Conjunctiva-associated lymphoid tissue in avian ...

Developmental and Comparative Immunology 36 (2012) 289–297

Contents lists available at ScienceDirect

Developmental and Comparative Immunology journal homepage: www.elsevier.com/locate/dci

Conjunctiva-associated lymphoid tissue in avian mucosal immunity F.W. van Ginkel a,⇑, S.L. Gulley a, A. Lammers b, F.J. Hoerr a, R. Gurjar a, H. Toro a a

Department of Pathobiology, College of Veterinary Medicine, Auburn University, Auburn, AL 36849, USA Adaptation Physiology Group, Department of Animal Sciences, Wageningen Institute of Animal Sciences, Wageningen University, Marijkeweg 40, PG 6709 Wageningen, The Netherlands

b

a r t i c l e

i n f o

Article history: Received 28 February 2011 Revised 26 April 2011 Accepted 27 April 2011 Available online 27 May 2011 Keywords: Conjunctiva-associated lymphoid tissue Paraocular immunity Avian immunity Mucosal immunity

a b s t r a c t Conjunctiva-associated lymphoid tissue’s (CALT) role in generating avian mucosal adaptive immunity was measured by analyzing cellular composition, expression of the polymeric immunoglobulin receptor (pIgR), and production of cytokines and antibodies in chickens ocular exposed to a replication-deficient adenovirus of serotype 5 (Ad5). These studies demonstrate that CALT contains B cells, cd T cells, T helper, and cytotoxic T cells, and a T lymphocyte composition, which more resembles Harderian glands than spleen. CALT-derived lymphocytes contain antigen-specific, IgA-secreting plasma cells and cytokineproducing lymphocytes after ocular Ad5 vaccination. The expression of the pIgR in the CALT’s lymphoepithelium emphasizes the importance of mucosal immune protection by paraocular lymphoid tissues. The CALT immune response after ocular Ad5 boosting was influenced by prior high dose in ovo Ad5 priming. Thus, both mucosal and systemic immunization influenced Ad5-induced IFN-c responses in CALT. Ó 2011 Elsevier Ltd. All rights reserved.

1. Introduction Understanding immune responses of the head-associated lymphoid tissues (HALTs) in avian species and particularly in domestic poultry is essential, because ocular, spray, aerosol, and may be even drinking water-delivered vaccines, as well as pathogens such as infectious bronchitis virus and avian influenza, etc., target mucosal surfaces of paraocular and upper respiratory tract tissues (Suarez and Schultz-Cherry, 2000; van Ginkel et al., 2008). The HALT includes major inductive sites such as the Harderian glands and conjunctiva-associated lymphoid tissues (CALT) as well as lymphoid follicles distributed throughout the mucosal surfaces (Maslak and Reynolds, 1995). Numerous publications are available on the role of the Harderian glands in generating specific immune responses (Albini et al., 1974;Ameiss et al., 2006; Bang and Bang, 1968; Bienenstock et al., 1973; Burns, 1976; Mansikka et al., 1989; van Ginkel et al., 2009); however, limited information is available on the role of CALT in generating specific immunity after ocular antigen exposure. We hypothesized, that CALT significantly contribute to mucosal immune protection of paraocular tissues by generating antigen-specific immune responses.

Abbreviations: CALT, conjunctiva-associated lymphoid tissue; HALT, headassociated lymphoid tissue. ⇑ Corresponding author. Address: Department of Pathobiology, College of Veterinary Medicine, Auburn University, 217 Scott Ritchey Research Center, Auburn, AL 36849, USA. Tel.: +1 334 844 0132; fax: +1 334 844 2652. E-mail address: [email protected] (F.W. van Ginkel). 0145-305X/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.dci.2011.04.012

The CALT is located within conjunctival folds and fissures of the lower eyelid, and together with the Harderian glands form the main paraocular-associated lymphoid structures. The chicken CALT, first described in 1991, can be readily detected within the first week post-hatch (Fix and Arp, 1991a). The early lymphoid structure contains lymphocytes, lymphoblasts, and macrophages. Two week-old chickens developed germinal centers in CALT, and by 4 weeks of age plasma cells are detected. The epithelium associated with these follicles contains M-like cells and intraepithelial lymphocytes, but lacks goblet cells (Fix and Arp, 1989, 1991a), resembling the follicle-associated epithelium of mucosal inductive sites in mammals. The turkey CALT epithelium also shows a close association of M-like cells with the underlying lymphoid tissue (Fix and Arp, 1989). M-cells are found on follicle-associated epithelium and are specialized in uptake and transcytosis of antigens and microorganisms (Hathaway and Kraehenbuhl, 2000). High endothelial venules were observed at the base of the lymphoid nodules containing lymphocytes but were not observed in non-lymphoid areas of the lower eyelid (Fix and Arp, 1989, 1991a). The CALT in the lower and upper eyelid appears similar in structure. The upper eyelid nodules are much smaller, and are located near the lacrimal duct. In adult birds CALT displayed morphologic characteristics normally observed in the mucosal immune system and, therefore, may have a role in mucosal immunity in chickens (Fix and Arp, 1991a). Carbon or iron oxide particle uptake in CALT in the lower eyelid of chickens was significantly higher compared with non-lymphoidepithelium of the lower eyelid. Uptake of particles by CALT

