Glucose-Dependent Insulinotropic Polypeptide Receptor Deficiency Leads to Impaired Bone Marrow Hematopoiesis

Fernanda Dana Mantelmacher, Sigal Fishman, Keren Cohen, Metsada Pasmanik Chor, Yuichiro Yamada, Isabel Zvibel and Chen Varol
J Immunol April 15, 2017, 198 (8) 3089-3098; DOI: https://doi.org/10.4049/jimmunol.1601441

Abstract

The bone marrow (BM) contains controlled specialized microenvironments, or niches, that regulate the quiescence, proliferation, and differentiation of hematopoietic stem and progenitor cells (HSPC). The glucose-dependent insulinotropic polypeptide (GIP) is a gut-derived incretin hormone that mediates postprandial insulin secretion and has anabolic effects on adipose tissue. Previous studies demonstrated altered bone microarchitecture in mice deficient for GIP receptor (Gipr−/−), as well as the expression of high-affinity GIP receptor by distinct cells constructing the BM HSPC niche. Nevertheless, the involvement of GIP in the process of BM hematopoiesis remains elusive. In this article, we show significantly reduced representation and proliferation of HSPC and myeloid progenitors in the BM of Gipr−/− mice. This was further manifested by reduced levels of BM and circulating differentiated immune cells in young and old adult mice. Moreover, GIP signaling was required for the establishment of supportive BM HSPC niches during HSPC repopulation in radioablated BM chimera mice. Finally, molecular profiling of various factors involved in retention, survival, and expansion of HSPC revealed significantly lower expression of the Notch-receptor ligands Jagged 1 and Jagged 2 in osteoblast-enriched bone extracts from Gipr−/− mice, which are important for HSPC expansion. In addition, there was increased expression of CXCL12, a factor important for HSPC retention and quiescence, in whole-BM extracts from Gipr−/− mice. Collectively, our data suggest that the metabolic hormone GIP plays an important role in BM hematopoiesis.

Introduction

The gut-derived incretin hormones glucose-dependent insulinotropic polypeptide (GIP) and glucagon-like peptide-1 are secreted by K and L cells, enteroendocrine cells present in the small intestine and duodenum. Their combined incretin effect accounts for ∼70% of the total insulin secreted after oral glucose administration and is mediated by signaling through their respective G protein–coupled receptors, GIP receptor (GIPR) and glucagon-like peptide-1 receptor, which are expressed in pancreatic β-cells (1–3). Recent studies indicated that both incretins have extrapancreatic effects in various organs (1–3), particularly in postprandial nutrient-dependent regulation of bone metabolism (4). An early study demonstrated that Gipr−/− mice have decreased bone size and mass, decreased bone mineral density and content, and lower trabecular bone volume (5), although in another study the reduction in trabecular bone volume was not significant (6). Osteoblast number is similar between wild-type (WT) and Gipr−/− mice, but the latter exhibit significantly increased numbers of multinucleated osteoclasts (6). In contrast, mice overexpressing GIP display significant increases in bone mass, as assessed by densitometry and histomorphometry (7). Overall, these studies indicate that GIP favors bone formation over resorption. However, in another Gipr−/− mouse model, there is, in fact, an increase in the volume and number of trabecular bone, with a reduction in bone matrix strength and quality (8, 9). This is accompanied by a reduction in the number of trabecular osteoclasts and an increase in osteoblast activity, as evidenced by an increased rate of bone formation. Although there is disagreement among these studies, all support a pivotal role for GIP in the remodeling of bone microarchitecture and matrix.

The presence of GIPR was identified in a variety of cells residing in the bone microenvironment, such as osteoblasts, osteoclasts, osteocytes, chondrocytes, and bone marrow (BM) pluripotent mesenchymal stem cells (BM-MSC) (6, 9–14). Highlighting the functional importance of GIP in the BM, its receptor in bone cells shows a similar degree of affinity with its respective receptor in pancreatic β-cells (15). In osteoblasts, GIP increases the expression of type 1 collagen and alkaline phosphatase activity and has an antiapoptotic effect (6, 9, 10, 12). Yet, in another study, GIPR deletion was accompanied by increased osteoblast activity (8). With respect to osteoclasts, there is a dispute whether GIP can directly regulate their bone-resorption activity (6, 13). Endothelial cells also express a functional GIPR (16–18), yet specific GIP-governed regulation of BM endothelial cells has not been established. Recently, it was shown that GIP also exerts a direct protective action against apoptosis in human BM-MSC (14).

The BM hematopoietic stem cell niche is a specialized microenvironment where multiple cell types support the key properties of the resident hematopoietic stem and progenitor cells (HSPC), such as self-renewal and long-term multilineage repopulation capacity. This niche and the hematopoietic cells within it form an ecosystem that governs the ability to maintain homeostatic hematopoiesis and to respond to stress and disease (19–21). Early studies located the BM niche in the region of trabecular bone at the endosteum (22, 23), an area lined with bone-forming osteoblasts and bone-resorbing osteoclasts. This region is filled with numerous arterioles transporting oxygen, nutrients, and growth factors, as well as with a sinusoidal network of venules composed of fenestrated endothelial cells allowing cell migration. For a long time, osteoblasts were considered the pivotal cells supporting hematopoiesis (24–26). However, studies using advanced in vivo imaging revealed that transplanted HSPC are not exclusively adjacent to osteolineage cells and engraft at small endothelial microdomains (27, 28). Moreover, osteoblast-specific depletion of factors critical for HSPC regulation, such as the secreted stromal cell–derived factor 1 (SDF1)/CXCL12 or stem cell factor (SCF), has no effect on BM and spleen hematopoiesis and the reconstitution ability of irradiated mice (29, 30). In the recent past, the vasculature, including endothelial and perivascular cells, emerged as a key structure for the maintenance of HSPC in the BM niche (19–21). The dormant, quiescent HSPC are found around arterioles, where secreted SDF1 or SCF promote their maintenance (31). In contrast, less quiescent or activated HSPC are located near sinusoidal niches, where sinusoidal endothelial cells promote their expansion and self-renewal through Notch signaling (32). The importance of Notch signaling in hematopoiesis was further underscored by studies in which endothelial cell–specific deletion of the Notch ligand Jagged 1 resulted in decreased hematopoiesis and exhaustion of the HSPC pool (33).

We recently discovered an immune-regulatory role for GIP in a model of obesity-induced insulin resistance; we showed that continuous treatment with a GIP analog reduced the levels of circulating and visceral adipose tissue–infiltrating proinflammatory innate and adaptive immune cells (34). Given the expression of GIPR in cells constituting the BM HSPC niche and the altered architecture and matrix of Gipr−/− bones, we endeavored to study the regulation of BM hematopoiesis by GIP. We show that mice deficient in the receptor for this metabolic hormone possess an altered BM hematopoiesis that is manifested by significantly reduced numbers and fractions of HSPC, myeloid progenitors (MP), and differentiated immune cells of various lineages. This phenotype appears in young and old adult mice. Moreover, we show that normal BM hematopoiesis depends on proper GIP signaling in BM niche cells rather than in immune cells. Finally, we show altered expression of specific factors affecting HSPC quiescence, retention, and expansion.

Materials and Methods

Animals

Male C57BL/6 WT mice and C57BL/6 Gipr−/− mice (35) were maintained in the animal facility of the Tel-Aviv Sourasky Medical Center. Mice had unrestricted access to food and water, were housed in temperature and humidity–controlled rooms, and were kept on a 12-h light/dark cycle. Use of animals was in accordance with the National Institutes of Health policy on the care and use of laboratory animals and was approved by the Tel-Aviv Sourasky Medical Center Animal Use and Care Committee.

