Immunomodulatory strategies prevent the development of autoimmune emphysema

The experimental protocol was approved by the Animal Care and Use Committee of the University of Colorado Denver. In our study we used 6 week old male Sprague-Dawley rats (body weight 200 g). Animals were divided into four groups (n = 6 per group): (1) control group (IP injection of phosphate buffered saline (PBS)), (2) HUVEC (H group) alone (10 × 10⁶ cells, single IP injection[14]), (3) intrathymic injection of HUVEC (H-T-H group) (200,000 cells) 2 weeks prior to immunization with HUVEC; (4) pristane (H-P group) (200 μl, single IP injection) 3 days prior to immunization with HUVEC. Pristane (2,6,10,14-tetramethyl-pentadecane) was purchased from Sigma-Aldrich Co. (St. Louis, MO). While in our previous study we performed immunization with adjuvant Titermax Gold, later we found out that HUVEC itself are strong immunogens and no adjuvant is needed for antibody production. In the present study we omitted this step of adjuvant administration.

Primary cultures of HUVEC were isolated in our laboratory following procedures described by Bruneel et al [24].

Intrathymic injections were performed under anesthesia induced with a combination of ketamine(100 mg/kg)/xylazine (15 mg/kg). The thymus gland was exposed through a small incision above the sternum and 200,000 HUVEC cells in 100 μl of PBS were injected into thymic lobes (50 μl/per lobe). The control group received only PBS injections. The incision was closed with a 4-0 nylon suture. The injections were performed into 6 week old rats (body weight 200 g) 2 weeks prior to immunization with HUVEC. Indeed, these are very young rats and prior literature [25‐27] supports intrathymic injection at 6 weeks.

All the animals were sacrificed 3 weeks after the IP HUVEC treatment or PBS injection [14]. Lung tissue was obtained by opening the chest wall. Lungs were ligated. The right lung was inflated with 0.5% low melting agarose at a constant pressure of 25 cmH₂O and fixed in 10% formalin for 48 hr. Lung sections were embedded in paraffin using standard techniques. The upper and lower lobes of the left lung were snap-frozen for protein extraction.

The 5 μm thick sections were stained with hematoxylin and eosin. Ten images per slide were acquired at x200 magnification with an AxioCam Color Camera using an Axioskop 2 Microscope (Carl Zeiss MicroImaging Inc., Thornwood, NY). The degree of emphysema was quantified by measuring alveolar airspaces (in pixels per square micrometer [14]) and mean linear intercept (MLI; [28]). The CD4+ cell accumulation areas were measured in pixels per square micrometers. A macro was created to automate the segmentation, processing, and quantification steps. For each animal, 10 fields at a magnification of x100 were captured in a blinded fashion using Carl Zeiss KS300 Imaging System software.

Terminal deoxynucleotidyl transferase-mediated dUTP nick end-labeling (TUNEL) was performed with TACS 2 TdT DAB kit (Trevigen, Gaithersburg, MD) following manufacturer's protocol. The ratio of TUNEL-positive cells to total cells (apoptotic index) was measured assessing more than 3,000 parenchymal cells of each lung sample in each group [29].

Active caspase-3 was assessed using a rabbit polyclonal antibody to cleaved caspase-3 (Cell Signaling Technology Inc., Beverly, MA). Immunohistochemical staining was performed using VECTASTAIN ABC kit (Vector Laboratories Inc., Burlingame, CA).

WB analysis was performed as previously described [14]. Anti-CD19 antibodies were from Santa Cruz Biotechnology Inc., Santa Cruz, CA, and anti-CD68 from Serotec Ltd, Oxford, UK.

Cytokine levels in whole lung tissue homogenates were determined using Quantibody Rat-1 Array (RayBiotech, Inc, Norcross, GA) that allows measurement of the following cytokines: CINC-2 (CXCL3/GROgamma/DCIP-1), CINC-3 (CXCL3/GRObeta/MIP-2), Fractalkine (CX3CL1), CNTF, GM-CSF, IL-1α, IL-4, LIX (CXCL5), β-NGF and VEGF.

Cells were isolated as described by Finotto et al., [30]. Negative selection of CD4+ T cells was performed using Rat CD4+ T Cell Enrichment Column from R&D Systems, Inc. (Minneapolis, MN). The aliquot of the sorted cell preparation (100 μl) was cytospinned, stained with Wright stain and counted. The CD4+ T cells isolated from the lungs were identified by flow cytometry. We also did the CD3 staining to confirm that the CD4+ cells were T cells and not, for example, strange macrophages.

