Effects of Cyclosporin A induced T-lymphocyte depletion on the course of avian Metapneumovirus (aMPV) infection in turkeys

Developmental and Comparative Immunology 34 (2010) 518–529

Effects of Cyclosporin A induced T-lymphocyte depletion on the course of avianMetapneumovirus (aMPV) infection in turkeys

Dennis Rubbenstroth a, Tina S. Dalgaard b, Sonja Kothlow c, Helle R. Juul-Madsen b, Silke Rautenschlein a,*a Clinic for Poultry, University of Veterinary Medicine Hannover, Bunteweg 17, 30559 Hannover, Germanyb Department of Animal Health, Welfare and Nutrition, Faculty of Agricultural Sciences, Blichers Alle, PB 50, 8830 Tjele, Denmarkc Institute for Animal Physiology, University of Munich, Veterinarstr. 13, 80539 Munich, Germany

A R T I C L E I N F O

Article history:

Received 28 October 2009

Received in revised form 22 December 2009

Accepted 22 December 2009

Available online 4 January 2010

Keywords:

T-lymphocytes

Flowcytometry

Proliferation assay

Cell-mediated immunity

Immunosuppression

CFSE

A B S T R A C T

The avian Metapneumovirus (aMPV) causes an economically important acute respiratory disease in

turkeys (turkey rhinotracheitis, TRT). While antibodies were shown to be insufficient for protection against

aMPV-infection, the role of T-lymphocytes in the control of aMPV-infection is not clear. In this study we

investigated the role of T-lymphocytes in aMPV-pathogenesis in a T-cell-suppression model in turkeys. T-

cell-intact turkeys and turkeys partly depleted of functional CD4+ and CD8+ T-lymphocytes by Cyclosporin

A (CsA) treatment were inoculated with the virulent aMPV subtype A strain BUT 8544. CsA-treatment

resulted in a significant reduction of absolute numbers of circulating CD4+ and CD8a+ T-lymphocytes by up

to 82 and 65%, respectively (P < 0.05). Proportions of proliferating T-cells within mitogen-stimulated

peripheral blood mononuclear cells were reduced by similar levels in CsA-treated birds compared to

untreated controls (P < 0.05). CsA-treated turkeys showed delayed recovery from aMPV-induced clinical

signs and histopathological lesions and a prolonged detection of aMPV in choanal swabs. The results of this

study show that T-lymphocytes play an important role in the control of primary aMPV-infection in turkeys.

� 2010 Elsevier Ltd. All rights reserved.

Contents lists available at ScienceDirect

Developmental and Comparative Immunology

journa l homepage: www.e lsev ier .com/ locate /dc i

Symbols

a alpha

b beta

g gamma

d delta

1. Introduction

The avian Metapneumovirus (aMPV) is an economicallyimportant pathogen of turkeys and chickens. It was first detectedin the late 1970s in South Africa and is now widely distributed in

Abbreviations: AIV, avian influenza virus; aMPV, avian Metapneumovirus; ANOVA,

analysis of variance; CD, cluster of differentiation; CD50, median ciliostatic dose; CEF,

chicken embryo fibroblasts; CFSE, carboxyfluorescein succinimidyl ester; CID50,

median cell culture infectious dose; ConA, concanavalin A; CsA, Cyclosporin A; ELISA,

enzyme linked immunosorbent assay; Exp, experiment; FITC, fluorescein isothiocy-

anate; Fig., figure; FS, forward scatter; IBDV, infectious bursal disease virus; IBV,

infectious bronchitis virus; IgA, immunoglobulin A; IgG, immunoglobulin G; IgM,

immunoglobulin M; MHC, major histocompatibility complex; OD, optical density;

PBMC, peripheral blood mononuclear cells; PBS, phosphate-buffered saline; PE,

phycoerythrin; PHA, phytohemagglutinin; PI, post-infection; RT-PCR, reverse

transcription polymerase chain reaction; SPRD, spectralred; SS, side scatter; S/P,

sample to positive ratio; TCR, T-cell receptor; TOC, tracheal organ culture; TRT, turkey

rhinotracheitis; VN, virus neutralizing; VNT, virus neutralization test.

* Corresponding author. Tel.: +49 511 9538763; fax: +49 511 9538580.

E-mail address: [email protected] (S. Rautenschlein).

0145-305X/$ – see front matter � 2010 Elsevier Ltd. All rights reserved.

doi:10.1016/j.dci.2009.12.011

many countries worldwide [1]. aMPV is a negative sense, singlestranded RNA-virus in the subfamily of Pneumovirinae within thefamily of Paramyxoviridae [1,2]. Strains of aMPV can be dividedinto subgroups based on the nucleotide sequence of theattachment (G) protein gene. To date four subgroups (A to D)have been classified [3–5]. In Europe subgroups A and B aredominating, whereas in North America only strains of subgroup Chave been detected. Subgroup D has been detected only in 1985 inFrench turkey flocks [1].

aMPV replicates in the upper respiratory tract epithelium of anumber of gallinaceous bird species [1,6,7]. In susceptible hosts thevirus causes an acute respiratory disease called turkey rhino-tracheitis (TRT) in turkeys or avian rhinotracheitis (ART) in otherbird species. Clinical signs of the disease are characterized byrespiratory symptoms such as sneezing, nasal and ocular dischargeand swelling of the infraorbital sinus [1].

