Prevention of mucosally induced uveitis with a HSP60-derived peptide linked to cholera toxin B subunit

2.1 Induction of uveitis

The onset of uveitis in the immunized rats was on day 10, with a rapid increase up to day 22 after which there was only a small rise up to day 28 (Fig. 1). At this time 48/73 (65.8%) rats developed uveitis, as assessed by the combined clinical and histological criteria.

2.2 Prevention of uveitis

To prevent the development of uveitis rats were pre-treated by gavage with CTB-linked p336–351, followed by the immunizing regimen of p336–351, in parallel with rats receiving only the immunizing peptide. The development of clinical and/or histological uveitis in the tolerized rats was delayed to day 16 with a slow progressive increase up to day 28 (Fig. 1). Of the 66 tolerized rats, only 11 (16.7%) developed uveitis after treatment with CTB-linked p336–351 and this was a very significant decrease in uveitis (χ²=34.2, p<0.0001). The results of 13 experiments are presented separately for clinical and histological development of uveitis, in order to appreciate the variations between the two modes of assessment and between the experiments (Table 1). The use of a control peptide 136–150 linked to CTB and administered by gavage, followed by the immunizing p336–351, failed to affect the development of uveitis, as 14/20 (70%) of the rats showed the disease (Table 2). Thus, prevention of uveitis was specific to p336–351-CTB. Protective efficacy was 74.7%, calculated as follows: 100×(frequency of uveitis in tolerized/frequency of uveitis in immunized animals).

2.3 Adoptive transfer of lymphocytes from tolerized rats

MLN cells were prepared from tolerized rats and 3×10⁷ cells were administered into the tail vein on day 1 and 8 of starting oral immunization with p336–351. Adoptive transfer of lymphocytes from the tolerized rats showed that 2/12 (16.6%) developed uveitis (Table 3). This compared with the development of uveitis in the control tolerized rats in 1/8 (12.5%) and in 10/11 (91%) of immunized animals carried out in the same experiment.

2.4 In situ cytokine expression

In situ hybridization studies of the MLN cells for mRNA expression of Th1 and Th2 cytokines were carried out in immunized and tolerized rats (Fig. 2). A raised mRNA expression of IFN-γ was found in the MLN of immunized (10.1±4.7) compared with the tolerized rats (3.97±1.6), though the 5% level of significance was not reached, because of variation in the immunized rats. However, mRNA expression of IL-10 was significantly increased in MLN cells from tolerized (6.1±2.2), as compared with immunized (2.3±0.9) animals (p<0.05). Examination of the posterior segment of the eye showed a significant reduction in IFN-γ mRNA expressing cells in the tolerized (0.08±0.08), as compared with the immunized (1.1±0.2) rats (p<0.05). Similar results were obtained with IL-12 (0.9±0.1 and 3.3±1.0; p<0.05), but surprisingly, IL-10 showed little or no difference between the two groups of animals (Fig. 2). However, tolerized rats showed a significant increase (p<0.001) in TGF-β expressing cells (2.9±0.35) compared with the immunized rats (0.7±0.1). Thus, tolerized animals which did not develop uveitis express a predominantly Th2 type cytokine pattern in MLN and Th3 in the uveal tract, whereas immunized rats with uveitis express predominantly Th1 type of cytokines in the uveal tract and MLN.

2.5 CD4⁺CD45RC^lowRT6⁺ regulatory T cells

Preliminary examination of the proportion of CD4⁺ and CD8⁺ T cells by flow cytometry in the immunized rats with uveitis and tolerized rats without uveitis showed no significant difference (data not presented). We examined the CD4⁺CD45RC^lowRT6⁺ T cells which are regulatory Th2 type memory cells, by flow cytometry in the spleen, mesenteric lymph nodes and Peyer's patches (Fig. 3). A significant increase in the proportion of CD4⁺CD45RC^lowRT6⁺ splenic cells was found in the tolerized rats without uveitis (61.0±15.8%), compared with the immunized rats with uveitis (43.7±2.1%; t=2.56, p<0.05; Table 4). The most significant increase in the proportion of CD4⁺CD45RC^lowRT6⁺ cells was found in the mesenteric lymph node cells in the tolerized rats without uveitis (74.6±4.3%), compared with the immunized rats with uveitis (50.5%±3.6; t=4.3, p<0.005). Although a similar trend was observed in the cells eluted from Peyer's patches this failed to reach the 5% level of significance because of large variation in the proportion of eluted cells from the intestinal tissues of different animals. Similar flow cytometric analysis confirmed that cells from rats that received MLN cells from tolerized rats showed significantly higher CD4⁺CD45RC^lowRT6⁺ MLN cells (75.4±2.0%) than those of the immunized rats (63.9±2.0%; t=4.0, p=0.01; Table 4]).