290

F.W. van Ginkel et al. / Developmental and Comparative Immunology 36 (2012) 289–297

increased significantly in chickens between 3 and 5 weeks of age and between 5 and 15 min of contact. Based on these observations it was concluded that CALT in chickens displays an age-dependent maturity, which may affect mucosal immunity (Fix and Arp, 1991c). Uptake of particles by CALT in turkeys resulted in similar findings (Fix and Arp, 1991b). Changes in lymphocyte subpopulations in chicken CALT were studied between 1 and 8 weeks after hatching. B-lymphocytes were observed within the germinal centers of CALT in 4-weekold birds. CALT T-lymphocytes surrounded the B-cell-rich germinal centers. CD3+ T-lymphocytes were the most prevalent lymphocytes in all age groups examined. Increasing concentrations of CD3+, CD4+, and CD8+ T-lymphocytes were observed in the CALT as chicks aged from 1 to 4 weeks. After 4 weeks these cell populations stabilized (Maslak and Reynolds, 1995), indicating that maturation of CALT takes 4 weeks. Ocular exposure of 1-day-old turkeys to Bordetella avium induced large germinal centers in CALT, indicating an active immune response in this tissue (Fix and Arp, 1989). We have previously shown, that in ovo and mucosal vaccination with an Ad5 vector expressing the H5 gene of avian influenza virus (Ad5–H5) protects chickens against highly pathogenic avian influenza challenge (Toro et al., 2007, 2008, 2011). Furthermore, we have demonstrated that mucosal delivery of Ad5–H5 elicits both mucosal and systemic immune responses (van Ginkel et al., 2009). Mucosal immunization confers protection even in chickens with low levels of plasma H5-specific antibody levels, indicating the importance of cell mediated immunity and/or local immune responses to control pathogens (Toro et al., in 2011). To understand the importance of CALT in paraocular immunity we characterized its cellular composition and structure, and antigen-specific immune responses after ocular and in ovo immunization. 2. Materials and methods 2.1. Chickens Specific pathogen free (SPF) white leghorn chicken eggs (Sunrise Farms, Catskill, NY) were incubated and hatched. Chickens were maintained in biosafety level 2 facilities in Horsfall-type isolation units throughout the experiment. All chickens were 4– 9 weeks old when ocularly immunized with Ad5. This age assured proper development of CALT. Experimental procedures and animal care were performed in compliance with federal and institutional animal care and use guidelines, and Auburn University College of Veterinary Medicine is an AAALAC-accredited institution. 2.2. Adenovirus vector vaccination The replication defective human Ad5 vector (Toro et al., 2007) (Vaxin Inc., Birmingham, AL) was used to track induction of specific immunity in the CALT. Chickens were inoculated via the ocular route with a volume of 80–100 ll containing 2.0–2.5 108 infectious units (ifu) of Ad5 per bird. Chickens were boosted with the same dose via the ocular route 14 days after priming. In ovo Ad5 vaccination was performed as previously described (van Ginkel et al., 2009). The effects of in ovo priming on day 18 of egg incubation on the mucosal immune responses to ocular Ad5 boosted chickens were evaluated. 2.3. Antibody measurements Antibody levels were measured in plasma and lacrimal fluids on days 7 and 10 after ocular Ad5 boosting. Blood samples were collected by brachial vein puncture in EDTA tubes. Plasma was