Virus replication in the respiratory epithelium results in influxof lymphoid cells and mucosal damage such as epithelialdesquamation and loss of ciliar activity [6,8]. A systemicimmunosuppression has been proposed as an additional conse-quence of aMPV-infection [9,10]. By these means aMPV-induceddisease supports secondary respiratory infections in chickens andturkeys, as experimentally demonstrated for a number of bacterialpathogens [11–13].

Vaccination is widely used to control aMPV-infection inturkey flocks. Current vaccination regimes are mainly based on

mailto:[email protected]

http://www.sciencedirect.com/science/journal/0145305X

http://dx.doi.org/10.1016/j.dci.2009.12.011

D. Rubbenstroth et al. / Developmental and Comparative Immunology 34 (2010) 518–529 519

attenuated live or inactivated vaccines and have proven to beuseful tools for the prevention of the disease [14]. Neverthelessthey remain to have considerable drawbacks. Mild disease dueto residual virulence of attenuated live vaccines has beenreported, as well as reversion to full virulence after severalin vivo passages of vaccine strains in turkey or chicken flocks[15–17]. The necessity of parenteral application makes inacti-vated vaccines inconvenient for use in commercial poultryoperations with high numbers of animals. Parenteral applicationof vaccines may also fail to induce sufficient cell-mediatedimmunity on respiratory surfaces [18,19]. Therefore, effortsare made to overcome these problems by development ofrecombinant and subunit vaccines [20–23]. The development ofnew and improved vaccines and vaccination regimes for themost parts depends on a broadened knowledge of the immunemechanisms responsible for protection against aMPV-infectionand disease.

Circulating antibodies have been shown to be insufficient forthe protection of turkeys against aMPV-infection. Field observa-tions and experimental results suggest only a poor correlationbetween vaccine-induced serum antibody levels and actualprotection of the flock [19,24,25]. High levels of maternallyderived antibodies [26] as well as passively transferred aMPV-specific antibodies [27] did not prevent aMPV replication andclinical disease in turkeys. Results of Jones et al. [28] demonstratedthat vaccination of chemically B-cell compromised turkeysresulted in full protection against challenge with virulent aMPVdespite the absence of antibodies. These findings suggest that otherimmune mechanisms than humoral immunity, such as cell-mediated immune mechanisms (CMI) may play a major role inprotection against aMPV.

Few studies have been published on antigen-specific T-cellactivity in chickens. Detection of virus-specific cytotoxicT-lymphocytes (CTL) in cytotoxicity assays has been establishedfor chickens infected with infectious bronchitis virus (IBV) andavian influenza virus (AIV) [29,30]. Adoptive transfer of thesecells to naıve chicks was shown to provide protection againstchallenge with virulent IBV and AIV [29–31]. Since inbred turkeylines with defined major histocompatibility complex (MHC)haplotypes are not available, these models are not applicablefor turkeys. Experimental in vivo T-lymphocyte depletionin chickens can be achieved by repeated injections of monoclo-nal antibodies specific for chicken T-lymphocytes [32,33].However, antibodies specific for turkey T-lymphocytes are notavailable.

Chemical T-cell-suppression with Cyclosporin A (CsA) hasbeen widely used for the investigation of T-lymphocytefunctions in chickens [34–40], turkeys [41–43] and pheasants[44]. CsA is a hydrophobic fungal metabolite which hasinhibiting effects on the early phase of T-lymphocyte activation[45,46]. In mammals this was attributed to a reduction ofinterleukin 2 (IL-2) production at the transcript level [47,48].CsA also abrogates mitogen stimulation of chicken T-lympho-cytes in vitro [49–52].

The objective of this study was to investigate the influence ofT-lymphocytes on the course of primary aMPV-infection inturkeys. In two consecutive experiments CsA-treated anduntreated turkey poults were inoculated with a virulent aMPVsubtype A strain. The CsA-induced T-lymphocyte suppressionwas confirmed by flowcytometric phenotyping of blood lym-phocytes and by measuring the ex vivo mitogen response ofT-cells. The development of clinical disease, histopathologicallesions of respiratory epithelia, induction of local and systemicantibodies and aMPV-clearance were compared between aMPV-infected immunocompetent and T-lymphocyte-compromisedturkeys.

2. Materials and methods

2.1. Turkeys

Two experiments were performed. For Exp. 1 Big 6 turkey eggswere obtained from a commercial hatchery and hatched at theClinic for Poultry, University of Veterinary Medicine Hannover. ForExp. 2 day-old commercial female Big 6 turkey poults wereobtained from a commercial hatchery. These turkeys had beenexposed to a commercial aMPV subtype B live vaccine in thehatchery as detected by subtype-specific reverse transcriptasepolymerase chain reaction (RT-PCR). Turkeys were housed onwood shaving litter in positive pressure isolation units of the Clinicfor Poultry, University of Veterinary Medicine, Hannover, followinganimal welfare guidelines. Water and commercial feed wereprovided ad libitum. Before the beginning of the experimentsturkeys were confirmed to be free of maternal antibodies againstaMPV by enzyme linked immunosorbent assay (ELISA) and virusneutralization test (VNT) and to be free of aMPV subtype A by RT-PCR.

