PTPN2 improves implant osseointegration in T2DM via inducing the dephosphorylation of ERK

Establishment of T2DM rat model

Twenty-four age-matched male Sprague Dawley rats (weight 200 ± 20 g) were provided by Beijing Vital River Laboratory Animal Technology Co., Ltd (China) and kept in Experimental Animal Center of Shandong University (Jinan, China). All experimental procedures were approved by Animal Ethics Committee of Shandong University (Jinan, China). All animal experiments were carried out in accordance with National Institutes of Health guide for the care and use of Laboratory animals (NIH Publications No. 8023, revised 1978). After one-week acclimation, all rats were randomly divided to control group (N group, n = 6) and type 2 diabetic group (n = 18). T2DM rats were induced by four-week high-fat and high-carbohydrate diet (Beijing Ke’ao Xieli Feed Co. Ltd, China), followed by intraperitoneal injection with streptozotocin (STZ, Sigma, USA) at a dose of 35 mg/kg in 0.1 M citrate-buffered saline (pH = 4.2) once. ¹⁵ The control group received normal chow and citrate buffer injection. Fasting blood glucose (FBG) ≥13.9 mmol/L one week after STZ injection was considered to be the standard for successful T2DM rat model. ¹⁶ Subsequently, diabetes rats with similar degrees of hyperglycemia were randomly divided into T2DM group (n = 6), T2DM+PTPN2 group (n = 6), and T2DM+Vehicle group (n = 6).

Implant placement

Animals were anesthetized by pentobarbital sodium (40 mg/kg body weight, Sigma, USA). Subsequently, every rat received two special designed sand-blasted and acid-etched (SLA) porous mini-Ti implants (1 mm in diameter and 10 mm in length, Wego, China) in the distal femora as previously reported. ¹⁷ A 10 mm longitudinal incision was made along the medial side of the knee joint, and the patella was dislocated laterally to expose the distal metaphysis of the femur. With the knee in flexion, a 1.0-mm diameter and 10.0-mm long hole was made from the intercondylar notch to the proximal end of femur with surgical stainless steel drills cooled under sterile saline solution. The implant was then inserted into the prepared hole via distal femoral metaphysis until the implant end was below the articular surface which was nearly parallel to the long axis of the femur. The patella was relocated and soft tissues were sutured. The animals received intramuscular injection of penicillin (800,000 IU/ml, Harbin Pharmaceutical Group Co., Ltd, China) for three postoperative days. All rats were euthanatized after three months. Five fresh femur samples were used to perform pull-out tests, while others were fixed in 4% polyformaldehyde and preserved in 70% alcohol at 4°C. Fasting blood glucose and body weight were measured every two weeks after implantation.

Local overexpression of PTPN2 in rats

2 × 10⁹ adeno-associated virus (AAV) particles constitutively expressing PTPN2 (Genechem, Shanghai, China) were injected into the knee joint capsule. The control group was injected with the vehicle. The injection was repeated every two weeks till the end of the experiment. The experimental design in vivo is shown in Figure 1.

OPEN IN VIEWER

Figure 1.

Illustration of the experimental design in vivo. SD rats: Sprague Dawley rats; STZ: streptozotocin; AAV: adeno-associated virus. (A color version of this figure is available in the online journal.)

Histological analysis

Histologic samples were dehydrated in a series of graded alcohol solutions and embedded in polymethyl methacrylate (PMMA, Merck AG, Germany). For each sample, the cross section of the implant with 100 µm thickness was prepared by hard tissue slicer (Leica, Germany). Next, the sections were stained with toluidine blue. The index of bone-implant contact (BIC) was calculated by Image-Pro Plus 6.0 as the length percentage of direct bone-implant interface to total implant circumference. ¹⁸

Microscopic computerized tomography analysis

Microscopic computerized tomography (Xradia 510 Versa, Zeiss, Germany) was used to evaluate osseointegration of the implant (60 kV, 5 w, 9.8454 µm pixel resolution, 3 s exposure time). The region of interest (ROI) was defined as the 200 µm region around the implant. Bone volume per total volume (BV/TV) and percentage of osseointegration (OI%) were calculated by Mimics software 17.0. BV represents the bone volume in the ROI and TV represents the total space volume of the ROI. OI% is the area proportion of direct bone-implant interface to total implant surface.

