Pathogenic effect of interleukin-17A in induction of Sjögren's syndrome-like disease using adenovirus-mediated gene transfer

SS non-susceptible C57BL/6J mice were bred and maintained under specific pathogen-free conditions. The animals were maintained on a 12-hr light-dark schedule and provided food and acidified water ad libitum. At times indicated in the text, mice were euthanized by cervical dislocation following deep anesthetization with isoflurane, after which organs were freshly explanted for analyses. Both the breeding and use of these animals for the present studies were approved by the University of Florida's IACUC and IBC. Salivary glands of mice were cannulated with mouse IL-17A-expressing Ad5-IL17A vector using retrograde injections at either 7 weeks (wks) of age (n = 11) or 16 wks of age (n = 8). In addition, mice at 6 wks (n = 4) and 15 wks (n = 4) were randomly selected and used as pre-treated or baseline analysis. Age- and sex-matched control C57BL/6J mice (n = 10 per age group) received the Ad5-LacZ control vector using the same protocols.

The recombinant adenovirus vectors used in this study were generously provided by Dr. Jay K. Kolls (Children's Hospital of Pittsburgh, Pittsburgh, PA, USA). These vectors are based on the first generation adenovirus serotype 5 (Ad5) and shown to produce their appropriate and functional mouse IL- 17A and LacZ products [22‐24]. To obtain sufficient viral vectors for the present studies, each recombinant vector was amplified in HEK293 cells, purified by two rounds of CsCl gradient centrifugation, then dialyzed against 100 mM Tris-HCl (pH 7.4), 10 mM MgCl₂ and 10% (v/v) glycerol, as described elsewhere [25].

Previous studies have demonstrated that retrograde salivary gland cannulation is an effective method to direct local gene expression in the salivary glands [26‐28]. In brief, prior to cannulation, each mouse was anesthetized with a ketamine:xylazine mixture (100 mg/mL, 1 mL/kg body weight; Fort Dodge Animal Health, Fort Dodge, IA, USA) and xylazine (20 mg/mL, 0.7 mL/kg body weight; Phoenix Scientific, St. Joseph, MO, USA)) intramuscularly. Stretched PE-10 polyethylene tubes were inserted into each of the two openings of the salivary ducts. After securing the cannulas, the mouse received an intramuscular injection of atropine (1 mg/kg), followed 10 minutes later by a slow, steady injection of viral vector. Each salivary gland received 50 μl of vector solution containing 10⁷ viral particles). The cannulas were removed five minutes later to ensure successful cannulation.

To measure stimulated saliva flow, individual non-anesthetized mice were weighed and given an intraperitoneal injection of 100 μl of phosphate-buffered saline (PBS) containing isoproterenol (0.02 mg/ml) and pilocarpine (0.05 mg/ml). Saliva was collected for 10 minutes from the oral cavity of individual mice using a micropipette starting 1 minute after injection of the secretagogue. The volume of each saliva sample was measured. Prior to vector cannulation and again at each time-point designated in the text, saliva and sera were collected from each mouse. Samples were stored at -80°C until analyzed.

Measurements of IL-6 and IL-17A cytokine levels in sera samples were performed by an independent contractor (Millipore, Billerica, MA, USA) using Luminex^® platform.

Spleens were freshly explanted, gently minced through stainless steel sieves, suspended in PBS and centrifuged (1,200 rpm for five minutes). Erythrocytes were lysed by seven-minute incubation in 0.84% NH4Cl. The resulting leukocyte suspensions were washed two times in PBS, counted and resuspended inculture media (RPMI 1640 medium, 10% FBS, 2 mM L-glutamine, 0.05 mM β-mercaptoethanol) at a density of 2 × 10⁶ cells/ml. One million cells were pipetted to individual wells of a 24-well microtiter plate pre-coated with anti-CD3 (10 μg/ml) and anti-CD28 antibodies (2 μg/ml) for T cell activation. Cells were incubated for five hours with Leukocyte Activation Cocktail containing GolgiPlug (2 μl/ml). Collected cells were fixed and permeabilized using Cytofix/CytopermFixation/Permeabilization. Flow cytometric acquisition for intracellular staining was performed following staining with PE-Cy5-conjugated anti-mouse CD4, FITC-conjugated anti-IFN-γ and PE-conjugated anti-IL-17AA. The cells were counted on a FACSCalibur (BD, Franklin Lakes, NJ, USA) and analyzed by FCS Express (De Novo Software, Los Angeles, CA, USA).

