DOUBLE INCRETIN RECEPTOR KNOCK-OUT (DIRKO) MICE PRESENT WITH ALTERATIONS OF TRABECULAR AND CORTICAL MICROMORPHOLOGY AND
BONE STRENGTH
Aleksandra Mieczkowska1, Sity Mansur3, Beatrice Bouvard1,4, Peter R Flatt3, Bernard Thorens5, Nigel Irwin3, Daniel Chappard1,2, Guillaume Mabilleau1,2
1 LUNAM Université, GEROM-LHEA, Institut de Biologie en Santé, Angers, France
2 LUNAM Université, SCIAM, Institut de Biologie en Santé, Angers, France
3 Ulster University, School of Biomedical Sciences, Coleraine, United Kingdom
4 Service de Rhumatologie, CHU d'Angers, Angers, France
5 University of Lausanne, Centre for integrative genomics, Lausanne, Switzerland
Running title: Double incretin receptor knock-out and bone quality
Please send all correspondence to:
Guillaume Mabilleau, PhD : +33(0) 244 688 349 GEROM-LHEA UPRES EA 4658 Fax : +33(0) 244 688 451
Institut de Biologie en Santé : [email protected] Université d’Angers
4 rue larrey
49933 Angers Cedex 09 France
Keywords: gut hormones, bone strength, nanoindentation,
Conflict of interest: Aleksandra Mieczkowska, Sity Mansur, Beatrice Bouvard, Peter R Flatt, Bernard Thorens, Nigel Irwin, Daniel Chappard and Guillaume Mabilleau have no conflict of interest to declare
SUMMARY
A role for gut hormone in bone physiology has been suspected. We evidenced alterations of microstructural morphology (trabecular and cortical) and bone strength (both at the whole-bone - and tissue level) in double incretin receptor knock-out (DIRKO) mice as compared to wild-type littermates.
These results support a role for gut hormones in bone physiology.
ABSTRACT
Purpose/Introduction: The two incretins, glucose-dependent insulinotropic polypeptide (GIP) and glucagon-like peptide-1 (GLP-1), have been shown to control bone remodeling and strength. However, lessons from single incretin receptor knock-out mice highlighted a compensatory mechanism induced by elevated sensitivity to the other gut hormone. As such, it is unclear whether the bone alterations observed in GIP or GLP-1 receptor deficient animals resulted from the lack of a functional gut hormone receptor, or by higher sensitivity for the other gut hormone. The aims of the present study were to investigate the bone micro-structural morphology, as well as bone tissue properties, in double incretin receptor knock-out (DIRKO) mice.
Methods: Twenty-six week old DIRKO mice were age- and sex-matched with wild-type (WT) littermates. Bone microstructural morphology was assessed at the femur by microCT and quantitative X-ray imaging, whilst tissue properties were investigated by quantitative backscattered electron imaging and Fourier-transformed infrared microscopy. Bone mechanical response was assessed at the whole-bone- and tissue-level by 3-point bending and nanoindentation, respectively.
Results: As compared to WT animals, DIRKO mice presented significant augmentations in trabecular bone mass and trabecular number whereas bone outer diameter, cortical thickness and cortical area were reduced. At the whole-bone-level, yield stress, ultimate stress and post-yield work to fracture were significantly reduced in DIRKO animals. At the tissue-level, only collagen maturity was reduced by 9% in DIRKO mice leading to reductions in maximum load, hardness and dissipated energy.
Conclusions: This study demonstrated the critical role of gut hormones in controlling bone microstructural morphology and tissue properties.
INTRODUCTION
In adulthood, bone is permanently remodeled in mass and architecture to adapt and repair damage induced by mechanical stress and ageing. Bone remodeling necessitates a spatio-temporal coupling between osteoclasts, the bone-resorbing cells, and osteoblasts, the bone-forming cells. Bone remodeling is traditionally considered to be regulated by hormones, autocrine/paracrine signals from the microenvironment and mechanical loading. A role for gastro-intestinal hormones in controlling bone remodeling has been suggested as changes in the profile of serum markers of bone remodeling after a meal coincides with peaks in gastro-intestinal hormones in plasma [1, 2].
Among all gut hormones, a specific interest has arisen in the incretin hormones, glucose-dependent insulinotropic polypeptide (GIP) and glucagon-like peptide-1 (GLP-1). GIP is synthesized and secreted into the blood stream by the duodenal endocrine K cells, whilst GLP-1 is produced by ileal endocrine L cells. Release of gut hormones is stimulated by the presence of nutrients and especially glucose in the intestinal lumen [3]. In order to induce a response from the target tissue, these hormones require binding to their specific G-coupled receptors, GIPr and GLP-1r respectively, that activate the adenylyl cyclase pathway [4, 5].
Previously, we and others have reported the presence of functional GIPrs on bone cells and alterations of bone microarchitecture and quality in genetically-modified GIPr animals [6-10]. Although the presence of a functional GLP-1r on bone cells is controversial, mice presenting with genetic deletion of GLP-1r exhibit profound alterations of bone microarchitecture and quality [11]. However, knock-out mice lacking the GLP-1r have a marked and significant upregulation of plasma GIP levels and an elevated sensitivity to GIP action. Similarly, lack of GIPr action leads to increased GLP-1 sensitivity [12-14]. As such, it is unclear whether the observed alterations of bone microarchitecture and quality observed in GIP and GLP-1 receptor deficiency results from the lack of a functional incretin receptor, or by a compensatory mechanism induced by elevated sensitivity to the other incretin hormone. Furthermore, it is unclear whether each gut hormone has a preferred site of action in bone, i.e., trabecular or cortical bone.