BM radiation chimeras

Six-week-old recipient male WT C57BL/6 mice or Gipr−/− mice were subjected to whole-body irradiation (950 rad) using a TrueBeam linear accelerator (Varian Medical Systems). The next day, 5 × 106 or 5 × 105 BM cells were harvested from the hind limbs (tibia and femur) of the appropriate donor mice (WT or Gipr−/−) and injected i.v. into the respective recipients. Mice were allowed to reconstitute their BM cells for 8 wk. Subsequently, HSPC and immune cell populations were analyzed in BM and blood.

Isolation of single-cell suspensions from peripheral blood and BM

Peripheral blood mononuclear and polymorphonuclear cells from WT and Gipr−/− mice were isolated using BD FACS Lysing Solution (cat. no. 349202; Becton Dickinson), according to the manufacturer’s instructions. For the isolation of BM cells, femurs of WT and Gipr−/− mice were carefully dissected and mechanically crushed using a mortar and pestle in Liver Digest Medium (cat. no. 17703034; Thermo Fisher Scientific) supplemented with 0.1% DNase 1 (Roche), followed by a 30-min digestion at 37°C with shaking (150 rpm). After incubation, cells were washed and underwent RBC lysis using BD FACS Lysing Solution before staining.

Flow cytometry analysis of immune cell populations

Abs used to identify HSPC and MP included CD45.2 (104), CD117 (cKit) (2B8), Sca1 (D7), and lineage mixture (17A2/RB6-8C5/RA3-6B2/Ter-119/M1/70) (all from BioLegend). HSPC were defined as CD45+Lin−Sca1+cKit+ cells, and MP were identified as CD45+Lin−Sca1−cKit+ cells. The following Abs were used to identify BM and peripheral blood–differentiated immune cells: CD115 (CSF-1R) (AFS98), CD11b (M1/70), Ly6C (HK1.4), Ly6G (1A8), CD3 (17A2), NK1.1 (PK136), B220 (RA3-6B2), and TCRb (H57-597) (all from BioLegend), and Siglec-F (E50-2440; BD Biosciences). The cell cycle in HSPC was analyzed as previously described (36). BM cells extracted from crushed bones were stained for HSPC surface markers, as described above, and were washed, fixed, and permeabilized, using a BD Cytofix/Cytoperm kit (cat. no. 554722). Subsequently, cells were stained for the expression of the proliferation marker Ki67 (SolA15; eBioscience) and washed. Cells were then resuspended in 1 ml of PBS containing 1 mg/ml RNase A, 0.1% saponin, and 25 μg/ml propidium iodide (to label genomic DNA). Cells were analyzed with a FACSCanto II flow cytometer using FACSDiva (both from Becton Dickinson) or FlowJo software.

Real-time PCR

For gene-expression profiling of total BM cells, BM single-cell suspensions were isolated, as described above, and total RNA was extracted using TRI Reagent (Sigma, St. Louis, MO). For gene expression of osteoblast-enriched preparations, femurs were flushed out of BM cells with PBS−/− and subsequently subjected to additional flushing with TRI Reagent. In this case, RNA is extracted out of cells that adhere to the periosteum and endosteum, which contain large numbers of osteoblasts. Total RNA was reverse transcribed using the High-Capacity cDNA Reverse Transcription Kit (cat. no. 4368814; Applied Biosystems). Quantitative real-time PCR was performed with Fast SYBR Green Master Mix (Applied Biosystems) using a Corbett rotor light cycler (Mortlake, Australia). The intron span primers were as follows: Tgfb, 5′-ATTCAGCGCTCACTGCTCTT-3′ (forward) and 5′-GTTGGTATCCAGGGCTCTCC-3′ (reverse); Vegfb, 5′-CCCTGTGTCCCAGTTTGATG-3′ (forward) and 5′-ACATTGCCCATGAGTTCCAT-3′ (reverse); Sdf1, 5′-ACCAGTCAGCCTGAGCTACC-3′ (forward) and 5′-TAATTTCGGGTCAATGCACA-3′ (reverse); Scf, 5′-TGGATGACCTCGTGTTATGC-3′ (forward) and 5′-TTTCTCGGGACCTAATGTTGA-3′ (reverse); Jagged 1 (Jag1), 5′-TGAAGGGGTGCGGTATATCT-3′ (forward) and 5′-AGGTGCCACCGTTTTTACAG-3′ (reverse); Jagged 2 (Jag2), 5′-TCTGTGAGGACCTGGTGGAT-3′ (forward) and 5′-GCAGTTCTTGCCACCAAAGT-3′ (reverse); angiopoietin1 (Angpt1), 5′-CCAGGAGTTGGAGAAGCAAC-3′ (forward) and 5′-TCTGCACAGTCTCGAAATGG-3′ (reverse); E-selectin (Sele), 5′-GCGCTTTCTCTCTGCTCTTG-3′ (forward) and 5′-TGAATTGCCACCAGATGTGT-3′ (reverse); osteopontin (Spp1), 5′-GGTGCCAAGAGTGTGTTTGA-3′ (forward) and 5′-TGAATTGCCACCAGATGTGT-3′ (reverse); alkaline phosphatase (Alpl), 5′-CTGACTGACCCTTCGCTCTC-3′ (forward) and 5′-TCATGATGTCCGTGGTCAAT-3′ (reverse); Osteocalcin (Bglap), 5′-AAGCAGGAGGGCAATAAGGT-3′ (forward) and 5′-TTTGTAGGCGGTCTTCAAGC-3′ (reverse); Runx2, 5′-CGCATTCCTCATCCCAGTAT-3′ (forward) and 5′-TGGCTCAGATAGGAGGGGTA-3′ (reverse); Collagen1α1 (Col1a1), 5′-GAGAGCATGACCGATGGATT-3′ (forward) and 5′-CCTTCTTGAGGTTGCCAGTC-3′ (reverse); osterix (Sp7), 5′-GATGGCGTCCTCTCTGCTT-3′ (forward) and 5′-AGCGTATGGCTTCTTTGTGC-3′ (reverse).

Data analysis using the ImmGen database

Raw microarray data were downloaded from the Gene Expression Omnibus from the GSE15907 study using Partek GS 6.6 (http://www.partek.com/pgs). HSPC and downstream BM progenitors were selected, and a total of 32 sample arrays was analyzed with at least two biological repeats per cell type. Raw gene expression data were obtained for Gipr. For normalization, we obtained the expression of the adiponectin gene (Adipoq), which is uniquely expressed by adipocytes and, hence, served to set up the background gene expression level. As a positive control, we compared GIPR gene expression for that of the SCF receptor (cKit, Kit) and the CD45 leukocyte common Ag (Ptprc); both are expressed by all HSPC and downstream BM progenitors. Data are presented as mean (± SEM).

Statistical analysis

Data were analyzed using an unpaired, two-tailed t test with GraphPad Prism 5.0b (San Diego, CA). Data are presented as mean ± SEM; the p values < 0.05 were considered statistically significant.

Results

Reduced representation and proliferation of HSPC and MP in the BM of young adult Gipr−/− mice

GIP plays an important role in the regulation of bone metabolism. Specifically, Gipr−/− mice exhibit altered bone microarchitecture and biomechanical properties (4, 5, 10, 18, 37). Nevertheless, whether GIP affects the BM hematopoietic activity remains elusive. WT and Gipr−/− mice maintain similar body weights when fed a regular diet (5, 8, 38). However, and in agreement with previous studies (5, 6), femurs isolated from 6-wk-old Gipr−/− mice were lower in length and mass in comparison with age-matched WT mice (Fig. 1A). Flow cytometry analysis revealed a significant reduction in the numbers of total BM cells, specifically CD45+ leukocytes, in the femurs of Gipr−/− mice. However, there was a similar representation of CD45− nonleukocyte BM stromal cells in WT and Gipr−/− femurs (Fig. 1B). In accordance with the reduced numbers of immune cells, Gipr−/− femurs contained significantly fewer Lin−cKit(CD117)+Sca1+ HSPC and Lin−cKit+Sca1− MP cells (Fig. 1C, 1D). Moreover, Gipr−/− mice had a 2-fold reduction in the fraction of HSPC out of total CD45+ living BM immune cells in comparison with WT mice (Fig. 1E).

