Abstract
Glucose-dependent insulinotropic polypeptide (GIP) has been recognized in the last decade as an important contributor of bone remodelling and is necessary for optimal bone quality. However, GIP receptors are expressed in several tissues in the body and little is known about the direct vs indirect effects of GIP on bone remodelling and quality. The aims of the present study were to validate two new GIP analogues, called [d-Ala2]-GIP-Tag and [d-Ala2]-GIP1–30, which specifically target either bone or whole-body GIP receptors, respectively; and to ascertain the beneficial effects of GIP therapy on bone in a mouse model of ovariectomy-induced bone loss. Both GIP analogues exhibited similar binding capacities at the GIP receptor and intracellular responses as full-length GIP1–42. Furthermore, only [d-Ala2]-GIP-Tag, but not [d-Ala2]-GIP1–30, was undoubtedly found exclusively in the bone matrix and released at acidic pH. In ovariectomized animals, [d-Ala2]-GIP1–30 but not [d-Ala2]-GIP-Tag ameliorated bone stiffness at the same magnitude than alendronate treatment. Only [d-Ala2]-GIP1–30 treatment led to significant ameliorations in cortical microarchitecture. Although alendronate treatment increased the hardness of the bone matrix and the type B carbonate substitution in the hydroxyapatite crystals, none of the GIP analogues modified bone matrix composition. Interestingly, in ovariectomy-induced bone loss, [d-Ala2]-GIP-Tag failed to alter bone strength, microarchitecture and bone matrix composition. Overall, this study shows that the use of a GIP analogue that target whole-body GIP receptors might be useful to improve bone strength in ovariectomized animals.
Introduction
Some evidences have emerged recently that the gut, and more specifically entero-endocrine cells, may play a role in maintaining optimal bone quality and bone mass (Henriksen et al. 2003, Xie et al. 2005, Tsukiyama et al. 2006, Walsh & Henriksen 2010, Gaudin-Audrain et al. 2013, Mabilleau et al. 2013, Mieczkowska et al. 2013, 2015b , Nissen et al. 2014, Torekov et al. 2014). Among the plethora of peptides secreted by the gastrointestinal tract, the glucose-dependent insulinotropic polypeptide (GIP), synthesized and secreted by entero-endocrine K cells, has emerged as a potential candidate. Indeed, whole-body GIP receptor (GIPr) deficiency led to alterations of trabecular and cortical bone microarchitectures, tissue mineral density and collagen maturity (Gaudin-Audrain et al. 2013, Mieczkowska et al. 2013). Furthermore, administration of stable GIP analogues improved bone matrix composition and biomechanics at the tissue level in healthy and diabetic rodent models (Mabilleau et al. 2014, Mansur et al. 2015, 2016).
In rodents, the GIPr is widely expressed in the body, and expression has been documented in the endocrine pancreas, gastrointestinal tract, adipose tissue, adrenal cortex, pituitary gland, vascular endothelium and several regions in the central nervous system (Baggio & Drucker 2007). Expression in bone has also been reported and the GIPr seems to be expressed in rodent and human osteoblasts, osteocytes and osteoclasts (Bollag et al. 2000, Mieczkowska et al. 2015a , Mabilleau et al. 2016). However, due to this wide variety of tissue expression, it is not clear whether the marked bone effects observed in previous rodent studies arise from inactivation/activation of bone-specific GIPr or extraskeletal GIPr.
The rapid degradation of GIP in plasma by dipeptidyl peptidase-4 (DPP-4) precludes to its use as a therapeutic approach. As such, a series of GIP modifications have previously been conducted and led to several GIP analogues with proven efficacy (Irwin & Flatt 2009). From these manipulations, it appears that the N-terminal extremity of GIP, and particularly, the first two amino acids, was particularly important in allowing receptor activation. Furthermore, only the first 30 amino acids are required to induce biological activity (Hinke et al. 2001). As such, we produced two new GIP analogues, namely [d-Ala2]-GIP1–30 and [d-Ala2]-GIP-Tag that possess a d-alanine in position two to confer DPP-4 resistance. Furthermore, [d-Ala2]-GIP-Tag possesses a tag of nine negatively charged amino acids at its C-terminal extremity that, according to previous published studies, should give a bone-specific affinity (Kasugai et al. 2000, Yokogawa et al. 2001). The current gold standard medication for treating post-menopausal osteoporosis is represented by bisphosphonates and as such, we thought to also ascertain how the two new molecules above compared with alendronate.
The main goals of this study were to (1) verify that the tag confers a bone-specific targeting, (2) ascertain the biological efficacy of these two new GIP analogues, [d-Ala2]-GIP1–30 and [d-Ala2]-GIP-Tag and (3) investigate their therapeutic potentials in ovariectomy-induced bone fragility as compared to alendronate.
Materials and methods
Reagents
All GIP analogues were purchased from GeneCust Europe with a purity >95% (Dudelange, Luxembourg). Purity has been verified by high-performance liquid chromatography and peptide composition validated by mass spectroscopy. Sequences are provided in Table 1. Macrophage-colony stimulating factor (M-CSF) and receptor activator of nuclear factor kB ligand (RANKL) were purchased from R&D Systems Europe. Fluo-4AM was purchased from Invitrogen. All other chemicals were obtained from Sigma-Aldrich unless otherwise stated.
Peptide sequences and characteristics.
| GIP analogues | Amino acid sequence | Purity (%) | Theoretical molecular weight (Da) | Measured molecular weight (Da) |
|---|---|---|---|---|
| GIP1–42 | Y[d-Ala]EGTFISDYSIAMDKIHQQDFVNWLLAQKGKKNDWKHNITQ | 96.5 | 4983.53 | 4983.64 |
| [d-Ala2]-GIP1–30 | Y[d-Ala]EGTFISDYSIAMDKIHQQDFVNWLLAQK | 97.9 | 3531.95 | 3532.02 |
| 5-FAM-[d-Ala2]-GIP1–30 | 5′Fam-Y[d-Ala]EGTFISDYSIAMDKIHQQDFVNWLLAQK | 96.9 | 3890.25 | 3890.34 |
| [d-Ala2]GIP-Tag | Y[d-Ala]EGTFISDYSIAMDKIHQQDFVNWLLAQKGAADDDDDD | 95.8 | 4421.68 | 4421.76 |
| 5-FAM-[d-Ala2]GIP-Tag | 5′Fam-Y[d-Ala]EGTFISDYSIAMDKIHQQDFVNWLLAQKGAADDDDDD | 95.6 | 4779.98 | 4780.08 |
In vitro mineral-binding assay
Carboxymethylated poly(2-hydroxyethylmethacrylate) (pHEMA) discs and their mineralization were performed as previously described (Filmon et al. 2002). Mineralized discs were incubated for 16 h with 5 nmol of 5-carboxyfluorescein (5-FAM), 5-FAM-[d-Ala2]-GIP1–30, 5-FAM-[d-Ala2]-GIP-Tag or calcein green. PHEMA discs were rinsed extensively with distilled water prior to observation with a Leica TCS SP8 confocal laser scanning microscope (Leica). Excitation was performed at 488 nm with an argon laser and emission was recorded in the range 510–550 nm. After observation, the mineral was dissolved with 0.2 M HCl overnight and fluorescence readings were performed with a M2 microplate reader (Molecular Devices, St Gregoire, France) set up at 480 nm for excitation and 530 nm for emission. Calcium concentrations were estimated as published previously (Degeratu et al. 2013).