The aims of the present study were to conduct a comprehensive investigation of trabecular and cortical bone microarchitecture, as well as investigation of bone quality, in double incretin receptor knock-out (DIRKO) mice in order to decipher the role of GIP and GLP-1 in bone physiology. Our
results indicated that DIRKO mice presented with significant modifications of trabecular and cortical microstructural morphology as well reduced bone strength at the whole-bone- and tissue-level.
MATERIAL AND METHODS
Animals
Twenty-six-week-old female mice presenting a deletion of both the GLP-1r and GIPr (DIRKO) were used in this study. This age was chosen based on our previous observations that greater difference in skeletal phenotype is observed in aged mice with incretin receptor deletion [6]. The generation of DIRKO mice used in this study are described in details elsewhere [15]. Briefly, Gipr-/-Glp-1r+/+ and Gipr+/+Glp-1r-/- mice, both on a C57BL/6 background [6, 11], were crossed in order to obtain Gipr+/- Glp-1r+/- mice. Then the double heterozygotes were crossed to generate Gipr-/- Glp-1r-/- (DIRKO) and Gipr+/+ Glp-1r+/+ (WT) littermates. Genotype was determined by genomic DNA preparation from tail snips and analyzed by southern blotting as described in details in [15, 16]. DIRKO and WT background was maintained by crossing siblings together. Age- and sex-matched animals were used.
A total of 8 DIRKO and 10 control mice were used in this study. Animals were maintained on a 12h:12h light-dark cycle in a temperature-controlled room (21.5 ± 1°C). Animals received standard rodent maintenance diet and water ad libitum. All experiments were conducted according to United Kingdom Office regulations (UK Animals Scientific Procedures Act 1986) and European Union laws.
Animals were sacrificed by lethal inhalation of CO2. At necropsy, animals were weighted and fat percentage was determined using a PIXImus system (Inside Outside sales, Wisconsin, USA). Femurs were then collected, cleaned of soft tissue and stored in 70% ethanol at 4°C until use.
Intraperitoneal glucose tolerance tests
Drinking water and a standard rodent maintenance diet (Trouw Nutrition, Cheshire, UK) were freely available until 18h before test. Mice received an intraperitoneal injection of glucose alone (18mmol/kg body weight) in a final volume of 8ml/kg body weight. Blood samples were collected from the cut tip on the tail vein of conscious mice into chilled fluoride/heparin glucose microcentrifuge tubes (Sarstedt, Nümbrecht, Germany) immediately prior to injection and at 15, 30 and 60 minutes post injection.
Plasma was aliquoted and stored at -20°C prior to glucose and insulin determination. Plasma glucose and insulin were assayed by an automated glucose oxidase procedure using a beckman glucose
analyzer II (Beckman Instruments, Galway, Ireland) and a modified dextran-coated charcoal immunoradioassay [17], respectively.
MicroCT
MicroCT analysis was performed in the distal metaphysis and at the midshaft of left femurs with a Skyscan 1172 microtomograph (Bruker MicroCT, Kontich, Belgium) equipped with an X-ray tube operating at 69 kV/100 µA. The pixel size was fixed at 3.75 µm, the rotation step at 0.25° and exposure was done with a 0.5-mm aluminum filter. Trabecular analysis was performed at the distal metaphysis. The trabecular volume of interest (VOI) was located 0.5 mm above the growth plate and extended on a 2-mm height in direction of the diaphysis. The following parameters were determined with the CTAn software (release 1.11.4.2, Bruker microCT): trabecular bone volume (BV/TV, in %), trabecular thickness (Tb.Th, in µm), trabecular number (Tb.N, mm-1), trabecular separation (Tb.Sp, in µm). Three-dimension models have been reconstructed from the image stack with the Ant software (release 2.5, Bruker MicroCT). Trabecular VOIs have been pseudo-colored for better visualization of the trabecular network. Cortical parameters were assessed at the midshaft femur. Cortical VOIs extended on 2-mm centered at the midshaft femur. External bone diameter (B.Dm, in mm), marrow diameter (Ma.Dm, in mm), cortical thickness (Ct.Th, in µm), and cross-sectional moment of inertia (CSMI, in mm4) were measured with a lab-based routine made with ImageJ 1.45s (NIH, Bethesda, MD). All these parameters were determined according to guidelines and nomenclature proposed by the American Society for Bone and Mineral Research [18].
Quantitative X-ray imaging (qXRI)
Bone mineral content at the midshaft femur was determined using qXRI as previously reported by Bassett et al. [19]. Briefly, digital X-ray images of the left femur were recorded at a 12-µm pixel resolution using a Faxitron MX20 device (Edimex, Angers, France) operating at 26 kV and a 4X magnification. The region of interest was located at the midshaft femur and represented a height of 2 mm. A 1.5-mm thick steel plate and a 1.5-mm thick polyester plate were used on each microradiograph and served as internal standards as previously reported [11, 19]. Before converting the 14-bit DICOM images into 8-bit tiff images, the histogram was stretched from the polyester (gray level 0) to the steel (gray level 16383) standards using ImageJ 1.45s. Increasing gradations of
mineralization density were represented in 16 equal intervals using the 16-colours lookup table in ImageJ 1.45s. The frequency of occurrence of an i grey level (Fi) was calculated as follows:
100
where Ni represents the number of pixels with the i grey level and Nt the total number of pixels. The frequency distribution as a function of grey level was plotted and the mean grey level (GLmean) of each bone was deduced from this distribution using the following formula:
100 Where GLi represents the value of the i grey level.
Bone histomorphometry
Left femurs were embedded, undecalcified in poly (methylmethacrylate) (pMMA) at 4°C to preserve enzyme activities. Sections (7-µm thickness) were performed on a heavy-duty microtome equipped with a 50° tungsten carbide knife. For each animal, four sections (50 um apart) were stained with toluidine blue for osteoblast counting (original magnification x400) and four additional sections were stained for the osteoclastic tartrate resistant acid phosphatase (TRAcP) and detection of osteoclasts.