FIGURE 1.
FIGURE 1.

Gipr−/− mice display reduced numbers and fractions of BM HSPC and MP. (A) Measurements of femoral bone length and mass in 6-wk-old WT and Gipr−/− mice. (B) Bar graph showing the numbers of total BM cells, CD45+ leukocytes, and CD45− nonleukocyte cells per femur of WT and Gipr−/− mice. (C) Representative flow cytometry image showing the gating strategy used to define HSPC and MP cells out of total CD45+Lin− BM cells isolated from femurs of 6-wk-old WT and Gipr−/− mice. (D) Bar graph showing the numbers of HSPC and MP cells per femur of WT and Gipr−/− mice. (E) Bar graph showing the percentage of HSPC cells out of BM CD45+ cells. Data are presented as mean ± SEM. Experiments were repeated two times, n ≥ 6 mice per group. ***p < 0.001, unpaired, two-tailed t test.

We next examined whether the reduced HSPC representation is due to less HSPC proliferation. HSPC cycling was measured on stained fixed and permeabilized cells for the Ki-67 nuclear marker of cell cycling (negative in quiescent cells) and for DNA content with propidium iodide (Fig. 2A), as previously described (36). There was a profound reduction in the proportions of proliferating BM HSPC in Gipr−/− versus WT femurs, which are in G1 (Ki-67+, 2n DNA) or S/G2/M (Ki-67+, > 2n DNA) phases of active cell cycle. In contrast, Gipr−/− femurs contained significantly higher proportions of quiescent HSPC at the G0 stage (Ki-67−, 2n DNA) (Fig. 2B). The numbers of HSPC per femur in all phases of the cell cycle also were reduced significantly in Gipr−/− femurs (Fig. 2C). Therefore, HSPC are more quiescent in bones deficient in GIP signaling.

FIGURE 2.
FIGURE 2.

Reduced proliferation of BM HSPC cells in Gipr−/− BM. (A) Representative flow cytometry images showing the gating strategy used to define BM HSPC distribution in cell cycle phases G0 (Ki-67−, DNA = 2n), G1 (Ki-67+, DNA = 2n), and S/G2/M (Ki-67+, DNA > 2n). (B) Bar graph showing the fraction of BM HSPC for each cell cycle phase. (C) Bar graph showing the number of BM HSPC for each cell cycle phase. Data are presented as mean ± SEM. Experiments were repeated two times, n ≥ 10 mice per group. *p < 0.05, **p < 0.01, ***p < 0.001, unpaired, two-tailed t test.

Reduced levels of differentiated BM and circulating immune cells in young adult Gipr−/− mice

We next examined whether the reduced numbers of BM HSPC and MP cells observed in Gipr−/− mice affect the levels of mature immune cells. Flow cytometry analyses were performed to assess the levels of differentiated myeloid and lymphoid immune cells in the BM (Fig. 3A) and peripheral blood (Fig. 3B), including CD11b+CD115 (M-CSFR)−/loLy6G+ neutrophils, Ly6Chi and Ly6Clo subsets of CD11b+Ly6G−CD115+ monocytes, SiglecF+CD115− eosinophils, CD11b−TCRβ−B220hi B cells, CD11b−B220−Nk1.1−TCRβ+ T cells, CD11b−/loTCRβ−NK1.1+ NK cells, and CD11b−TCRβ+NK1.1+ NKT cells. In comparison with WT femurs, those extracted from Gipr−/− mice contained significantly lower numbers of differentiated CD45+ immune cells, including the myeloid and lymphoid lineages, such as neutrophils, Ly6Chi and Ly6Clo monocytes, T cells, NK cells, and NKT cells, whereas B cells were similarly represented (Fig. 3C). However, in the blood of Gipr−/− mice, CD11b+ myeloid cells, but not CD11b− lymphocytes, were reduced significantly (Fig. 3D). In particular, there was a profound reduction in defined myeloid cell populations, such as neutrophils, monocyte subsets, and eosinophils, whereas there were similar levels of T cells, B cells, and NK cells in comparison with WT mice (Fig. 3E). Therefore, GIPR deficiency results in impaired generation of differentiated immune cells, in particular those belonging to the myeloid lineage.

FIGURE 3.
FIGURE 3.

Gipr−/− mice have reduced numbers of BM and circulating differentiated immune cells. Representative flow cytometry images showing the gating strategy used to define differentiated immune cells in the BM (A) and peripheral blood (B). Ly6G+CD115− neutrophils, SiglecF+CD115− eosinophils, and Ly6Chi and Ly6Clo subsets of CD115+ monocytes were gated on CD45+CD11b+ BM myeloid cells. TCRβ+NK1.1−B220− T cells, B220hiTCRβ− B cells, NK1.1+TCRβ− NK cells, and NK1.1+TCRβ+ NKT cells were gated on CD45+CD11b−/lo BM cells. (C) Graph showing the number of cells for various immune cell subsets in the femurs of 6-wk-old WT (●) and Gipr−/− (▽) mice. Graphs showing the number of differentiated immune cells in 50 μl of peripheral blood from 6-wk-old WT (●) and Gipr−/− (▽) mice, including total CD45+ immune cells, CD11b+ myeloid cells, and CD11b− lymphoid cells (D) and specifically Ly6Chi and Ly6Clo monocytes, neutrophils, eosinophils, B cells, T cells, and NK cells (E). Data are presented as mean ± SEM. Experiments were repeated at least two times, n ≥ 12 mice per group. *p < 0.05, **p < 0.01, ***p < 0.001, unpaired, two-tailed t test.

Impaired BM hematopoiesis persists in old adult Gipr−/− mice

It reported previously that, over time, compensatory mechanisms are initiated to restore bone architecture (5). Therefore, we investigated whether the impaired hematopoiesis phenotype observed in young Gipr−/− mice persisted in 6-mo-old mice. There was a prominent reduction in circulating total CD45+ immune cells in old Gipr−/− mice, including CD11b− lymphoid and CD11b+ myeloid lineages (Fig. 4A) and specifically neutrophils and monocytes (Fig. 4B). Moreover, old Gipr−/− mice had reduced numbers (Fig. 4C) and percentages (Fig. 4D) of BM HSPC. Of note, at this age there was similarity in the length (Fig. 4E) and mass (Fig. 4F) of WT versus Gipr−/− femurs. Collectively, these results uncover impaired hematopoiesis in the BM of mature Gipr−/− mice that persists through adulthood.

FIGURE 4.
FIGURE 4.

Impaired hematopoiesis remains at 18 wk of age. Bar graphs for 18-wk-old WT and Gipr−/− mice showing numbers (per 50 μl of blood) of circulating total CD45+ immune cells and specifically those belonging to CD11b+ myeloid and CD11b− lymphoid lineages (A) and of circulating monocytes and neutrophils (B). Bar graphs showing cell numbers (C) and fractions (D) of HSPC in WT and Gipr−/− femurs. Bar graphs showing femur bone length (E) and mass (F) in WT and Gipr−/− mice. Data are presented as mean ± SEM (n = 5 mice per group). *p < 0.05, unpaired, two-tailed t test.