obtained by centrifugation for 1 min at 4500g. Lacrimal fluids were obtained as previously described (Toro et al., 1993). Plasma and tears were combined with a 10 protease inhibitor cocktail containing 4-(2-aminoethyl)benzenesulfonyl fluoride hydrochloride, aprotinin, bestatin hydrochloride, N-(trans-epoxysuccinyl)-L-leucine 4-guanidinobutylamide, EDTA, and leupeptin (Sigma, Saint Louis, MO) prior to short-term storage at 4 °C or for long-term storage at 80 °C. 2.4. Chicken B cell enzyme-linked immunospot (ELISPOT) assay CALT lymphocytes were collected from chickens on days 8, 9, and 11 after the third immunization to measure the number of IgA and IgG antibody secreting cells specific for Ad5 using an ELISPOT assay as previously described (van Ginkel et al., 2009). In brief, CALTs were mechanically disrupted and lymphocytes were isolated by centrifugation over a 1.077 g/ml Histopaque-ficoll density gradient. Lymphocytes were loaded at 5 106 lymphocytes/ml and/or 10-fold dilutions thereof onto nitrocellulose backed, 96-well microplates coated with heat-killed Ad5 virus (108 particles/well) and blocked with complete RPMI-1640 medium containing 10% fetal calf serum (FCS). The cells were incubated for 18 h at 41 °C. The plates were washed 5 with PBS-Tween 20 (0.05%) and incubated overnight at 4 °C with goat–anti-chicken IgA conjugated to biotin (Gallus Immunotechnology Inc., Fergus, ON, Canada) followed by 5 washes and 30 min incubation with neutralite–avidin–HRP (Southern Biotechnology Associates, Birmingham, AL). The plates were washed and incubated at room temperature for 15–30 min with peroxidase substrate (Moss Inc., Pasadena, MD) prior to stopping the reaction by washing the plates with water. 2.5. Immunohistochemistry Tissues were fixed in chilled acetate–alcohol (5% acetic acid and 95% ethanol) for 2 h at 4 °C and subsequently incubated in 30% sucrose at 4 °C. Tissues were submerged in Neg-50 embedding medium (Richard-Allen Scientific, Kalamazoo, MI) in an embedding mold (Fisher Scientific Inc., Suwanee, GA) and frozen in 2-methylbutane cooled with liquid nitrogen. Cryostat tissue sections (5 lm) were placed on SuperfrostÒ Plus microscope slides (Labsco, Inc., Louisville, KY) and allowed to dry. Lipids were removed with icecold acetone and slides were air-dried. The 5 lm cryostat sections were blocked with 10% fetal calf serum (FCS) and incubated overnight at 4 °C with rabbit–anti-chicken pIgR affinity purified antibody, diluted (1:200) in PBS with 10% FCS. This antibody was generated by immunization of a rabbit with a synthetic peptide that was fused to Keyhole Limpet Hemocyanin (KLH). The sequence of the synthetic peptide was GLS NRV SLD ISE GP, which is located in the first immunoglobulin-like domain of the pIgR (Wieland et al., 2004). Characterization of the antibody by ELISA and Western blotting showed that the affinity-purified antibody specifically binds recombinant SC (secretory component) as well as SC in bile and fecal extracts. The rabbit’s pre-immunization serum was used as the negative control. Sections were then incubated for 4 h at room temperature with donkey–anti-rabbit IgG conjugated to fluorescein isothiocyanate (FITC) (1:2000) (Novus Biologicals, LLC, Littleton, CO). Between steps slides were extensively washed with PBS and were mounted using Fluormount G (Southern Biotechnology Associates Inc., Birmingham, AL). All images were captured with a Retiga 2000R camera using a Nikon Eclipse E800 fluorescence microscope using a 60 objective and NIS Elements BR 3.0 software. IgA- and IgG-positive B cells in CALT were detected by immunohistochemistry as previously described (van Ginkel et al., 2009). In brief, tissues were fixed, embed, and frozen as outlined above. Five micrometer sections were placed on SuperfrostÒ Plus microscope


slides (Labsco, Inc., Louisville, KY) and allowed to dry. Lipids were removed with ice-cold acetone and slides were air-dried. Sections were either stained with hematoxylin and eosin for histological analysis, or blocked with 10% FCS and incubated overnight at 4 °C with 1:2000 diluted FITC-conjugated goat–anti-chicken IgA (Novus Biologicals, LLC) or biotinylated mouse–anti-chicken IgG antibodies, followed by 1:200 dilution of phycoerythrin (PE)-conjugated neutralite avidin (Southern Biotechnology Associates Inc., Birmingham, AL) for immunofluorescence. All images were captured with a Retiga 2000R camera using a Nikon Eclipse E800 fluorescence microscope using a 60 objective and NIS Elements BR 3.0 software. 2.6. Isolation of CALT lymphocytes and flow cytometry CALTs were dissected from the lower eye-lid and the exposed lymphoid conjunctiva-associated tissue was removed (Fig. 1). CALT and Harderian gland derived lymphocytes were isolated as previously reported for Harderian glands and are routinely over 90% pure (van Ginkel et al., 2008). The FACS analyses were gated for chicken lymphocytes. CALT lymphocytes were stained with 1.0– 0.5 lg of the following monoclonal antibodies for FACS analysis: mouse–anti-chicken IgM-biotin, mouse–anti-chicken CD4-biotin, mouse–anti-chicken CD3-PE, mouse–anti-chicken CD8a-FITC, and mouse–anti-chicken CD8b-FITC (Southern Biotechnology Associ-