Only TRAcP-positive nucleated cells in contact with bone were counted as osteoclasts. For trabecular analysis, the region of interest (ROI) was located in the secondary spongiosa 0.5 mm below the growth plate on a height of 2 mm. Cortical analysis was performed at the midshaft femur with an ROI of 2mm centered at the midshaft. Standard bone histomorphometrical nomenclatures, symbol and units were used as described in the report of the American Society for Bone and Mineral Research [20].
Serum levels of soluble mediators
Just before sacrificing the animal, blood samples were collected by intracardiac collection. Blood samples were spun at 2000 g for 15 min at 4°C and aliquoted by 100 ul. As an average, ≈ 200ul were recovered from each animal. Serum levels of c-terminal telopeptide of collagen type I (CTX-I) and osteocalcin were assessed with commercially available kit (Ratlaps®-Immunodiagnostic systems and mouse osteocalcin kit-Immutopics, respectively) according to the manufacturer recommendations.
Three-point bending
Three-point bending experiments were performed on the right femur. Prior to mechanical testing, femurs were rehydrated in saline overnight at 4°C. The femur length was measured with digital calipers between the two groups of animals. No modification in this parameter was evidenced. Three- point bending strength was measured with a constant span length of 10 mm on an Instron 5942 (Instron, Elancourt, France). The press head as well as the two support points were rounded to avoid shear load and cutting. Femurs were positioned horizontally with the anterior surface facing upward, centered on the support and the pressing force was applied vertically to the midshaft of the bone.
Each bone was tested with a loading speed of 2 mm.min-1 until failure with a 500N load cell. The load- displacement curve was acquired with the Bluehill 3 software (Instron) and converted into stress-strain curve. Bending properties such as ultimate stress, yield stress (0.2% offset method), bending modulus and post-yield work to fracture were calculated according to previously published equations [21-23]
and represented material properties at the whole-bone level.
Fourier transformed infrared microscopy (FTIRM)
After three-point bending experiments, femurs were cross-sectionally cut at the midshaft using a diamond saw (Accutom, Struers, Champigny sur Marne, France) and embedded undecalcified in polymethylmethacrylate at 4°C as previously reported [24]. Cross-sections (4µm thickness) of the midshaft femur were cut dry on a heavy duty microtome equipped with tungsten carbide knives (Leica Polycut S) and sandwiched between BaF2 optical windows. Spectral analysis were obtained on a Bruker Vertex 70 spectrometer (Bruker optics, Ettlingen, Germany) interfaced with a Bruker Hyperion 3000 infrared microscope equipped with a standard single element Mercury Cadmium Telluride (MCT) detector (750-4000cm-1). Infrared spectra were recorded at a resolution of 4cm-1, with an average of 32 scans in transmission mode. Background spectral images were collected under identical conditions from the same BaF2 windows at the beginning and end of each experiment to ensure instrument stability. For FTIR analysis, 12 spectra were acquired randomly on cortical bone and analyzed with the Opus Software (release 6.5, Bruker). The contribution of the embedding polymethylmethacrylate (pMMA) and water vapor were corrected for each spectrum prior to baseline correction. Individual spectra were then subjected to curve fitting using a commercially available software package (Grams/AI 8.0, Thermofisher scientific, Villebon sur Yvette, France). Every absorption band was
characterized by its area using the Levenberg-Maquardt algorithm. The second derivative spectrum was used to determine the number and the position of the bands constituting every spectral interval.
All bands were positioned at maximal intensities with a Gaussian shape. Position, height, width at half intensity and area were obtained. Peaks corresponding to amide I bands (1585-1725 cm-1) were considered for collagen analysis. Collagen maturity, determined as the relative ratio of subbands located at 1660 cm-1 and 1690 cm-1 of the amide I peak, was assessed. Although multiple theories exists about the significance of this ratio (pyridinium trivalent to dehydrodihydroxylysinonorleucine divalent collagen cross-links, modifications of secondary structure of collagen molecules with ageing of the matrix), this ratio indicates the maturity of the collagen of the bone matrix [25, 26].
Nanoindentation
Nanoindentation tests evaluated the mechanical properties of the bone matrix. As nanoindentation assesses volume of material at a length scale less than that of individual microstructural features in bone, this technique avoids confounding factors such as bone microarchitecture and porosity that affect material properties at larger length scales such as 3-point bending. PMMA blocks used for FTIRM sections were polished to a 1-µm finish with diamond particles (Struers, France) and the same subregions as used for FTIRM were analyzed by nanoindentation. Prior to nanoindentation testing, blocks were rehydrated overnight in saline. Twelve indentations, at distance from canals, osteocyte lacunae and/or microcracks were randomly positioned in cortical bone with a NHT-TTX system (CSM, Peseux, Switzerland) equipped with a Berkowitch diamond probe. Indentations were made up to a depth of 900 nm with a loading/unloading rate of 40mN/min. At maximum load, a holding period of 15 seconds was applied to avoid creeping of the bone material. The following material properties at the tissue-level, maximum load, indentation modulus, hardness and dissipated energy, were determined according to Oliver and Pharr [27].