GIPR deficiency alters the BM HSPC niche

The impaired BM hematopoiesis in Gipr−/− mice may be the consequence of a deficiency in GIPR signaling in HSPC and their immune cell descendants or in various types of nonimmune cells that constitute the BM HSPC niche. Transplantation of whole BM into radioablated hosts leads to engraftment of HSPC and the subsequent restoration of hematopoiesis. To study the importance of GIPR signaling in the recovering immune cells, lethally irradiated WT recipients were engrafted with five million WT or Gipr−/− BM cells (groups A and B, respectively) (Fig. 5A). Flow cytometry analysis at 8 wk following BM engraftment revealed similar numbers and fractions of HSPC (Fig. 5B, 5C) and MP (Fig. 5D, 5E) between groups A and B. Concomitantly with the establishment of groups A and B, Gipr−/− recipient mice were lethally irradiated and reconstituted with WT or Gipr−/− BM cells (groups C and D, respectively) (Fig. 5A). Strikingly, there was a significant reduction in HSPC (Fig. 5B, 5C) and MP (Fig. 5D, 5E) levels, regardless of whether the BM graft was from WT or Gipr−/− mice. In agreement with this, there was a significant reduction in groups C and D, but not in groups A and B, in total circulating CD45+ immune cells and in CD11b+ myeloid and CD11b− lymphocyte cells (Fig. 5F), including representatives of both compartments, such as neutrophils, monocytes, and T cells (Fig. 5G). Therefore, these results show that the number or quality of specialized BM niches available to the donor HSPC is not sufficient in Gipr−/− mice. They also suggest that there may be HSPC-supportive BM niche cells that are not replaced by donor BM graft and require GIP signaling for their correct support of BM hematopoiesis.

FIGURE 5.
FIGURE 5.

The altered BM hematopoiesis in Gipr−/− mice originates from impaired support of BM HSPC niche cells. (A) Four groups of BM chimera mice were established: WT BM into WT recipients (group A), Gipr−/− BM into WT recipients (group B), WT BM into Gipr−/− recipients (group C), and Gipr−/− BM into Gipr−/− mice (group D). Lethally irradiated recipients were reconstituted with 5 × 106 BM cells and analyzed 8 wk later. Bar graphs showing HSPC cell numbers (B) and fractions (C) in femurs extracted from the various BM chimera groups. Bar graphs showing MP cell numbers (D) and fractions (E) in femurs extracted from the various BM chimera groups. Bar graphs showing the number of differentiated immune cells in 50 μl of peripheral blood extracted from the various BM chimera groups, including CD45+ total immune cells, CD11b+ myeloid cells, and CD11b− lymphoid immune cells (F) and specifically neutrophils, monocytes, and T cells (G). Bar graphs showing MP and HSPC numbers per femur (H) and their fractions out of total CD45+ BM immune cells (I) at 8 wk following reconstitution of lethally irradiated WT recipients with 5 × 105 WT or Gipr−/− BM grafts. Bar graphs showing the numbers of differentiated immune cells per femur (J) or per 50 μl of peripheral blood (K) at 8 wk following reconstitution of lethally irradiated WT recipients with 5 × 105 WT or Gipr−/− BM grafts. Data are presented as mean ± SEM. Experiments were repeated once with n = 5 mice per group (A–G) or n ≥ 7 mice per group (H–K). An unpaired, two-tailed t test was used to compare group B, group C, and group D with group A (A–G) or to compare lethally irradiated WT recipients reconstituted with WT or Gipr−/− BM grafts (H–K). *p < 0.05, **p < 0.01, ***p < 0.001. ns, not significant.

We next compared the hematopoietic reconstitution ability of Gipr−/− versus WT BM cells in a more stringent transplantation scheme in which only 500,000 cells were engrafted into lethally irradiated WT recipients. Flow cytometry analysis 8 wk later revealed no significant difference in the numbers and fractions of HSPC and MP cells between WT recipients engrafted with WT or Gipr−/− BM cells (Fig. 5H, 5I). Nevertheless, those reconstituted by a Gipr−/− BM graft exhibited a considerable reduction in total BM CD45+ immune cells, which was significant in the myeloid CD11b+ immune cell compartment, including neutrophils and Ly6Chi monocytes. There also was a significant reduction in T cells, NK cells, and NKT cells, whereas B cells were present in similar amounts in both groups (Fig. 5J). This impaired hematopoiesis was further reflected in the circulation; myeloid and lymphoid immune cells were significantly reduced in WT recipients reconstituted by Gipr−/− BM graft (Fig. 5K).

Although these results may be the outcome of the reduced numbers of transplanted HSPC in the Gipr−/− BM graft (Fig. 1D, 1E), they may also imply that HSPC coming from the Gipr−/− BM niche maintain cell-intrinsic defects that are manifested under challenging conditions. Another possible explanation is that GIP acts directly on HSPC or MP. Therefore, we analyzed the gene expression of its receptor (Gipr) in BM HSPC and their downstream precursors using published data sets available from the ImmGen Consortium database (GSE15907) (Fig. 6). To set up the background expression level for each cell population, Gipr expression was compared with that of adiponectin (Adipoq), which is uniquely expressed by adipocytes. As a positive control, Gipr expression was compared with that of the SCF receptor cKit (Kit) and the CD45 leukocyte common Ag (Ptprc), both of which are ubiquitously expressed by all examined BM precursors. In all examined BM precursor cell populations, with the exception of the common myeloid precursors, Gipr expression was significantly higher in comparison with Adipoq (significance marked by *). Nevertheless, it was tremendously lower than that of Kit (significance marked by #) or Ptprc (significance marked by +) (Fig. 6). Therefore, the relatively poor GIPR gene expression renders it less probable that GIP acts directly through HSPC and their downstream BM precursors.

FIGURE 6.
FIGURE 6.

BM HSPC and their downstream precursors exhibit poor GIPR gene expression. Bar graph showing the raw gene expression of GIPR (Gipr) in comparison with that of adiponectin (Adipoq; statistical significance marked by *), cKit (Kit; statistical significance marked by #), and CD45 (Ptprc; statistical significance marked by +) in BM HSPC and immune cell precursors. Immune cell precursors included: long-term HSPC (LT-HSPC), short-term HSPC (ST-HSPC), multipotent progenitors (MPP), common lymphoid progenitors (CLP), common myeloid progenitors (CMP), megakaryocyte-erythroid progenitors (MEP), granulocyte-macrophage progenitors (GMP) and macrophage-dendritic cell progenitors (MDP). Gene-expression data were extracted from the ImmGen Consortium database (GSE15907). There were at least two biological repeats for each cell population. Data were analyzed by unpaired, two-tailed t test and presented as mean ± SEM. #,+p < 0.05, **,##,++p < 0.01, ***,###,+++p < 0.001.

GIPR deficiency alters BM expression of factors affecting HSPC quiescence, retention, and expansion

Given that Gipr−/− BM HSPC niche cells yielded impaired hematopoiesis, we next assessed the gene expression of various factors affecting HSPC quiescence, retention, and expansion in total bone extracts and in osteoblast-enriched extracts. Among all examined factors, there was a unique 2-fold increase in the expression of the Sdf1 gene encoding the CXCL12 HSPC retention chemokine in total bone extracts of Gipr−/− mice (Fig. 7A). In contrast, in the osteoblast-enriched fraction of Gipr−/− mice, there was a significant reduction in the gene expression of Jag1 and Jag2, encoding for the Jagged 1 and Jagged 2 ligands of Notch-1 receptor, as well as an increase in the expression of osteopontin (OPN) (Spp1) (Fig. 7B). These factors regulate HSPC expansion (20, 33, 39). We also determined the gene expression of various markers associated with osteoblast differentiation and function, including alkaline phosphatase (Alpl), osteocalcin (Bglap), and collagen I (Col1a1), as well as the transcription factors runt related transcription factor 2 (Runx2) and osterix (Sp7). There were no changes in the gene expression of these osteoblast-associated molecules in the osteoblast-enriched bone preparations from Gipr−/ −and WT mice (Fig. 7C). Collectively, these results uncover an important role for GIP in instructing BM hematopoiesis through its regulation of BM HSPC niche cells.

FIGURE 7.
FIGURE 7.