291

ates Inc., Birmingham, AL), FITC-conjugated goat–anti-chicken IgA (Gallus Immunotech Inc., Fergus, Canada), mouse–anti-chicken IgM-biotinylated (Southern Biotechnology Associates Inc.), TCR1PE (chicken cd T cell receptor), TCR2-FITC (Vb1 ab T cell receptor), TCR3 (Vb2 ab T cell receptor) (Southern Biotechnology Associates, Inc., Birmingham, AL) in PBS, 1% bovine serum albumin (BSA), 0.01% sodium azide at 4 °C. Cells were washed and those that were stained with biotinylated antibodies were subsequently incubated with neutralite avidin-PE or streptavidin-Alexa 660 diluted 1:2000. Washed cells were fixed in 1% paraformaldehyde in PBS overnight at 4 °C. Prior to analysis, cells were filtered through a 50 lm nylon mesh (Small Parts Inc., Miami Lakes, FL). Cell populations were counted in a MoFlo high-performance cell sorter (Dako Colorado, Inc., Fort Collins, CO). Data were analyzed using Summit software v4.3. 2.7. RT-PCR chicken polymeric immunoglobulin receptor and cytokines Total RNA was isolated from the CALT using Tri-reagent (Molecular Research, Inc., Cincinnati, OH) according to the manufacturer’s protocols. The RNA quality and quantity were measured with a NanoDrop (Thermo Scientific, Wilmington, DE). One microgram of total RNA was reverse transcribed and amplified by 35 PCR cycles of 94 °C, 1 min., 58 °C, 1 min to detect mRNA expression of the chicken polymeric immunoglobulin receptor (pIgR). pIgR mRNA amplification resulted in a 400 bp amplicon as previously reported (van Ginkel et al., 2009). Total RNA was isolated from CALT and analyzed for IFN-c mRNA expression. Relative presence of IFN-c and IL-2 mRNA copy numbers in total RNA was measured by quantitative reverse transcription polymerase chain reaction (qRT-PCR) in CALT relative to those obtained from unvaccinated controls using a Biorad CFX96 thermal cycler (Biorad, Hercules, CA) with a one-step SYBR green master mix (Quanta Biosciences, Gaithersburg, MD) and qScript one-step RT reaction in a final volume of 25 ll. The RT reaction was performed at 50 °C for 10 min followed by 40 PCR cycles. PCR primers used for IFN-c amplification have been reported previously (Hong et al., 2006; Kaiser et al., 2000) and the primers for IL-2 by Liu et al. (2010). The b-actin gene was used as the housekeeping gene using primers described by others (Abdul-Kareem et al., 2006). b-Actin amplification was as follows: 30 s at 95 °C, 30 s at 50 °C, and 30 s at 72 °C; IFN-c: 30 s at 95 °C, 30 s at 62 °C, and 30 s at 72 °C; and IL-2: 10 s at 95 °C, 20 s at 56 °C, and 15 s at 72 °C. Cytokine expression values were normalized to the b-actin gene and expressed as fold-increase of expression of the unvaccinated control using the cycle threshold (Ct) DDCt analysis. Samples were run in 96 well plates (VWR International, West Chester, PA). The relative expression of cytokine genes was determined using the 2 DDCt method (Livak and Schmittgen, 2001). 2.8. IFN-c ELISA

Fig. 1. Conjunctiva-associated lymphoid tissue (CALT) located on the lower and upper chicken eyelid of control chicken. (A) Arrows indicated CALT located on the lower eyelid. (B) Arrow indicates the considerably smaller CALT on the upper eyelid. (C) Histology of the lower eyelid CALT (H&E; 4 objective). D and E are enlarged sections (H&E; 60 objective) from C to illustrate the presence of germinal centers and eyelid epithelium respectively.

The chicken IFN-c ELISA kit (BioSource International, Inc., Camarillo, CA) was performed according to the manufacturer’s guidelines. In brief, NUNC MaxiSorp ELISA plates were coated with 2 lg/ml of capture antibody overnight at 4 °C. The plates were blocked with assay buffer, i.e., 0.5% BSA, 0.01% Tween 20, 0.02% sodium azide in PBS, for 1 h at room temperature. Chicken IFN-c was loaded onto the plate as a standard at various concentrations starting at 5.0 ng/ml. Tears and serum samples were loaded at various dilutions in 100 ll assay buffer onto these plates to be evaluated for IFN-c content. The biotinylated detection antibody was immediately added in 50 ll at a dilution recommended by the manufacturer. The plates were incubated for 2 h at room temperature while shaking. The plates were washed and incubated for 30 min with streptavidin– HRP. The washed plates were developed using ultra-TMB substrate

292


(Thermo Scientific, Rockford, IL). The reaction was stopped after 30 min and the absorbance was determined at 450 nm using the Powerwave XS (BioTek Instruments, Inc., Winooski, VT). 2.9. Statistical analysis All statistical analyses were performed using the Student’s unpaired t test or a non-parametric Mann–Whitney test. Differences were considered significant at P < 0.05. 3. Results 3.1. Chicken CALT The present studies focused on CALT obtained from the inside of the lower eyelid (Fig. 1A) but not from the upper eyelid (Fig. 1B). The lower eyelid’s CALT constitutes the majority of CALT tissue in chickens, and is considered part of the mucosal immune system (Fix and Arp, 1991a). Histological examination of CALT obtained from 6 week-old chickens, shows a lymphoepithelium (Fig. 1C) with germinal centers (Fig. 1D) located on the mucosa of the eyelid, while the external, cutaneous portion of the eyelid is covered by keratinized epithelium (Fig. 1E). 3.2. pIgR expression Using RT-PCR specific for the chicken pIgR (van Ginkel et al., 2009) we demonstrated that pIgR mRNA is indeed expressed in chicken CALT (Fig. 2A). Expression of pIgR protein on the CALTassociated epithelium was confirmed by immunohistochemistry. The rabbit pre-bleed serum did not stain the CALT epithelium (Fig. 2B), while the epithelial cell layer shows strong pIgR-specific staining in Fig. 2C. In the enlarged section in Fig. 2D, it is shown that the pIgR-specific antiserum stained secreted pIgA on the apical surface of the epithelial cells as well as the basolateral side of the epithelial cells, although with lower intensity (see arrows Fig. 2D). The basolateral side of the epithelium is the site where the pIgR is normally expressed on epithelial cells in mammals and where it associates with pIgA destined for transport across the epithelium (Kaetzel, 2005). In unstained CALT tissue (Fig. 2E) and in the control serum stained CALT (Fig. 2B) some autofluorescence was detected but this was not located in the epithelium. 3.3. IgA-expressing B cells in CALT As is shown in Fig. 3, IgA-positive B cells were very abundant in some germinal centers (Fig. 3B), while few, more scattered IgG-positive B cells were detected in CALT (Fig. 3C). Consistent with active transport of IgA across the epithelium, detection of IgA on the mucosal surface of CALT coincided frequently with the presence of IgA-positive B cells in the lamina propria (Fig. 3A). Some germinal centers within CALT stained exclusively positive for IgA (Fig. 3B), demonstrating that active switching to IgA is occurring in this tissue. In contrast, IgG positive cells were more dispersed and less frequent (Fig. 3C). 3.4. CALT lymphocyte numbers after immunization The number of CALT lymphocytes isolated from individual chickens was counted after ocular and in ovo immunization and compared to unimmunized, age-matched control chickens (Table 1). These analyses were performed to better understand how immunization and route of immunization would affect CALT lymphocyte numbers. While control birds and in ovo only immunized birds averaged 2.0 105 and 2.4 105 lymphocytes per chicken in