Quantitative backscattered electron imaging (qBEI)
Quantitative backscattered electron imaging was employed to determine the bone mineral density distribution (BMDD) as previously reported [28]. QBEI experiments were performed on the same blocks and same regions as nanoindentation and FTIRM. Polymethylmethacrylate blocks were carbon-coated and observed with a scanning electron microscope (EVO LS10, Carl Zeiss Ltd,
Nanterre, France) equipped with a five quadrant semi-conductor backscattered electron detector. The microscope was operated at 20 keV with a probe current of 250 pA and a working distance of 15 mm.
The backscattered signal was calibrated using pure carbon (Z=6, mean grey level = 25), pure aluminum (Z=13, mean grey level =225) and pure silicon (Z=14, mean grey level =253) standards (Micro-analysis Consultants Ltd, St Ives, UK). For these contrast/brightness settings, the BSE grey level histogram was converted into weight percentage of calcium. Eventual changes in brightness and contrast due to instrument instabilities were checked by monitoring the current probe and imaging the reference material (C, Al and Si) every 15 min. The cortical bone area was imaged at a 200X nominal magnification, corresponding to a pixel size of 0.5 µm per pixel. Six images per samples were taken.
The grey levels distribution of each image was analyzed with Image J. Three variables were obtained from the bone mineral density distribution: Capeak as the most frequently observed calcium concentration, Camean as the average calcium concentration and Cawidth as the width of the histogram at half maximum of the peak.
Statistical analysis
Results were expressed as mean ± standard error of the mean (SEM). Non-parametric Mann-Whitney U-test was used to compare the differences between the groups using the Systat statistical software release 13.0 (Systat software Inc., San Jose, CA). Differences at p<0.05 were considered to be significant.
RESULTS
DIRKO mice exhibit a mild impairment in glucose tolerance
As represented Figure 1, no significant differences were evidenced between WT and DIRKO mice in term of body weight and fat mass (p=0.149 and p=0.177, respectively). On the other hand, as compared to WT animals, DIRKO mice exhibited a significant increase in glucose plasma level (p=0.034) after an IPGTT. Although reduced in DIRKO mice, the insulin response did not reach statistical significance as compared with WT animals (p= 0.127).
DIRKO mice have a significant increase in trabecular bone mass
Trabecular microstructural morphology of the distal metaphysis of the femur was assessed by microCT. As represented Figure 2, trabecular bone mass appeared higher in DIRKO mice as compared with WT animals. Indeed, trabecular analysis (Table 1) revealed that BV/TV was significantly increased by 210% in DIRKO animals (p=0.003). Furthermore, this increase in BV/TV was accompanied by an increase in Tb.N (+207%, p=0.012) and a reduction in Tb.Sp (-22%, p=0.038). On the other hand, Tb.Th was unaffected in DIRKO mice. Furthermore, the number of osteoclasts lying down on trabecular bone was significantly reduced by 46% (p=0.002). On the other hand, the number of osteoblasts was unchanged between the two groups of animals. Serum levels of CTX were also significantly reduced by 27% (p=0.012) whilst osteocalcin levels were unchanged.
DIRKO mice have reduced cortical bone mass and cortical strength
In order to further understand the skeletal phenotype of DIRKO mice, we also investigated the cortical microstructural morphology and mechanical response at the whole-bone -level. As represented Figure 3, the bone mineral content at the midshaft femur was markedly reduced in DIRKO mice as indicated by a 5% reduction in GLmean (p=0.004). A careful analysis of the cortical morphology at the midshaft femur (Table 2) highlighted a significant decrease in B.Dm (-9%, p=0.028) in DIRKO animals whilst Ma.Dm was unchanged as compared with WT animals. As a result, Ct.Th and Ct.Ar were significantly reduced by 17% and 14% in DIRKO as compared with WT mice (p=0.004 and p=0.012 respectively).
The cross-sectional moment of inertia was also decreased in DIRKO mice as compared to WT animals by 25% (p=0.012). Investigations of bone strength by 3-point bending exhibited significant reductions in yield stress (-48%, p=0.032), ultimate stress (-44%, p=0.034) and post-yield work to fracture (-31%, p=0.034) in DIRKO animals. On the other hand, the bending modulus was unchanged between the two groups of animals.
DIRKO mice exhibited a deterioration of mechanical response at the tissue-level and altered collagen maturity
To further elucidate the mechanism underlying the reduction in cortical bone strength, we investigated the mechanical response of the bone matrix at the tissue-level by nanoindentation (Table 3). This methodology avoids confounding factors, like intracortical porosity for example, that could bias the interpretation of the mechanical response at the whole-bone -level determined by 3-point bending.
Nanoindentation experiments highlighted that DIRKO mice presented with significant reductions in maximum load (-21%, p=0.027), hardness (-29%, p=0.014) and dissipated energy (-25%, p=0.014).
On the other hand, the elastic response of the bone matrix, as determined by the indentation modulus, was unchanged between the two groups of animals.
These modifications of mechanical properties at the tissue-level led us to investigate the bone mineral density distribution. Histograms representing the distribution of calcium in the bone matrix in WT and DIRKO mice were almost superimposed (data not shown). Indeed, no differences in any of the calcium parameter (Capeak, Camean and Cawidth) were evidenced indicating no defect in bone mineral density of the cortical bone matrix (Table 3). On the other hand, the assessment of collagen maturity by infrared microscopy revealed a significant reduction by 9% in the 1660/1690 cm-1 ratio (p=0.011, Table 3) in DIRKO mice.
DISCUSSION
The strength of bone to resist fracture is not only dependent on its microstructural morphology but also on material properties at the tissue-level. A role for gut hormones, especially GIP and GLP-1, in controlling bone microarchitecture and material properties has been suggested recently [6, 7, 11].