4.1 Peptides, CTB, and antibodies

Peptide aa 336–351 (QPHDLGKVGEVIVTKD) was derived from the sequence of the human mitochondrial 60-kDa HSP and was prepared at the Hansen's Disease Laboratory (Centre for Disease Control, Atlanta, GA) as described 7. Recombinant cholera toxin B subunit (rCTB) was produced in a mutant strain of Vibrio cholerae in which the cholera toxin genes were deleted and this was transfected with a multicopy high expression plasmid encoding CTB 34. The rCTB was purified to apparent homogeneity from the culture filtrate by precipitation with hexametaphosphate followed by adsorption chromatography on hydroxyapatite and gel filtration chromatography through Sephacryl S-100 (Pharmacia, Uppsala, Sweden). Peptide 336–351 was covalently conjugated to CTB using N-succinimidyl (3-[2-pyridyl])-dithio) propionate (SPDP) bifunctional coupling reagent 35, according to the manufacturer's instructions (Pharmacia). Briefly, p336–351 and CTB were separately derivatized with SPDP by incubation (23°C, 30 min) at molar ratios of 1:5. The SPDP-derivatized CTB was reduced with dithiothreitol and the resulting preparation was freed ofexcess DTT and pyridine-2-thione by Sephadex G-25 chromatography. SPDP-derivatized p336–351 was mixed in a 2.5-fold molar excess with the derivatized and reduced CTB and incubated for 16 h at 23°C.The resulting peptide-CTB conjugate was isolated by gel filtration through a Sephadex G-25 column (Pharmacia) in PBS. Control peptide 136–150 (aa NPVEIRRGVMLAVDA) was similarly linked to CTB.

Anti-rat mAb used in this work were as follows: RPE-Cy5-conjugated OX-8 (anti-CD8, Serotec, Kidlington, Oxford, GB), PE conjugated OX-22 (anti-CD45RC; Serotec), cy-chrome-conjugated OX-35 (anti-CD4; PharMingen, Heidelberg, Germany) PE-conjugated OX-39 (anti-CD25; Serotec), FITC-conjugated anti-CD28 (Serotec), FITC-conjugated anti-rat IgG (PharMingen), and anti-rat RT6.1 (Serotec).

4.2 Immunization and tolerization

Male Lewis rats weighing 200–220 g were housed in groups of five and given food and water ad libitum. For the induction of uveitis by the oral mucosal route animals were given five consecutive doses on alternate days of 65 μg/ml of p336–351, dissolved in 1 ml of PBS containing type II soy bean trypsin inhibitor (10 μg/ml, Sigma, Poole, GB) by oral gavage (days 0, 2, 4, 6, 8). The optimum concentration of the peptide was determined previously 2.

To induce tolerance animals were treated with five consecutive doses on alternate days of 15 μg/ml of p336–351 conjugated to recombinant CTB and dissolved in 1 ml of 0.35 M sodium bicarbonate buffer prior to immunization with the peptide alone (days –10, –8, –6, –4, –2). Control groups were dosed with 15 μg/ml of CTB followed by either sodium bicarbonate buffer or 65 μg/ml ofp336–351. The optimum tolerizing concentration of the CTB linked p336–351 was determined by prior experiments with different doses of the CTB-peptide. Altogether 13 tolerizing experiments were carried out with the CTB linked p336–351 in parallel with immunization in groups of 4–10 rats. Tolerization was also carried out with CTB linked to a control peptide (136–150) in 4 experiments with 20 rats. Attempts were then made to prevent uveitis by tolerization at the same time as immunization or 1, 3, 5 and 7 days after starting the immunizing regime.