Expression of mediators regulating HSPC quiescence, maintenance, retention, and expansion in total BM cells and in osteoblast-enriched cells of WT and Gipr−/− mice. Bar graphs showing quantitative real-time PCR–based gene-expression profiling of HSPC BM niche–associated factors in total BM cells (A) and osteoblast-enriched cells (B) extracted from 6-wk-old WT and Gipr−/− mice. (C) Expression of genes associated with osteoblast differentiation and function in osteoblast-enriched cells extracted from 6-wk-old WT and Gipr−/− mice. Data are presented as mean ± SEM. Experiments were repeated two times, n ≥ 6 mice per group. *p < 0.05, **p < 0.01, unpaired, two-tailed t test.

Discussion

Our study shows for the first time, to our knowledge, a predominant role for the metabolic hormone GIP and its receptor in BM hematopoiesis. We demonstrate reduced numbers, fractions, and proliferation of HSPC and MP in BM of young and old Gipr−/− mice. This further develops into significantly reduced numbers of BM and circulating differentiated immune cells. Moreover, we show that lack of GIP signaling in stromal cells composing the BM HSPC niche impairs their ability to support the reconstitution of BM hematopoiesis by engrafted HSPC following radioablation. Specifically, Gipr−/− mice exhibit reduced levels of the Notch receptor ligands Jagged 1 and Jagged 2 and increased expression of OPN in osteoblast-enriched BM preparations, as well as increased levels of CXCL12 in total BM extracts.

Bone is a dynamic tissue that is continuously remodeled throughout life to adapt to the mechanical stresses and needs of the developing human skeleton. Previous studies demonstrated the existence of an entero–osseous axis and established a major role for incretins in the nutrient-dependent regulation of bone metabolism (4). With respect to GIP, transgenic mouse models genetically manipulated to obtain GIP overexpression or deficiency of its receptor showed significant alterations in the bone phenotype of adult animals, with contradictory findings (5–9). Earlier studies supported a role for GIP in favoring bone formation over resorption (5–7). In these studies, Gipr−/− mice displayed decreased bone size and mass, reduced trabecular volume and connectivity, impaired biomechanical properties, and reduced bone formation. In contrast, in another Gipr−/− mouse model, there was increased trabecular bone volume and extension of the marrow area diameter (8, 9). Notably, all of these studies agree that there is decreased bone strength and quality and abnormalities in bone mineral density and content, as well as in the collagenous matrix, of Gipr−/− mice. Our BM chimerism approach reveals the incompetence of Gipr−/− bones to support adequate reconstituting hematopoiesis of WT engrafted HSPC. Therefore, the alterations in bone microarchitecture and matrix in Gipr−/− mice may disturb the number, quality, and function of BM HSPC niches. They may also impair the storage capacity of the bone, as well as the differentiation, proliferation, and migration properties of HSPC and their downstream precursors and mature immune cell descendants. Alterations in trabecular bone volume and number are of unique interest given the established importance of this bone compartment to BM hematopoiesis (22, 23). Therefore, further studies are required to assess whether GIP positively regulates BM hematopoiesis through its postprandial remodeling of bone architecture and matrix. Interestingly, we and other investigators (5) observed normalization in femur length and mass in old Gipr−/− mice, yet we show that these mice still exhibit impaired hematopoiesis. To be cautious, although these macro-architecture parameters seem to achieve normality in old Gipr−/− bones, microarchitecture and matrix alterations may still be present.

Our results also suggest that GIP may affect BM hematopoiesis through its direct regulation of the hematopoiesis-supportive activity of stromal cells composing the BM HSPC niche. Indeed, we show a significant decline in the numbers and fraction of HSPC cells that are in an active cell cycle phase and an increase in those that are quiescent. Moreover, molecular profiling of various factors reported to be profoundly involved with the regulation of quiescence, maintenance, and expansion of HSPC (19–21) revealed a marked reduction only in the expression of two of the Notch ligands, Jagged 1 and Jagged 2, in RNA extracted from osteoblast-enriched fractions of Gipr−/− femurs but not from total bone extracts. Emerging evidence highlights Notch signaling, especially that induced by Jagged 1, as a key mediator of HSPC self-renewal and differentiation (reviewed in Refs. 19–21). Therefore, our results offer a possible mechanism by which GIP positively regulates osteoblast-governed support of BM hematopoiesis through their production of Jagged 1 and Jagged 2. Indeed, osteoblasts express a functional GIPR and induce their expression of type 1 collagen and alkaline phosphatase activity in response to its activation (6, 10, 12). Osteoblast numbers are similar between WT and Gipr−/− mice (6), and we show no difference between WT and Gipr−/− mice in terms of their expression of markers characteristic of osteoblast differentiation. Hence, GIP may regulate osteoblast-governed support of hematopoiesis. In this manner, activated osteoblasts produce higher levels of Jagged 1, which supports the self-renewal of Notch-expressing HSPC (24), and this pathway is negatively regulated by OPN (40). We show in this article that hampered GIPR signaling results in increased OPN expression and reduced Jagged 1 expression. Therefore, GIP may positively regulate the expression of Jagged 1 and Jagged 2 and negatively regulate that of OPN. Of note, Jagged 1 promotes bone tissue formation and induces osteoblast differentiation (41, 42), suggesting that some of the observed structural alterations in Gipr−/− bones could be due to the cell-autonomous reduced expression of Jagged 1 in osteoblasts.

CXCL2 was an additional factor with altered expression in Gipr−/− mice; it was increased in total bone extracts but not in the osteoblast-enriched fractions. This immune mediator was associated with effects on HSPC retention (43), quiescence (44, 45), and repopulating activity (45). Various types of BM HSPC niche cells can produce CXCL12 (reviewed in Refs. 19–21). Using a Cxcl12-GFP reporter mouse, perivascular cells surrounding sinusoidal endothelial cells or located near the endosteum were demonstrated as the main source of CXCL12. Hence, these cells were named CXCL12-abundant reticular cells and were shown to be important for the maintenance of the quiescent HSPC pool (46). However, targeted deletion of CXCL12 from osterix-expressing stromal cells, including CXCL12-abundant reticular cells, has no major effect on HSPC function. Deletion of CXCL12 in endothelial cells results in a modest loss of long-term HSPC repopulating activity. Strikingly, its deletion from nestin-negative mesenchymal progenitors is associated with a marked loss of HSPC, long-term repopulating activity, and hematopoietic stem cell quiescence (47). The cellular source and the consequence of CXCL12 elevation in the BM of Gipr−/− mice remain elusive. Moreover, it may be a compensatory response to better retain HSPC in the BM of Gipr−/− mice; perhaps in the future it can be related to the higher proportions of quiescent HSPC in these mice.

We show decreased levels of mature immune cells in the BM and circulation of Gipr−/− mice. In the BM, immune cells belonging to myeloid and lymphoid lineages were reduced, including neutrophils, monocytes, T cells, NK cells, and NKT cells. Yet, in the circulation, only myeloid cells were significantly affected by GIPR deficiency; B cell, T cell, and NK cell numbers were comparable to WT mice. The relatively short half-life of myeloid cells, such as neutrophils, monocytes, and eosinophils, in comparison with effector or memory T or B cells may explain these results. Nevertheless, it may also be that MP are more affected by GIPR deficiency in comparison with lymphoid precursors. When lethally irradiated mice were reconstituted with 500,000 BM cells, GIPR deficiency resulted in a significant impairment of hematopoiesis that was manifested by decreased levels of myeloid and lymphoid BM and blood immune cells. Although this may be the result of reduced HSPC levels in the Gipr−/− BM graft, it may also imply on cell-intrinsic deficits in immune cells. Gene-expression analyses of BM HSPC and their downstream precursors revealed very low relative expression of GIPR on these cells, suggesting that it is less likely that direct GIP-associated cell-intrinsic mechanisms instruct their hematopoiesis activity. Of note, gene-expression levels are not enough to exclude that option.