Pull-out test

The maximum pull-out force represents the maximum retention force obtained by the implant in the bone. Immediately after collection, five samples of each group were subjected to pull-out test by the universal material testing instrument (Shimadzu, Japan) at a constant speed of 1 mm/min until the bone-implant interface ruptured.

Scanning electron microscopy

The surface of SLA porous mini-Ti implant was observed by SEM. After three months healing, implants were pulled out to observe the adhesion and microstructure of the bone on the implant surface under a scanning electron microscope (Inspect F, FEI, Nederland) at an accelerated voltage of 1 kV.

Immunohistochemistry staining

IHC staining was performed on prepared decalcified sections to investigate the expression of PTPN2 (Abcam, ab180764, 1:150 dilution) and ALP (Abcam, ab108337, 1:300 dilution) in peri-implant osseous tissues according to the procedure. For quantitative assessment, five slices of each sample were analyzed in every group by Image J software (National Institutes of Health, Bethesda, MD, USA). The mean integrated optical density (IOD) for PTPN2 and ALP were also calculated.

Cell culture, treatment, transfection, and osteogenic induction

Rat bone marrow stem cells (RBMSCs) were isolated from rat femur bone marrow as reported previously. ¹⁹ Cells at the third to fifth passages were used for experiments. Cells were cultured in Dulbecco’s modified Eagle’s medium (DMEM, Hyclone, USA) supplemented with 10% fetal bovine serum (FBS, Hyclone, USA) at 37°C in a 5% CO₂ incubator. Cells were treated as following groups: (1) normal control group (NC group), cells were cultured in complete culture medium containing 5.5 mmol/L D-glucose; (2) high glucose group (HG group), cells were incubated with complete culture medium containing 30 mmol/L glucose; (3) high glucose with PTPN2 overexpression group (HG+PTPN2 group), cells were infected with lentivirus constitutively expressing PTPN2 and cultured in complete culture medium with 30 mmol/L glucose ²⁰ ; (4) high glucose with vehicle group (HG+Vehicle group), cells were treated with vehicle and cultured in complete culture medium with 30 mmol/L glucose. For osteogenic induction, RBMSCs were cultured in an osteogenic induction medium containing DMEM supplemented with 10% FBS, 50 mg/L ascorbic acid (Sigma, USA), 10 nM dexamethasone (Solarbio, China), and 10 mM β-glycerophosphate (Solarbio, China).

Alkaline phosphatase assay

After seven days of osteogenic induction, ALP activity was measured by Alkaline Phosphatase Assay Kit (Nanjing Jiancheng Bioengineering Institute, China) according to the manufacturer’s instructions. The OD absorbance at 520 nm wavelength was measured by microplate spectrophotometer.

Alizarin red staining

After 28 days of osteogenic induction, cells were fixed in 4% paraformaldehyde for 30 min, rinsed with ddH₂O three times, and stained with 0.1% Alizarin Red (Sigma, USA) at room temperature for 10 min. Next, the mineralized nodule was observed under light microscope (Olympus IX73, Tokyo, Japan). For quantifying the relative amount of calcium, 300 µL of 10% (w/v) cetylpyridinium chloride(Sigma-Aldrich) and 10 mm sodium phosphate (pH 7.0) solution were added to the stained dishes and the absorbance of extracted dye was determined at 562 nm.

RNA isolation and quantitative real-time PCR

After 28 days of induction, total RNA of RBMSCs with different treatments was extracted with Trizol® reagent (TaKaRa Bio Inc., Tokyo, Japan), and 1 µg RNA was transcribed with reverse transcriptase (TaKaRa Bio Inc.) according to the manufacturer’s recommendation. Quantitative Real-time PCR (qRT-PCR) was performed using 1 µL of cDNA in a 20 µL reaction volume with Roche 480 in triplicate. The relative expression level of the housekeeping gene β-Actin was used to normalize gene expression in each sample in different groups. PCR conditions were as follows: hot-start enzyme activation at 95°C for 30 s; 55 cycles of denaturation at 95°C for 5 s, annealing at 60°C for 35 s and extension at 72°C for 1 min; finally at 40°C for 30 s. The sequences of the primers for amplification of rats β-Actin, PTPN2, Runx2, Osx and Ocn were as follows: β-Actin, 5′-GAGACCTTCAACACCCCAGCC-3′ and 5′-GGCCATCTCTTGCTCGAAGTC-3′; PTPN2, 5′-ATCGAGCGGGAGTTCGA-3′ and 5′-TCTGGAAACTTGGCCACTC-3′; Runx2, 5′-AGTGTGTGTGTCCGCATGAT-3′ and 5′-CCACTTGGGGTCTAAGAACG-3′; Osx, 5′-GGAAAGGAGGCACAAAGAAGC-3′ and 5′-CCCCTTAGGCACTAGGAGC-3′; Ocn, 5′-ACCCAGAATTACCTGATCC-3′ and 5ʹ- AGACCACCCAGCACAAC-3′; The amount of mRNA was calculated for each sample based on the standard curve using the LightCycler® Software 4.0 (Roche, Basel, Switzerland).

