A total of 60 female SD rats (7 weeks old, weighting 160±10 g) were obtained from Animal Centre of the Chinese Academy of Medical Sciences. Animals were maintained in a temperature (25.0±0.2˚C) and humidity (45%±5%) controlled specific pathogen free laboratory, under a 12-light/dark cycle, fed with standard rodent pellets and allowed free access to filtered water. The animals were acclimatized for three days. All experimental procedures were approved by the Ethics Committee of the Institute of Medicinal Plant Development, Chinese Academy of Medical Sciences and Peking Union Medical College (approval no. SLXD-20170705016).

DHT (cat. no. C10033864), Fe-Dex (cat. no. 50807A-1), PMX (cat. no. 1405-20-5) and CFA (cat. no. 1001646446) were obtained from MilliporeSigma. A Tissue Membrane Protein Extraction Kit (cat. no. KGP3100-2) was obtained from Nanjing KeyGen Biotech Co., Ltd. The ELISA kits used for the detection of rat anti-histidyl tRNA synthetase (Jo-1) antibody (cat. no. QS49908), anti-melanoma differentiation-associated protein 5 (MDA5) antibody (cat. no. QS49938), tumor necrosis factor-α (TNF-α; cat. no. QS41721), and interferon-γ (IFN-γ; cat. no. QS44104) were purchased from Beijing Qisong Biotechnology Co., Ltd. Among all the rats, 12 rats were used as donors for allogeneic plasma membrane antigen preparation.

The plasma membrane antigen preparation was modified from a method described previously (17) under aseptic conditions. Briefly, membrane proteins were obtained from 12 wild-type female SD rats. The rats were euthanized under anesthesia by intraperitoneal injection of sodium pentobarbital (50 mg/kg). After the blood was taken from abdominal aorta, normal saline (20 ml/rat) was perfused to eliminate remaining blood. Death was confirmed when the spontaneous breathing ceased. The trapezius, gastrocnemius, and their nearby subcutaneous connective tissue were then collected. The tissue samples were rinsed with pre-cooled phosphate buffer saline, and adipose tissue was removed. The samples were then cut to small pieces (0.2x0.2x0.2 mm) and homogenized in lysis buffer by sonication(power 500W, pulses of 15 sec on 10 sec off, with 30-40 cycles). The membrane protein was extracted from the samples using a membrane protein extraction kit, according to the manufacturer's instructions. The extracted samples were quantified using a BCA assay and adjusted to the concentration of 10.0 mg/ml. Lastly, the membrane protein (10 mg/ml) was emulsified with an equal volume of CFA.

The rats were randomly divided into 6 groups (n=8/group), including a control group and five immunized groups. Rats in all five immunized groups received 10 mg/kg membrane antigen subcutaneously once weekly from d₀₁ to d₂₈ for sensitization immunization. A total of 14 days after the last sensitization immunization, these rats received membrane antigen challenge immunization with subcutaneous administration once weekly from d₄₂ to d₆₃. Rats in the control group were subcutaneously administered normal saline (1.00 ml/kg). At d₇₇, four groups of antigen-immunized rats received 10 mg/kg DHT intragastrically, 10 mg/kg Fe-Dex intraperitoneally and 20 mg/kg PMX administration, or a combination of these three toxins (DHT10 mg/kg + 10 mg/kg Fe-Dex + 20 mg/kg PMX) to mimic calciphylaxis, respectively. At d₈₄, the severity of DM manifestations in each rat was scored according to the assessment method described in Lennon et al (18). At d₉₁, rats were euthanized by exsanguination under anesthesia with intraperitoneal injection of sodium pentobarbital (50 mg/kg), and were immobilized in the supine position with spontaneous breathing, the bloods were collected, and skin and skeletal muscles were collected from the limbs for further analysis (Fig. 1).

DM manifestation in each animal was scored as described by assessment of muscle weakness as described by Lennon et al (18), to mimic that used climactically for patients. The scoring criteria are shown in Table I.