Our results may suggest GIP as ancillary treatment in conditions in which accelerated hematopoiesis is indicated and required, such as postchemotherapy, with or without BM transplantation, postdrug-induced myelosuppression conditions, or postradioablation. Moreover, GIP agonists and antagonists are being experimentally and clinically tested as novel therapeutic agents, especially for type 2 diabetes (48). Given the emerging extrapancreatic effects of GIP, its therapeutic modulation might be considered for other diabetes-related complications, such as cardiovascular diseases, retinopathy, nephropathy, neuropathy, cognitive impairment, and bone fractures (49). Interestingly, type 2 diabetic patients are relatively immune-compromised and may be prone to various infections (50, 51). The question of whether a decline in the response to GIP, as happens in the pancreas (52), also occurs in the BM and contributes to this susceptibility should be addressed in further studies. Previous studies established a pivotal role for GIP in the remodeling of bone architecture and matrix (4–10, 18, 37). Therefore, its long-term pharmacological manipulation may alter BM HSPC niche availability and function, as well as the storage capacity, differentiation, and migration of immune cells. Moreover, GIP manipulation may alter the hematopoietic supportive activity of various BM HSPC niche cells in which a functional GIPR was documented, such as osteoblasts (6, 9, 10, 12), BM stromal cells (11), and endothelial cells (16, 18). Thus, clinical studies involving GIP agonists/antagonists will have to take into consideration the long-term impact of GIP augmentation or inhibition on BM hematopoiesis. We show that an absence of GIP signaling results mainly in myeloid cell deficiency, whereas adaptive immune cells, such as T and B cells, as well as NK cells, are normally presented in the circulation. Thus, one may expect that long-term GIP manipulation would not pose a dramatic challenge to adaptive immunity but in the long run may lead to myeloid cell deficiency. Yet, the effect of inducible GIP deficiency remains elusive. Collectively, we demonstrate that GIP is an important regulator of the BM HSPC niche function.

Disclosures

The authors have no financial conflicts of interest.

Acknowledgments

We thank Michail Kaplan (Sourasky Medical Center) for animal care and Dr. Nathan Strauss (Sourasky Medical Center) for radiation services. Finally, we thank medical student Or Touval for assistance with gene-expression analyses.

Footnotes

  • ↵1 I.Z. and C.V. are cosenior authors.

  • This work was supported by Israel Science Foundation Grants 35/12 and 1146/16 (to C.V. and S.F.).

  • Abbreviations used in this article:

    BM
    bone marrow
    BM-MSC
    BM pluripotent mesenchymal stem cell
    GIP
    glucose-dependent insulinotropic polypeptide
    GIPR
    GIP receptor
    HSPC
    hematopoietic stem and progenitor cell
    MP
    myeloid progenitor
    OP
    osteopontin
    SCF
    stem cell factor
    SDF-1
    stromal cell–derived factor-1
    WT
    wild-type.

  • Received August 18, 2016.
  • Accepted February 1, 2017.