However, in single incretin receptor knock-out mice, there are always possibilities that the second gut hormone compensates for the lack of biological action of the first. As such, to decipher the role of gut hormones in skeletal biology, in the present study, we investigated the bone phenotype of double incretin receptor knock-out mice. Previous investigations in this animal model failed to highlight modifications of either body weight or composition in DIRKO animals as compared with WT littermates [15] and our observations totally agree with these observations. Furthermore, previous investigations suggested a mild impairment in glucose tolerance and no significant modification in insulin secretion and here again, our observations, in the present study, are in agreement with published work by others [15, 29]. However, in the present study, we evidenced that DIRKO animal present with significant modifications of bone morphology at the micro-level. Indeed, DIRKO mice had increases in trabecular bone mass and trabecular number whilst reductions in the outer bone diameter and cortical thickness were observed. Furthermore, alterations of bone strength at the whole-bone (yield stress, ultimate stress, post-yield work to fracture) and tissue (maximum load, hardness, dissipated energy)
levels were encountered. These alterations of bone strength were also associated with modifications of tissue properties with reduction in collagen maturity.
Controversies on the exact role of the GIPr in trabecular bone exist. The functional GIPr is a G protein- coupled receptor belonging to the B1 subfamily that is composed of a large extracellular N terminus responsible for GIP binding, a serpentine domain with seven transmembrane helices and a short intracellular C terminus [30, 31]. Xie et al reported a reduction in trabecular bone volume as early as 4- weeks of age [9] in an animal model of GIPr deficiency where only a portion of the extracellular domain of the GIPr was deleted. In the same GIPr-deficiency model, Tsukiyama et al. reported similar effects at an older age (8 weeks old) [8]. We recently reported an increase in trabecular bone volume in a second animal model of GIPr deficiency, where the entire extracellular domain and a portion of the first transmembrane helix of the GIPr were genetically deleted [6]. This increase in trabecular bone volume was linked to modifications of the adipokine network at 16 weeks of age and reduction in age- related bone loss in older animals. In the present study, we evidenced a significant reduction in the number of osteoclasts in trabecular bone and a reduction in serum CTX levels in DIRKO mice to a similar extent observed in single GIPr KO mice. As such it is possible that the observed higher trabecular bone volume observed in DIRKO animals might also be due to the same mechanism and a reduction in age-related bone loss.
Furthermore, trabecular bone microarchitecture in GLP-1r deficient animal has also been investigated.
Yamada et al reported a trend for a reduction in trabecular bone mass in these animals at 6 weeks of age, accompanied by increases in eroded surface and number of multinucleated osteoclasts [32]. Our unpublished observations performed on the same model of Glp-1r-/- animals but at 16 weeks of age, support the studies of Yamada, where we observed a 20% reduction in BV/TV at the distal metaphysis of the femur accompanied by a 26% reduction in trabecular number.
The DIRKO mice used in the present study were generated by crossing Gipr-/- and Glp1r-/- mice. The model of Gipr-/- mice used for this crossing was the one used in our previous investigations [6]. The results obtained in the present study in trabecular bone of DIRKO mice are similar to those obtained with single Gipr-/- mice suggesting that the action of the GIPr might be predominant over GLP-1r in trabecular bone.
In cortical bone, fewer controversies exist on the effects of GIPr and GLP-1r on bone microarchitecture and bone strength. The midshaft femur has been chosen for the assessment of cortical bone as this
location is almost exclusively made of cortical bone in C57BL/6 mice [33, 34]. GIPr-deficient animals exhibit a reduction in cortical thickness due to larger marrow diameter associated with unchanged outer bone diameter [7]. In contrast although GLP-1r-deficient animals also present with a decrease in cortical thickness, this reduction is due to a decrease in the outer bone diameter with unchanged marrow diameter [11]. In the present study, DIRKO mice exhibited modifications of cortical bone microarchitecture with reduction in cortical thickness due to decrease in bone outer diameter with unchanged marrow diameter. These data suggest a reduction of bone expansion at the periosteal site.
However, these mice were not labeled with calcein or tetracyclins and we were not in a position to assess osteoblast activity. Nevertheless, collectively these data support alterations of cortical bone morphology in DIRKO mice similar to those observed in Glp1r-/- mice.
Cortical bone strength was also dramatically altered in both models of single receptor deletions at the whole-bone and tissue levels. Gipr-/- mice presented with a reduction in post-yield work to fracture (- 14%) whilst ultimate stress, yield stress and bending modulus were unchanged [7]. Glp-1r-/- mice exhibited reductions in yield stress (-7%) and post-yield work to fracture (-20%) whilst ultimate stress and bending modulus were unchanged [11]. In the present study, DIRKO mice present with a more severe phenotype with considerable reductions in yield and ultimate stresses (-48% and -44%
respectively) and post-yield work to fracture (-31%). Furthermore, at the tissue level, DIRKO mice exhibited a significant reduction in collagen maturity as also observed in Gipr-/- and Glp-1r-/- mice, but interestingly, the bone mineral density distribution was unchanged, a feature only observed previously in Glp-1r-/- mice. Bone strength at the tissue level was also drastically altered in DIRKO mice with significant reductions in maximum loading, hardness and dissipated energy but not in indentation modulus. Hardness represents the plastic response of the bone matrix whilst indentation modulus represents the elastic response. Taken together, it is plausible that the decrease in collagen maturity led to the decrease in hardness and hence in maximum loading. Overall, the cortical bone phenotype in DIRKO mice is closer to Glp-1r-/- mice than Gipr-/- animals.