Fig. 2. Expression of the pIgR in chicken CALT. (A) Expression of the pIgR in CALT was analyzed in three chickens with or without reverse transcription (RT). The PCR product was present in all three RT-PCR analyzed CALT samples. To confirm the pIgR expression in CALT at the protein level immunofluorescent staining specific for the chicken pIgR was performed (panels B–E). CALT tissue was stained with (B) preimmune rabbit serum followed by donkey–anti-rabbit conjugated to FITC or (C) pIgR-specific rabbit serum followed by donkey–anti-rabbit conjugated to FITC, or (E) without any staining. Panel D is an enlarged section of the epithelium seen in panel C. The arrows point to the basolateral side of the epithelium, where specific staining for the pIgR is observed. This demonstrates that CALT-associated epithelium expresses the pIgR.

CALT respectively, ocularly primed and boosted birds had a mean of 7.11 105 lymphocytes, and the in ovo immunized bird boosted ocularly contained 5.32 105 lymphocytes. The ocularly immunized birds displayed a significant increase in CALT lymphocytes over in ovo immunized birds (P < 0.0001), as well as over control birds (P = 0.0025). The mean number of lymphocytes increased 3-fold after ocular immunization when compared to control or in ovo immunized birds. Birds immunized by the in ovo route followed by an ocular boost immunization also displayed increased numbers of lymphocytes in CALT, and the number of lymphocytes isolated was significantly higher than for in ovo immunized chickens (P = 0.0434). The increase of lymphocytes in CALT after ocular exposure to Ad5 is consistent with its involvement in immune protection after ocular exposure to pathogens.

3.5. B cell composition of CALT The FACS analysis (Fig. 4) measured the frequency of IgA+ and IgM+ B cells in 6 week-old chickens. Fig. 4A shows the staining profile for IgA+ and IgM+ positive cells. Fig. 4B depicts the average number of IgA+ and IgM+ positive lymphocytes from 5 chickens. Unstained lymphocytes displayed lower than 1.0% positive cells in FACS analysis (data not shown). From the total lymphocytes derived from CALT 5.6 ± 1.7% were IgA+ B cells and 34.2 ± 9.6% were IgM+ B cells. Thus, CALT contains approximately 40% B cells. To measure the number of Ad5-specific antibody secreting cells after

293


Fig. 3. IgA positive B cells in CALT of control chickens. Five micrometer tissue section of CALT was stained with anti-chicken-IgA-FITC (B), anti-chicken IgG-PE (C) or was stained with both antibodies (A). All images were captured with a Retiga 2000R camera using a Nikon Eclipse E800 fluorescence microscope using a 60 objective and NIS Elements BR 3.0 software.

Table 1 Lymphocytes recovered from CALT in Ad5 immunized chickens. Average number of lymphocytes ± SEM

Number of birds

Immunization route

Day of immunization

Final dose Ad5 (ifu)

2.0 ± 0.74 105 7.11 ± 1.26 105

9 6

None Ocular

0 2 108

2.4 ± 0.30 105 5.32 ± 1.84 105

19 10

In ovo In ovo + ocular

– Ocular primed and boosted 7–10 days prior to sacrifice day 18 in ovo d18 in ovo, ocular boosted 7–10 days prior to sacrifice

1 109 2 108

Fig. 4. IgA secreting B cells in CALT. (A) FACS profiles of lymphocytes isolated from CALT and stained for IgM (white) IgA (grey) or negative control (black) in 6-week-old unimmunized control chickens. (B) The average number of IgA and IgM positive CALT lymphocytes from five chickens. Unstained controls contained 0.05). The concentration of IFN-c in plasma was inversely proportional to the in ovo priming dose on day 7 post challenge (Fig. 8D) and was significantly higher in the 105 in ovo group compared to the 107 (P = 0.0076) or the 109 in ovo group (P = 0.0008), while the 107 and 109 in ovo groups did not significantly differ in IFN-c levels (P > 0.05). The in ovo vaccination dose did not significantly change IFN-c levels in lacrimal fluids on day 7


