Adenoviral gene transfer of a single-chain IL-23 induces psoriatic arthritis-like symptoms in NOD mice

Abstract

Previously, we demonstrated that intratumoral delivery of adenoviral vector encoding single-chain (sc)IL-23 (Ad.scIL-23) was able to induce systemic antitumor immunity. Here, we examined the role of IL-23 in diabetes in nonobese diabetic mice. Intravenous delivery of Ad.scIL-23 did not accelerate the onset of hyperglycemia but instead resulted in the development of psoriatic arthritis. Ad.scIL-23-treated mice developed erythema, scales, and thickening of the skin, as well as intervertebral disc degeneration and extensive synovial hypertrophy and loss of articular cartilage in the knees. Immunological analysis revealed activation of conventional T helper type 17 cells and IL-17-producing γδ T cells along with a significant depletion and suppression of T cells in the pancreatic lymph nodes. Furthermore, treatment with anti-IL-17 antibody reduced joint and skin psoriatic arthritis pathologies. Thus, these Ad.scIL-23-treated mice represent a physiologically relevant model of psoriatic arthritis for understanding disease progression and for testing therapeutic approaches.-Flores, R. R., Carbo, L., Kim, E., Van Meter, M., De Padilla, C. M. L., Zhao, J., Colangelo, D., Yousefzadeh, M. J., Angelini, L. A., Zhang, L., Pola, E., Vo, N., Evans, C. H., Gambotto, A., Niedernhofer, L. J., Robbins, P. D. Adenoviral gene transfer of a single-chain IL-23 induces psoriatic arthritis-like symptoms in NOD mice.

Keywords: adenovirus; interleukin-17; interleukin-23; psoriasis; type 1 diabetes.

Figure 1

NOD mice infected with Ad.scIL-23 experience a significant delay in the development of diabetes and reduction of insulitis. A) Eight-week-old NOD mice were injected once with Ad.psi5 or with Ad.scIL-23 virus through tail-vein injection. Blood glucose was assessed every week. Mice were considered diabetic with 2 consecutive daily readings ≥250 mg/dl. Arrow indicates age of mice at injection with virus. B) Pancreases were fixed in 10% formalin for at least 24 h, and H&E staining was done following standard procedures on paraffin-embedded tissues, which were cut into 5-μm sections. For control mice, 188 islets were scored for insulitis and 170 islets were scored for Ad.scIL-23 mice (1-way ANOVA with Tukey’s multiple comparison test; ***P < 0.001). The results shown are from 1 experiment (control mice, n = 17; Ad.scIL-23 mice, n = 8) (A); 2 cumulative independent experiments (10 mice per experiment) are also shown (B).

Figure 2

NOD mice treated with Ad.scIL-23 exhibit signs of psoriasis. A, B) Mice infected with Ad.scIL-23 displayed differences in appearance when compared with Ad.psi5 mice, particularly 1) the ears, 2) spine, 3) tail, and 4) feet. C–F) Two days following the injection of virus, and every 2–3 d thereafter, mice were tracked for inflammation by assessing the skin for erythema (C), scales (D), and thickness (E), from which a cumulative score was calculated (F). The depths of these symptoms were scored using the following scale: 0 = none, 1 = slight, 2 = moderate, 3 = marked, 4 = very marked or severe. These mice were tracked for 4 wk. Results shown are cumulative of 2 independent experiments (control, n = 15 mice; Ad.scIL-23, n = 15 mice; experiment 1, n = 10; experiment 2, n = 5).

Figure 3

Histological analysis of the ears (A), skin (B), and tail (C) of NOD mice infected with Ad.scIL-23 reveals characteristics of psoriasis. A) Loss of cartilage (a); thickening of the dermal (b) and epidermal (c) layers; and keratosis (d) of the epidermis. B) Erythyma of the muzzle (a), paws (b), and torso (c); significant hair loss also indicated by c; and increased thickening of the dermal (e) and epidermal (f) layers. C) Epidermal retes (a, b); inflamed tendons (c, e); and hyperkeratosis (d) outside the tail. Tissue specimens were collected 4 wk following infection and fixed in 10% formalin for at least 4 h, which was followed by incubation in 30% sucrose at 4°C overnight. H&E staining was done following standard procedures on paraffin-embedded tissues, which were cut into 5-μm sections. The results shown are representative of 2 independent experiments (control, n = 15 mice; Ad.scIL-23 n = 15 mice; experiment 1, n = 10; experiment 2, n = 5).

Figure 4

Histological analysis reveals damage to the NP in the spinal discs of NOD mice treated with Ad.scIL-23. A, C) Spines were collected 4 wk following infection, decalcified, embedded in paraffin, and sagittally sliced into 5-μm sections for H&E (A) and aggregan (C) immunohistochemistry staining using standard procedures. The asterisk indicates the NP, whereas the arrow indicates the annulus fibrosus. B) NP tissues were isolated by microdissection using a dissection microscope from 5 lumbar discs of each mouse, pooled and digested with papain at 60°C for 2 h. The DNA concentration of each sample was measured using the PicoGreen assay and used to normalize the glycosaminoglycan values (control, n = 5; Ad.scIL-23, n = 5; Unpaired t test; *P < 0.05). The results shown in C are from 2 independent experiments (control, n = 5 mice; Ad.scIL-23, n = 5 mice; experiment 1, n = 3; experiment 2, n = 2).

Figure 5

Effect of Ad.scIL-23 on the knee joints of NOD mice. Decalcified sections were stained with H&E (A, B) or SOFG (C, D). Knee joints of animals receiving Ad.scIL-23 (bottom panels) showed synovial hypertrophy accompanied by neovascularization and loss of articular cartilage from both tibial and femoral surfaces. D) Red arrow indicates loss of proteoglycan and the black arrow indicates thickness of cartilage lesion. The results shown are representative of 2 independent experiments (both control and Ad.scIL-23, n = 5 mice; experiment 1, n = 2; experiment 2, n = 3).

Figure 6

Anti–IL-17 mAb treatment reduces psoriatic symptoms in Ad.scIL-23–infected mice. Eight-week-old mice infected with Ad.scIL-23 were treated with either anti–IL-17 mAb on a weekly basis for up to 3 wk, isotype-control Ab, or left alone. A) Mice were monitored for symptoms of psoriasis utilizing the PASI system (2-way ANOVA with Sidak’s multiple comparison test; *P < 0.05, **P < 0.01). B) Epidermal thickening was determined by measuring the length of the layer from top to bottom as shown in the representative image and reported for the ear, torso skin, and tail in micrometers (unpaired Student’s t test; ****P < 0.0001). C, D) Histological comparison of knee joints from mice treated with anti–IL-17 mAb or control Ab is shown. H&E staining (C) showed reduced skin thickness (short black arrows) and reduced synovial hyperplasia (thin blue arrows) in mice treated with anti–IL-17. SOFG staining (D) reveals retention of articular cartilage in mice treated with anti–IL-17. The arrows identify a large cartilagenous lesion in the knee joint of control animals (right panel) that is absent from animals treated with anti–IL-17 (left panel). The results shown in this figure are from 1 experiment. For both control and Ad.scIL-23, n = 6 mice per group for all tissues analyzed.