2.2. aMPV strains

The virulent aMPV subtype A strain BUT 8544 [53] was kindlyprovided by R.C. Jones, Liverpool, UK. The strain was propagatedand titrated in chicken tracheal organ culture (TOC) followingstandard protocols [54]. Titres were calculated as medianciliostatic doses (CD50) by the method of Reed and Muench [55].

An aMPV subtype A strain attenuated to chicken embryofibroblasts (CEFs) and designated BUT/CEF was kindly provided byE.F. Kaleta, Gießen, Germany. The strain was propagated andtitrated on CEF cultures. Titres were calculated as median cellculture infectious doses (CID50) by the method of Reed and Muench[55]. This strain was used for VNT.

2.3. CsA-treatment

For CsA-treatment Sandimmun 100 mg capsules (Novartis,Numberg, Germany) were used. The contents of the capsules,containing 100 mg CsA per ml, were aspired into syringes. Turkeyswere treated with a dose of 100 mg CsA per kg body weight byintramuscular injection into the calf muscles. The treatment wasrepeated at intervals of 3–4 days throughout the experiments. Asimilar treatment regime was described previously for the CsA-treatment of turkeys [42].

2.4. Isolation of peripheral blood mononuclear cells (PBMCs)

Peripheral blood mononuclear cells (PBMCs) were isolated bysucrose gradient centrifugation. Heparinised blood samples of 1 mlwere diluted with 1 ml phosphate-buffered saline (PBS), pH 7.4.Diluted blood was carefully underlaid with 1 ml Biocoll 1.090 g/ml(Biochrom, Berlin, Germany). The samples were then centrifugedat 755 � g for 20 min. The interphase, containing the lymphocytesand monocytes, was carefully collected with a Pasteur pipette andwashed twice with PBS by centrifugation at 319 � g for 5 min toremove residual Biocoll. The cell pellet was resuspended in PBS andadjusted to 107 cells/ml. Cells were directly used for phenotypingor staining with carboxyfluorescein succinimidyl ester (CFSE).

2.5. Antibodies used for flowcytometric analysis

Purified PBMC, diluted whole blood samples and stimulatedPBMC cultures were stained with fluorescence-labelled monoclo-nal antibodies directed against the cell surface antigens CD4,CD8a, T-cell receptor (TCR) ab and MHC class II (MHC-II).

D. Rubbenstroth et al. / Developmental and Comparative Immunology 34 (2010) 518–529520

Antibodies were conjugated to phycoerythrin (PE), fluoresceinisothiocyanate (FITC) or spectralred (SPRD). The following con-jugates were used: anti-TCRab-PE (Clone TCR2), anti-MHC-II-PE(Clone 2G11), anti-CD4-PE or anti-CD4-FITC (Clone CT4), anti-CD8a-FITC (Clone CT8). Anti-CD8a-Biotin (Clone CT8) was pre-incubated with Streptavidin-SPRD for 45 min before use and isdesignated anti-CD8a-SPRD. All antibodies and Streptavidin-SPRDwere obtained from Southern Biotech (Birmingham, AL, USA).Appropriate antibody concentrations were determined by titrationfor each cell source and each antibody vial used. Anti-chickenantibodies used in this study were shown to cross-react withturkey cells [8,56,57].

2.6. Flowcytometric analysis of relative lymphocyte populations in

PBMC

In Exp. 1 purified PBMC were stained with anti-TCRab-PE, anti-MHC-II-PE or a combination of anti-CD4-PE and anti-CD8a-FITCand subsequently fixed with paraformaldehyde using previouslydescribed protocols [8]. Flowcytometric analysis was performedwith an EPICS-XL flowcytometer (Beckmann-Coulter, Galway,Ireland) and CXP analysis software (Beckmann-Coulter) was usedfor data analysis. An electronic gate was set by forward scatter (FS)and side scatter (SS) characteristics to exclude cell debris anderythrocytes. Stained lymphocyte subpopulations were discrimi-nated by increased fluorescence signal. The percentage oflymphocyte subpopulations within gated cells was calculated.Results are presented as x-fold change compared to the untreated,virus-free control group. The anti-MHC-II-antibody was used as amarker for B-lymphocytes, because B-lymphocyte-specific anti-bodies cross-reacting with turkey cells were not available for ourlaboratory. MHC-II-positive lymphocytes were discriminated frommonocytes by low FS and SS signals. Double straining of chickenPBMC with anti-MHC-II and the chicken B-lymphocyte-specificanti-Bu-1 antibody (clone AV20, Southern Biotech) or the chickenmonocyte-specific antibody KUL-01 (Southern Biotech) hadconfirmed that B-lymphocytes and monocytes can be distin-guished in chickens by FS vs. SS gating (unpublished observations).

2.7. Flowcytometric analysis of absolute lymphocyte subpopulation

numbers

In Exp. 2 lymphocytes from diluted, EDTA-treated bloodsamples were stained with fluorescence-labelled monoclonalantibodies directed against CD4, CD8a and MHC-II and analysedby flowcytometry [58].