Cell culture and activity of GIP analogues
MC3T3-E1 cells were purchased from American type culture collection (ATCC). Cells were grown and expanded in propagation medium containing alpha minimum essential medium (αMEM) supplemented with 5% foetal bovine serum (FBS), 5% bovine calf serum, 100 U/mL penicillin and 100 μg/mL streptomycin in a humidified atmosphere enriched with 5% CO2 at 37°C.
Competitive whole cell binding studies were performed in cold αMEM supplemented with 0.1% bovine serum albumin, protease inhibitors (Halt protease inhibitor cocktail, ThermoFisher Scientific, Villebon sur Yvette, France), 8 × 10−9 M FAM-GIP1–42 and appropriate peptide concentrations. Equilibrium binding was achieved overnight at 4°C. Cells were then washed twice with cold assay buffer, solubilized in 0.1 M NaOH and transferred to opaque microplate for fluorescence readings.
cAMP stimulation experiment was performed in response to 100 pM GIP analogues in MC3T3-E1 cells with a fluorometric commercially available kit (reference KGE002B, R&D Systems Europe) (Mieczkowska et al. 2015a ). Assessment of the cell phospho-proteome was assessed with the Proteome profiler anti-phosphokinase assay (reference ARY003b, R&D Systems Europe).
MC3T3-E1 cells were seeded in 96-well plate with clear bottom and opaque edges (ibidi GmbH, Martinsried, Germany). Cells were incubated with 4 µM Fluo-4-AM for 45 min at 37°C in the dark and washed with pre-warmed HEPES buffered saline. The plate was placed in a M2 microplate reader (Molecular Devices) and signals were acquired at 37°C with an excitation wavelength of 490 nm and an emission wavelength set at 515 nm for 5 min. Cells were then stimulated with 100 pM GIP analogues for 15 min and signals were again acquired with the microplate reader. Autofluorescence was measured in unloaded cells, and this value was subtracted from all measurements.
Collagen maturity assay was performed as described in detail elsewhere (Mieczkowska et al. 2015a ).
In order to generate mature human osteoclasts, peripheral mononuclear blood cells were isolated from buffy coat (Etablissement français du sang, Angers, France) and cultured in the presence of 25 ng/mL M-CSF and 30 ng/mL soluble human RANKL as described previously (Mabilleau et al. 2011).
Animals
BALB/c (BALB/cJRj) mice were obtained from Janvier Labs (Saint-Berthevin, France). All animal experiments were approved by Ethical committee in animal use of the Pays de la Loire under the animal licence CEEA-PdL06-01740.01. Mice were housed four animals per cage in the institutional animal lab (Agreement E49007002) at 24°C ± 2°C with a 12-h light/darkness cycle and were provided with tap water and normal diet (Diet A04, Safe, Augy, France) ad libitum until killing by cervical dislocation. All procedures were conducted according to the French Animal Scientific Procedures Act 2013-118.
In vivo localization of fluorescently labelled GIP analogues
Intraperitoneal injections of saline or fluorescent GIP analogues (50 nmol/kg body weight) were performed at 4 weeks of age in 15 female BALB/c mice (n = 5/group). This dose of fluorescent GIP analogues was chosen to ensure detection in the investigated tissues. Twenty-four hours after injection, visceral adipose tissue, adrenal gland, bladder, left femur and tibia, brain, heart, small intestine, kidney, liver, lung, pancreas, skeletal muscle, spleen and stomach were collected, immediately snap-frozen in liquid nitrogen and stored at −80°C until use. Then, frozen tissues were powdered, suspended in Tris 0.1 M pH 7.4 and fluorescence readings with a microplate reader as detailed above were performed. Fluorescence readings were normalized by the concentration of proteins measured with the bicinchoninic acid assay (Pierce Biotechnology). Right femurs of 4-week-old mice were collected at necropsy, fixed in buffered formalin and embedded in polymethylmethacrylate (pMMA) at low temperature (Chappard 2009). Thick cross-sections at the mid-diaphysis of all femurs were cut with a low speed precision saw (Minitom, Struers, Champigny sur Marne, France). Femur sections were grinded up to a thickness of 50 µm and subsequently imaged with the confocal microscope as explained earlier.
Additionally, right tibias of 5-FAM-[d-Ala2]-GIP-Tag-injected mice were collected at necropsy, fixed in buffered formalin and embedded in polymethylmethacrylate (pMMA) at low temperature. Thick cross-sections (500 µm-thick) at the mid-diaphysis were cut with a low speed precision saw and incubated in saline or 0.1 M acetic acid (pH 4.5) for 24 h. The resulting solution was buffered with 1 M Tris and fluorescence readings were performed with the M2 microplate reader as explained earlier.
Long-term effects of GIP analogues in ovariectomy-induced bone loss
Bilateral ovariectomy (OVX) was performed in 32 BALB/c mice at 12 weeks of age under general anaesthesia supplemented with a β2 adrenergic receptor agonist. At 16 weeks of age, mice were randomly allocated into four groups: vehicle daily (OVX + Veh, n = 8), 25 nmol/kg/day intraperitoneally (i.p.) [d-Ala2]-GIP1–30 (OVX + GIP1–30, n = 8), 25 nmol/kg/day ip [d-Ala2]-GIP-Tag (OVX + GIP-Tag, n = 8) and 10 µg/kg alendronate twice a week i.p. (OVX + Aln, n = 8). These doses and regimens of GIP analogues and alendronate were based on previous published studies where these molecules were proven active with beneficial effects on bone or equivalent to approved clinical dose (Mabilleau et al. 2014, Shao et al. 2017). Eight sham-operated female BALB/c mice with the same age and injected daily with saline were used as controls (Sham + Veh). All mice from the second study were also administered with calcein (10 mg/kg; i.p.) 10 and 2 days before being culled at 24 weeks of age. At necropsy, blood was collected by intracardiac aspiration (~250 µL). Non-fasting glucose level were evaluated with an Accu-Chek mobile glucometer (Roche Diabetes Care GmbH). Then, blood was spun at 15,600 g for 15 min at 4°C and serum was aliquoted, snap-frozen in liquid nitrogen and stored at −80°C until use. After necropsy, tibias, femurs and uterus were collected and cleaned of soft tissues. Femur length was measured with a digital caliper (Mitutoyo, Roissy en France, France).
ELISA
Serum levels of C-terminal telopeptide of collagen type I (CTx-I) and N-terminal propeptide of type I collagen (P1NP) were measured with the RatLaps and Rat/mouse P1NP ELISA kits, respectively (Immunodiagnostic Systems Ltd, Boldon, UK), according to the manufacturer recommendations.
Microcomputed tomography
X-ray microcomputed tomography (MicroCT) analyses of the abdomen were performed to measure abdominal fat volume, that represents a good indicator of whole-body fat mass (Judex et al. 2010). Anaesthetised animals were placed in a Skyscan 1076 microtomograph (Bruker MicroCT, Kontich, Belgium) and the region localized between L1 and the hip was selected for fat depot evaluation. Acquisitions were performed at 40 kV, 250 µA, 100 ms integration time. The isotropic pixel size was fixed at 35 µm, the rotation step at 0.6° and exposure was done with a 0.5 mm aluminium filter. Tibias were scanned with a Skyscan 1172 microtomograph (Bruker MicroCT) operated at 70 kV, 100 µA, 340 ms integration time. The isotropic pixel size was fixed at 4 µm, the rotation step at 0.25° and exposure was done with a 0.5 mm aluminium filter. Each 3D reconstruction image dataset was binarized using global thresholding. Cortical volume of interest extended on 1 mm centred at the midshaft tibia. All histomorphometrical parameters were determined according to guidelines and nomenclature proposed by the American Society for Bone and Mineral Research (Bouxsein et al. 2010).