295

Fig. 6. T cell receptor (TCR) expression in CALT, Harderian glands, and spleens. FACS analyses of CALT, Harderian glands, and spleens of 3 individually analyzed control chickens not challenged with Ad5 were analyzed for expression of the cd TCR (TCR1), the ab TCR containing Vb1 (TCR2) and the ab TCR containing Vb2 (TCR3). The cd TCR is significantly lower for CALT compared to Harderian glands and spleen, while the latter two do not differ. The number of ab TCR (both TCR2 and TCR3) positive lymphocytes in CALT and Harderian glands is significantly lower than that observed in the spleen, while CALT and Harderian glands do not significantly differ from each other. Expressed are the means ± one standard deviation.

Fig. 7. IFN-c and IL-2 mRNA expression in CALT, Harderian glands, and spleens after ocular Ad5 immunization. Fourteen days after priming with Ad5 the chicken were ocularly boosted with 2 108 ifu of Ad5. CALT, Harderian glands, and spleens were harvested at 7, 8, 9, and 11 days after boosting. RNA was isolated and analyzed for (A) IFNc and (B) IL-2 expression relative to that observed in control chickens. Depicted are the means ± one standard error of the fold increased of IFN-c or IL-2 expression for 3 chickens per time point and 6 naïve control chickens. The (⁄) indicates significant increased IFN-c expression compared to the controls (P < 0.05).

after ocular boosting (P > 0.05) (Fig. 8C). All in ovo immunized groups displayed a significant increase of IFN-c in tears after ocular Ad5 administration compared to pre-challenge levels (P < 0.05) (Fig. 8C). IFN-c levels in tears 16 days after ocular Ad5 immunization declined and did no longer significantly differ from controls (data not shown).

4. Discussion CALT is located on the eyelids of chickens (Fix and Arp, 1991a) and together with the Harderian glands constitute relevant avian head-associated lymphoid structures. RT-PCR analyses of CALT

and immunostaining of CALT tissue sections with chicken pIgRspecific antiserum demonstrated that pIgR is expressed in CALTassociated epithelium. Expression of pIgR allows active transportation of pIgA across the epithelium and contributes to the immune protection against pathogens affecting mucosal tissues including the eye (Kaetzel, 2005; Knop and Knop, 2007). The demonstration of pIgR in chickens (Wieland et al., 2004) and the prevalence of pIgA in chicken tears (van Ginkel et al., 2009) supports the hypothesis that mucosal immune responses are playing a role in protecting the paraocular and associated tissues such as nostrils, sinuses, and upper respiratory tract against invading pathogens. The presence of IgA+ B cells in CALT, and the ability of CALT lymphocytes to generate Ad5-specific antibody responses as measured by

296


Fig. 8. IFN-c levels in tears and plasma of ocular Ad5 immunized chickens. Chickens were ocularly immunized with Ad5 (2 108). Tears (A) and plasma (B) were collected at the indicated days post immunization. Chickens previously immunized with Ad5 in ovo were ocularly challenged with Ad5 at 4 weeks of age. Indicated are the IFN-c levels detected in lacrimal fluid (C) and plasma (D) 7 and 10 days after ocular Ad5 immunization. Depicted are the mean ± one standard error (n = 5 to 6 chickens per data point). The (⁄) indicates significant increased IFN-c expression compared to the controls (P < 0.05). In panel D the IFN-c expression in the 105 in ovo group is significantly higher than that observed in the other two in ovo groups.

ELISPOT indicates that CALT contributes IgA to the lacrimal fluid. Direct measurement of antigen-specific B cell and T cell responses in CALT of chickens after ocular immunization has not been reported before. Thus, these analyses provide new insight into the role of CALT in mucosal immunity in chickens. In the current study the peak of antibody-secreting cell responses in the Harderian glands occurred on day 11 after ocular exposure. Although about 40% of CALT lymphocytes are B cells the maximum antigen-specific IgA secreting cell response per 106 lymphocytes was 3-fold lower than the observed response in the Harderian glands after ocular immunization. This indicates that the Harderian glands may play a more important role in the production of lacrimal IgA. The at times close association of IgA+ B cells in the lamina propria of the eyelid and the presence of IgA on the overlaying mucosal surface is consistent with pIgA produced in the lamina propria by plasma cells and pIgA transported across the epithelium by pIgR located on the basolateral side of the epithelium. This indicates that CALT contributes to the protection of mucosal surfaces of the eye. Whether induction of immune responses in CALT will contribute to mucosal immune responses at other mucosal sites in chickens is not known at this time. In addition to B cells, CALT contains a considerable population of T cells. Based on CD3 staining approximately 37% of the total lymphocyte population are T cells, of which 16.5% CD4+ T cells, 6.0% CD8b+ T lymphocytes, and 6.0% cd T cells. Chicken CD8b expression has been reported to be dependent on co-expression of the chicken CD8a chain (Tregaskes et al., 1995). Our observations in CALT-derived lymphocytes are consistent with those observations in that all CD8b+ T cells contained the a-chain. Intraepithelial lymphocytes at mucosal surfaces in mammals can be