Blood samples were collected with a syringe and wereimmediately transferred to S-Monovette EDTA-tubes (Sarstedt,Numbrecht, Germany). A total volume of 400 ml EDTA-treatedblood was mixed with 80 ml fixation reagent Transfix (Cytomark,Buckingham, UK), resulting in a 1.2-fold dilution of the sample.Samples were then stored for up to 1 day until further analysis.Before cell-staining, fixed blood samples were further diluted 50-fold in flow-buffer (PBS with 1% bovine serum albumin), resultingin total in a 60-fold dilution of the blood sample. An amount of50 ml of the diluted blood was transferred to a flowcytometrytube, which then contained 0.83 ml of the original blood sample. Avolume of 50 ml antibody dilutions was added, containing acombination of the following antibodies: anti-CD4-FITC, anti-MHC-II-PE and anti-CD8a-SPRD. Samples were incubated at roomtemperature for 45 min, before 375 ml cold flow-buffer wasadded. Immediately before analysis with an EPICS-XL flowcyt-ometer (Beckmann-Coulter, Galway, Ireland) 25 ml of fluorescentbeads (FlowCount, Batch 7548032, 960 beads/ml; Beckmann-Coulter, Galway, Ireland) were added. Analysis was done usingCXP analysis software (Beckmann-Coulter). Absolute lymphocyte

subpopulation numbers, presented as number of cells per mlblood, were calculated as follows: number of stained lymphocytesdivided by the number of detected FlowCount beads andmultiplied by the number of FlowCount beads added to one mlblood (29,010 beads).

2.8. CFSE staining, ex vivo mitogen stimulation and flowcytometric

proliferation assay

Staining of purified PBMC with CFSE was performed followingpreviously published methods [59,60]. Briefly, CFSE (Sigma–Aldrich, Steinheim, Germany) was added to PBMC suspensionscontaining 5 � 106 cells/ml to give a final CFSE-concentration of0.75 mM. Following incubation of the cells with CFSE for 30 min at37 8C, RPMI 1640 medium (Biochrom, Berlin, Germany) supple-mented with 10% fetal calf serum (FCS; PAA, Pasching, Austria) wasadded at twice the initial volume, to neutralize remaining CFSE.Cells were pelleted at 319 � g for 5 min and subsequently washedtwice with PBS. After the last washing step, pelleted cells wereresuspended in RPMI-5 medium (RPMI 1640 with L-glutamine,100 U/ml penicillin, 100 mg/ml streptomycin and 5% FCS).Replicates of 106 cells per well were cultured in 200 ml RPMI-5in 96-well round bottom cell culture plates (Sarstedt, Numbrecht,Germany). Duplicate wells were supplemented with eitherconcanavalin A (ConA; Sigma, Saint Louis, USA) or phytohemag-glutinin (PHA; Sigma, Saint Louis, USA) at final concentrations of5 mg/ml. Cells cultured with mitogen-free medium were used asunstimulated controls. Following incubation at 41 8C and 5% CO2

for 72 h, 20 ml PBS supplemented with 20 mM EDTA were added toeach well to release adherent cells from the wall. After incubationfor 5 min, cells were completely transferred to a fresh 96-wellround bottom plate (Sarstedt, Numbrecht, Germany). Cells werestained with anti-CD4-PE or anti-CD8a-PE or a combination ofanti-CD4-PE and anti-CD8a-SPRD, as described for leukocytephenotyping, but cells were not fixed with paraformaldehyde.Propidium iodide (Sigma–Aldrich, Steinheim, Germany) was addedto the antibody dilutions for discrimination of life and dead cells.Flowcytometry was started immediately after staining. Analysiswas done using CXP analysis software (Beckmann-Coulter). Cellswere gated by FS vs. SS characteristics, excluding cell debris anderythrocytes. Dead cells were excluded by increased propidiumiodide uptake. Proliferated cells were identified by reduced CFSE-intensity. Results are either presented as percentage of proliferatedCD4 and CD8a positive cells within live cells, as percentage of CD4or CD8a positive cells within live cells or as percentage ofproliferated cells within CD4 or CD8a positive lymphocytesubpopulations.

2.9. In vitro detection of CsA-mediated inhibition of turkey

lymphocytes

PBMC of five untreated turkeys were isolated and stained withCFSE as described above. Duplicates of cell suspensions weresupplemented with either ConA or PHA (5 mg/ml) alone or incombination with 0.125–2 mg/ml CsA (Cicloral, Hexal; Holz-kirchen, Germany). After cultivation for 72 h cells were harvested,stained with CD4-PE, CD8a-SPRD and propidium iodide andanalysed by flowcytometry as described above.

2.10. Clinical score

Clinical signs were recorded as individual scores per animalfollowing the scoring system of Rubbenstroth and Rautenschlein[27]. A score of 0 (no signs) to 3 (severe signs) was assigned to eachof the following respiratory symptoms: nasal exudate, oculardischarge and infraorbital swelling. The sum of these scores

Table 1Experimental design.

Group a CsA-treatment aMPV challenge Number of birds

per group

Exp. 1 Exp. 2

CC �b � 7 16

TC +c � 5 12

CA � + 8 16

TA + + 8 16

Exposed to aMPV subtype B vaccine � +

Beginning of CsA-treatment at day of life 6 4

Challenge infection at day of life 30 27

a C: control; T: treated; A: inoculated with aMPV subtype A.b Negative.c Positive.


resulted in a total score of 0–9 for every turkey. Results arepresented as mean scores per day and group.