It is worth noting that the expression of a functional GLP-1r in bone is still controversial. The GLP-1r has been detected at the transcript levels in several osteoblast cell lines [35], but we previously failed to demonstrate its expression in bone marrow-derived osteoblast and osteoclast cultures extracted
from normal mice [11]. Also, Nuche-Berenguer reported the presence of a second new functional receptor expressed in the murine MC3T3-E1 osteoblast cell line [36], although this data has yet to be confirmed in primary bone cells. Nonetheless, we have to bear in mind that the observed cortical phenotype observed in Glp-1r-/- and DIRKO mice might not be solely due to deletion of the conventional GLP-1r but might result from the action of GLP-1 on other undefined receptors.
It is intriguing to observe an augmentation in trabecular bone mass whilst the cortical bone mass is reduced. This difference between the two bone envelopes might be due to a plethora of events and seem suggest differential effects of incretin hormones on each of the bone compartment. However, as the GLP-1r is not express in primary bone cells, it is unlikely that this phenotype arise from a direct effect of GLP-1r deletion. GIPr and GLP-1r are also widely expressed in other tissues and especially in the pancreas where they are both present at the surface of beta cells. One of the functions of GIP and GLP-1 is to potentiate glucose-dependent insulin release. DIRKO mice, with the same genetic background as used in the present study, have a significant reduction in insulin secretion after an oral glucose tolerance test as compared to single receptor knock-out mice [15]. As such it is plausible that the observed skeletal phenotype could also result from lower circulating insulin levels. Furthermore, we cannot rule out other extra-pancreatic consequences of double incretin receptor knock-out on calcitonin and glucocorticoid secretions and modifications of adipokine network as previously observed [6, 37-39]. Interestingly, the GIPr but not the GLP-1r is expressed in skeletal muscle cells [40, 41]. It is also possible that the observed discrepancy between trabecular and cortical bone results from changes in the local skeletal muscle.
Undoubtedly, DIRKO mice present a reduction in bone strength, but this reduction seemed to be more marked at the whole-bone than at the tissue level. A parameter that was not investigated in our study is the impact of incretin receptors deletion on cortical porosity. Cortical porosity is comprised of canal and vascular networks and osteocyte lacunar system [42-44]. In mice, canal diameter is ranged between 7-9 microns [45]. It is difficult to assess accurately this parameter with desktop microCT and often require the use of nanoCT or synchrotron radiation CT that allows a much lower nominal resolution (~500-700nm nominal resolution). Nevertheless, it would be interesting to assess whether
Gipr-/-, Glp-1r-/- and DIRKO mice exhibit changes in cortical porosity and if this could account for the dramatic reduction in bone strength observed at the whole-bone level.
In conclusion, to our knowledge, this is the first study investigating the bone microarchitecture and quality in DIRKO mice. DIRKO mice exhibited dramatic and profound alterations of bone microarchitecture and bone strength suggesting the important role of gut hormone in regulating bone quality.
ACKNOWLEDGEMENTS
The authors are grateful to N Gaborit and G Brossard for their help with microCT. This work was supported by grants from the Bioregos2 Program and the University of Ulster Proof of Principle Funding Program.
REFERENCES
1. Elnenaei MO, Musto R, Alaghband-Zadeh J, Moniz C, Le Roux CW (2010) Postprandial bone turnover is independent of calories above 250 kcal. Ann Clin Biochem 47:318-320
2. 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. J Bone Miner Res 18:2180-2189
3. Baggio LL, Drucker DJ (2007) Biology of incretins: GLP-1 and GIP. Gastroenterology 132:2131-2157
4. Thorens B (1992) Expression cloning of the pancreatic beta cell receptor for the gluco-incretin hormone glucagon-like peptide 1. PNAS 89:8641-8645
5. Usdin TB, Mezey E, Button DC, Brownstein MJ, Bonner TI (1993) Gastric inhibitory polypeptide receptor, a member of the secretin-vasoactive intestinal peptide receptor family, is widely distributed in peripheral organs and the brain. Endocrinology 133:2861-2870
6. Gaudin-Audrain C, Irwin N, Mansur S, Thorens B, Flatt PR, Basle MF, Chappard D, Mabilleau G (2013) Glucose-dependent insulinotropic polypeptide receptor deficiency leads to modifications of trabecular bone mass and quality in mice. Bone 53:221-230
7. 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
8. Tsukiyama K, Yamada Y, Yamada C, et al. (2006) Gastric inhibitory polypeptide as an endogenous factor promoting new bone formation after food ingestion. Mol Endocrinol 20:1644-1651
9. Xie D, Cheng H, Hamrick M, et al. (2005) Glucose-dependent insulinotropic polypeptide receptor knockout mice have altered bone turnover. Bone 37:759-769
10. Xie D, Zhong Q, Ding KH, et al. (2007) Glucose-dependent insulinotropic peptide- overexpressing transgenic mice have increased bone mass. Bone 40:1352-1360
11. 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. J Endocrinol 219:59-68
12. Flamez D, Van Breusegem A, Scrocchi LA, Quartier E, Pipeleers D, Drucker DJ, Schuit F (1998) Mouse pancreatic beta-cells exhibit preserved glucose competence after disruption of the glucagon-like peptide-1 receptor gene. Diabetes 47:646-652
13. Pamir N, Lynn FC, Buchan AM, et al. (2003) Glucose-dependent insulinotropic polypeptide receptor null mice exhibit compensatory changes in the enteroinsular axis. Am J Physiol Endocrinol Metab 284:E931-939