rich in CD8aa+ T cells, which are considered important in protecting these surfaces (Gangadharan and Cheroutre, 2004; Leishman et al., 2002). In CALT approximately 8% of the cells are CD8aa+ and about 25% of these T cells or CD4+CD8aa+ (2%). The CD4+CD8aa+ T cell subset has previously been characterized for chicken lymphocytes in the spleen, peripheral blood, and intestinal tract intraepithelial lymphocytes (Tregaskes et al., 1995; Luhtala et al., 1997). This CD4+CD8aa+ T cell population has been reported in mammals to represent effector/memory CD4+ T cells (Macchia et al., 2006) and seems to have maintained normal CD4+ T cell functions in chickens (Luhtala et al., 1997). This is consistent with observations in mammals that CD8aa is expressed independent of MHC class restriction of the T cell receptor (Gangadharan and Cheroutre, 2004). Whether or not this population can display other activities in addition to CD4+ T helper function is not known at this time. The paraocular T cell responses after Ad5 administration are reflected in the IFN-c response detected in tears and the induction of IFN-c mRNA in CALT. Thus, T cells and likely other IFN-c secreting cells, such as NK cells, contribute to the CALT cytokine responses after ocular Ad5 exposure. Interestingly, prior in ovo priming did not alter the IFN-c response in tears, but did affect the degree of involvement of the systemic immune compartment based on plasma IFN-c levels. The magnitude of the systemic immune response after ocular Ad5 boosting was inversely proportional to the in ovo priming dose. Ocularly administered Ad5 is presumably controlled by the memory immune response induced by a 109 in ovo Ad5 dose. This immunization group presumably generated a local memory response after ocular Ad5 boosting with no measurable systemic immune involvement based on IFN-c production in plasma and lacrimal fluids. Conversely, the 105 in ovo group has a


stronger systemic immune involvement in addition to a strong IFN-c response in lacrimal fluids, while the 107 in ovo vaccinated group showed intermediate IFN-c response in serum between those observed in the 109 and 105 in ovo groups after ocular boost. These observations indicate that the highest vaccine dose delivered in ovo induced a memory response in ocular lymphoid tissues and reduced systemic involvement upon subsequent ocular Ad5 boost. Alternatively, the in ovo Ad5 dose has induced a dose-dependent systemic unresponsiveness or suppression upon subsequent Ad5 exposure. In conclusion, CALT is an important contributor to ocular mucosal immunity in chickens based on the composition of its lymphocytes, the induction of antigen-specific IgA antibody secreting cells after ocular exposure, the expression of the pIgR, the production of IFN-c mRNA, and IFN-c levels in lacrimal fluids. In addition, the HALT seems to be activated by high dose in ovo Ad5 vector vaccination. Acknowledgements We wish to thank Cassandra Breedlove for her technical assistance and Lisa Parsons for assistance in animal care. This study was supported by an Animal Health and Disease Research grant. References Abdul-Kareem, M.F., Hunter, B.D., Sarson, A.J., Mayameei, A., Zhou, H., Sharif, S., 2006. Marek’s disease virus-induced transient paralysis is associated with cytokine gene expression in the nervous system. Viral Immunology 19, 167– 176. Albini, B., Wick, G., Rose, E., Orlans, E., 1974. Immunoglobulin production in chicken Harderian glands. International Archives of Allergy and Applied Immunology 47, 23–34. Ameiss, K.A., Attrache, J.E.l., Barri, A., McElroy, A.P., Caldwell, D.J., 2006. Influence of orally administrered CpG-ODNs on the humoral response to bovine serum albumin (BSA) in chickens. Veterinary Immunology and Immunopathology 110, 257–267. Bang, B.G., Bang, F.B., 1968. Localized lymphoid tissues and plasma cells in paraocular and paranasal organ systems in chickens. American Journal of Pathology 53, 735–751. Bienenstock, J., Gauldie, J., Perey, D.Y., 1973. Synthesis of IgG, IgA, IgM by chicken tissues: immunofluorescent and 14C amino acid incorporation studies. Journal of Immunology 111, 1112–1118. Burns, R.B., 1976. Specific antibody production against a soluble antigen in the Harderian gland of the domestic fowl. Clinical and Experimental Immunology 26, 371–374. Fix, A.S., Arp, L.H., 1989. Conjunctiva-associated lymphoid tissue (CALT) in normal and Bordetella avium-infected turkeys. Veterinary Pathology 26, 222–230. Fix, A.S., Arp, L.H., 1991a. Morphologic characterization of conjunctiva-associated lymphoid tissue in chickens. Journal of Veterinary Research 52, 1852–1859. Fix, A.S., Arp, L.H., 1991b. Particle uptake by conjunctiva-associated lymphoid tissue (CALT) in turkeys. Avian Diseases 35, 100–106. Fix, A.S., Arp, L.H., 1991c. Quantification of particle uptake by conjunctivaassociated lymphoid tissue (CALT) in chickens. Avian Diseases 35, 174–179. Gangadharan, D., Cheroutre, H., 2004. The CD8 isoform CD8aa is not a functional homologue of the TCR co-receptor CD8ab. Current Opinion in Immunology 16, 264–270.