2.11. Histopathology

Samples of trachea and nasal turbinates were fixed in 10%phosphate-buffered formalin and embedded in paraffin. Tissuesections were stained with haematoxylin and eosin (H&E). Theidentity of the sections was blinded before analysis by lightmicroscopy. Mononuclear and heterophilic cell infiltrations of themucosa and deciliation and desquamation of respiratory epithelialcells were considered as conspicuous patho-histological lesions[6,8].

2.12. Detection of aMPV by RT-PCR

For detection of aMPV-specific RNA a subtype-specific RT-PCRwas used [61]. RNA was isolated from choanal swabs with TrifastGOLD (Peqlab, Erlangen, Germany). The RT reaction was performedusing the ImProm-II� RT system (Promega, Madison, USA). For PCRsteps SAWADY Taq-DNA-Polymerase (Peqlab, Erlangen, Germany)was used. PCR products were separated by agarose gel electro-phoresis and visualized by ethidium bromide staining andultraviolet transillumination. Detailed procedures have beenpublished elsewhere [27].

2.13. Serology

aMPV-specific IgG antibodies were detected by the AvianRhinotracheitis Antibody ELISA Test Kit (BioChek, Gouda, Nether-lands) following the manufacturers’ instructions. Serum sampleswere diluted 500-fold in the provided dilution buffer. Lacrimalfluid was collected with filter paper discs (6 mm, Schleicher andSchull, Dassel-Einbeck, Germany) carefully placed underneath theeyelid of the turkey for several seconds until the disc wascompletely soaked. Two discs containing 24 ml fluid each werecollected from each sampled animal and were stored together in asingle tube with 240 ml dilution buffer. This resulted in a 6-folddilution of the lacrimal fluid, which was used in the ELISA withoutfurther dilution. Optical density (OD) values of samples andpositive controls were corrected by subtraction of the mean ODvalue of negative controls. ELISA-results are presented as ratioof corrected sample OD to mean corrected positive control OD(S/P ratio).

Virus neutralizing (VN) antibodies were detected by VNT aspreviously described [27]. Briefly, replicates of 50 ml seriallydiluted sera were mixed with an equal volume of mediumcontaining 100 CID50 of aMPV-strain BUT/CEF and incubated for1 h at 37 8C in 96-well cell culture plates. Subsequently 100 mlmedium containing 7.5 � 104 CEFs were added to each well. After 7days of culture at 37 8C and 5% CO2 cytopathic effects wererecorded. VN titres were calculated using the method of Reed andMuench [55] and are presented as log 2 titres.

Both methods applied here have been shown to detectantibodies directed against aMPV-A and aMPV-B [8].

2.14. Experimental design

Experimental designs are summarized in Table 1.

2.14.1. Experiment 1

Twenty-eight day-old turkey poults were randomly assigned tofour groups of 5–8 animals. Turkeys of two groups (TC-1, 5 birdsand TA-1, 8 birds) were treated with CsA (100 mg/kg bodyweight)by intramuscular injection into the calf muscle beginning at theage of 6 days. The medication was repeated every 3–4 days until

the end of the experiment. Turkeys of groups CC-1 (7 birds) and CA-1 (8 birds) remained untreated. At the age of 30 days, groups CA-1and TA-1 were challenged with 103 CD50 per bird of strain BUT8544 by oculonasal route. At day 14 post-inoculation (PI) theexperiment was terminated and all turkeys were necropsied.

Clinical signs were recorded daily after aMPV-inoculation.Serum samples (n = 5–8) were collected at days 7, 10 and 14 PI.Choanal swabs (n = 5–8) were collected at days 4, 7, 10 and 14 PIfor aMPV detection by RT-PCR. Lacrimal fluid (n = 5–8) for aMPVantibody detection was collected at days 7 and 14 PI. PBMC wereisolated for flowcytometric phenotyping (n = 5–6) at days 13, 16,23 and 30 post-hatch and days 8 and 14 PI. Heparinised blood for ex

vivo lymphocyte stimulation was collected from groups CC-1 andTC-1 (n = 5–6) at day 41 post-hatch. At necropsy at day 14 PIsamples of nasal turbinates and trachea were collected forhistopathological examination (n = 5–8).

2.14.2. Experiment 2

A repeat experiment with a similar experimental design wasperformed with 60 turkey poults. These turkeys had been exposedto a commercial live aMPV subtype B vaccine in the hatchery, asdetected by subtype-specific nested RT-PCR. Groups TC-2 (12birds) and TA-2 (16 birds) were CsA-treated as described for Exp. 1beginning at 4 days of age, whereas groups CC-2 and CA-2 (16 birdseach) remained untreated. At the age of 27 days, groups CA-2 andTA-2 were oculonasally challenged with strain BUT 8544 (103 CD50

per bird). At days 7 and 14 PI turkeys were sacrificed for necropsy(n = 5–8 per group).

