Introduction

Type 2 diabetes (T2D) is a metabolic disease, with an increasing worldwide prevalence, characterized by chronic hyperglycemia and adverse effects on multiple organ systems, including bones [1]. Patients with T2D have an increased fracture risk, particularly at the hip, compared to individuals without diabetes. A recent meta-analysis reported that individuals with T2D have 1.27 relative risk (RR) of hip fracture compared to non-diabetic controls [2]. Fragility fractures in patients with T2D occur at normal or even higher bone mineral density compared to healthy subjects, implying compromised bone quality in diabetes. T2D is associated with a reduced bone turnover [3], as shown by lower serum levels of biochemical markers of bone formation, such as procollagen type1 amino-terminal propeptide (P1NP) and osteocalcin, and bone resorption, C-terminal cross-linked telopeptide (CTX) in diabetic patients compared to nondiabetic individuals [4–7]. Accordingly, dynamic bone histomorphometry of T2D postmenopausal women showed a lower bone formation rate, mineralizing surface, osteoid surface, and osteoblast surface[8]. Our group recently demonstrated that T2D is also associated with increased SOST and decreased RUNX2 genes expression, compared to non-diabetic subjects[9]. Moreover, we have proved in a diabetic model that a sclerostin-resistant Lrp5 mutation, associated with high bone mass, fully protected bone mass and strength even after prolonged hyperglycemia [10]. Sclerostin is a potent inhibitor of the canonical Wnt signaling pathway, a key pathway that regulates bone homeostasis [11].

Diabetes and chronic hyperglycemia are also characterized by increased advanced glycation end-products (AGEs) production and deposition [12]. AGEs may interfere with osteoblast differentiation, attachment to the bone matrix, function, and survival [13,14]. AGEs also alter bone collagen structure and reduce the intrinsic toughness of bone, thereby affecting bone material properties [9,15,16]. In this work,we hypothesized that T2D and AGEs accumulation downregulate Wnt canonical signaling and negatively affect bone strength.. Results confirmed that T2D downregulates Wnt beta/catenin signaling and reduces collagen mRNA levels and bone strength, in association with AGEs accumulation.

Materials and Methods

Study subjects

We enrolled a total of 36 postmenopausal women (15 with T2D and 21 non-diabetic controls) undergoing hip arthroplasty for osteoarthritis, consecutively screened to participate in this study between 2020 and 2022. Diabetes status was confirmed by the treating diabetes physician. Participants were diagnosed with diabetes when they had fasting plasma glucose (FPG) ≥126 mg/dl or 2-h plasma glucose (2-h PG) ≥200 mg/dl during a 75-g oral glucose tolerance test (OGTT); or hemoglobin A1c (HbA1c) ≥6.5% in accordance with the American Diabetes Association diagnostic criteria. Eligible participants were ≥60 years of age. Exclusion criteria were any diseases affecting bone or malignancy. Additionally, individuals treated with medications affecting bone metabolism such as estrogen, raloxifene, tamoxifen, bisphosphonates, teriparatide, denosumab, thiazolidinediones, glucocorticoids, anabolic steroids, and phenytoin, and those with hypercalcemia or hypocalcemia, hepatic or renal disorder, hypercortisolism, current alcohol or tobacco use were excluded. The study was approved by the Ethics Committee of the Campus Bio-Medico University of Rome and all participants provided written informed consent. All procedures were conducted in accordance with the Declaration of Helsinki.

Specimen preparation

Femoral head specimens were obtained during hip arthroplasty. As described previously [9], trabecular bone specimens were collected fresh and washed multiple times in sterile PBS until the supernatant was clear of blood. Bone samples were stored at – 80 °C until analysis.

Bone histomorphometry

Trabecular bone from femur heads werefixed in 10% neutral buffered formalin for 24 h prior to storage in 70% ethanol. Tissues were embedded in methylmethacrylate and sectioned sagittally by the Washington University Musculoskeletal Histology and Morphometry Core. Sections were stained with Goldner’s trichrome. Then, a rectangular region of interest containing trabecular bone was chosen below the cartilage-lined joint surface and primary spongiosa. This region had an average dimension of 45 mm². Tissue processing artifacts, such as folding and edges, were excluded from the ROI. A threshold was chosen using the Bioquant Osteo software to automatically select trabeculae and measure bone volume. Finally, Osteoid was highlighted in the software and quantified semi-automatically using a threshold and correcting with the brush tool. Unstained and TRAP-stained (Sigma) slides were imaged at ×20 high resolution using a NanoZoomer 2.0 with bright field and FITC/TRITC (Hamamatsu Photonics). Images were then analyzed via Bioquant Osteo software according to the manufacturer’s instructions and published standards (v18.2.6, Bioquant Image Analysis Corp., Nashville, TN).