297

Hathaway, L.J., Kraehenbuhl, J.P., 2000. The role of M cells in mucosal immunity. Cellular and Molecular Life Sciences 57, 323–332. Hong, Y.H., Lillehoj, H.S., Lillehoj, E.P., Lee, S.H., 2006. Changes in immune-related gene expression and intestinal lymphocyte subpopulations following Eimeria maxima infection of chickens. Veterinary Immunology and Immunopathology 114, 259–272. Kaiser, P., Rothwell, L., Galyov, E.E., Barrow, P.A., Burnside, J., Wigley, P., 2000. Differential cytokine expression in avian cells in response to invasion by Salmonella typhimurium, Salmonella enteritidis and Salmonella gallinarum. Microbiology 146, 3217–3226. Kaetzel, C.S., 2005. The polymeric immunoglobulin receptor: bridging innate and adaptive immune responses at mucosal surfaces. Immunological Reviews 206, 83–99. Knop, E., Knop, N., 2007. Anatomy and immunology of the ocular surface. Chemical Immunology and Allergy 92, 36–49. Leishman, A.J., Gapin, L., Capone, M., Palmer, E., MacDonald, H.R., Kronenberg, M., Cheroutre, H., 2002. Precursors of functional MHC class I- or class II-restricted CD8aa+ T cells are positively selected in the thymus by agonist self-peptides. Immunity 16, 355–364. Liu, H., Zhang, M., Han, H., Yuan, J., Li, Z., 2010. Comparison of the expression of cytokine genes in the bursal tissues of the chickens following challenge with infectious bursal disease viruses of varying virulence. Virology Journal 7, 364– 372. Livak, K.J., Schmittgen, T.D., 2001. Analysis of the relative gene expression data using realtime quantitative PCR and the 2(-deltadelta C(T)) method. Methods 25, 402–408. Luhtala, M., Lassila, O., Toivanen, P., Vainio, O., 1997. A novel peripheral CD4+ CD8+ T cell population: inheritance of CD8alpha expression on CD4+ T cells. European Journal of Immunology 27, 189–193. Macchia, I., Gauduin, M.C., Kaur, A., Johnson, J.P., 2006. Expression of CD8alpha identifies a distinct subset of effector memory CD4+ T lymphocytes. Immunology 119, 232–242. Mansikka, A., Sandberg, M., Veromaa, T., Vainio, O., Granfors, K., Toivanen, P., 1989. B cell maturation in the chicken Harderian gland. Journal of Immunology 142, 1826–1833. Maslak, D.M., Reynolds, D.L., 1995. B cells and T-lymphocyte subsets of the headassociated lymphoid tissues of the chicken. Avian Diseases 39, 736–742. Toro, H., Tang, D.C., Suarez, D.L., Zhang, J., Shi, Z., 2008. Protection of chickens against avian influenza with non-replicating adenovirus-vectored vaccine. Vaccine 26, 2640–2646. Toro, H., Tang, D.-C., Suarez, D.L., Sylte, M.J., Pfeiffer, J., Van Kampen, K.R., 2007. Protective avian influenza in ovo vaccination with non-replicating human adenovirus vector. Vaccine 25, 2886–2891. Toro, H., Suarez, D.L., Tang, D.C., van Ginkel, F.W., Breedlove, C., 2011. Avian influenza mucosal vaccination in chickens with replication-defective recombinant adenovirus vaccine. Avian Diseases 55, 43–47. Toro, H., Gonzalez, P., Vallejos, P., Ferreira, A., 1993. Transfer of IgG from serum to lachrymal fluid in chickens. Avian Diseases 37, 60–66. Tregaskes, C.A., Kong, F., Paramithiotis, E., Chen, C.L.H., Ratcliffe, M.J.H., Davison, T.F., Young, J.R., 1995. Identification and analysis of the expression of CD8ab and CD8aa isoforms in chicken reveals a major TCRcd CD8ab subset of intestinal intraepithelial lymphocytes. Journal of Immunology 154, 4485–4493. van Ginkel, F.W., Tang, D.C., Gulley, S.L., Toro, H., 2009. Induction of mucosal immunity in the avian Harderian gland with a replication-deficient Ad5 vector expressing avian influenza H5. Developmental and Comparative Immunology 33, 28–34. van Ginkel, F.W., van Santen, V.L., Gulley, S.L., 2008. Infectious bronchitis virus in the chicken Harderian gland and lachrymal fluid: viral load, infectivity, immune cell responses, and effects of viral immunodeficiency. Avian Diseases 52, 608– 617. Wieland, W.H., Orzáez, D., Lammers, A., Parmentier, H.K., Verstegen, M.W., Schots, A., 2004. A functional polymeric immunoglobulin receptor in chicken (Gallus gallus) indicates ancient role of secretory IgA in mucosal immunity. The Biochemical Journal 380, 669–676.