Marrow adipose tissue assessment
After microCT scans, tibias were embedded undecalcified in pMMA at 4°C. Longitudinal sections were cut and stained with toluidine blue. The extent of marrow adipose tissue (Ad.Ar/Ma.Ar) was computerized with a routine in ImageJ (release 1.51s, National Institutes of Health, Bethesda, MA, USA). The nomenclature proposed by the American Society for Bone and Mineral Research was used in this study (Dempster et al. 2013).
Bone strength assessment
At necropsy, femurs were cleaned of soft tissue and immediately frozen in a saline-soaked gauze at −20°C. Three-point bending experiments were performed on femurs after thawing bones at 4°C overnight. Measurements were done with an Instron 5942 (Instron, Elancourt, France) as reported previously (Mieczkowska et al. 2015b ). The load-displacement curve was acquired with the Bluehill 3 software (Instron). Ultimate load, ultimate displacement, stiffness and total absorbed energy were computerized (Turner & Burr 1993).
After three-point bending experiments, femurs were embedded undecalcified in pMMA at 4°C and cross-sections were made at the midshaft using a diamond saw (Accutom, Struers, Champigny sur Marne, France). Blocks were polished to a 1 µm finish with diamond particles (Struers, France) and subjected to rehydration in saline 24 h prior to nanoindentation testing. Twelve indentations, at distance from canals, osteocyte lacunae and/or microcracks were randomly positioned in cortical bone with a NHT-TTX system (Anton Paar, Les Ulis, France) as previously detailed (Aguado et al. 2017). At maximum load, a holding period of 15 s was applied to avoid creeping of the bone material. The following material properties at the tissue level: maximum load (Force max), indentation modulus (EIT), indentation hardness (HIT) and dissipated energy (Wplast) were determined according to Oliver and Pharr (1992).
Fourier-transform infrared microscopy
Four micrometres cross-sectional sections of the midshaft femur were sandwiched between BaF2 optical windows and Fourier-transform infrared microscopy (FTIRM) assessment was performed at bone formation site by recording infrared spectra only between double calcein labelling. A Bruker Vertex 70 spectrometer (Bruker optics, Ettlingen, Germany) interfaced with a Bruker Hyperion 3000 infrared microscope were used as previously reported (Pereira et al. 2017). Each spectrum was corrected for Mie scattering with the RMieS-EMSC_v5 algorithm (kind gift of Prof Peter Gardner, University of Manchester, UK) prior to be subjected to pMMA subtraction. Second derivative spectroscopy was applied to find the position of underlying peaks and curve fitting was performed with a routine script in Matlab (The Mathworks, Natick, USA) as previously reported (Mansur et al. 2015). The evaluated infrared spectral parameters were (1) mineral-to-matrix ratio, calculated as the ratio of integrated areas of the υ1, υ3 phosphate band at 900–1200/cm to the amide I band at 1585–1725/cm (Boskey et al. 2005); (2) mineral maturity calculated as the area ratio of the subbands at 1020/cm and 1030/cm of the phosphate band (Gadaleta et al. 1996); (3) carbonate-to-phosphate ratio, calculated as the ratio of the υ2 carbonate band at 850–900/cm to the υ1, υ3 phosphate band (Paschalis et al. 1996); (4) carbonate substitution type by integrating the area of subbands located at 866/cm (labile), 871/cm (type B) and 878/cm (type A) over the υ2 carbonate band (Rey et al. 1989); (5) acid phosphate content, calculated as the area ratio of the 1127/cm and 1096/cm subbands (Spevak et al. 2013) and (6) collagen maturity, determined as the relative ratio of subbands located at 1660/cm (trivalent cross-links) and 1690/cm (divalent cross-links) of the amide I peak (Paschalis et al. 2001).
Bone mineral density distribution evaluation
Quantitative backscattered electron imaging (qBEI) experiments were performed on the same blocks and same regions as nanoindentation. A full description of qBEI preparation, calibration and analysis has already been extensively described elsewhere (Roschger et al. 1998, Mabilleau et al. 2013, Mieczkowska et al. 2015b ). Cortical bone area was imaged at a 200× nominal magnification, corresponding to a pixel size of 0.5 µm. Four images per samples were taken. Two variables were obtained from the bone mineral density distribution: Camean as the average calcium concentration and Cawidth as the width of the histogram at half maximum of the peak. Following this, the blocks were imaged at a 200× magnification with a confocal microscope (Leica SP8, Leica) equipped with an argon laser at 488 nm and a hybrid GaAs detector (Leica) to find bone surface with double labels. Confocal images were superimposed on qBEI images in order to delineate new bone matrix formed during the time-course of the study. Using ImageJ 1.51s, a straight line (4 pixel width) perpendicular to the mineralization front across the new bone structural unit with a step size of 0.5 µm was drawn on qBEI image. The calcium content was plotted vs distance of mineralization front. These plots show a biphasic aspect with fast mineralization process close to the mineralization front followed by a slow mineralization process. The two mineralization processes were then analysed by linear curve fitting with a lab-made routine in Excel 2010 (Microsoft, Issy-les-Moulineaux, France). Caturn was determined as the calcium concentration where the fast mineralization process was changing to the slow mineralization process as described by Roschger et al. (2008).
Statistical analysis
All data were analysed using Prism 6.0 (GraphPad Software Inc.). Mineral binding was analysed by a one-way ANOVA followed by post hoc Dunnett’s multiple comparisons tests. Tissue distribution of both fluorescent analogues was analysed by a two-way ANOVA with Sidak’s multiple comparisons tests. GIPr-binding assay was analysed by non-linear regression analysis. Intracellular signalling (cAMP, intracellular calcium and phospho-proteins) as well as in vitro collagen maturity and extent of osteoclast formation and resorption in vitro were analysed with the non-parametrical Kruskal–Wallis test.
Due to the adaptive nature of bone, bone strength, bone microarchitecture and bone compositional parameters have been adjusted for body size (body mass × femur length) using a linear regression method as reported in detail elsewhere (Jepsen et al. 2015). One-way ANOVA followed by post hoc Dunnett’s multiple comparisons tests were employed to analyse the differences between OVX + Veh and all the other groups of mice in any of the body size-adjusted parameters. Differences at P equal to or less than 0.05 were considered significant.
Results
[d-Ala2]-GIP-Tag but not [d-Ala2]-GIP1–30 is capable of binding to hydroxyapatite and targeting bone tissue
Microscopic examinations of calcospherites grown at the surface of carboxymethylated pHEMA revealed that 5-FAM-[d-Ala2]-GIP-Tag and calcein green, but neither 5-FAM-[d-Ala2]-GIP1–30 nor 5-FAM alone, were significantly bound to hydroxyapatite (Fig. 1A). Tissue distribution of the two fluorescently labelled analogues highlighted differences between the two molecules. Indeed, 5-FAM-[d-Ala2]-GIP1–30 was mainly observed in adipose tissue, adrenal gland, bone, brain, intestine, liver and pancreas, whilst 5-FAM-[d-Ala2]-GIP-Tag was exclusively found in bone (Fig. 1B). Microscopic examinations of femur midshaft cross-sections in 5-FAM-[d-Ala2]-GIP-Tag-injected mice revealed the presence of fluorescent bands, suggesting the incorporation of this analogue in the bone mineral (Fig. 1C). On the other hand, such bone distribution was not observed in 5-FAM-[d-Ala2]-GIP1–30-injected animals (Fig. 1C). Furthermore, incubation of thick tibia slices in acidic conditions (pH 4.5), but not in neutral solution, was capable of releasing 5-FAM-[d-Ala2]-GIP-Tag (Fig. 1D).