Lung sections were stained with an anti-CD4 antibody and CD4+ cells were counted in 5 different lung fields. The CD4+ cell accumulation areas were measured in pixels per square micrometers. A macro was created to automate the segmentation, processing, and quantification steps using a Carl Zeiss KS300 Imaging System software. A CD4+ T cell count was also performed on cells isolated from the lungs. A CD4+ T cell count was also performed on cells isolated from the lungs by flow cytometry. Both methods gave comparable results.

Isolated CD4+ T cells were washed in PBS supplemented with 1% bovine serum albumin (BSA) (FACS buffer) and surface stained with mAbs directed against CD3 (FITC, Invitrogen) and CD4 (APC, Invitrogen or BD) along with one of the following mAbs to TCR V beta (8.5, 10, or 16.1 (AbD Serotec) or CD25 (PE, Biolegend) for 30 min at 4°C as described previously [31]. Cells were washed in FACS buffer and stained with secondary antibodies to each TCR V beta (R-PE, Invitrogen). For intranuclear staining with anti-Foxp3 (PE; eBiosciences), surface stained cells were washed twice, fixed and permeabilized prior to staining with anti-Foxp3. Cells were washed and resuspended in 1% formaldehyde. The number of events collected ranged between 1 and 3 million. Samples were acquired on a FACSCalibur equipped with 488 nm argon and 633 nm red-diode laser for four color detection. Analysis was performed using FlowJo (Treestar) software. Lymphocytes were gated using forward scatter versus side scatter.

All data are expressed as mean ± SEM. Statistical analysis was performed with the Statview software package (SAS Institute Inc., Cary, NC). The comparison between two groups was performed by employing the Student's unpaired t test for independent samples. The values obtained in the different groups of rats were compared using one-way ANOVA. Statistical difference was accepted at p < 0.05.

As previously reported [14], when compared to lungs from PBS-injected control animals (Figure 1A), histological examination of lung tissue samples demonstrated diffusely enlarged airspaces in HUVEC-immunized rats (Figure 1B). In contrast, there was no alveolar space enlargement in the lungs from animals treated with either intrathymic HUVEC-injection (Figure 1C) or pristane (Figure 1D) in association with intraperitoneal HUVEC injection. The quantitative data of alveolar airspace surface area and mean linear intercept (MLI) from all treatment groups are shown in Figure 1E and Figure 1F. The sections evaluated were selected in a random fashion.

Both of the immunomodulatory treatments protected lungs against cell death as shown by the reduced numbers of TUNEL positive cells. Figure 2A shows PBS-treated control rat lung. When compared to lungs from HUVEC-immunized rats (Figure 2B), intrathymic injection of HUVEC followed by HUVEC immunization (Figure 2C) and pristane treatment (Figure 2D) significantly reduced TUNEL-positive cells in the rat lungs. The quantitative analysis of TUNEL positive cells is shown as an apoptotic index (Figure 2E).

Earlier we have reported [14] that during the first week of the immunization there was no difference in the numbers of CD19 or CD68 cells in immunized or control rat lungs. However, the accumulation of B cells (CD19+; Figure 3A) was observed in rat lungs at 1.5-3 weeks after HUVEC-injection. The number of these cells in the lung tissue was attenuated by intrathymic HUVEC-injection (Figure 3B) and pristane treatment (Figure 3C). The accumulation of macrophages as shown by immunohistochemistry for CD68 in HUVEC-immunized rat lungs (Figure 3D) was reduced by intrathymic HUVEC-injection (Figure 3E) and pristane-treatment (Figure 3F).

Because TUNEL staining detects both apoptotic and necrotic cells, to identify apoptotic events in the lung tissue, we performed immunohistochemistry (IHC) for active caspase-3 (Figure 3G - HUVEC-immunized; Figure 3H - intrathymic HUVEC-injection and Figure 3I pristane-treated rat lungs). The number of active caspase-3 positive cells was markedly reduced by both of the tolerizing strategies.

We confirmed the IHC findings by performing quantitative Western blot (WB) analysis of whole lung tissue extracts. The levels of activated caspase-3, CD68 and CD19 were significantly higher in the HUVEC-immunized rats compared to the PBS-treated controls (Figure 3J). The induction of central tolerance with intrathymic HUVEC-injection and the immunomodulatory strategy using pristane attenuated the expression of caspase-3 as well as the expression of CD68 and CD19.