Bone compression tests

We used cylindrical bone specimens of trabecular core (with a diameter of 10 mm and a length of 20 mm) from 10 T2D and 21 non-diabetic subjects to measure bone mechanical parameters (Young’s modulus, ultimate strength and yield strength), as previously described [9].

RNA extraction and gene expression by RT-PCR

Total RNA from trabecular bone samples was extracted using TRIzol (Invitrogen) following the manufacturer’s instructions. The concentration and purity of the extracted RNA were assessed spectrophotometrically (TECAN, InfiniteM200PRO), and only samples with 260/280 absorbance ratio between 1.8 and 2 were used for reverse transcription using High-Capacity cDNA Reverse Transcription Kit (Applied Biosystems, Carlsbad, CA) according to the manufacturer’s recommendations. Transcription products were amplified using TaqMan real-time PCR (Applied Biosystems, Carlsbad, CA) and a standard protocol (95°C for 10 minutes; 40 cycles of 95°C for 15 seconds and 60°C for 1 minute; followed by 95°C for 15 seconds, 60°C for 15 seconds, and 95°C for 15 seconds). β-Actin expression was used as an internal control (housekeeping gene). Relative expression levels of Sclerostin (SOST), Dickkopf-1 (DKK-1), Wnt ligands (WNT5A and WNT10B), T-cell factor/ lymphoid enhancer factor 1 (LEF-1), collagen type I alpha 1 chain (COL1A1), glycogen synthase kinase 3 beta (GSK3B), axis inhibition protein 2 (AXIN2), beta-catenin (BETA-CATENIN) and secreted frizzled-related protein 5 (SFRP5) were calculated using the 2^-ΔCt method.

Statistical analysis

Data were analyzed using GraphPad Prism 9.0 (GraphPad Software, San Diego, CA). Patients’ characteristics were described using means and standard deviations or medians and 25th-75th percentiles, as appropriate, and percentages. Group data are presented in boxplots with median and interquartile range; whiskers represent maximum and minimum values. We assessed data for normality and Mann-Whitney test was used to compare variables between groups. Data were analyzed using nonparametric Spearman correlation analysis and the correlation coefficients (r) were used to assess the relationship between variables. We used Grubbs’ Test to assess and exclude outliers.

Results

Subject characteristics

Clinical characteristics of study subjects are presented in Table 1. T2D and non-diabetic subjects did not differ in age, BMI and menopausal age. As expected, fasting glucose was significantly higher in T2D compared to non-diabetic subjects [112.00 mg/dl (104.0-130.0 ) mg/dl, vs. 94.00 (87.2-106.3), respectively; p=0.009]. Median HbA1c was determined in all T2D subjects within three months before surgery [(6.95% (6.37-7.37)]. Median disease duration in T2D subjects was [14.50 years (7.25-19.25)]. Diabetes medications included monotherapy with metformin (n=12) and combination therapy with metformin plus insulin and glinide (n=3). There were no differences in serum calcium, eGFR (CKD-EPI equation) and serum blood urea nitrogen.

Table 1.

Bone Histomorphometry

Bone samples of 8 T2D and 9 non-diabetic subjects were used for histomorphometry analysis. We found no significant differences in BV/TV and osteoid volume, while mineralized volume/total volume (Md.V/TV) trended lower in T2D subjects relative to controls [0.249% (0.156-0.336) vs 0.352% (0.269-0.454); p=0.053] (Table 2).

Table 2.

Bone compression tests

Young’s modulus was lower in T2D compared to non-diabetic subjects [21.6 MPa (13.46-30.10) vs. 76.24 MPa (26.81-132.9); p=0.0025), while ultimate strength and yield strength were not different between the two groups (Table 3).

Table 3.

Gene Expression

SOST mRNA was significantly higher in T2D than in non-diabetic subjects (Fig. 1A, p<0.0001), whereas there was no difference in DKK1 gene expression between the two groups (Fig. 1B). Of note, SOST mRNA transcript was very low in the majority of non-diabetic subjects (Fig. 1A). LEF-1 (Fig. 1C, p=0.0136), WNT10B (Fig. 1D, p=0.0302) and COL1A1 (Fig.1F, p=0.0482) mRNA transcripts were significantly lower in T2D compared to non-diabetic subjects. Conversely, WNT5A was higher in T2D relative to non-diabetics (Fig. 1E, p=0.0232). Moreover, GSK3B was significantly increased in T2D compared to non-diabetic subjects (Fig.1G, p=0.0456), but we did not find any significant difference in gene expression of AXIN2, BETA-CATENIN and SFRP5 (Fig. 1H-J) between our groups.