Following euthanasia, whole salivary glands containing submandibular, sublingual, and parotid glands were surgically removed from each mouse and placed in 10% phosphate-buffered formalin for 24 hrs. Fixed tissues were embedded in paraffin and sectioned at 5 μm thickness. Paraffin-embedded sections were de-paraffinized by immersing in xylene, followed by dehydrating in ethanol. The tissue sections were prepared and stained with hematoxylin and eosin (H&E) dye. Stained sections were observed under a microscope for glandular structure and leukocyte infiltration determination. A double-blinded procedure was used to enumerate leukocytic infiltrations (lymphocytic foci) in the histological sections of salivary glands. Lymphocytic foci (LF) were defined as aggregates of >50 leukocytes quantified per each histological section. Calculations were based on one histological section per mouse.

Histological sections of salivary glands were incubated with rat anti-mouse B220 (BD Pharmingen, San Jose, CA, USA) and goat anti-mouse CD3 (Santa Cruz Biotechnology, Santa Cruz, CA, USA), followed by incubation with Texas Red-conjugated rabbit anti-rat IgG (Biomeda, Foster City, CA, USA) and FITC-conjugated rabbit anti-goat IgG (Sigma-Aldrich, St. Louis, MO, USA). The slides were mounted with DAPI-mounting medium (Vector Laboratories, Burlingame, CA, USA). Sections were observed at 200X magnification using a Zeiss Axiovert 200 M microscope.and images were obtained with AxioVs40 software (Ver. 4.7.1.0, Zeiss) (Carl Zeiss, Thornwood, NY, USA). Enumeration of B, T cells and total number of nuclei in the LF were performed using Mayachitra imago software (Mayachitra, Inc, Santa Barbara, CA, USA).

Immunohistochemical staining for IL17A were carried out as previously described [8]. In brief, paraffin-embedded salivary glands were deparaffinized by immersion in xylene, followed by antigen retrieval with 10 mM citrate buffer, pH 6.0. Tissue sections were incubated overnight at 4°C with anti-IL-17A antibody (Santa Cruz Biotechnology). Isotype controls were done with rabbit IgG. The slides were incubated with biotinylated goat anti-rabbit IgG followed by horseradish peroxidase-conjugated strepavidin incubation using the Vectastain ABC kit. The staining was developed by using diaminobenzidine substrate (Vector Laboratories), and counterstaining was performed with hematoxylin. Sections were observed at 200X magnification using a Zeiss Axiovert 200 M microscope. And images were obtained with AxioVs40 software (Ver. 4.7.1.0, Zeiss) (Carl Zeiss). Enumeration of IL17A-positive cells was performed on the entire histological sections of the whole salivary glands using Mayachitra imago software (Mayachitra, Inc.), although lymphocytic infiltrations are normally seen only in the submandibular glands.

ANA in the sera of mice were detected using HEp-2 ANA kit (INOVA Diagnostics, Inc., San Diego, CA, USA). All procedures were performed per manufacturer's instructions. In brief, HEp-2 fixed substrate slides were overlaid with appropriate mouse sera diluted 1:40, 1:80 and 1:160. Slides were incubated for one hour at room temperature in a humidified chamber. After three washes for five minutes with PBS, the substrate slides were covered with Alexa 488-conjugated goat anti-mouse IgG (H/L) (Invitrogen Inc, Carlsbad, CA, USA) diluted 1:100 for 45 minutes at room temperature. After three washes, fluorescence was detected by fluorescence microscopy at 200X magnification using a Zeiss Axiovert 200 M microscope and all images were obtained with AxioVs40 software with constant exposure of 0.3 seconds (Carl Zeiss). Negative controls are secondary antibody only and positive controls are standard serum with nuclear speckled pattern provided with the kits. Data presented in the results are from slides using 1:40 dilutions of sera from each experimental group.

Adenoviral vectors have been reported to show peak gene expressions around Day 5 post-infusion and then persist for approximately two weeks [29]. In the current study, immunohistochemical staining for the presence of LacZ protein in the infused salivary glands demonstrated that optimal transduction efficiency was approximately 26 ± 5% at two weeks post-infusion which decreased to 15 ± 3% by nine weeks post-infusion. The cells within the salivary glands positive for LacZ expression were predominantly ductal cells, as expected, and acinar cells (data not shown), indicating the virus was capable of passing through the ducts.