Clinical signs were recorded daily after aMPV-inoculation.Serum samples (n = 7–8) were collected at days 7, 11 and 14 PI.Choanal swabs (n = 6–8) for aMPV detection by RT-PCR werecollected at days 4, 7, 11 and 14 PI. Lacrimal fluid (n = 7–8) foraMPV antibody detection was collected at days 7 and 14 PI. Bloodsamples for detection of absolute lymphocyte counts (n = 7–10)were collected at days 11, 13, 20 and 27 post-hatch and at days 7and 14 PI. Heparinised blood for ex vivo lymphocyte stimulationwas collected from groups CC-2 and TC-2 (n = 5) at day 40 post-hatch. Samples of nasal turbinates and trachea were collected forhistopathological examination (n = 5–8) at necropsy at days 7 and14 PI.

2.15. Statistical analysis

Statistical analysis was performed using Statistix 9 (AnalyticalSoftware, Tallahassee, USA). Results of serological tests (ELISA,VNT) and relative and absolute peripheral lymphocyte subpopula-tions were analysed by one-way analysis of variance (ANOVA) andTukey’s comparison of means. Body weights were analysed byfactorial ANOVA. Analysis of aMPV-PCR results and histopatholog-ical examination was done by Fisher’s exact test. The paired T-test

Fig. 1. In vitro effect of CsA on the mitogen response of CD4+ T-lymphocytes. CFSE-stained turkey peripheral blood mononuclear cells (PBMCs) were cultured for 72 h in the

presence of the T-lymphocyte mitogens ConA or PHA (5 mg/ml) with or without CsA (1 mg/ml). CD4+ cells and proliferated cells (indicated by reduced CFSE-intensity) were

detected by flowcytometry. (A–C) Representative results of one turkey are presented as CFSE vs. CD4-PE density plots of live cells and as CFSE histograms of CD4+ cells. (A)

Stimulated with ConA. (B) Stimulated with ConA in the presence of CsA. (C) Stimulated with PHA with or without CsA. (D) Summary of the results of 5 turkeys. Asterisks

indicate significantly reduced CD4+ T-cell proliferation in the presence of CsA compared to stimulation with the respective mitogen without CsA (Paired T-test, P < 0.05).


was used for analysis of CsA in vitro effects on mitogen stimulationof lymphocytes. Effects of CsA-treatment on ex vivo lymphocytestimulation were analysed by Two-sample T-test for unequalvariances. Clinical scores were analysed with Wilcoxon rank sumtest. For all tests P-values lower than 0.05 were considered toindicate significant differences.

3. Results

3.1. In vitro effects of CsA on mitogen-induced lymphocyte

proliferation

CFSE-stained turkey PBMCs were stimulated with ConA orPHA (5 mg/ml) in the presence of different CsA-concentrations(0.125–2 mg/ml). After incubation for 72 h the proliferation ofCD4+ and CD8a+ T-lymphocytes was determined by flowcytometry.

CsA-concentrations of 0.25–1 mg/ml completely abolishedmitogen-induced proliferation of CD4+ T-cell, as shown for 1 mg/ml CsA in Fig. 1. Similar results were obtained for CD8a+ T-lymphocytes (data not shown). Concentrations lower than0.25 mg/ml did not sufficiently inhibit proliferation, whereasconcentrations higher than 1 mg/ml caused enhanced cell death(data not shown). These results show that CsA successfully inhibitsmitogen-induced proliferation of turkey T-lymphocytes in vitro.

3.2. Effects of CsA-treatment and aMPV-infection on lymphocyte

subpopulations in peripheral blood

Relative proportions (Exp. 1; Fig. 2) and absolute numbers (Exp.2; Fig. 3) of CD4+, CD8a+ and TCRab+ T-lymphocytes and MHC-II+

B-lymphocytes were determined by flowcytometric analysis.As shown by the analysis of absolute cell counts in Exp. 2,

absolute numbers of circulating T- and B-lymphocytes increasedwith the age of the turkeys (Exp. 2; Fig. 3). Relative and absolutenumbers of all T-lymphocyte subpopulations were significantlyreduced in CsA-treated turkeys (TC-1 and TC-2) compared tountreated turkeys (CC-1 and CC-2) as early as 7 days after the

beginning of treatment and remained low throughout theexperiments. A significant reduction of relative B-cell numbersin Exp. 1 occurred first at day 16 of life (representing 10 days afterthe beginning of treatment, data not shown) and was also observedat days 0 and 14 PI in this experiment (Fig. 2D). Absolute numbersof B-cells in Exp. 2 were significantly reduced only at day 7 PI (30days after the beginning of treatment; P < 0.05; Fig. 3C).

Infection with aMPV did not influence relative or absolutenumbers of CD4+ T-cells or MHC-II-positive B-cells. Relativenumbers of circulating CD8a+ T-cells were significantly reduced inthe aMPV-infected group CA-1 compared to the control group CC-1at day 7 PI (P < 0.05; Fig. 2B), whereas no effect of aMPV-Ainfection on absolute numbers of CD8a+ cells was observed in Exp.2 (Fig. 3B).

3.3. Effects of CsA-treatment on ex vivo mitogen response of PBMC

In both experiments CsA-treatment of turkeys resulted in asignificant reduction (P < 0.05) of the percentage of proliferatedCD4+ and CD8a+ T-lymphocytes within live cells followingstimulation with ConA and PHA (Fig. 4A; Exp. 2 shown as arepresentative experiment). This was mainly due to significantlyreduced (P < 0.05) proportions of T-lymphocytes in stimulated andnon-stimulated wells obtained from CsA-treated turkeys (TC-2), ascompared to untreated turkeys (CC-2; Fig. 4B). This is incongruence with reduced T-lymphocyte numbers in PBMC. CsA-treatment did not significantly decrease the proportion ofproliferated cells within the CD4+ and CD8a+ T-lymphocytesubpopulations (Fig. 4C).