Western blot analysis

Proteins were extracted in RIPA buffer (Solarbio, China) containing 1% PMSF (Solarbio, China) and 1% phosphatase inhibitor (Solarbio, China). The protein concentration was determined by BCA assay kit (Beyotime, China); 20 µg proteins were separated by 10% SDS-polyacrylamide gels and transferred onto polyvinylidene difluoride (PVDF) membranes. The membranes were blocked with 5% fat-free milk, and incubated overnight at 4°C with primary antibodies against PTPN2 (1:1000, Santa Cruz Biotechnology, USA), Runt-related transcription factor 2 (RUNX2, 1:1000, CST, USA), Osterix (OSX, 1:1000, CST, USA), Osteocalcin (OCN, 1:1000, CST, USA), ERK (1:1000, CST, USA), JNK (1:1000, CST, USA), p38(1:1000, CST, USA), p-ERK (1:1000, CST, USA), p-JNK (1:1000, CST, USA), p-p38 (1:1000, CST, USA), and β-Actin (1:1000, CST, USA). The membranes were washed three times with TBST and incubated with anti-mouse or anti-rabbit HRP-conjugated secondary antibodies (1:20,000) for 1 h at room temperature. The protein bands were detected by enhanced chemiluminescence under Amersham Imager 600 (Millipore, USA) and the gray values were analyzed by Image J software (National Institutes of Health, Bethesda, MD, USA).

Statistical analysis

All data are presented as the mean ± standard deviation (SD) of at least three independent experiments. Data normality was assessed by Shapiro–Wilk test. In case of normal distribution, one-way ANOVA combined with S-N-K test were used to analyze group-to-group significant differences in SPSS software (SPSS 20.0). Nonparametric tests were used when the data had no normal distribution. Differences were considered statistically significant for P value < 0.05.

Local overexpression of PTPN2 did not change fasting blood glucose level and body weight in T2DM rats

All rats survived during the whole experimental period and no infection of the implant was observed. After STZ injection, fasting blood glucose level of T2DM rats increased to over 20 mmol/L and remained at this level for three months (Table 1). Of note, there were no significant differences in fasting blood glucose level between T2DM group, T2DM+PTPN2 group, and T2DM+Vehicle group.

OPEN IN VIEWER

Table 1.

Effect of PTPN2 local overexpression on body weights and fasting blood glucose levels in experimental rats (n =6/group).

		Control	T2DM	T2DM+PTPN2	T2DM+Vehicle
Body weight (g)	0 day	274 ± 11	290 ± 10	277 ± 11	279 ± 12
	1 week	294 ± 13	285 ± 12	286 ± 12	285 ± 16
	2 weeks	313 ± 11*	289 ± 9	299 ± 9	297 ± 9
	4 weeks	343 ± 9*	274 ± 11	273 ± 13	279 ± 9
	6 weeks	376 ± 14*	264 ± 13	265 ± 13	268 ± 13
	8 weeks	398 ± 10*	254 ± 12	254 ± 12	255 ± 13
	10 weeks	431 ± 5*	245 ± 9	245 ± 12	243 ± 9
	12 weeks	462 ± 13*	237 ± 9	236 ± 11	234 ± 9
Fasting blood glucose (mmol/L)	0 day	4.5 ± 0.1	4.6 ± 0.2	4.5 ± 0.3	4.6 ± 0.2
	1 week	4.8 ± 0.1*	21.7 ± 2.1	20.4 ± 1.1	22.2 ± 2.3
	2 weeks	4.7 ± 0.1*	25.1 ± 1.2	25.3 ± 1.3	25.4 ± 1.0
	4 weeks	4.7 ± 0.2*	24.1 ± 1.4	25.4 ± 1.0	26.0 ± 0.6
	6 weeks	4.6 ± 0.1*	23.8 ± 1.1	24.1 ± 1.1	24.1 ± 1.3
	8 weeks	4.6 ± 0.2*	25.7 ± 0.8	25.2 ± 1.0	25.9 ± 1.5
	10 weeks	4.6 ± 0.2*	25.0 ± 0.9	25.7 ± 1.3	25.4 ± 1.2
	12 weeks	4.6 ± 0.2*	25.7 ± 1.5	25.4 ± 1.2	25.7 ± 1.2

Note: 0 day represented the day before STZ inducement. 1 week represented the day after STZ injection. 2–12 weeks represented the corresponding time after implantation.