Mineral-binding capacity of GIP analogues. (A) 5-FAM-[d-Ala2]-GIP-Tag and FAM-[d-Ala2]-GIP1–30 were incubated for 24 h with discs of pHEMA that had been previously mineralized. 5-FAM and Calcein were used as negative and positive controls, respectively. 5-FAM-[d-Ala2]-GIP-Tag but not 5-FAM-[d-Ala2]-GIP1–30 was capable of significantly binding to the mineralized discs. Values are means ± s.e.m. #P < 0.05 vs 5-FAM. (B) Tissue distribution of 5-FAM, 5-FAM-[d-Ala2]-GIP1–30 and 5-FAM-[d-Ala2]-GIP-Tag. Fluorescence, in arbitrary units (a.u.) was weighted by the protein mass and detected in several tissues harvested from animals injected with 5-FAM-[d-Ala2]-GIP1–30. On the other hand, fluorescence due to 5-FAM-[d-Ala2]-GIP-Tag was exclusively found in bone. *P < 0.05 vs 5-FAM, #P < 0.05 vs 5-FAM-[d-Ala2]-GIP1–30. Values are means ± s.e.m. (C) 5-FAM-[d-Ala2]-GIP1–30 and 5-FAM-[d-Ala2]-GIP-Tag were injected into young mice and the extent of GIP analogue binding in bone was assessed after 16 h. A significant fluorescent line was clearly visible in the bone matrix of animals injected with 5-FAM-[d-Ala2]-GIP-Tag but not in animals injected with 5-FAM-[d-Ala2]-GIP1–30. Arrowheads indicate the fluorescence line. CtB, cortical bone; BM, bone marrow. (D) [d-Ala2]-GIP-Tag at acidic but not neutral pH was released from the bone slice as demonstrated by significant higher level of fluorescence. *P < 0.05 vs pH 7.0. Values are means ± s.e.m. A full colour version of this figure is available at https://doi.org/10.1530/JOE-18-0214.
Citation: Journal of Endocrinology 239, 2; 10.1530/JOE-18-0214
Cellular and molecular activities of [d-Ala2]-GIP-Tag are not affected by the C-terminal modification
Next, we thought to investigate the biological activity of both GIP analogues. As represented in Fig. 2A, [d-Ala2]-GIP1–30 and [d-Ala2]-GIP-Tag did not show any differences in their capacity to bind to the GIPr with IC50 of 65.5 ± 2.5 pM and 72.9 ± 2.7 pM, respectively. More importantly, their binding activity was similar to GIP1–42, with IC50 of 65.3 ± 1.7 pM. Both GIP analogues were capable of inducing cAMP production and rise in intracellular calcium to the same level as observed with GIP1–42 (Fig. 2A). Phospho-proteome analysis showed that osteoblasts stimulated with GIP1–42 also activated p38α, CREB, AMPKα2 and STAT2 in addition to cAMP (Fig. 2B). [d-Ala2]-GIP1–30 and [d-Ala2]-GIP-Tag showed similar actions on all these intracellular pathways (Fig. 2B). Finally, we tested whether [d-Ala2]-GIP1–30 and [d-Ala2]-GIP-Tag were capable of improving collagen maturity as observed with GIP1–42 and indeed, this parameter was significantly augmented by 32 and 37%, with [d-Ala2]-GIP1–30 and [d-Ala2]-GIP-Tag, respectively as compared with untreated cells (Fig. 2C). As suspected, both GIP analogues were also capable to reduce osteoclast formation and osteoclast-mediated bone resorption in vitro in a similar extent to GIP1–42 (Fig. 2D).
Biological activity of GIP analogues. (A) Receptor binding properties and activation of cAMP and intracellular calcium. Receptor binding properties, cAMP and intracellular calcium responses of [d-Ala2]-GIP1–30 and [d-Ala2]-GIP-Tag were not significantly different to those of native GIP1–42. Values are means ± s.e.m. (B) Activation of intracellular pathways in MC3T3-E1 cells. [d-Ala2]-GIP1–30, [d-Ala2]-GIP-Tag and GIP1–42 significantly increased the phosphorylation of p38α, CREB, AMPKα2 and STAT2 in a similar manner. *P < 0.05 vs vehicle. Values are means ± s.e.m. (C) [d-Ala2]-GIP1–30, [d-Ala2]-GIP-Tag and GIP1–42 significantly increased collagen maturity in vitro and (D) reduced the number of newly generated osteoclast per well and the extent of osteoclast resorption. Values are means ± s.e.m. *P < 0.05 vs vehicle and #P < 0.05 vs M-CSF + RANKL. A full colour version of this figure is available at https://doi.org/10.1530/JOE-18-0214.
Citation: Journal of Endocrinology 239, 2; 10.1530/JOE-18-0214
Effects of [d-Ala2]-GIP1–30 vs [d-Ala2]-GIP-Tag in OVX-induced bone loss
We next examined the biological effects of GIP analogues in the OVX mouse model. As compared with Sham + Veh animals and shown in Table 2, OVX + Veh mice presented with higher abdominal fat volume and CTx-I levels and lower uterus mass. Treatment with [d-Ala2]-GIP1–30 significantly reduced CTx-I levels whilst treatment of OVX animals with [d-Ala2]-GIP-Tag significantly reduced abdominal fat volume, marrow adipose tissue and CTx-I levels. Alendronate administration only significantly reduced CTx-I levels.
Body weight, composition and metabolic properties.
| Sham + Veh | OVX + Veh | OVX + GIP1–30 | OVX + GIP-Tag | OVX + Aln | |
|---|---|---|---|---|---|
| Body mass (g) | 23.5 ± 0.4 (0.062) | 25.6 ± 0.9 | 26.1 ± 0.5 (0.945) | 23.3 ± 0.6 (0.053) | 26.6 ± 0.6 (0.645) |
| Abdominal fat volume (%) | 14.6 ± 0.6 (<0.001) | 24.7 ± 2.4 | 23.1 ± 1.6 (0.695) | 10.8 ± 1.0 (<0.001) | 25.6 ± 1.8 (0.695) |
| Uterus mass (g) | 0.14 ± 0.01 (<0.001) | 0.04 ± 0.01 | 0.05 ± 0.01 (0.915) | 0.03 ± 0.01 (0.674) | 0.05 ± 0.01 (0.915) |
| Femur length (mm) | 13.9 ± 0.1 (0.967) | 14.0 ± 0.1 | 14.1 ± 0.1 (0.980) | 14.0 ± 0.1 (0.999) | 14.2 ± 0.1 (0.898) |
| Marrow adipose tissue (%) | 0.6 ± 0.2 (0.498) | 1.0 ± 0.4 | 0.4 ± 0.1 (0.123) | 0.2 ± 0.1 (0.043) | 1.2 ± 0.2 (0.960) |
| Non-fasting glucose (mmol/L) | 9.7 ± 0.4 (0.574) | 10.5 ± 0.6 | 10.6 ± 0.6 (>0.999) | 10.7 ± 0.3 (0.627) | 10.6 ± 0.3 (>0.999) |
| CTx-I (ng/mL) | 8.9 ± 0.5 (<0.001) | 14.7 ± 1.0 | 9.2 ± 0.7 (<0.001) | 9.4 ± 1.0 (0.02) | 10.8 ± 1.0 (<0.001) |
| P1NP (ng/mL) | 20.4 ± 1.7 (0.062) | 26.4 ± 1.3 | 23.6 ± 1.9 (0.680) | 21.3 ± 2.1 (0.120) | 22.5 ± 1.3 (0.340) |
Data are presented as mean ± s.e.m. (P value). Data have been analysed by one-way ANOVA followed by post hoc Dunnett’s multiple comparison test using OVX + Veh group as the control group.
