moser B.
        • Guo J.
        • Rong L.L.
        • Lu Y.
        • et al.
        Receptor for advanced glycation end products and its ligands: a journey from the complications of diabetes to its pathogenesis.
        Ann N Y Acad Sci. 2005; 1043: 553-561
        • Manolagas S.C.
        • Almeida M.
        Gone with the Wnts: beta-catenin, T-cell factor, forkhead box O, and oxidative stress in age-dependent diseases of bone, lipid, and glucose metabolism.
        Mol Endocrinol. 2007; 21: 2605-2614
        • Owen M.
        Lineage of osteogenic cells and their relationship to the stromal system.
        in: Bone and mineral research. vol. 3. Elsevier, Amsterdam1985: 1-25
        • Friedenstein A.J.
        • Chailakhjan R.K.
        • Lalykina K.S.
        The development of fibroblast colonies in monolayer cultures of guinea-pig bone marrow and spleen cells.
        Cell Tissue Kinet. 1970; 3: 393-403
        • Friedenstein A.J.
        • Deriglasova U.F.
        • Kulagina N.N.
        • Panasuk A.F.
        • Rudakowa S.F.
        • Luriá E.A.
        • et al.
        Precursors for fibroblasts in different populations of hematopoietic cells as detected by the in vitro colony assay method.
        Exp Hematol. 1974; 2: 83-92
        • Maniatopoulos C.
        • Sodek J.
        • Melcher A.H.
        Bone formation in vitro by stromal cells obtained from bone marrow of young adult rats.
        Cell Tissue Res. 1988; 254: 317-330
        • Herbertson A.
        • Aubin J.E.
        Dexamethasone alters the subpopulation make-up of rat bone marrow stromal cell cultures.
        J Bone Miner Res. 1995; 10: 285-294
        • Keila S.
        • Kelner A.
        • Weinreb M.
        Systemic prostaglandin E2 increases cancellous bone formation and mass in aging rats and stimulates their bone marrow osteogenic capacity in vivo and in vitro.
        J Endocrinol. 2001; 168: 131-139
        • Tabuchi C.
        • Simmons D.J.
        • Fausto A.
        • Russell J.E.
        • Binderman I.
        • Avioli L.V.
        Bone deficit in ovariectomized rats. Functional contribution of the marrow stromal cell population and the effect of oral dihydrotachysterol treatment.
        J Clin Invest. 1986; 78: 637-642
        • Dresner-Pollak R.
        • Gelb N.
        • Rachmilewitz D.
        • Karmeli F.
        • Weinreb M.
        Interleukin 10-deficient mice develop osteopenia: decreased bone formation, and mechanical fragility of long bones.
        Gastroenterology. 2004; 127: 792-801
        • Keila S.
        • Pitaru S.
        • Grosskopf A.
        • Weinreb M.
        Bone marrow from mechanically unloaded rat bones expresses reduced osteogenic capacity in vitro.
        J Bone Miner Res. 1994; 9: 321-327
        • Weinreb M.
        • Suponitzky I.
        • Keila S.
        Systemic administration of an anabolic dose of PGE2 in young rats increases the osteogenic capacity of bone marrow.
        Bone. 1997; 20: 521-526
        • Marie P.J.
        Human osteoblastic cells: a potential tool to assess the etiology of pathologic bone formation.
        J Bone Miner Res. 1994; 9: 1847-1850
        • Bain S.
        • Ramamurthy N.S.
        • Impeduglia T.
        • Scolman S.
        • Golub L.M.
        • Rubin C.
        Tetracycline prevents cancellous bone loss and maintains near-normal rates of bone formation in streptozotocin diabetic rats.
        Bone. 1997; 21: 147-153
        • Shyng Y.C.
        • Devlin H.
        • Sloan P.
        The effect of streptozotocin-induced experimental diabetes mellitus on calvarial defect healing and bone turnover in the rat.
        Int J Oral Maxillofac Surg. 2001; 30: 70-74
        • Kemink S.A.
        • Hermus A.R.
        • Swinkels L.M.
        • Lutterman J.A.
        • Smals A.G.
        Osteopenia in insulin-dependent diabetes mellitus; prevalence and aspects of pathophysiology.
        J Endocrinol Invest. 2000; 23: 295-303
        • Gunczler P.
        • Lanes R.
        • Paz-Martinez V.
        • Martins R.
        • Esaa S.
        • Colmenares V.
        • et al.
        Decreased lumbar spine bone mass and low bone turnover in children and adolescents with insulin dependent diabetes mellitus followed longitudinally.
        J Pediatr Endocrinol Metab. 1998; 11: 413-419
        • Locatto M.E.
        • Abranzon H.
        • Caferra D.
        • Fernandez M.C.
        • Alloatti R.
        • Puche R.C.
        Growth and development of bone mass in untreated alloxan diabetic rats. Effects of collagen glycosylation and parathyroid activity on bone turnover.
        Bone Miner. 1993; 23: 129-144
        • Botolin S.
        • Faugere M.C.
        • Malluche H.
        • Orth M.
        • Meyer R.
        • McCabe L.R.
        Increased bone adiposity and peroxisomal proliferator-activated receptor-gamma2 expression in type I diabetic mice.
        Endocrinology. 2005; 146: 3622-3631
        • Bouillon R.
        • Bex M.
        • Van Herck E.
        • Laureys J.
        • Dooms L.
        • Lesaffre E.
        • et al.
        Influence of age, sex, and insulin on osteoblast function: osteoblast dysfunction in diabetes mellitus.