Data were presented as mean  ±  SD, *P < 0.05, Control group vs. other three groups.

Measurements of body weight are shown in Table 1. The rats in T2DM group showed significant decrease in body weight compared with control group (P < 0.05), while no differences were found between T2DM group, T2DM+PTPN2 group, and T2DM+Vehicle group.

PTPN2 improved implant osseointegration in T2DM rats

Toluidine blue staining showed that the peri-implant bone mass was decreased and scattered in T2DM group. Local overexpression of PTPN2 increased bone mass and density around the implant in T2DM rats (Figure 2(a)). As shown in Figure 2(d), the BIC of T2DM group was significantly decreased compared with control group. Moreover, the contact ratio restored to almost normal level after PTPN2 local overexpression. Approximately 15% more BIC rate was found in implants with PTPN2 overexpression than in vehicle group.

OPEN IN VIEWER

Figure 2.

Evaluation of the implant osseointegration after three-month healing. (a) Histological evaluation with toluidine blue staining showed bone microarchitecture at bone-implant interface. (b) Osteogenesis of implants was analyzed by micro-computed tomography three months after surgery. (c) The red rectangle region showed the 3D reconstruction area and the blue rectangle region showed the region of interest to evaluated osseointegration. (d) Bone-implant contact ratio (BIC, %). Quantitative calculation of (e) OI% and (f) BV/TV. Data were presented as mean ± SD, n = 5 specimens/group, *P < 0.05. (A color version of this figure is available in the online journal.)

Next, Micro-CT analysis was performed to examine the peri-implant trabecular microstructures (Figure 2(b), (c), (e), and (f)). The red rectangle region in Figure 2(c) showed the 3D reconstruction area. The red circle in Figure 2(b) and the blue rectangle region in Figure 2(c) showed the region of interest to evaluated osseointegration. In the control group, the implant was almost fully covered by trabecular, but there were relatively few trabecular microstructures around the titanium implant in T2DM group. Local overexpression of PTPN2 could ameliorate the injuries of trabecular significantly. Furthermore, the OI% for T2DM and control group were 24.78 ± 0.87% and 52.17 ± 1.27%, respectively, and the OI% was significantly higher in T2DM+PTPN2 group than in T2DM+vehicle group. The qualitative analysis of BV/TV showed the same tendency among the four groups (Figure 2(f)).

The biomechanical properties were detected by pull-out test. As shown in Figure 3, the maximum forces decreased in T2DM model, but we observed significant increase of maximum forces in T2DM+PTPN2 group (45.04 ± 1.26 N) compared to T2DM+vehicle group (36.56 ± 0.81 N).

OPEN IN VIEWER

Figure 3

Results of pull-out test three months after implantation. Data were presented as mean ± SD, n = 5 specimens/group, *P < 0.05.

In addition, the microstructure of SLA porous mini-Ti implant and peri-implant bone was analyzed by SEM. The surface of implant was almost occupied by well-organized trabecular pores in control group after three-month healing. In T2DM group, the bone mass was reduced and the porosity of the trabecular bone was increased. However, T2DM+PTPN2 group showed a better bone regeneration not only in bone mass but also in bone density than T2DM+vehicle group (Figure 4). Taken together, the above data indicate that PTPN2 could improve implant osseointegration in T2DM rats.

OPEN IN VIEWER

Figure 4.

SEM images of SLA porous mini-Ti implants and peri-implant bone surfaces after three-month healing.

What’s more, immunohistochemistry staining was complete to confirm whether PTPN2 protein was actually overexpressed and detect the expression of the osteogenic marker ALP protein around the implant. Figure 5(a) and (c) indicated PTPN2 and ALP expression in the different groups. Similar to previous results, T2DM rats showed lower PTPN2 and ALP concentrations compared to normal rats. And PTPN2 was surely overexpressed in T2DM+PTPN2 group compared to that in vehicle group. Consistent with PTPN2 protein, ALP expression was obviously increased in T2DM+PTPN2 group. The value of IOD/area (Figure 5(b) and (d)) further illustrated that overexpression of PTPN2 can increase the expression of ALP around the implant in T2DM rats.