Figure 1.

Correlation analysis of Wnt target genes, AGEs and glycemic control

As shown in figure 2, AGEs were inversely correlated with LEF-1 (Fig. 2A, p=0.0255) and COL1A1 mRNA abundance (Fig. 2B, p=0.0004), whereas they were positively correlated with SOST (Fig. 2C, p<0.0001) and WNT5A mRNA (Fig. 2D, p=0.0322). There was no correlation between AGEs content and WNT10B (Fig. 2E; p=0.1938) or DKK1 gene expression (Fig. 2F; p=0.9349). Likewise, we did not find any significant correlation between LEF-1, WNT5A, WNT10B, DKK-1, COL1A1 expression in bone and glycemic control in T2D individuals (Supplemental Figure 1A-D). However, there were positive correlations between SOST and fasting glucose levels (Fig. 3A, p=0.0043), SOST and disease duration (Fig. 3B, p=0.00174), WNT5A, GSK3B and fasting glucose levels (Fig. 3C, p=0.0037; Fig. 3D, p=0.0051).

Figure 2.

Figure 3.

Correlation analysis of Wnt target genes and bone mechanical parameters

As shown in Figure 4, Young’s Modulus was negatively correlated with SOST (Fig.4A, p=0.0011), AXIN2 (Fig. 4D, p=0.0042) and SFRP5 (Fig 4F, p=0.0437), while positively correlated with LEF -1 (Fig. 4B, p=0.0295) and WNT10B (Fig. 4C, p=0.0001). Ultimate strength was associated with WNT10B (Fig. 4F, p=0.0054) and negatively correlated with AXIN2 (Fig. 4G, p=0.0472). Finally, yield strength was associated with LEF-1 (Fig. 4H, p=0.0495) and WNT10B (Fig. 4I, p=0.0020) and negatively correlated with GSK3B (Fig. 4J, p=0.0245), AXIN2 (Fig. 4K, p=0.0319), and SFRP5 (Fig. 4L, p=0.0422). Non-significant correlations are reported in Supplementary figure 2 (A-Q).

Figure 4.

Discussion

We show that key components of the Wnt/beta-catenin signaling are abnormally expressed in the bone of postmenopausal women with T2D and they are associated with AGEs and reduced bone strength. LEF-1, a transcription factor that mediates responses to Wnt signal and Wnt target genes itself, and WNT10B, an endogenous regulator of Wnt/β-catenin signaling and skeletal progenitor cell fate, are both downregulated in bone of postmenopausal women with T2D. Consistently, in this group, the expression of the Wnt inhibitor, SOST is increased, suggesting suppression of Wnt/β-catenin signaling. Interestingly, our data suggest that sclerostin expression is very low in healthy postmenopausal women not affected by osteoporosis. Moreover, we reported an increase in the expression level of bone GSK3B, in line with downregulated Wnt/β-catenin signaling in T2D. Our data also show that the expression of WNT5A, a non-canonical ligand linked to inhibition of Wnt/beta-catenin signaling was increased, whereas COL1A1 was decreased. These findings are consistent with reduced bone formation and suppression of Wnt signaling in T2D. We have previously reported upregulation of SOST and downregulation of RUNX2 mRNA in another cohort of postmenopausal women with T2D [9]. Of note, the cohort of T2D subjects studied here had glycated hemoglobin within therapeutic targets, implying that the changes in gene transcription we identified persist in T2D bone despite good glycemic control.

High circulating sclerostin has been reported in diabetes [17,18], and increased sclerostin is associated with fragility fractures [19]. Aside from confirming higher SOST expression, we also show that other Wnt/β-catenin osteogenic ligands are abnormally regulated in the bone of T2D postmenopausal women. WNT10B is a positive regulator of bone mass; transgenic overexpression in mice results in increased bone mass and strength [20], whereas genetic ablation of WNT10B is characterized by reduced bone mass [21,22], and decreased number and function of osteoblasts [21]. More to the point, WNT10B expression is reduced in the bone of diabetic mice [23]. Therefore, the reduced WNT10B in human bone we found in the present study further supports the hypothesis of reduced bone formation in T2D. Accordingly, LEF-1 gene expression was also downregulated confirming that Wnt/beta-catenin pathway is decreased in T2D. Importantly, the overexpression of LEF-1 induces the expression of osteoblast differentiation genes (osteocalcin and COL1A1) in differentiating osteoblasts [24]. In fact, in this study we also demonstrated that a downregulation of LEF-1 in T2D bone goes along with a downregulation of COL1A1, strengthen data of a reduced production of bone matrix most likely as the result of reduced osteoblasts synthetic activity in diabetes [8,25]. Reduced RUNX2 in T2D postmenopausal women also confirms previous findings [9] and further supports the notion of reduced osteoblast differentiation or function in diabetes. On the other hand, the contribution of upregulated WNT5A in diabetic bone is more complex. WNT5A regulates Wnt/beta-catenin signaling depending on the receptor availability [26]. Non-canonical WNT5A activates β-catenin-independent signaling, including the Wnt/Ca⁺⁺[27] and planar cell polarity pathways [28].