4.3 Clinical and histological examination

All animals were observed daily from day 8 for signs of uveitis by means of slit-lamp microscopy, without any prior knowledge of their treatment. The clinical signs of uveitis were enlarged and tortuous vessels in the iris, constricted pupils and inflammatory cells found in the anterior chamber of the eye as has been reported previously 1. The histological features of anterior uveitis were a mononuclear cell infiltration in and around the ciliary body and iris, with an occasional nodular mononuclear cell infiltrate of the iris 1. Both clinical and histological features were scored using the system described 36. Posterior synechiae between the iris and the capsule of the lens were indicative of uveitis that had cleared. Focal loss of photoreceptors, in the absence of mononuclear cells was also recorded. Animals were killed by cervical dislocation either during clinical disease or 28 days after the first immunization and their eyes were removed for histological examination or frozen in liquid nitrogen prior to sectioning for in situ hybridization. Spleens, mesenteric lymph nodes and Peyer's patches were used to examine phenotypic expression of cell surface molecules by flow cytometry.

4.4 Analysis of cell surface phenotypic expression

Cells isolated from the spleen, mesenteric lymph nodes and Peyer's patches were incubated with the appropriate FITC-, PE- or cy-chrome conjugated mAb, as described in Sect. 4.1. Flow cytometric analysis was performed on a FACScan flow cytometer (Coulter).

4.5 Detection of cytokine mRNA in mesenteric lymph nodes

Expression of IL-10 and IFN-γ mRNA was examined by in situ hybridization as described 37. Briefly, 200-μl aliquots containing 4×10⁵ MLN mononuclear cells were applied to the wells of round-bottom 96-well microtiter plates (Nunc), in triplicate. Aliquots of 10 μg/ml of p336–351 were added to the cell cultures and after 24 h the cells were washed, counted, and 1×10⁵ cells were dried onto electrically charged glass ProbeOn slides (Fisher Scientific, Pittsburgh, PA). Synthetic oligonucleotide probes (Scandinavian Gene Synthesis AB, Koping, Sweden) were labeled using ³⁵S deoxyadenosine-5′-α-(thio)-triphosphate (New England Nuclear, USA) with terminal deoxynucleotidyl transferase (Amersham, Little Chalfont, GB). A mixture of two different oligonucleotide probes was employed for the cytokine. The oligonucleotide sequences were obtained from GenBank using MacVector software. Cells were hybridized with 10⁶ cpm labeled probe/100 μl of hybridization mixture. After emulsion autoradiography, development and fixation, the coded slides were examined by dark field microscopy for positive cells defined as containing >30 grains/cell in a star-like pattern. The intracellular grains were counted by light microscopy. A negative control probe was used in parallel with the specific cytokine probes, with cells from each specimen and these did not show any staining of cells. The results were expressed as the number of labeled cells/10⁵ cells counted.

4.6 Detection of cytokine mRNA in the uveal tract

Cryostat sections (14 μm) of the whole eye were prepared and mounted onto ProbeOn slides. Mixtures of four different labeled synthetic oligonucleotide probes were employed for each of the tested cytokines and the oligonucleotide sequences were obtained from GenBank and probes were designed using MaxVector software. These were as follows: IFN γ (J00219) with complementary bases 507–556, 4682–4729, 4660–4707, 4641–4688 38; IL-12 (M65272, M38444, M65271, M38443) with bases 157–204, 544–591, 191–238, 279–326 39; IL-10 (M57627) with bases 3178, 97–144, 340–387, 376–423 40; and TGF-β (X02812) with bases 1363–1410, 1457–1504, 1766–1813, 1953–2000) 41. In situ hybridization was performedas described above to detect cytokine mRNA expression in cells of the uveal tract. The results are expressed as the number of labeled cells per tissue section using image analysis to measure the tissue section.

4.7 Adoptive transfer of regulatory cells

Cells were isolated from the mesenteric lymph nodes of rats on day 12 of starting the tolerizing regimen of CTB-linked p336–351, administered five times on alternate days. They were resuspended in saline at a concentration of 6×10⁷ cells/ml and 0.5 ml of viable cells were injected intravenously into the tail veins of rats. The cells were injected on the first day of immunization with p336–351 and on day 8. The rats were observed from day 8 daily for signs of uveitis by means of slit lamp microscopy and were killed either during clinical disease or on day 30.

4.8 Statistical method

The χ² test, Student's t-test or Mann Whitney test was used, as appropriate to analyze the results.