CTx-I, C-terminal telopeptide of type I collagen; P1NP, N-terminal propeptide of type I procollagen.
After the 8-week experimental treatment period, structural mechanical properties were assessed by three-point bending (Fig. 3A, B, C, D, E and F). As expected, OVX + Veh mice presented with significant reductions in ultimate force (−18%, P = 0.0005), yield load (−27%, P < 0.0001) and stiffness (−34%, P < 0.0001). Treatments with alendronate or [d-Ala2]-GIP1–30, but not [d-Ala2]-GIP-Tag, significantly augmented by 33 (P < 0.0001) and 25% (P = 0.0013) stiffness, respectively. Bone strength was also investigated at the tissue level by nanoindentation (Fig. 3G, H, I and J). As compared with sham animals, OVX + Veh mice presented no significant alterations in any of the studied parameters. The use of alendronate significantly augmented HIT by 29% (P = 0.0191). Neither [d-Ala2]-GIP1–30 nor [d-Ala2]-GIP-Tag significantly modified strength at the tissue level.
Effects of GIP analogues on bone strength in ovariectomy-induced bone loss. (A, B, C, D, E and F) Bone strength has been assessed at the whole-body-level by three-point bending and (G, H, I and J) at the tissue level by nanoindentation. Values are means ± s.e.m. HIT, indentation hardness; EIT, indentation modulus; force max, maximum load to reach 900 nm in depth; Wplast, dissipated energy. *P < 0.05 vs OVX + Veh. A full colour version of this figure is available at https://doi.org/10.1530/JOE-18-0214.
Citation: Journal of Endocrinology 239, 2; 10.1530/JOE-18-0214
As compared with Sham + Veh animals, significant microarchitectural alterations of cortical bone were evidenced as expected in OVX + Veh animals and represented by lower total cross-sectional area (Tt.Ar, −10%, P = 0.0249), marrow area (Ma.Ar, −14%, P = 0.0046) and cortical area (Ct.Ar, −9%, P = 0.0248) (Table 3). On the other hand, cortical thickness (Ct.Th), moment of inertia about the anteroposterior (Iap) or mediolateral (Iml) axis and polar moment of inertia (J) were not significantly different between the two groups of animals. As compared with OVX + Veh animals, treatment with [d-Ala2]-GIP1–30 significantly increased Tt.Ar, Ma.Ar, Ct.Ar and J by 10 (P = 0.0417), 16 (P = 0.0041), 9 (P = 0.0430) and 18% (P = 0.0246), respectively. Treatment with [d-Ala2]-GIP-Tag did not result in significant modifications of cortical microarchitecture although a trend to similar effects as observed with [d-Ala2]-GIP1–30 was noted (Table 3). Treatment with alendronate resulted only in lower values for Iml (−21%, P = 0.0277).
Cortical bone microarchitectural parameters at the midshaft tibia.
| Sham + Veh | OVX + Veh | OVX + GIP1–30 | OVX + GIP-Tag | OVX + Aln | |
|---|---|---|---|---|---|
| Tt.Ar (mm2) | 1.72 ± 0.04 (0.025) | 1.54 ± 0.04 | 1.70 ± 0.03 (0.042) | 1.66 ± 0.03 (0.221) | 1.51 ± 0.08 (0.988) |
| Ma.Ar (mm2) | 0.71 ± 0.02 (0.005) | 0.61 ± 0.02 | 0.71 ± 0.01 (0.004) | 0.67 ± 0.01 (0.208) | 0.63 ± 0.03 (0.873) |
| Ct.Ar (mm2) | 1.00 ± 0.02 (0.038) | 0.90 ± 0.01 | 0.98 ± 0.02 (0.044) | 0.96 ± 0.03 (0.378) | 0.87 ± 0.05 (0.747) |
| Ct.Th (µm) | 245 ± 3 (0.411) | 236 ± 5 | 230 ± 5 (0.630) | 235 ± 4 (0.997) | 225 ± 6 (0.740) |
| Iap (mm4) | 0.23 ± 0.01 (0.814) | 0.22 ± 0.02 | 0.24 ± 0.01 (0.538) | 0.22 ± 0.01 (0.990) | 0.24 ± 0.03 (0.815) |
| Iml (mm4) | 0.26 ± 0.00 (0.648) | 0.24 ± 0.01 | 0.27 ± 0.01 (0.297) | 0.28 ± 0.01 (0.081) | 0.19 ± 0.02 (0.028) |
| J (mm4) | 0.49 ± 0.03 (0.347) | 0.44 ± 0.02 | 0.52 ± 0.01 (0.049) | 0.48 ± 0.02 (0.648) | 0.41 ± 0.04 (0.800) |
Data are presented as mean ± s.e.m. (P value). Data have been body size adjusted with a linear regression method and analysed by one-way ANOVA followed by post hoc Dunnett’s multiple comparison test using OVX + Veh group as the control group.
Ct.Ar, cortical bone area; Ct.Th: cortical thickness; Iap, moment of inertia about the anteroposterior axis; Iml, moment of inertia about the mediolateral axis; J, polar moment of inertia; Ma.Ar, medullary area; Tt.Ar, total cross-sectional area.
Alterations of bone matrix composition was also evidenced in OVX + Veh animals as compared with Sham + Veh (Fig. 4). Indeed, at site of bone formation, collagen maturity and mineral-to-matrix ratio were significantly lowered by 25 (P = 0.0261) and 35% (P = 0.0070), respectively in OVX + Veh animals. As compared with OVX + Veh animals, treatment with [d-Ala2]-GIP1–30 or [d-Ala2]-GIP-Tag significantly lowered the overall mean calcium distribution in the bone matrix (Camean) by 7 (P = 0.0002) and 4% (P = 0.0217), respectively. These two molecules also reduced the Caturn value by 6 (P = 0.005) and 7% (P = 0.002), respectively. At site of bone formation, none of these molecules modified the bone matrix composition. On the other hand, treatment with alendronate significantly reduced calcium distribution heterogeneity (Cawidth) by 11% (P = 0.0044) and augmented Caturn values by 7% (P < 0.001) in the bone matrix. At the site of bone formation, alendronate resulted in higher carbonate-to-phosphate ratio by 16% (P = 0.0204), mainly by reduction in loosely bound carbonate (−47%, P = 0.0053) and increase in type B carbonate substitution (31%, P = 0.0008).
Effects of GIP analogues on bone matrix composition. (A) Tissue mineral density distribution has been studied by qBEI at the midshaft tibia and revealed significant lower values of Camean and Caturn in the presence of [d-Ala2]-GIP1–30 or [d-Ala2]-GIP-Tag and a significant lower heterogeneity and higher Caturn in the presence of alendronate. Values are means ± s.e.m. *P < 0.05 vs OVX + Veh. (B) Bone matrix composition has been investigated at site of bone formation and revealed the lack of effects of both GIP analogues. Treatment with alendronate resulted in higher values for carbonate-to-phosphate ratio and type B carbonate substitution and a lower value of labile carbonate substitution. Values are means ± s.e.m. *P < 0.05 vs OVX + Veh. A full colour version of this figure is available at https://doi.org/10.1530/JOE-18-0214.