To determine if transduction of salivary glands with IL-17A alters the serum cytokine profiles, serum preparations were assessed for temporal changes in pro-inflammatory cytokine levels. Sera of treated mice were collected at Days 5 and 12 post-treatment to determine the efficacy of the IL-17A expressing viral vectors to affect cytokine secretions. As shown in Figure 1, C57BL/6J mice treated with the Ad5-IL17A vector at 10⁷ viral particles per salivary gland exhibited a marked increase in the levels of serum IL-17A compared to baseline levels or with C57BL/6J mice receiving the control Ad5-LacZ vector at 10⁷ viral particles per salivary gland, demonstrating the efficacy of this viral vector to produce IL-17A. In addition, Ad5-IL17A-treated C57BL/6J mice also secreted elevated amounts of the IL-17A-related cytokine IL-6 following cannulation. Thus, the vectors gain access into the glands and apparently secrete IL-17A in quantities that elevate systemic levels.

As mentioned previously, salivary glands were cannulated with Ad5-IL17A vector at either 7 wks or 16 wks of age. The time points chosen are based on extensive studies of the development and onset of disease in our C57BL/6.NOD-Aec1Aec2 mouse model of SS [1‐3, 30, 31]. The two time points selected represent the innate and adaptive immune response phases, respectively, in the disease model, thus they were chosen to mimic these changes in the parental C57BL/6 mouse. Microarray analyses examined the temporal differential gene expression of salivary and lacrimal glands of C57BL/6 mice revealed gradual change in pathophysiological related genes from 16 to 20 wks of age, concomitantly, leukocyte infiltration in the exocrine glands is often observed at these ages [32, 33]. Thus, it is important to examine the role of IL17A in the development of SS prior and post to any pathophysiological changes.

Mice treated with Ad5-IL17A or Ad5-LacZ at either 7 wks or 16 wks of age were euthanized at 26 and 27 wks of age, that is, 19 wks and 11 wks post-treatment, respectively. Splenocytes were isolated from individual mice and examined for the number of IFN-γ and IL-17A secreting CD4+T cells. Representative data, presented in Figure 2b, c, revealed that the number of IL-17A secreting CD4+T cells in the spleens of mice receiving the Ad5-IL17A vector at seven weeks of age was approximately two-fold higher than mice receiving the control Ad5-LacZ vector, while the number of IFN-γ secreting CD4+T cells was approximately half at time of analysis. Similarly, the number of IL-17A secreting CD4+T cells in the spleens of mice receiving the Ad5-IL17A vector at 16 wks of age was approximately seven-fold higher than mice receiving the control Ad5-LacZ vector, while the number of IFN-γ secreting CD4+T cells was less than 50% at time of analysis (Figure 2e, f). Results of a similar analysis with untreated mice performed one week prior to vector cannulations are presented in Figures 2a, d. These data suggest that even though the Ad5 vector is considered locally restricted, the effect in C57BL/6 J mice appeared systematic. More importantly, the systemic effects of IL17A in Ad5 appears to be correlated with the duration of gene expression after vector cannulation as evidenced by the two-fold increase in the levels of IL-17A secreting cells at 19 wks post-treatment in younger mice but a seven-fold increase at 11 wks post-treatment in the older group. However, one cannot rule out the possibility that different efficacies are achieved based on the status of disease development in different ages of mice.

Lymphocyte infiltration of the salivary and/or lacrimal glands is a critical criterion for identification of the autoimmune phase of SS in both human and animal models. Although the number of LF present in the salivary and lacrimal glands does not often correlate directly with disease or its severity, SS patients and NOD-derived mouse strains exhibiting SS-like disease typically have lymphocytic infiltrates in their salivary glands. IL-17A appears to play a critical role in the development of LF and has recently been found to be present within LF in both SS patients and animal models [8]. Salivary glands of C57BL/6J mice following cannulation with Ad5-IL17A vector were examined for the presence of infiltrating leukocytes. Salivary glands retrieved from C57BL/6J mice treated with Ad5-LacZ vector at either 7 or 16 wks of age revealed that 10% (1 of 10) in each group had evidence of glandular infiltrations (Figure 3a, b, g, h; Table 1). This observation is consistent with the number of healthy, untreated C57BL/6J mice expected to have infiltration of the salivary glands [8]. In contrast, salivary glands from C57BL/6J mice treated with Ad5-IL17A vector at seven weeks of age showed infiltrations in 91% (10 of 11) with the mean LF per histological section numbering 4 ± 1.32, while salivary glands from C57BL/6J mice treated with Ad5-IL17A vector at 16 wks of age revealed infiltrations in 75% (6 of 8) with a mean LF number per histological section of 2 ± 0.83 (Table 1).