        J Clin Endocrinol Metab. 1995; 80: 1194-1202
        • Shires R.
        • Teitelbaum S.L.
        • Bergfeld M.A.
        • Fallon M.D.
        • Slatopolsky E.
        • Avioli L.V.
        The effect of streptozotocin-induced chronic diabetes mellitus on bone and mineral homeostasis in the rat.
        J Lab Clin Med. 1981; 97: 231-240
        • Lozano D.
        • de Castro L.F.
        • Dapía S.
        • Andrade-Zapata I.
        • Manzarbeitia F.
        • Alvarez-Arroyo M.V.
        • et al.
        Role of parathyroid hormone-related protein in the decreased osteoblast function in diabetes-related osteopenia.
        Endocrinology. 2009; 150: 2027-2035
        • Botolin S.
        • McCabe L.R.
        Bone loss and increased bone adiposity in spontaneous and pharmacologically induced diabetic mice.
        Endocrinology. 2007; 148: 198-205
        • Fowlkes J.L.
        • Bunn R.C.
        • Liu L.
        • Wahl E.C.
        • Coleman H.N.
        • Cockrell G.E.
        • et al.
        Runt-related transcription factor 2 (RUNX2) and RUNX2-related osteogenic genes are down-regulated throughout osteogenesis in type 1 diabetes mellitus.
        Endocrinology. 2008; 149: 1697-1704
        • Gao Y.
        • Xue J.
        • Li X.
        • Jia Y.
        • Hu J.
        Metformin regulates osteoblast and adipocyte differentiation of rat mesenchymal stem cells.
        J Pharm Pharmacol. 2008; 60: 1695-1700
        • Molinuevo M.S.
        • Schurman L.
        • McCarthy A.D.
        • Cortizo A.M.
        • Tolosa M.J.
        • Gangoiti M.V.
        • et al.
        Effect of metformin on bone marrow progenitor cell differentiation: in vivo and in vitro studies.
        J Bone Miner Res. 2010; 25: 211-221
        • Motyl K.J.
        • McCauley L.K.
        • McCabe L.R.
        Amelioration of type I diabetes-induced osteoporosis by parathyroid hormone is associated with improved osteoblast survival.
        J Cell Physiol. 2012; 227: 1326-1334
        • Liang J.
        • Chen H.
        • Pan W.
        • Xu C.
        Puerarin inhibits caspase-3 expression in osteoblasts of diabetic rats.
        Mol Med Rep. 2012; 5: 1419-1422
        • Alikhani M.
        • Alikhani Z.
        • Boyd C.
        • MacLellan C.M.
        • Raptis M.
        • Liu R.
        • et al.
        Advanced glycation end products stimulate osteoblast apoptosis via the MAP kinase and cytosolic apoptotic pathways.
        Bone. 2007; 40: 345-353
        • Mercer N.
        • Ahmed H.
        • Etcheverry S.B.
        • Vasta G.R.
        • Cortizo A.M.
        Regulation of advanced glycation end product (AGE) receptors and apoptosis by AGEs in osteoblast-like cells.
        Mol Cell Biochem. 2007; 306: 87-94
        • Gopalakrishnan V.
        • Vignesh R.C.
        • Arunakaran J.
        • Aruldhas M.M.
        • Srinivasan N.
        Effects of glucose and its modulation by insulin and estradiol on BMSC differentiation into osteoblastic lineages.
        Biochem Cell Biol. 2006; 84: 93-101
        • Stolzing A.
        • Coleman N.
        • Scutt A.
        Glucose-induced replicative senescence in mesenchymal stem cells.
        Rejuvenation Res. 2006; 9: 31-35
        • Weinberg E.
        • Maymon T.
        • Weinreb M.
        AGEs induce caspase-mediated apoptosis of rat BMSCs via TNFα production and oxidative stress.
        J Mol Endocrinol. 2013; (in press)
        • Ahmed N.
        Advanced glycation endproducts – role in pathology of diabetic complications.
        Diabetes Res Clin Pract. 2005; 67: 3-21
        • Yamagishi S.
        Role of advanced glycation end products (AGEs) in osteoporosis in diabetes.
        Curr Drug Targets. 2011; 12: 2096-2102
        • Maiese K.
        • Chong Z.Z.
        • Shang Y.C.
        Mechanistic insights into diabetes mellitus and oxidative stress.
        Curr Med Chem. 2007; 14: 1729-1738
        • Chen R.M.
        • Wu G.J.
        • Chang H.C.
        • Chen J.T.
        • Chen T.F.
        • Lin Y.L.
        • et al.
        2,6-Di-isopropylphenol protects osteoblasts from oxidative stress-induced apoptosis through suppression of caspase-3 activation.
        Ann N Y Acad Sci. 2005; 1042: 448-459
        • Hamada Y.
        • Fujii H.
        • Kitazawa R.
        • Yodoi J.
        • Kitazawa S.
        • Fukagawa M.
        Thioredoxin-1 overexpression in transgenic mice attenuates streptozotocin-induced diabetic osteopenia: a novel role of oxidative stress and therapeutic implications.
        Bone. 2009; 44: 936-941
        • Mordwinkin N.M.
        • Meeks C.J.
        • Jadhav S.S.
        • Espinoza T.
        • Roda N.
        • diZerega G.S.
        • et al.
        Angiotensin-(1-7) administration reduces oxidative stress in diabetic bone marrow.
        Endocrinology. 2012; 153: 2189-2197