Citation: Journal of Endocrinology 239, 2; 10.1530/JOE-18-0214
Discussion
With respect to its important role in maintaining bone strength in animal models of receptor deletion, GIP has promises as a therapeutic agent in treating bone fragility. In the present study, we investigated bone-targeting capacities and biological activities as well as therapeutical potencies of two new GIP analogues in ovariectomy-induced bone loss. The bone-targeting capacity of [d-Ala2]-GIP-Tag, as opposed to [d-Ala2]-GIP1–30, was evident and emphasized the importance of acidic amino acids in promoting bone affinity. Acidic amino acid tag mimics the observed aspartic acid repetition in the noncollagenous bone protein osteopontin. In bone, after secretion, osteopontin rapidly binds to hydroxyapatite and sequence analysis of osteopontin identified the aspartic acid repetition as a putative mineral-binding site (Oldberg et al. 1986, Butler 1989). Similar to what is observed with bisphosphonate, a molecule bound to the bone mineral is thought to be released upon bone resorption. The first evidence suggesting such properties of the acidic amino acid tag was reported by Kasugai et al. (2000). Since their discovery, at least six distinct molecules have been developed so far with bone-targeting properties using an acidic amino acid tag (Yokogawa et al. 2001, Nishioka et al. 2006, Miller et al. 2008, Montano et al. 2008, Takahashi et al. 2008, Hsieh et al. 2014). In 2007, Murphy et al. reported the higher efficacy of acidic amino acid tags in comparison to the bisphosphonate structural group (Murphy et al. 2007). We deliberately chose to fuse this tag at the C-terminal end of [d-Ala2]-GIP1–30 because only the first 30 amino acids are important for GIP helicoidal secondary structure and hence its receptor binding and biological properties (Manhart et al. 2003, Alana et al. 2006). However, in this study, we also provided clear evidence that [d-Ala2]-GIP1–30 and [d-Ala2]-GIP-Tag presented the same receptor-binding affinities as full-length GIP1–42 and that the same intracellular signalling pathways were activated in osteoblasts in response to these GIP analogues. Previously GIP1–42 has been reported to enhance collagen maturity in osteoblast cultures (Mieczkowska et al. 2015a ) and to reduce cell differentiation and activity in osteoclast cultures (Mabilleau et al. 2016). In the present study, we provided clear evidence that the two new GIP analogues, [d-Ala2]-GIP1–30 and [d-Ala2]-GIP-Tag, exhibited similar actions in osteoblast and osteoclast cultures.
However, when administered in vivo, these two molecules presented differences. Indeed, [d-Ala2]-GIP1–30 localizes in several tissues that could potentially affect bone physiology whilst as discussed earlier, [d-Ala2]-GIP-Tag localizes almost exclusively in bone. In the ovariectomy-induced bone fragility model, [d-Ala2]-GIP1–30, but not [d-Ala2]GIP-Tag, was proven potent to improve bone strength, mainly by modifying the cortical microarchitecture. However, caution should be taken for interpretation of these observations. Firstly, the activity of [d-Ala2]-GIP-Tag has been tested in isolated cell culture, but not in vivo after incorporation into the bone mineral. Our release assay demonstrated that at acidic pH, close to pH obtained during osteoclast resorption, the fluorescent peptide could be released from bone. However, it was not possible to assess its biological activity. Furthermore, due to the low concentration given to the animals, it was not possible to assess the presence of [d-Ala2]-GIP-Tag in blood or urine. As such, we cannot rule out that the observed lack of effects of [d-Ala2]-GIP-Tag could be due to either low bioavailability or degradation of the peptide after osteoclast resorption. Another explanation, and in addition to GIPr tissue targeting, could suggest that to be beneficial for bone health, extraskeletal GIPr have to be targeted rather than bone-specific GIPr. However, a limitation to this study is that we did not generate tissue-specific invalidation or extraskeletal tissue-specific activation of GIPr to ascertain how the GIP/GIPr pathway controls bone physiology.
The mechanism of action of [d-Ala2]-GIP1–30 was also compared with alendronate. In the present study, alendronate, given at a dose comparable to what is used in humans in the treatment of post-menopausal osteoporosis (i.e. 70 mg/week orally), improved bone strength by acting mostly on bone matrix composition (HIT, Cawidth, carbonate-to-phosphate ratio) rather than restoring cortical bone microarchitecture. On the other hand, [d-Ala2]-GIP1–30 acted preferentially on cortical bone microarchitecture and had almost no action on bone matrix composition, except a small decrease in tissue mineral density. This indicates that the molecular mechanisms of action of these two pharmacological interventions are probably different and in the future, administration of both molecules jointly should be envisaged.
In conclusion, we developed two new GIP analogues that target whole-body GIPr or only bone-specific GIPr. In ovariectomized animals, only [d-Ala2]-GIP1–30 was potent in ameliorating bone strength by restoring cortical bone microarchitecture rather than acting on bone matrix composition in opposition to what was observed with alendronate. This study brought new evidences that targeting the GIP/GIPr pathway might be valuable in bone disorders although further studies will be needed before translating these findings to human post-menopausal osteoporosis.
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.
Funding
This work was supported by a grant from the Société Française de Rhumatologie.
Author contribution statement
Guillaume Mabilleau: Conception and design, acquisition of data, analysis and interpretation of data, drafting and revising the manuscript. Benoit Gobron: Acquisition of data, analysis and interpretation of data, revising the manuscript. Aleksandra Mieczkowska: Acquisition of data, analysis and interpretation of data, revising the manuscript. Rodolphe Perrot: Acquisition of data, analysis and interpretation of data, revising the manuscript. Daniel Chappard: Analysis and interpretation of data, revising the manuscript. All authors approved the current version of the manuscript. Guillaume Mabilleau takes responsibility for the integrity of the data and analysis.
Acknowledgements
The authors are grateful to Nadine Gaborit and Stéphanie Lemière for their help with microCT. The authors also thank the personnel of the animal care facility (University of Angers-SCAHU) for their help with animal handling and injection.
References
Aguado E, Mabilleau G, Goyenvalle E & Chappard D 2017 Hypodynamia alters bone quality and trabecular microarchitecture. Calcified Tissue International 100 332–340. (https://doi.org/10.1007/s00223-017-0235-x)
Alana I, Parker JC, Gault VA, Flatt PR, O’Harte FP, Malthouse JP & Hewage CM 2006 NMR and alanine scan studies of glucose-dependent insulinotropic polypeptide in water. Journal of Biological Chemistry 281 16370–16376. (https://doi.org/10.1074/jbc.M510414200)
Baggio LL & Drucker DJ 2007 Biology of incretins: GLP-1 and GIP. Gastroenterology 132 2131–2157. (https://doi.org/10.1053/j.gastro.2007.03.054)
Bollag RJ, Zhong Q, Phillips P, Min L, Zhong L, Cameron R, Mulloy AL, Rasmussen H, Qin F, Ding KH, et al.2000 Osteoblast-derived cells express functional glucose-dependent insulinotropic peptide receptors. Endocrinology 141 1228–1235. (https://doi.org/10.1210/endo.141.3.7366)
Boskey AL, DiCarlo E, Paschalis E, West P & Mendelsohn R 2005 Comparison of mineral quality and quantity in iliac crest biopsies from high- and low-turnover osteoporosis: an FT-IR microspectroscopic investigation. Osteoporosis International 16 2031–2038. (https://doi.org/10.1007/s00198-005-1992-3)
Bouxsein ML, Boyd SK, Christiansen BA, Guldberg RE, Jepsen KJ & Muller R 2010 Guidelines for assessment of bone microstructure in rodents using micro-computed tomography. Journal of Bone and Mineral Research 25 1468–1486. (https://doi.org/10.1002/jbmr.141)
Butler WT 1989 The nature and significance of osteopontin. Connective Tissue Research 23 123–136. (https://doi.org/10.3109/03008208909002412)
Chappard D 2009 Technical aspects: how do we best prepare bone samples for proper histological analysis? In Bone Cancer: Progression and Therapeutic Approaches, pp 203–210. Ed Heymann D. London, UK: Academic Press/Elsevier.