Usually, in normal rat lung, very few CD4+ T cells can be found (Figure 4A). Conversely, in HUVEC-injected rat lungs there is a significant accumulation of CD4+ T cells (Figure 4B). Tolerizing strategies such as thymic injection of antigen (Figure 4C) and pristane treatment (Figure 4D) reduced the CD4+ cell numbers to the normal level. We have quantified the accumulation of CD4+ cells on IHC slides using the Carl Zeiss KS300 Imaging System software. As shown in Figure 4E, there was almost a 3-fold increase in the CD4+ T cell numbers in the HUVEC-immunized rat lungs when compared to untreated controls or rats treated with thymic injection of antigen or with pristane lungs. CD4+ cells from 4 rats per group were counted.

Pristane treatment decreased numbers of CD4+ T cells in the lung tissue. Here we have tested whether pristane treatment affected T cell receptor expression in the lung by testing the expression of TCR Vβ16.1, TCR Vβ8.5 and TCR Vβ10 in the treated and untreated HUVEC-immunized rat lungs. There was a higher expression of TCR Vβ8.5 in lungs and spleens from pristane-treated and HUVEC-immunized rats when compared to normal controls. The expression of TCR Vβ10 was similar in all 3 groups. While there was no significant difference in TCR Vβ expression on CD4+ T cells obtained from lung and spleen of normal controls and HUVEC-immunized animals (n = 4 rats per group), there was a significantly decreased expression of Vβ16.1 in both lungs (p ≤ 0.008) and spleens (p ≤ 0.003) of pristane treated rats compared to rats receiving only HUVEC (Table 1). Similar findings were seen with the intrathymic injection of HUVEC prior to IP immunization in lungs and spleen (9.03 ± 0.94% and 8.83 ± 0.47% respectively) when compared to HUVEC-immunized rats (11.5 ± 0.88% and 12.1 ± 0.4%; p ≤ 0.03 and p ≤ 0.007). Since HUVEC-immunized rats have 3-times more of such CD4+ T cells in the lungs, these findings might suggest that pristane and intrathymic injection of HUVEC targeted a specific CD4+ T cell subset and raise the possibility that CD4+ T cells expressing TCR Vβ16.1 may represent the pathogenic T cell population.

Recent findings indicate that targeted deletion of Treg cells causes spontaneous autoimmune disease in mice, whereas augmentation of Treg-cell function can prevent the development of - or alleviate - variants of experimental autoimmune encephalomyelitis [32]. As shown in Table 2 levels of Foxp3+/CD25+/CD4+ Treg cells in the normal control rat lungs and spleen are extremely low. However, intrathymic injection of antigen prior to IP immunization resulted in a significantly increased percentage of Foxp3-expressing CD4+CD25+ Treg cells when compared with HUVEC-immunized rats, suggesting that the induction of central tolerance results in an expansion of naturally-occurring Treg cells (n = 4 rats per group).

To examine whether induction of immune deviation by pristane can affect lung cytokine levels in lung tissue homogenates, we used Quantibody Rat-1 Array that allows measurements of the following cytokines: cytokine induced neutrophil chemoattractant-2 (CINC-2 also known as CXCL3/GROgamma/DCIP-1); cytokine induced neutrophil chemoattractant-3 (CINC-3 also known as CXCL3/GRObeta/MIP-2); Granulocyte-Macrophage Colony Stimulating Factor (GM-CSF); lipopolysaccharide-induced CXC chemokine (LIX also known as CXCL5); interleukin-4 (IL-4); interleukin 1alpha (IL-1α); ciliary neurotrophic factor (CNTF); Fractalkine (CX3CL1), nerve growth factor beta (β-NGF) and vascular endothelial growth factor (VEGF). HUVEC-immunization resulted in a greater than 40-fold increase in IL-1α, and a decrease in vascular endothelial growth factor (VEGF) and CNTF, levels (Figure 5A), whereas pretreatment with pristane preserved the normal VEGF and CNTF levels in the lung and significantly decreased IL-1α, levels in the HUVEC-immunized animals.

Earlier we have demonstrated [14], that serum from HUVEC-immunized animals contains antibodies that bind to several HUVEC proteins. Induction of central tolerance achieved by intrathymic injection of HUVEC prior to IP immunization or immunomodulatory therapy with pristane abolished antibody production in the treated rats (Figure 5B). Pre-immune serum did not cross-react with HUVEC proteins. Thus, these findings suggest that the central and peripheral immunomodulatory strategies resulted in a reduction in lung inflammation characterized by a reduced recruitment of macrophages, CD4+ T cells and B cells to the rat lung in response to intraperitoneal HUVEC injection.