14. Pederson RA, Satkunarajah M, McIntosh CH, Scrocchi LA, Flamez D, Schuit F, Drucker DJ, Wheeler MB (1998) Enhanced glucose-dependent insulinotropic polypeptide secretion and insulinotropic action in glucagon-like peptide 1 receptor -/- mice. Diabetes 47:1046-1052 15. Preitner F, Ibberson M, Franklin I, et al. (2004) Gluco-incretins control insulin secretion at
multiple levels as revealed in mice lacking GLP-1 and GIP receptors. J Clin Invest 113:635- 645
16. Scrocchi LA, Brown TJ, MaClusky N, Brubaker PL, Auerbach AB, Joyner AL, Drucker DJ (1996) Glucose intolerance but normal satiety in mice with a null mutation in the glucagon-like peptide 1 receptor gene. Nature Med 2:1254-1258
17. Flatt PR, Bailey CJ (1981) Abnormal plasma glucose and insulin responses in heterozygous lean (ob/+) mice. Diabetologia 20:573-577
18. 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. J Bone Miner Res 25:1468-1486
19. Bassett JH, van der Spek A, Gogakos A, Williams GR (2012) Quantitative X-ray imaging of rodent bone by Faxitron. Methods in molecular biology 816:499-506
20. 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. J Bone Miner Res 28:2-17
21. Ritchie RO, Koester KJ, Ionova S, Yao W, Lane NE, Ager JW, 3rd (2008) Measurement of the toughness of bone: a tutorial with special reference to small animal studies. Bone 43:798-812 22. Turner CH, Burr DB (1993) Basic biomechanical measurements of bone: a tutorial. Bone
14:595-608
23. McNamara LM, Ederveen AG, Lyons CG, Price C, Schaffler MB, Weinans H, Prendergast PJ (2006) Strength of cancellous bone trabecular tissue from normal, ovariectomized and drug- treated rats over the course of ageing. Bone 39:392-400
24. Blouin S, Basle MF, Chappard D (2008) Interactions between microenvironment and cancer cells in two animal models of bone metastasis. Br J Cancer 98:809-815
25. Farlay D, Duclos ME, Gineyts E, et al. (2011) The ratio 1660/1690 cm(-1) measured by infrared microspectroscopy is not specific of enzymatic collagen cross-links in bone tissue.
PloS one 6:e28736
26. Paschalis EP, Verdelis K, Doty SB, Boskey AL, Mendelsohn R, Yamauchi M (2001) Spectroscopic characterization of collagen cross-links in bone. J Bone Miner Res 16:1821- 1828
27. Oliver WC, Pharr GM (1992) An improved technique for determining hardness and elastic modulus using load and displacement sensing indentation experiments. J Mater Res 7:1564- 1583
28. 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
29. Hansotia T, Baggio LL, Delmeire D, Hinke SA, Yamada Y, Tsukiyama K, Seino Y, Holst JJ, Schuit F, Drucker DJ (2004) Double incretin receptor knockout (DIRKO) mice reveal an essential role for the enteroinsular axis in transducing the glucoregulatory actions of DPP-IV inhibitors. Diabetes 53:1326-1335
30. Gremlich S, Porret A, Hani EH, Cherif D, Vionnet N, Froguel P, Thorens B (1995) Cloning, functional expression, and chromosomal localization of the human pancreatic islet glucose- dependent insulinotropic polypeptide receptor. Diabetes 44:1202-1208
31. Yaqub T, Tikhonova IG, Lattig J, Magnan R, Laval M, Escrieut C, Boulegue C, Hewage C, Fourmy D (2010) Identification of determinants of glucose-dependent insulinotropic polypeptide receptor that interact with N-terminal biologically active region of the natural ligand. Mol Pharmacol 77:547-558
32. Yamada C, Yamada Y, Tsukiyama K, Yamada K, Udagawa N, Takahashi N, Tanaka K, Drucker DJ, Seino Y, Inagaki N (2008) The murine glucagon-like peptide-1 receptor is essential for control of bone resorption. Endocrinology 149:574-579
33. Bouxsein ML, Myers KS, Shultz KL, Donahue LR, Rosen CJ, Beamer WG (2005) Ovariectomy-induced bone loss varies among inbred strains of mice. J Bone Miner Res 20:1085-1092
34. Judex S, Garman R, Squire M, Donahue LR, Rubin C (2004) Genetically based influences on the site-specific regulation of trabecular and cortical bone morphology. J Bone Miner Res 19:600-606
35. Pacheco-Pantoja EL, Ranganath LR, Gallagher JA, Wilson PJ, Fraser WD (2011) Receptors and effects of gut hormones in three osteoblastic cell lines. BMC Physiol 11:12
36. Nuche-Berenguer B, Portal-Nunez S, Moreno P, Gonzalez N, Acitores A, Lopez-Herradon A, Esbrit P, Valverde I, Villanueva-Penacarrillo ML (2010) Presence of a functional receptor for GLP-1 in osteoblastic cells, independent of the cAMP-linked GLP-1 receptor. J Cell Physiol 225:585-592
37. Bates HE, Campbell JE, Ussher JR, Baggio LL, Maida A, Seino Y, Drucker DJ (2012) Gipr is essential for adrenocortical steroidogenesis; however, corticosterone deficiency does not mediate the favorable metabolic phenotype of Gipr(-/-) mice. Diabetes 61:40-48
38. Bjerre Knudsen L, Madsen LW, Andersen S, et al. (2010) Glucagon-like Peptide-1 receptor agonists activate rodent thyroid C-cells causing calcitonin release and C-cell proliferation.
Endocrinology 151:1473-1486
39. Madsen LW, Knauf JA, Gotfredsen C, et al. (2012) GLP-1 receptor agonists and the thyroid:
C-cell effects in mice are mediated via the GLP-1 receptor and not associated with RET activation. Endocrinology 153:1538-1547
40. Bullock BP, Heller RS, Habener JF (1996) Tissue distribution of messenger ribonucleic acid encoding the rat glucagon-like peptide-1 receptor. Endocrinology 137:2968-2978
41. Zhou H, Yamada Y, Tsukiyama K, et al. (2005) Gastric inhibitory polypeptide modulates adiposity and fat oxidation under diminished insulin action. Biochem Biophys Res Comm 335:937-942
42. Bousson V, Peyrin F, Bergot C, Hausard M, Sautet A, Laredo JD (2004) Cortical bone in the human femoral neck: three-dimensional appearance and porosity using synchrotron radiation.