References

  1. ↵
    1. Baggio L. L.,
    2. D. J. Drucker
    . 2007. Biology of incretins: GLP-1 and GIP. Gastroenterology 132: 2131–2157.
    OpenUrlCrossRefPubMed
    1. Holst J. J.
    2007. The physiology of glucagon-like peptide 1. Physiol. Rev. 87: 1409–1439.
    OpenUrlAbstract/FREE Full Text
  3. ↵
    1. Paschetta E.,
    2. M. Hvalryg,
    3. G. Musso
    . 2011. Glucose-dependent insulinotropic polypeptide: from pathophysiology to therapeutic opportunities in obesity-associated disorders. Obes. Rev. 12: 813–828.
    OpenUrlCrossRefPubMed
  4. ↵
    1. Yavropoulou M. P.,
    2. J. G. Yovos
    . 2013. Incretins and bone: evolving concepts in nutrient-dependent regulation of bone turnover. Hormones (Athens) 12: 214–223.
    OpenUrlCrossRefPubMed
  5. ↵
    1. Xie D.,
    2. H. Cheng,
    3. M. Hamrick,
    4. Q. Zhong,
    5. K. H. Ding,
    6. D. Correa,
    7. S. Williams,
    8. A. Mulloy,
    9. W. Bollag,
    10. R. J. Bollag,
    11. et al
    . 2005. Glucose-dependent insulinotropic polypeptide receptor knockout mice have altered bone turnover. Bone 37: 759–769.
    OpenUrlCrossRefPubMed
  6. ↵
    1. Tsukiyama K.,
    2. Y. Yamada,
    3. C. Yamada,
    4. N. Harada,
    5. Y. Kawasaki,
    6. M. Ogura,
    7. K. Bessho,
    8. M. Li,
    9. N. Amizuka,
    10. M. Sato,
    11. et al
    . 2006. Gastric inhibitory polypeptide as an endogenous factor promoting new bone formation after food ingestion. Mol. Endocrinol. 20: 1644–1651.
    OpenUrlCrossRefPubMed
  7. ↵
    1. Xie D.,
    2. Q. Zhong,
    3. K. H. Ding,
    4. H. Cheng,
    5. S. Williams,
    6. D. Correa,
    7. W. B. Bollag,
    8. R. J. Bollag,
    9. K. Insogna,
    10. N. Troiano,
    11. et al
    . 2007. Glucose-dependent insulinotropic peptide-overexpressing transgenic mice have increased bone mass. Bone 40: 1352–1360.
    OpenUrlCrossRefPubMed
  8. ↵
    1. Gaudin-Audrain C.,
    2. N. Irwin,
    3. S. Mansur,
    4. P. R. Flatt,
    5. B. Thorens,
    6. M. Baslé,
    7. D. Chappard,
    8. G. Mabilleau
    . 2013. Glucose-dependent insulinotropic polypeptide receptor deficiency leads to modifications of trabecular bone volume and quality in mice. Bone 53: 221–230.
    OpenUrlCrossRefPubMed
  9. ↵
    1. Mieczkowska A.,
    2. B. Bouvard,
    3. D. Chappard,
    4. G. Mabilleau
    . 2015. Glucose-dependent insulinotropic polypeptide (GIP) directly affects collagen fibril diameter and collagen cross-linking in osteoblast cultures. Bone 74: 29–36.
    OpenUrl
  10. ↵
    1. Bollag R. J.,
    2. Q. Zhong,
    3. P. Phillips,
    4. L. Min,
    5. L. Zhong,
    6. R. Cameron,
    7. A. L. Mulloy,
    8. H. Rasmussen,
    9. F. Qin,
    10. K. H. Ding,
    11. C. M. Isales
    . 2000. Osteoblast-derived cells express functional glucose-dependent insulinotropic peptide receptors. Endocrinology 141: 1228–1235.
    OpenUrlCrossRefPubMed
  11. ↵
    1. Ding K. H.,
    2. X. M. Shi,
    3. Q. Zhong,
    4. B. Kang,
    5. D. Xie,
    6. W. B. Bollag,
    7. R. J. Bollag,
    8. W. Hill,
    9. W. Washington,
    10. Q. S. Mi,
    11. et al
    . 2008. Impact of glucose-dependent insulinotropic peptide on age-induced bone loss. J. Bone Miner. Res. 23: 536–543.
    OpenUrlCrossRefPubMed
  12. ↵
    1. Pacheco-Pantoja E. L.,
    2. L. R. Ranganath,
    3. J. A. Gallagher,
    4. P. J. Wilson,
    5. W. D. Fraser
    . 2011. Receptors and effects of gut hormones in three osteoblastic cell lines. BMC Physiol. 11: 12.
    OpenUrlCrossRefPubMed
  13. ↵
    1. Zhong Q.,
    2. T. Itokawa,
    3. S. Sridhar,
    4. K. H. Ding,
    5. D. Xie,
    6. B. Kang,
    7. W. B. Bollag,
    8. R. J. Bollag,
    9. M. Hamrick,
    10. K. Insogna,
    11. C. M. Isales
    . 2007. Effects of glucose-dependent insulinotropic peptide on osteoclast function. Am. J. Physiol. Endocrinol. Metab. 292: E543–E548.
    OpenUrlAbstract/FREE Full Text
  14. ↵
    1. Berlier J. L.,
    2. I. Kharroubi,
    3. J. Zhang,
    4. A. Dalla Valle,
    5. S. Rigutto,
    6. M. Mathieu,
    7. V. Gangji,
    8. J. Rasschaert
    . 2015. Glucose-dependent insulinotropic peptide prevents serum deprivation-induced apoptosis in human bone marrow-derived mesenchymal stem cells and osteoblastic cells. Stem Cell Rev. 11: 841–851.
    OpenUrl
  15. ↵
    1. McIntosh C. H.,
    2. M. B. Wheeler,
    3. R. W. Gelling,
    4. J. C. Brown,
    5. R. A. Pederson
    . 1996. GIP receptors and signal-transduction mechanisms. Acta Physiol. Scand. 157: 361–365.
    OpenUrlCrossRefPubMed
  16. ↵
    1. Ding K. H.,
    2. Q. Zhong,
    3. C. M. Isales
    . 2003. Glucose-dependent insulinotropic peptide stimulates thymidine incorporation in endothelial cells: role of endothelin-1. Am. J. Physiol. Endocrinol. Metab. 285: E390–E396.
    OpenUrlAbstract/FREE Full Text
    1. Ding K. H.,
    2. Q. Zhong,
    3. J. Xu,
    4. C. M. Isales
    . 2004. Glucose-dependent insulinotropic peptide: differential effects on hepatic artery vs. portal vein endothelial cells. Am. J. Physiol. Endocrinol. Metab. 286: E773–E779.
    OpenUrlAbstract/FREE Full Text
  18. ↵
    1. Zhong Q.,
    2. R. J. Bollag,
    3. D. T. Dransfield,
    4. J. Gasalla-Herraiz,
    5. K. H. Ding,
    6. L. Min,
    7. C. M. Isales
    . 2000. Glucose-dependent insulinotropic peptide signaling pathways in endothelial cells. Peptides 21: 1427–1432.
    OpenUrlCrossRefPubMed
  19. ↵
    1. Boulais P. E.,
    2. P. S. Frenette
    . 2015. Making sense of hematopoietic stem cell niches. Blood 125: 2621–2629.
    OpenUrlAbstract/FREE Full Text
  20. ↵
    1. Mendelson A.,
    2. P. S. Frenette
    . 2014. Hematopoietic stem cell niche maintenance during homeostasis and regeneration. Nat. Med. 20: 833–846.
    OpenUrlCrossRefPubMed
  21. ↵
    1. Morrison S. J.,
    2. D. T. Scadden
    . 2014. The bone marrow niche for haematopoietic stem cells. Nature 505: 327–334.
    OpenUrlCrossRefPubMed
  22. ↵
    1. Gong J. K.
    1978. Endosteal marrow: a rich source of hematopoietic stem cells. Science 199: 1443–1445.
    OpenUrlAbstract/FREE Full Text
  23. ↵
    1. Lord B. I.,
    2. N. G. Testa,
    3. J. H. Hendry
    . 1975. The relative spatial distributions of CFUs and CFUc in the normal mouse femur. Blood 46: 65–72.
    OpenUrlAbstract/FREE Full Text
  24. ↵
    1. Calvi L. M.,
    2. G. B. Adams,
    3. K. W. Weibrecht,
    4. J. M. Weber,
    5. D. P. Olson,
    6. M. C. Knight,
    7. R. P. Martin,
    8. E. Schipani,
    9. P. Divieti,
    10. F. R. Bringhurst,
    11. et al
    . 2003. Osteoblastic cells regulate the haematopoietic stem cell niche. Nature 425: 841–846.
    OpenUrlCrossRefPubMed
    1. Taichman R. S.,
    2. S. G. Emerson
    . 1994. Human osteoblasts support hematopoiesis through the production of granulocyte colony-stimulating factor. J. Exp. Med. 179: 1677–1682.
    OpenUrlAbstract/FREE Full Text
  26. ↵
    1. Zhang J.,
    2. C. Niu,
    3. L. Ye,
    4. H. Huang,
    5. X. He,
    6. W. G. Tong,
    7. J. Ross,
    8. J. Haug,
    9. T. Johnson,
    10. J. Q. Feng,
    11. et al
    . 2003. Identification of the haematopoietic stem cell niche and control of the niche size. Nature 425: 836–841.
    OpenUrlCrossRefPubMed
  27. ↵
    1. Lo Celso C.,
    2. H. E. Fleming,
    3. J. W. Wu,
    4. C. X. Zhao,
    5. S. Miake-Lye,
    6. J. Fujisaki,
    7. D. Côté,
    8. D. W. Rowe,
    9. C. P. Lin,
    10. D. T. Scadden
    . 2009. Live-animal tracking of individual haematopoietic stem/progenitor cells in their niche. Nature 457: 92–96.
    OpenUrlCrossRefPubMed
  28. ↵
    1. Sipkins D. A.,
    2. X. Wei,
    3. J. W. Wu,
    4. J. M. Runnels,
    5. D. Côté,
    6. T. K. Means,