3.4. Development of clinical signs and histopathological lesions

following CsA-treatment and aMPV-inoculation

Turkeys treated with 100 mg CsA per kg body weight byintramuscular injection every 3–4 days did not show clinical signsattributable to CsA-treatment. However, CsA-treated turkeystransiently showed a slightly, but significantly reduced body

Fig. 2. Relative proportions of lymphocyte subpopulations in PBMC after CsA-treatment and aMPV-inoculation in Exp. 1. (A) CD4+ T-lymphocytes. (B) CD8a+ T-lymphocytes.

(C) TCRab+ T-lymphocytes. (D) MHC-II+ B-lymphocytes. CC-1: untreated, aMPV-free turkeys; TC-1: CsA-treated, aMPV-free turkeys; CA-1: untreated and aMPV-inoculated

turkeys; TA-1: CsA-treated and aMPV-inoculated turkeys. Values marked with different superscript letters at the same experimental day are significantly different from each

other (one-way ANOVA and Tukey’s comparison of means, P < 0.05); n = 5–6; dol: day of life; ND: not done.


weight in both experiments (P < 0.05, data not shown). Interest-ingly, CsA-treated turkeys exhibited enhanced feather growth atneck and head (Fig. 5), which was most prominent at the foreheadand the wattles. This was first observed about 4 weeks after thebeginning of treatment.

Following aMPV-infection, turkeys of all inoculated groupsexpressed respiratory symptoms, such as watery eyes, nasalexudate and swelling of infraorbital sinus. Clinical signs were firstobserved at day 3 PI and mean clinical scores peaked at day 6 PI(Fig. 6). In untreated groups (CA-1 and CA-2) clinical signs wanedalmost completely during the two following days. At day 9–10 PIno or only single animals of these groups expressed low clinicalscores. Clinical scores of CsA-treated, aMPV-inoculated groups(TA-1 and TA-2) decreased less rapidly and remained considerablyhigher than those of untreated groups at days 8–10 PI in bothexperiments. At day 12 PI symptoms had completely disappearedin all aMPV-inoculated groups. One turkey of group CA-1 died 1day before the end of Exp. 1. The cause of death could not beidentified by pathological and microbiological investigations.

Histopathological lesions of the respiratory mucosa wereobserved in nasal turbinates and tracheae of aMPV-inoculatedturkeys. The lesions were characterized by lymphoid andheterophilic infiltration of the mucosa and deciliation anddesquamation of the respiratory epithelium. Prominent lesionswere detected in 63–100% of nasal turbinates and tracheaecollected at day 7 PI from the aMPV-inoculated groups CA-2 andTA-2 (Exp. 2; Table 2), with no significant differences between thegroups (P > 0.05). At day 14 PI mucosal lesions had becomeconsiderably milder in the untreated groups CA-1 and CA-2. Mildlesions were observed in 38 to 71% of nasal turbinates, whereastracheal lesions were found only in one turkey of group CA-2. Incontrast, prominent lymphocellular infiltrations were still presentin the majority of birds of the CsA-treated groups TA-1 and TA-2 atday 14 PI. Lesions were detected in all nasal turbinates and in 75–100% of the tracheal samples collected from these groups at day 14PI. Frequencies of aMPV-induced histopathological lesions intrachea and nasal turbinates were significantly higher at day 14 PI

in CsA-treated groups compared to untreated groups in bothexperiments (P < 0.05; Table 2).

Uninfected groups CC-1, TC-1, CC-2 and TC-2 remained free ofrespiratory symptoms and mucosal lesions throughout bothexperiments (data not shown).

3.5. Effect of CsA-treatment on aMPV detection

aMPV was detected in choanal swabs by RT-PCR (Fig. 7). At days4 and 7 PI, 88–100% of the swabs collected from aMPV-inoculatedgroups (CA-1, TA-1 and CA-2, TA-2) were positive for aMPVsubtype A. Thereafter, detection rates in all infected groupsdecreased. At day 14 PI only one out of eight swabs was foundpositive in group CA-2 (Exp. 2; Fig. 7B) and samples from group CA-1 were all aMPV-negative (Exp. 1; Fig. 7A). In contrast detectionrates in groups TA-1 and TA-2 were still 38 and 63%, respectively.Statistical analysis of the combined data of both experimentsrevealed significantly higher (P < 0.05) numbers of aMPV-positiveswabs in the CsA-treated groups compared to the untreated groupsat days 10 and 11 PI (TA: 7/15; CA: 2/16) and day 14 PI (TA: 8/16;CA: 1/15). No aMPV-A was detected in the uninfected groups CC-1,TC-1, CC-2 and TC-2 throughout both experiments (data notshown).

In Exp. 2 an aMPV-B strain was detected by RT-PCR in samplescollected between days 7 and 20 post-hatch. The virus most likelyoriginated from contamination with an aMPV-B live vaccineroutinely used in the commercial hatchery from which the turkeysfor this experiment were delivered.