Degeratu CN, Mabilleau G, Cincu C & Chappard D 2013 Aluminum inhibits the growth of hydroxyapatite crystals developed on a biomimetic methacrylic polymer. Journal of Trace Elements in Medicine and Biology 27 346–351. (https://doi.org/10.1016/j.jtemb.2013.05.004)
Dempster DW, Compston JE, Drezner MK, Glorieux FH, Kanis JA, Malluche H, Meunier PJ, Ott SM, Recker RR & Parfitt AM 2013 Standardized nomenclature, symbols, and units for bone histomorphometry: a 2012 update of the report of the ASBMR Histomorphometry Nomenclature Committee. Journal of Bone and Mineral Research 28 2–17. (https://doi.org/10.1002/jbmr.1805)
Filmon R, Grizon F, Basle MF & Chappaard D 2002 Effects of negatively charged groups (carboxymethyl) on the calcification of poly(2-hydroxyethyl methacrylate). Biomaterials 23 3053–3059. (https://doi.org/10.1016/S0142-9612(02)00069-8)
Gadaleta SJ, Paschalis EP, Betts F, Mendelsohn R & Boskey AL 1996 Fourier transform infrared spectroscopy of the solution-mediated conversion of amorphous calcium phosphate to hydroxyapatite: new correlations between X-ray diffraction and infrared data. Calcified Tissue International 58 9–16. (https://doi.org/10.1007/BF02509540)
Gaudin-Audrain C, Irwin N, Mansur S, Flatt PR, Thorens B, Basle M, Chappard D & Mabilleau G 2013 Glucose-dependent insulinotropic polypeptide receptor deficiency leads to modifications of trabecular bone volume and quality in mice. Bone 53 221–230. (https://doi.org/10.1016/j.bone.2012.11.039)
Henriksen DB, Alexandersen P, Bjarnason NH, Vilsboll T, Hartmann B, Henriksen EE, Byrjalsen I, Krarup T, Holst JJ & Christiansen C 2003 Role of gastrointestinal hormones in postprandial reduction of bone resorption. Journal of Bone and Mineral Research 18 2180–2189. (https://doi.org/10.1359/jbmr.2003.18.12.2180)
Hinke SA, Manhart S, Pamir N, Demuth HR, Gelling W, Pederson RA & McIntosh CH 2001 Identification of a bioactive domain in the amino-terminus of glucose-dependent insulinotropic polypeptide (GIP). Biochimica et Biophysica Acta 1547 143–155. (https://doi.org/10.1016/S0167-4838(01)00181-9)
Hsieh KC, Kao CL, Feng CW, Wen ZH, Chang HF, Chuang SC, Wang GJ, Ho ML, Wu SM, Chang JK, et al.2014 A novel anabolic agent: a simvastatin analogue without HMG-CoA reductase inhibitory activity. Organic Letters 16 4376–4379. (https://doi.org/10.1021/ol501486b)
Irwin N & Flatt PR 2009 Therapeutic potential for GIP receptor agonists and antagonists. Best Practice and Research in Clinical Endocrinology and Metabolism 23 499–512. (https://doi.org/10.1016/j.beem.2009.03.001)
Jepsen KJ, Silva MJ, Vashishth D, Guo XE & van der Meulen MC 2015 Establishing biomechanical mechanisms in mouse models: practical guidelines for systematically evaluating phenotypic changes in the diaphyses of long bones. Journal of Bone and Mineral Research 30 951–966. (https://doi.org/10.1002/jbmr.2539)
Judex S, Luu YK, Ozcivici E, Adler B, Lublinsky S & Rubin CT 2010 Quantification of adiposity in small rodents using micro-CT. Methods 50 14–19. (https://doi.org/10.1016/j.ymeth.2009.05.017)
Kasugai S, Fujisawa R, Waki Y, Miyamoto K & Ohya K 2000 Selective drug delivery system to bone: small peptide (Asp)6 conjugation. Journal of Bone and Mineral Research 15 936–943. (https://doi.org/10.1359/jbmr.2000.15.5.936)
Mabilleau G, Chappard D & Sabokbar A 2011 Role of the A20-TRAF6 axis in lipopolysaccharide-mediated osteoclastogenesis. Journal of Biological Chemistry 286 3242–3249. (https://doi.org/10.1074/jbc.M110.150300)
Mabilleau G, Mieczkowska A, Irwin N, Flatt PR & Chappard D 2013 Optimal bone mechanical and material properties require a functional glucagon-like peptide-1 receptor. Journal of Endocrinology 219 59–68. (https://doi.org/10.1530/JOE-13-0146)
Mabilleau G, Mieczkowska A, Irwin N, Simon Y, Audran M, Flatt PR & Chappard D 2014 Beneficial effects of a N-terminally modified GIP agonist on tissue-level bone material properties. Bone 63 61–68. (https://doi.org/10.1016/j.bone.2014.02.013)
Mabilleau G, Perrot R, Mieczkowska A, Boni S, Flatt PR, Irwin N & Chappard D 2016 Glucose-dependent insulinotropic polypeptide (GIP) dose-dependently reduces osteoclast differentiation and resorption. Bone 91 102–112. (https://doi.org/10.1016/j.bone.2016.07.014)
Manhart S, Hinke SA, McIntosh CH, Pederson RA & Demuth HU 2003 Structure-function analysis of a series of novel GIP analogues containing different helical length linkers. Biochemistry 42 3081–3088. (https://doi.org/10.1021/bbib26868e)
Mansur SA, Mieczkowska A, Bouvard B, Flatt PR, Chappard D, Irwin N & Mabilleau G 2015 Stable incretin mimetics counter rapid deterioration of bone quality in type 1 diabetes mellitus. Journal of Cellular Physiology 230 3009–3018. (https://doi.org/10.1002/jcp.25033)
Mansur SA, Mieczkowska A, Flatt PR, Bouvard B, Chappard D, Irwin N & Mabilleau G 2016 A new stable GIP-oxyntomodulin hybrid peptide improved bone strength both at the organ and tissue levels in genetically-inherited type 2 diabetes mellitus. Bone 87 102–113. (https://doi.org/10.1016/j.bone.2016.04.001)
Mieczkowska A, Irwin N, Flatt PR, Chappard D & Mabilleau G 2013 Glucose-dependent insulinotropic polypeptide (GIP) receptor deletion leads to reduced bone strength and quality. Bone 56 337–342. (https://doi.org/10.1016/j.bone.2013.07.003)
Mieczkowska A, Bouvard B, Chappard D & Mabilleau G 2015a Glucose-dependent insulinotropic polypeptide (GIP) directly affects collagen fibril diameter and collagen cross-linking in osteoblast cultures. Bone 74 29–36. (https://doi.org/10.1016/j.bone.2015.01.003)
Mieczkowska A, Mansur S, Bouvard B, Flatt PR, Thorens B, Irwin N, Chappard D & Mabilleau G 2015b Double incretin receptor knock-out (DIRKO) mice present with alterations of trabecular and cortical micromorphology and bone strength. Osteoporosis International 26 209–218. (https://doi.org/10.1007/s00198-014-2845-8)