J Bone Miner Res 19:794-801
43. Cooper DM, Turinsky AL, Sensen CW, Hallgrimsson B (2003) Quantitative 3D analysis of the canal network in cortical bone by micro-computed tomography. Anat Rec Part B, New anatomist 274:169-179
44. Currey JD (2003) The many adaptations of bone. J Biomech 36:1487-1495
45. Schneider P, Stauber M, Voide R, Stampanoni M, Donahue LR, Muller R (2007) Ultrastructural properties in cortical bone vary greatly in two inbred strains of mice as assessed by synchrotron light based micro- and nano-CT. J Bone Miner Res 22:1557-1570
FIGURE LEGENDS
Figure 1: Morphometrical and metabolic parameters in wild-type (WT) and DIRKO animals. (A) Body weight and (B) fat mass are similar between WT and DIRKO mice. Plasma glucose (C) is higher in DIRKO mice in response to an IPGTT but no significant modifications of plasma insulin (D) were evidenced between the two groups of mice. *:
p<0.05 vs. WT.
Figure 2: 3D models of the femur distal metaphysis in wild-type and DIRKO mice. Trabecular VOIs have been pseudo-colored for better visualization of trabecular bone. DIRKO animals exhibited higher trabecular bone mass as compared with wild-type animals.
Figure 3: Bone mineral content at the midshaft femur determined by qXRI. (A) Cortical ROIs have been pseudo-colored for better visualization of the grey level distribution in wild- type (WT) and DIRKO animals. (B) GLmean was significantly reduced in DIRKO animals as compared with wild-type. **: p<0.01 vs. WT
TABLES
Table 1: Trabecular bone features at the distal femoral metaphysis in wild-type (n=10) and DIRKO (n=8) mice.
WT DIRKO P values
BV/TV (%) 1.02 ± 0.11 3.16 ± 0.70 0.003 Tb.Th (µm) 68.9 ± 1.9 71.1 ± 1.3 0.567 Tb.N (mm-1) 0.14 ± 0.02 0.43 ± 0.10 0.012
Tb.Sp (µm) 648 ± 22 508 ± 39 0.038
Tb.Pf 43.6 ± 3.3 35.1 ± 2.6 0.167
SMI 3.4 ± 0.1 3.2 ± 0.2 0.452
N.Oc/B.Pm (/mm) 1.3 ± 0.2 0.7 ± 0.1 0.002 Serum CTX (ng/ml) 9.3 ± 0.7 6.8 ± 0.5 0.012
N.Ob/B.Pm (/mm) 12 ± 1 15 ± 4 0.631
Serum osteocalcin (ng/ml) 15 ± 8 22 ± 10 0.487
Table 2: Cortical bone microstructural morphology and cortical strength in wild-type (n=10) and DIRKO (n=8) mice.
WT DIRKO P values
B.Dm (mm) 1.04 ± 0.01 0.95 ± 0.01 0.028 Ma.Dm (mm) 0.79 ± 0.01 0.77 ± 0.01 0.163
Ps.Pm (mm) 5.4 ± 0.1 5.1 ± 0.1 0.029
Es.Pm (mm) 4.2 ± 0.1 4.1 ± 0.1 0.343
Ct.Th (µm) 233 ± 5 200 ± 5 0.004
Ct.Ar (mm²) 0.85 ± 0.02 0.73 ± 0.02 0.012 CSMI (mm4) 0.035 ± 0.001 0.026 ± 0.001 0.012 N.Oc/Es.Le (/mm) 0.5 ± 0.2 0.5 ± 0.1 0.882 Yield stress (MPa) 0.59 ± 0.01 0.31 ± 0.02 0.032 Ultimate stress (MPa) 0.89 ± 0.05 0.50 ± 0.04 0.034 Bending modulus (GPa) 17.70 ± 2.30 18.85 ± 5.71 0.513 Post-yield work to fracture (kJ/m²) 0.48 ± 0.02 0.33 ± 0.04 0.034
Table 3: Tissue-level material properties measured at the midshaft femur in wild-type (n=10) and DIRKO (n=8) mice.
WT DIRKO P values
Maximum load (mN) 11.9 ± 0.9 9.4 ± 0.3 0.027
Hardness (MPa) 687 ± 51 490 ± 19 0.014
Indentation modulus (GPa) 13.8 ± 1.1 11.9 ± 0.3 0.221 Dissipated energy (mN.nm) 2982 ± 174 2225 ± 95 0.014 Capeak (% Ca) 25.1 ± 0.4 25.8 ± 0.8 0.522 Camean (%Ca) 24.3 ± 0.4 24.9 ± 0.8 0.522
Cawidth (%Ca) 2.7 ± 0.2 2.8 ± 0.2 0.575
Collagen maturity 3.4 ± 0.1 3.1 ± 0.1 0.011
WTDIRKO 0 20 Body weight (g)
0 10 20
WTDIRKO
Fat mass (%)
Plasma glucose AUC (mM/min)
0 200 400
0 0.4 0.8
Plasma insulin AUC (µg/l/min)
0204060 0 10 20Plasma glucose (mM)
Minutes
Plasma insulin (µg/l)
0 0.02 0.04
0204060Minutes (A)(B) (C)(D)
60 120 180
GLmean
**
(A) (B)