    7. A. D. Luster,
    8. D. T. Scadden,
    9. C. P. Lin
    . 2005. In vivo imaging of specialized bone marrow endothelial microdomains for tumour engraftment. Nature 435: 969–973.
    OpenUrlCrossRefPubMed
  29. ↵
    1. Ding L.,
    2. S. J. Morrison
    . 2013. Haematopoietic stem cells and early lymphoid progenitors occupy distinct bone marrow niches. Nature 495: 231–235.
    OpenUrlCrossRefPubMed
  30. ↵
    1. Ding L.,
    2. T. L. Saunders,
    3. G. Enikolopov,
    4. S. J. Morrison
    . 2012. Endothelial and perivascular cells maintain haematopoietic stem cells. Nature 481: 457–462.
    OpenUrlCrossRefPubMed
  31. ↵
    1. Kunisaki Y.,
    2. I. Bruns,
    3. C. Scheiermann,
    4. J. Ahmed,
    5. S. Pinho,
    6. D. Zhang,
    7. T. Mizoguchi,
    8. Q. Wei,
    9. D. Lucas,
    10. K. Ito,
    11. et al
    . 2013. Arteriolar niches maintain haematopoietic stem cell quiescence. Nature 502: 637–643.
    OpenUrlCrossRefPubMed
  32. ↵
    1. Butler J. M.,
    2. D. J. Nolan,
    3. E. L. Vertes,
    4. B. Varnum-Finney,
    5. H. Kobayashi,
    6. A. T. Hooper,
    7. M. Seandel,
    8. K. Shido,
    9. I. A. White,
    10. M. Kobayashi,
    11. et al
    . 2010. Endothelial cells are essential for the self-renewal and repopulation of Notch-dependent hematopoietic stem cells. Cell Stem Cell 6: 251–264.
    OpenUrlCrossRefPubMed
  33. ↵
    1. Poulos M. G.,
    2. P. Guo,
    3. N. M. Kofler,
    4. S. Pinho,
    5. M. C. Gutkin,
    6. A. Tikhonova,
    7. I. Aifantis,
    8. P. S. Frenette,
    9. J. Kitajewski,
    10. S. Rafii,
    11. J. M. Butler
    . 2013. Endothelial Jagged-1 is necessary for homeostatic and regenerative hematopoiesis. Cell Rep. 4: 1022–1034.
    OpenUrlCrossRefPubMed
  34. ↵
    1. Varol C.,
    2. I. Zvibel,
    3. L. Spektor,
    4. F. D. Mantelmacher,
    5. M. Vugman,
    6. T. Thurm,
    7. M. Khatib,
    8. E. Elmaliah,
    9. Z. Halpern,
    10. S. Fishman
    . 2014. Long-acting glucose-dependent insulinotropic polypeptide ameliorates obesity-induced adipose tissue inflammation. J. Immunol. 193: 4002–4009.
    OpenUrlAbstract/FREE Full Text
  35. ↵
    1. Miyawaki K.,
    2. Y. Yamada,
    3. H. Yano,
    4. H. Niwa,
    5. N. Ban,
    6. Y. Ihara,
    7. A. Kubota,
    8. S. Fujimoto,
    9. M. Kajikawa,
    10. A. Kuroe,
    11. et al
    . 1999. Glucose intolerance caused by a defect in the entero-insular axis: a study in gastric inhibitory polypeptide receptor knockout mice. Proc. Natl. Acad. Sci. USA 96: 14843–14847.
    OpenUrlAbstract/FREE Full Text
  36. ↵
    1. Winkler I. G.,
    2. V. Barbier,
    3. B. Nowlan,
    4. R. N. Jacobsen,
    5. C. E. Forristal,
    6. J. T. Patton,
    7. J. L. Magnani,
    8. J. P. Lévesque
    . 2012. Vascular niche E-selectin regulates hematopoietic stem cell dormancy, self renewal and chemoresistance. Nat. Med. 18: 1651–1657.
    OpenUrlCrossRefPubMed
  37. ↵
    1. Bollag R. J.,
    2. Q. Zhong,
    3. K. H. Ding,
    4. P. Phillips,
    5. L. Zhong,
    6. F. Qin,
    7. J. Cranford,
    8. A. L. Mulloy,
    9. R. Cameron,
    10. C. M. Isales
    . 2001. Glucose-dependent insulinotropic peptide is an integrative hormone with osteotropic effects. Mol. Cell. Endocrinol. 177: 35–41.
    OpenUrlCrossRefPubMed
  38. ↵
    1. Miyawaki K.,
    2. Y. Yamada,
    3. N. Ban,
    4. Y. Ihara,
    5. K. Tsukiyama,
    6. H. Zhou,
    7. S. Fujimoto,
    8. A. Oku,
    9. K. Tsuda,
    10. S. Toyokuni,
    11. et al
    . 2002. Inhibition of gastric inhibitory polypeptide signaling prevents obesity. Nat. Med. 8: 738–742.
    OpenUrlCrossRefPubMed
  39. ↵
    1. Mancini S. J.,
    2. N. Mantei,
    3. A. Dumortier,
    4. U. Suter,
    5. H. R. MacDonald,
    6. F. Radtke
    . 2005. Jagged1-dependent Notch signaling is dispensable for hematopoietic stem cell self-renewal and differentiation. Blood 105: 2340–2342.
    OpenUrlAbstract/FREE Full Text
  40. ↵
    1. Stier S.,
    2. Y. Ko,
    3. R. Forkert,
    4. C. Lutz,
    5. T. Neuhaus,
    6. E. Grünewald,
    7. T. Cheng,
    8. D. Dombkowski,
    9. L. M. Calvi,
    10. S. R. Rittling,
    11. D. T. Scadden
    . 2005. Osteopontin is a hematopoietic stem cell niche component that negatively regulates stem cell pool size. J. Exp. Med. 201: 1781–1791.
    OpenUrlAbstract/FREE Full Text
  41. ↵
    1. Dishowitz M. I.,
    2. F. Zhu,
    3. H. G. Sundararaghavan,
    4. J. L. Ifkovits,
    5. J. A. Burdick,
    6. K. D. Hankenson
    . 2014. Jagged1 immobilization to an osteoconductive polymer activates the Notch signaling pathway and induces osteogenesis. J. Biomed. Mater. Res. A 102: 1558–1567.
    OpenUrl
  42. ↵
    1. Hill C. R.,
    2. M. Yuasa,
    3. J. Schoenecker,
    4. S. L. Goudy
    . 2014. Jagged1 is essential for osteoblast development during maxillary ossification. Bone 62: 10–21.
    OpenUrl
  43. ↵
    1. Peled A.,
    2. I. Petit,
    3. O. Kollet,
    4. M. Magid,
    5. T. Ponomaryov,
    6. T. Byk,
    7. A. Nagler,
    8. H. Ben-Hur,
    9. A. Many,
    10. L. Shultz,
    11. et al
    . 1999. Dependence of human stem cell engraftment and repopulation of NOD/SCID mice on CXCR4. Science 283: 845–848.
    OpenUrlAbstract/FREE Full Text
  44. ↵
    1. Nie Y.,
    2. Y. C. Han,
    3. Y. R. Zou
    . 2008. CXCR4 is required for the quiescence of primitive hematopoietic cells. J. Exp. Med. 205: 777–783.
    OpenUrlAbstract/FREE Full Text
  45. ↵
    1. Tzeng Y. S.,
    2. H. Li,
    3. Y. L. Kang,
    4. W. C. Chen,
    5. W. C. Cheng,
    6. D. M. Lai
    . 2011. Loss of Cxcl12/Sdf-1 in adult mice decreases the quiescent state of hematopoietic stem/progenitor cells and alters the pattern of hematopoietic regeneration after myelosuppression. Blood 117: 429–439.
    OpenUrlAbstract/FREE Full Text
  46. ↵
    1. Sugiyama T.,
    2. H. Kohara,
    3. M. Noda,
    4. T. Nagasawa
    . 2006. Maintenance of the hematopoietic stem cell pool by CXCL12-CXCR4 chemokine signaling in bone marrow stromal cell niches. Immunity 25: 977–988.
    OpenUrlCrossRefPubMed
  47. ↵
    1. Greenbaum A.,
    2. Y. M. Hsu,
    3. R. B. Day,
    4. L. G. Schuettpelz,
    5. M. J. Christopher,
    6. J. N. Borgerding,
    7. T. Nagasawa,
    8. D. C. Link
    . 2013. CXCL12 in early mesenchymal progenitors is required for haematopoietic stem-cell maintenance. Nature 495: 227–230.
    OpenUrlCrossRefPubMed
  48. ↵
    1. Finan B.,
    2. T. D. Müller,
    3. C. Clemmensen,
    4. D. Perez-Tilve,
    5. R. D. DiMarchi,
    6. M. H. Tschöp
    . 2016. Reappraisal of GIP pharmacology for metabolic diseases. Trends Mol. Med. 22: 359–376.
    OpenUrl
  49. ↵
    1. Yabe D.,
    2. Y. Seino
    . 2013. Incretin actions beyond the pancreas: lessons from knockout mice. Curr. Opin. Pharmacol. 13: 946–953.
    OpenUrlCrossRefPubMed
  50. ↵
    1. Gupta S.,
    2. J. Koirala,
    3. R. Khardori,
    4. N. Khardori
    . 2007. Infections in diabetes mellitus and hyperglycemia. Infect. Dis. Clin. North Am. 21: 617–638, vii.
    OpenUrlCrossRefPubMed
  51. ↵
    1. Muller L. M.,
    2. K. J. Gorter,
    3. E. Hak,
    4. W. L. Goudzwaard,
    5. F. G. Schellevis,
    6. A. I. Hoepelman,
    7. G. E. Rutten
    . 2005. Increased risk of common infections in patients with type 1 and type 2 diabetes mellitus. Clin. Infect. Dis. 41: 281–288.
    OpenUrlAbstract/FREE Full Text
  52. ↵
    1. Drucker D. J.
    2013. Incretin action in the pancreas: potential promise, possible perils, and pathological pitfalls. Diabetes 62: 3316–3323.
    OpenUrlAbstract/FREE Full Text
View Abstract
Back to top

In this issue

Print
Download PDF
Article Alerts
Email Article
Citation Tools

Jump to section