Heterozygous Wnt5a null mice have low bone mass with impaired osteoblast and osteoclast differentiation [29]. Wnt5a inhibits Wnt3a protein by downregulating beta/catenin-induced reporter gene expression [26]. In line with these findings, we showed that there was an increased gene expression of WNT5A in bone of T2D postmenopausal women, confirming a downregulated Wnt/beta-catenin signaling and impaired osteoblasts function. Moreover, GSK3B is a widely expressed serine/threonine kinase involved in multiple pathways regulating immune cell activation and glucose metabolism. Preclinical studies reported that GSK3B is a negative regulator of Wnt/beta-catenin signaling and bone metabolism [30,31], and its increase is associated with T2D and alterations in insulin secretion and sensitivity [32,33]. Our data, confirmed that GSK3B is increased in T2D postmenopausal women and it is associated with reduced yield strength. In fact, we also showed an impaired bone mechanical plasticity in T2D, in line with other studies showing a reduced bone strength [16,34,35]. In addition, this study reported significant correlations of bone mechanical parameters and Wnt target genes, which might reflect the biological effect of downregulated Wnt signaling and AGEs accumulation on bone mechanical properties in diabetes.

We have previously shown that AGEs content is higher in T2D bone compared to non-diabetic bone, even in patients with well-controlled T2D [9]. Here we show that AGEs accumulation is positively correlated with SOST, WNT5A and GSK3B gene expression, and negatively correlated with LEF-1, WNT10B, and COL1A1 mRNA. These findings are consistent with the hypothesis that AGEs accumulation is associated with impaired Wnt signaling and low bone turnover in T2D. We did not find any abnormalities in histomorphometric parameters in our subjects with T2D, consistent with our previous report [9]. Reduced osteoid thickness and osteoblast number were reported in premenopausal T2D women with poor glycemic control compared to non-diabetic subjects but not in the group with good glycemic control [36]. Therefore, good glycemic control appears to prevent or rescue any changes in static histologic parameters of bone turnover that might be caused by uncontrolled diabetes.

Our study has some limitations. One is the cross-sectional design; another one is the relatively small number of T2D subjects enrolled. Moreover, we measured the mRNA abundance of the genes of interest, and we cannot assume that the differences we found reflect differences in protein abundance. Although osteoarthritis may affect some of the genes we studied [37], all study subjects were affected by variable degree of osteoarthritis, and the effect of such potential confounder is not likely to be different between T2D and control subjects. Finally, we did not use the tetracycline double-labeled technique to investigate dynamic bone parameters.

The main strength of our study is that this study is the first to explore the association of AGEs on Wnt pathway in postmenopausal T2D women. Moreover, we measured the expression of several Wnt genes directly on bone samples of postmenopausal T2D women.

In conclusion, our data show that, despite good glycemic control, T2D decreases expression of COL1A1 and Wnt genes that regulate bone turnover, in association with increased AGEs content and reduced bone strength. These results may help understand the mechanisms underlying bone fragility in T2D.

Data Availability

All data are available upon request to the corresponding authors

Disclosures

All authors have nothing to disclose relevant to this work.

Acknowledgements

This work was supported by an internal Grant of Campus Bio-Medico University of Rome.

Author Contributions

Conceptualization, N.N, R.P., M.M and R.S.; methodology, G.L., F.C., M.F., V.V., F.T., N.P., A.B., S.T..; software, G.L., C.P., A.B., and E.S.; validation, N.N. and G.L.; formal analysis, N.N. and G.L.; investigation, N.N., R.S., M.M. and R.P.; resources, N.N, G.V., R.P. M.M., and R.S.; data curation, G.L. and N.P..; writing—original draft preparation, G.L., M.F., N.P., R.C. and N.N.; writing—review and editing, G.L., N.N., C.P., N.P., F.Z., E.S., R.S. and R.C.; visualization, N.N. and R.S..; supervision, N.N., R.P., and M.M.; project administration, N.N. and G.L.; funding acquisition, N.N. and R.S. All authors have read and agreed to the published version of the manuscript.

Legend of supplementary figures

Supplementary figure 1.

Supplementary figure 2.