OPEN IN VIEWER

Figure 5.

Expression of PTPN2 and ALP protein in the peri-implant bone area after three-month healing. Immunohistochemical images for (a) PTPN2 and (c) ALP, the magnifications of upper and lower images are 20× and 40×, respectively. Quantitative analysis of mean integral optical density (IOD) for PTPN2 (b) and ALP (d) were also presented. Data were presented as mean ± SD, n = 5 specimens/group, *P < 0.05. (A color version of this figure is available in the online journal.)

PTPN2 promoted osteogenic differentiation of RBMSCs exposed to high glucose

To further confirm the role of PTPN2 in implant osseointegration, we established RBMSCs exposed to high glucose (HG) as the model. After 28 days of culture, we found fewer and smaller mineralized nodules in HG group compared with NC group, but cells in HG+PTPN2 group showed similar nodules in size and number compared with NC group (Figure 6). As shown in Figures 7(a) and 8(a) and (b), the expression of PTPN2 mRNA and protein was both decreased in HG group compared to NC group and increased in HG+PTPN2 group compared to HG+Vehicle group. To evaluate osteogenesis capacity of RBMSCs, we detected ALP activity, and found that high glucose significantly reduced ALP activity, while the overexpression of PTPN2 could increase ALP activity compared to vehicle group (Figure 8(i)). We also quantified the expression of some osteogenic markers after 28 days of culture by qRT-PCR and Western blot. We found that high glucose significantly reduced mRNA and protein levels of OCN, OSX, and RUNX2, while the overexpression of PTPN2 could increase the levels of these markers compared to vehicle group (Figures 7(b) to (d) and 8(c) to (e)).

OPEN IN VIEWER

Figure 6.

The overexpression of PTPN2 enhanced mineral deposition of RBMSCs after 28 days of osteogenic induction. (a) Representative photographs and low-lens pictures of Alizarin Red staining in different groups. (b) The concentration of calcium deposition was quantified by absorbance at 562 nm. Data were presented as mean ± SD, n = 5 specimens/group, *P < 0.05. (A color version of this figure is available in the online journal.)

OPEN IN VIEWER

Figure 7.

PTPN2 up-regulated the gene expression levels of Runx2, Osx, and Ocn of RBMSCs exposed to high glucose. (a) Gene expression levels of PTPN2 after 28 days of induction. (b) Gene expression levels of Runx2 after 28 days of induction. (c) Gene expression levels of Osx after 28 days of induction. (d) Gene expression levels of Ocn after 28 days of induction. Data were presented as mean ± SD, n = 5 specimens/group, *P < 0.05.

OPEN IN VIEWER

Figure 8.

PTPN2 promoted osteogenesis of RBMSCs exposed to high glucose via inducing the dephosphorylation of ERK. (a) The protein levels of PTPN2, RUNX2, OSX, OCN, p38, p-p38, JNK, p-JNK, ERK, and p-ERK were assessed by Western blot analysis after 28 days of osteogenic induction. (b–h) The quantification analysis of protein levels. (i) Early osteogenic differentiation activity was detected by ALP activity at seven days. Data were presented as mean ± SD, n = 5 specimens/group, *P < 0.05.

PTPN2 promoted the dephosphorylation of ERK in RBMSCs exposed to high glucose

To understand how osteogenic abilities of RBMSCs in high glucose were enhanced by PTPN2 overexpression, we detected the levels of ERK, p-ERK, JNK, p-JNK, p38, and p-p38. We found that the level of p-p38 increased in HG group compared to NC group but did not show significant differences between HG+PTPN2 group and HG+Vehicle group (Figure 8(a) and (f)). The level of p-JNK did not show significant differences between HG group and NC group but decreased in HG+PTPN2 group compared to HG+Vehicle group (Figure 8(a) and (j)). However, p-ERK increased in HG group compared to NC group and decreased in HG+PTPN2 group compared to HG+Vehicle group (Figure 8(a) and (h)). Collectively, these data suggest that PTPN2 selectively inhibits the activation of ERK branch of MAPK pathway in RBMSCs exposed to high glucose.