Besides the number of LF detected in the salivary glands of the experimental animals, immunofluorescent staining to detect B and T cells revealed further differences in the cellular composition of the infiltrations between mice administered Ad5-IL17A at an early or late stage. At time of euthanasia, C57BL/6J mice treated with Ad5-IL17A vector at 7 wks of age generally exhibited smaller foci containing fewer IL-17 positive cells compared to mice receiving the vector at 16 wks of age (Figure 3c-f, i-l). Consistent with previous observation, the smaller foci in mice treated at 7 wks of age may have resulted from the longer duration of time after cannulation (19 wks) reflecting the decreases in IL-17A serum levels and IL-17A- positive cell numbers. Detailed examination of IL-17A-positive cells revealed that a majority of IL-17A cells are present in the LF and ductal cells with smaller percentage of positive cells found in the epithelium and acinar cells. Nevertheless, these data support the concept that formation and maintenance of LF are due, in part, to the expression levels of IL17A in the salivary glands.

With the appearance of B and T lymphocytes within the salivary glands of Ad5-IL17A treated C57BL/6 mice, plus the significant changes within their splenic T_H17 and T_H1 cell populations, the presence of circulating autoantibodies, specifically ANA, detectable by staining of HEp-2 cells was examined. To identify the presence of ANA, the sera prepared from blood samples collected from each C57BL/6J mouse both pre- and post-cannulation were tested for reactivity on HEp-2 cells. As presented in Figure 4a, the sera collected from C57BL/6J mice at six weeks of age or one week prior to vector treatment showed a general weakly diffused cytoplasmic and nuclear background staining of the individual target cells. However, sera collected 19 wks post-treatment from mice treated with Ad5-IL17A vector at 7 wks of age showed no cytoplasmic staining with course speckled staining and negative nucleoli, while Ad5-LacZ treated mice exhibited diffused cytoplasmic staining, weak but fine speckled nucleoplasmic staining with negative nucleoli (Figures 4b, c). Similar results were seen in C7BL/6J mice whose salivary glands were transduced with Ad5-IL17A vector at 16 wks of age in which the pattern was pronounced course speckled staining with no cytoplasmic staining and negative nucleoli at 29 wks of age, or 11 wks post-treatment (Figures 4d-f). Considering the functions of IL-17A, it is interesting to see a gradual and subtle change in ANA profile from diffused cytoplasmic/nuclear pattern to a distinct course nuclear speckled pattern, suggesting influence of IL-17A on the B cells repertoire.

To determine if the expression of exogenous IL-17A can induce salivary gland dysfunction, saliva volumes for each mouse were measured at one week prior to treatment, then at three- to five-week intervals post-cannulation. C57BL/6J mice that received control Ad5-LacZ vector at seven weeks of age exhibited stable stimulated saliva volumes at seven weeks post treatment with a statistically non-significant increase in saliva volumes at 11 weeks post treatment. Nevertheless, C57BL/6J mice whose salivary glands were cannulated at seven weeks of age with Ad5-IL17A exhibited a significant and relatively rapid decrease in stimulated saliva volumes that was most pronounced at seven weeks post treatment, and this observation is seen even if the saliva volumes are converted to saliva flow rates based on weights of the mice. After seven weeks post treatment, these mice showed a slight recovery (Figure 5a). Similar results were observed with C57BL/6J mice cannulated at 16 wks of age with Ad5-LacZ and Ad5-IL17A vectors; however, no saliva volume recovery was observed at time of euthanization (that is, 11 wks post-treatment) (Figure 5b). Whether a reversal of this inhibition would occur in these older animals will require further studies. Thus, saliva secretions of mice receiving the Ad5-IL17A vector were significantly decreased one to two months post-treatment when compared to secretions of mice receiving the Ad5-LacZ vector.