The skin and skeletal muscle tissue samples were fixed in PBS-buffered 10% formalin. Following paraffin embedding, the tissue sections (6 µm) were stained with hematoxylin and eosin (H&E) or von Kossa using standard protocols. Briefly, the HE staining procedures were as follows: Slides were deparaffinized using xylene, hydrated using a gradient of alcohols solutions (100, 95 and 80%) followed by distilled water. Secondly, the slides were stained with hematoxylin solution for 5 min, rinsed with tap water, differentiated using 10% acetic acid in 95% alcohol for 1 min and rinsed with tap water. Finally, the slides were stained with eosin Y solution for 2~3 min, rinsed with 80% alcohol, dehydrated using a gradient of alcohols solutions (95, 100%), and then cleared with xylene. All these steps were performed at room temperature. Images were acquired using the Digital Pathology system (KFBIO), and pathological changes were quantified using Image Pro-Plus 7.0.1 (Media Cybernetics).

To quantify T lymphocyte infiltration, the lymphocyte infiltration areas (TA) and observed areas (OA) were measured under x20 magnification in skeletal muscle slides. The volume ratio of infiltrated T lymphocytes (ITL)was calculated using the formula: [(TA/OA)^3/2x100%] (14).

To quantify calcified nodules, the skeletal muscle tissue sections were stained at room temperature using the von Kossa method standard protocols. Briefly, after deparaffinizing and dehydrating, the slide was treated with 5% silver nitrate solution for 30 min, exposed to incandescent light for 60 min, and then treated with 5% sodium thiosulfate for 1 min. Finally, the slides were washed and stained with eosin for 10 min to made the plasma exhibit thin red. Calcified nodules area (CA) in arteriole and arteriole areas (AA) were analyzed under x40 magnification. The volume ratio of calcified nodules in arteriole was calculated using the formula: [(CA/AA)^3/2x100%] (15).

At d₁₄, one rat was randomly selected from the antigen-immunized rats, and the blood was obtained from its tail vein (1 ml) (19). The blood sample was centrifuged at 3,000 x g for 15 min at 4˚C, and the serum was collected and stored at -80˚C until analysis. The specific antibodies against plasma membrane antigen were detected using the agar immunodiffusion method. Briefly, an agar plate was prepared, and holes (diameter, 3 mm) were drilled into the agar ~5 mm apart. The prepared allogeneic plasma membrane antigen from the aforementioned donor animals or serum was added to the antigen well or sample well, respectively, and the plates were incubated at 37˚C for 72 h. A sample well was filled with normal saline as a negative control. Lastly, the plates were stained with Coomassie brilliant blue to display the immunocomplexes. Images were acquired using a digital camera (Canon, Inc.).

The serum levels of antibodies (anti-Jo-1 and anti-MDA5 antibodies) or cytokines (TNF-α and IFN-γ) from all rats were measured by ELISA using the double-antigen (‘antibody sandwich’) method according to the manufacturer's protocols (11). The concentration of pre-coated rat Jo-1 or MDA5 antigen was 1.5 or 2.5 µg/ml, respectively. The concentration of pre-coated anti-TNF-α or anti-IFN-γ antibody was both 2.0 µg/ml. The optical density at 450 nm was measured using a microplate reader (M1000; Tecan Group, Ltd.), and the concentrations were calculated by converting the optical density using a standard curve.

The data are presented as means ± SD. The differences between groups were analyzed with SPSS 16.0 software (SPSS, Inc.) with non-parametrical test using a Kruskal-Wallis' test followed by a Dunn's post-hoc test. Correlation analysis between two data was performed using Spearman's correlation analysis. P<0.05 was considered to indicate a statistically significant difference.

The clinical score in DM rats showed that 2 weeks after the antigen immunization challenge, rats that received membrane antigen immunization without the toxins presented local cutaneous swelling and enlarged lymph nodes caused by the injected skin drainage. Following toxin administration to induce calciphylaxis (DHT, Fe-Dex, PMX or all three combined), rats exhibited marked weakness, ear hyperemia, and limb edema, especially in the PMX and DHT + Fe-Dex + PMX groups (Fig. 2A). The clinical scores in the PMT and DHT + Fe-Dex + PMX groups were higher than those in the antigen-immunized, DHT, and Fe-Dex groups (Fig. 2B). These results demonstrated that the lesions in rats that received with PTM and DHT + Fe-Dex + PMX were most similar to the clinical manifestations seen in patients.