Miller SC, Pan H, Wang D, Bowman BM, Kopeckova P & Kopecek J 2008 Feasibility of using a bone-targeted, macromolecular delivery system coupled with prostaglandin E(1) to promote bone formation in aged, estrogen-deficient rats. Pharmacological Research 25 2889–2895. (https://doi.org/10.1007/s11095-008-9706-0)
Montano AM, Oikawa H, Tomatsu S, Nishioka T, Vogler C, Gutierrez MA, Oguma T, Tan Y, Grubb JH, Dung VC, et al.2008 Acidic amino acid tag enhances response to enzyme replacement in mucopolysaccharidosis type VII mice. Molecular Genetics and Metabolism 94 178–189. (https://doi.org/10.1016/j.ymgme.2008.01.007)
Murphy MB, Hartgerink JD, Goepferich A & Mikos AG 2007 Synthesis and in vitro hydroxyapatite binding of peptides conjugated to calcium-binding moieties. Biomacromolecules 8 2237–2243. (https://doi.org/10.1021/bm070121s)
Nishioka T, Tomatsu S, Gutierrez MA, Miyamoto K, Trandafirescu GG, Lopez PL, Grubb JH, Kanai R, Kobayashi H, Yamaguchi S, et al.2006 Enhancement of drug delivery to bone: characterization of human tissue-nonspecific alkaline phosphatase tagged with an acidic oligopeptide. Molecular Genetics and Metabolism 88 244–255. (https://doi.org/10.1016/j.ymgme.2006.02.012)
Nissen A, Christensen M, Knop FK, Vilsboll T, Holst JJ & Hartmann B 2014 Glucose-dependent insulinotropic polypeptide inhibits bone resorption in humans. Journal of Clinical Endocrinology and Metabolism 99 E2325–E2329. (https://doi.org/10.1210/jc.2014-2547)
Oldberg A, Franzen A & Heinegard D 1986 Cloning and sequence analysis of rat bone sialoprotein (osteopontin) cDNA reveals an Arg-Gly-Asp cell-binding sequence. PNAS 83 8819–8823. (https://doi.org/10.1073/pnas.83.23.8819)
Oliver WC & Pharr GM 1992 An improved technique for determining hardness and elastic modulus using load and displacement sensing indentation experiments. Journal of Materials Research 7 1564–1583. (https://doi.org/10.1557/JMR.1992.1564)
Paschalis EP, DiCarlo E, Betts F, Sherman P, Mendelsohn R & Boskey AL 1996 FTIR microspectroscopic analysis of human osteonal bone. Calcified Tissue International 59 480–487. (https://doi.org/10.1007/BF00369214)
Paschalis EP, Verdelis K, Doty SB, Boskey AL, Mendelsohn R & Yamauchi M 2001 Spectroscopic characterization of collagen cross-links in bone. Journal of Bone and Mineral Research 16 1821–1828. (https://doi.org/10.1359/jbmr.2001.16.10.1821)
Pereira M, Gohin S, Roux JP, Fisher A, Cleasby ME, Mabilleau G & Chenu C 2017 Exenatide improves bone quality in a murine model of genetically inherited type 2 diabetes mellitus. Frontiers in Endocrinology 8 327. (https://doi.org/10.3389/fendo.2017.00327)
Rey C, Collins B, Goehl T, Dickson IR & Glimcher MJ 1989 The carbonate environment in bone mineral: a resolution-enhanced Fourier Transform Infrared Spectroscopy Study. Calcified Tissue International 45 157–164. (https://doi.org/10.1007/BF02556059)
Roschger P, Fratzl P, Eschberger J & Klaushofer K 1998 Validation of quantitative backscattered electron imaging for the measurement of mineral density distribution in human bone biopsies. Bone 23 319–326. (https://doi.org/10.1016/S8756-3282(98)00112-4)
Roschger P, Paschalis EP, Fratzl P & Klaushofer K 2008 Bone mineralization density distribution in health and disease. Bone 42 456–466. (https://doi.org/10.1016/j.bone.2007.10.021)
Shao Y, Hernandez-Buquer S, Childress P, Stayrook KR, Alvarez MB, Davis H, Plotkin LI, He Y, Condon KW, Burr DB, et al.2017 Improving combination osteoporosis therapy in a preclinical model of heightened osteoanabolism. Endocrinology 158 2722–2740. (https://doi.org/10.1210/en.2017-00355)
Spevak L, Flach CR, Hunter T, Mendelsohn R & Boskey A 2013 Fourier transform infrared spectroscopic imaging parameters describing acid phosphate substitution in biologic hydroxyapatite. Calcified Tissue International 92 418–428. (https://doi.org/10.1007/s00223-013-9695-9)
Takahashi T, Yokogawa K, Sakura N, Nomura M, Kobayashi S & Miyamoto K 2008 Bone-targeting of quinolones conjugated with an acidic oligopeptide. Pharmacological Research 25 2881–2888. (https://doi.org/10.1007/s11095-008-9605-4)
Torekov SS, Harslof T, Rejnmark L, Eiken P, Jensen JB, Herman AP, Hansen T, Pedersen O, Holst JJ & Langdahl BL 2014 A functional amino acid substitution in the glucose-dependent insulinotropic polypeptide receptor (GIPR) gene is associated with lower bone mineral density and increased fracture risk. Journal of Clinical Endocrinology and Metabolism 99 E729–E733. (https://doi.org/10.1210/jc.2013-3766)
Tsukiyama K, Yamada Y, Yamada C, Harada N, Kawasaki Y, Ogura M, Bessho K, Li M, Amizuka N, Sato M, et al.2006 Gastric inhibitory polypeptide as an endogenous factor promoting new bone formation after food ingestion. Molecular Endocrinology 20 1644–1651. (https://doi.org/10.1210/me.2005-0187)
Turner CH & Burr DB 1993 Basic biomechanical measurements of bone: a tutorial. Bone 14 595–608. (https://doi.org/10.1016/8756-3282(93)90081-K)
Walsh JS & Henriksen DB 2010 Feeding and bone. Archives of Biochemistry and Biophysics 503 11–19. (https://doi.org/10.1016/j.abb.2010.06.020)
Xie D, Cheng H, Hamrick M, Zhong Q, Ding KH, Correa D, Williams S, Mulloy A, Bollag W, Bollag RJ, et al.2005 Glucose-dependent insulinotropic polypeptide receptor knockout mice have altered bone turnover. Bone 37 759–769. (https://doi.org/10.1016/j.bone.2005.06.021)
Yokogawa K, Miya K, Sekido T, Higashi Y, Nomura M, Fujisawa R, Morito K, Masamune Y, Waki Y, Kasugai S, et al.2001 Selective delivery of estradiol to bone by aspartic acid oligopeptide and its effects on ovariectomized mice. Endocrinology 142 1228–1233. (https://doi.org/10.1210/endo.142.3.8024)