To evaluate autoimmune injury, the pathological changes in the skin and skeletal muscles of DM model rats were observed using HE staining. In the skin tissue sections, rats in all five autoimmune groups showed dyskeratotic keratinocytes, basement membrane thickening, lymphocytes and few neutrophils infiltrated in the dermis. These changes were obvious in Fe-Dex and DHT + Fe-Dex + PMX groups (Fig. 3).

In the skeletal muscle tissue, compared with the control group, all the autoimmune groups showed edema and necrotic of muscle fibers, perifascicular atrophy, as well as perifascicular and perivascular lymphocytes infiltration (Fig. 4A). The morphometry results showed that the volume ratio of ITL was significantly increased in all the autoimmune groups. Moreover, the ITL volume ratio in the DHT + Fe-Dex + PMX group was higher than those in the antigen-immunized (P<0.05), DHT (P<0.01) and Fe-Dex (P<0.05) groups (Fig. 4B). This indicated that antigen immunization followed by DHT + Fe-Dex + PMX administration may represent the suitable model for DM.

Calciphylaxis was analyzed following calcified nodule staining using the von Kossa method. The calcified nodules stained as black deposits, which were most apparent in the DHT + Fe-Dex + PMX group (Fig. 4C). The severity of cutaneous calciphylaxis was measured as the volume ratio of calcified nodules in arterioles. Compared with the control, the calciphylaxis ratio in the PMX (P<0.05) and DHT + Fe-Dex + PMX groups was significantly increased (P<0.01). The rats in the DHT + Fe-Dex + PMX group had the highest ratio of calcified nodules in arterioles (Fig. 4D).

Autoimmune injury was demonstrated to have occurred based on the production of special antigens. Positivity for antibodies against plasma membrane antigens, as well as anti-Jo-1 antibody, anti-MDA5 antibody, TNF-α and IFN-γ were used to evaluate autoimmune injury.

In the agar immunodiffusion assay, of one randomly selected rats from the antigen-immunized animals, positive precipitate line was observed, which indicated production of specific antigens against membrane antigens (Fig. 5A).

Anti-MDA5 and anti-Jo-1 antibodies are important biomarkers for DM diagnosis (20). The levels of anti-Jo-1 antibody in the PMX and DHT + Fe-Dex + PMX groups increased significantly compared with those in the control (P<0.01). Among the four calciphylaxis groups, anti-Jo-1 antibody levels in DHT + Fe-Dex + PMX group was greater than those in the antigen-immunized (P<0.05), DHT (P<0.05), and Fe-Dex (P<0.05) groups (Fig. 5B). The serum anti-MDA5 antibody levels in all five autoimmune groups were increased significantly compared with those of the control (Fig. 5C).

As markers for the severity and prognosis of DM (21), the levels of IFN-γ and TNF-α were measured. IFN-γ is the signature cytokine of T helper 1 cells, and its levels increased in patients with DM in parallel with the progression of muscle weakness (22). In the present study, IFN-γ levels increased significantly in the PMX (P<0.05) and DHT + Fe-Dex + PMX (P<0.05) groups compared with the control group. Among all autoimmune groups, IFN-γ levels in the DHT + Fe-Dex + PMX group were the highest (Fig. 5D). TNF-α directly regulates muscle metabolism and regeneration (23). The levels of TNF-α significantly increased in the PMX (P<0.05) and DHT + Fe-Dex + PMX (P<0.01) groups compared with the control group (Fig. 5E).

Correlation analysis was performed (Table II). The results indicated that the clinical scores correlated with indicators of autoimmune injury (IFN-γ, ITL, TNF-α, anti-Jo-1 and anti-MDA5 antibody) and cutaneous calciphylaxis. Moreover, cutaneous calciphylaxis was also correlated with indicators of autoimmune injuries (IFN-γ, ITL, TNF-α, and anti-Jo-1 antibody).