3.6. Effect of CsA-treatment on the development of aMPV-specific

antibodies

aMPV-specific antibodies were detected in sera by VNT and IgG-ELISA and in lacrimal fluids by IgG-ELISA (Fig. 8). Beginning at day 7PI, aMPV-inoculated turkeys (CA-1, TA-1 and CA-2, TA-2) hadsignificantly increased VN titres in sera, and ELISA antibody levelsin sera and lacrimal fluids compared to the respective uninfected

Fig. 3. Effect of CsA-treatment on absolute numbers of lymphocyte subpopulations in peripheral blood of aMPV-inoculated and virus-free turkeys (Exp. 2). Lymphocytes in

diluted blood samples were stained with fluorescent antibodies and acquired by flowcytometry together with a defined number of fluorescent beads. Results are presented as

mean values per group and day with standard deviations. Values marked with different superscript letters at the same experimental day are significantly different from each

other (one-way ANOVA and Tukey’s comparison of means, P < 0.05); n = 7–10; dol: day of life; ND: not done. CC-2: untreated, aMPV-free turkeys; TC-2: CsA-treated, aMPV-

free turkeys; CA-2: untreated and aMPV-inoculated turkeys; TA-2: CsA-treated and aMPV-inoculated turkeys. (A) CD4+ T-lymphocytes. (B) CD8a+ T-lymphocytes. (C) MHC-

II+ B-lymphocytes; (D) Density blot of ungated events; forward scatter (FS) vs. sideward scatter (SS), region set on lymphocyte population. (E, F, G) dot blot of gated

lymphocytes, FS vs. fluorescence channels, regions set on lymphocyte subpopulations as indicated. (H) Dot blot of ungated events, FS vs. fluorescence channel, region set on

FlowCount beads population. (I, J, K) dot blots of gated lymphocytes, (I) CD4-FITC vs. CD8a-SPRD; (J) CD4-FITC vs. MHC-II-PE; (K) CD8a-SPRD vs. MHC-II-PE.


control group (CC-1, TC-1 and CC-2, TC-2; P < 0.05). Antibodylevels peaked at day 10 or 11 PI and were already slightly reducedat day 14 PI. Surprisingly the CsA-treated, aMPV-inoculated groups(TA-1 and TA-2) developed significantly higher serum antibodylevels (P < 0.05) compared to untreated, aMPV-inoculated turkeys(CA-1 and CA-2). This was detected beginning at day 10 or 11 PI byVNT and ELISA in both experiments. ELISA-IgG-antibodies inlacrimal fluids of CsA-treated, aMPV-inoculated turkeys in Exp. 2(TA-2) were significantly higher than those of group CA-2 at day 14PI (P < 0.05; Fig. 8F)

4. Discussion

In order to shed light on the role of T-lymphocytes in the controlof aMPV-infection, we investigated the course of primary aMPV-

infection in turkeys in a T-lymphocyte suppression model. CsA-treated and untreated commercial turkey poults were inoculatedwith virulent aMPV subtype A. Lymphocyte subpopulations inPBMC were counted and the T-lymphocyte mitogen response wasmeasured ex vivo to confirm the CsA-mediated T-cell suppression.Following aMPV-inoculation clinical signs, histopathologicallesions of respiratory epithelia, detection of aMPV by RT-PCRand antibody response of immunocompetent and CsA-treatedturkeys were compared.

Preliminary in vitro experiments demonstrated that CsAeffectively blocks ConA- and PHA-induced stimulation of CD4+

and CD8a+ turkey T-lymphocytes in a CFSE-based proliferationassay. CsA-concentrations of 0.25–1 mg/ml caused completeinhibition of T-lymphocyte proliferation. This is in agreementwith previous results obtained with chicken lymphocytes [49–52].

Fig. 4. Effect of CsA-treatment on the ex vivo mitogen response of turkey

lymphocytes (Exp. 2, presented as a representative experiment). CFSE-stained

peripheral blood mononuclear cells (PBMC) of CsA-treated (TC) and untreated (CC)

turkeys were collected at day 40 of life (36 days after beginning of treatment) and

cultivated in the presence of the T-lymphocyte mitogens ConA or PHA (5 mg/ml).

CD4+ and CD8a+ lymphocytes and proliferation of cells were detected by

flowcytometry. (A) Percentage of proliferated CD4+ or CD8a+ lymphocytes

within live cells. (B) Percentage of CD4+ or CD8a+ lymphocytes within live cells.

(C) Percentage of proliferated cells within CD4+ or CD8a+ lymphocyte

subpopulations. Asterisks indicate significantly lower values of group TC

compared to group CC (Two-Sample T-test for unequal variances, P < 0.05); n = 5.


Absolute numbers and relative proportions of differentlymphocyte subpopulations in peripheral blood of CsA-treatedand untreated turkeys were determined by flowcytometry toevaluate the in vivo T-cell suppression. For absolute cell counting asingle-step flowcytometric technique was adapted to the turkey[58]. We found that absolute numbers of both circulating T-lymphocytes and B-lymphocytes in peripheral blood increasedwith age, which has also been shown for juvenile chickens [58].CsA-treatment of turkeys significantly decreased relative andabsolute numbers of CD4+, CD8a+ and TCRab+ T-lymphocytesubpopulations as early as 7 days after the beginning of treatment.Previous studies on CsA-treatment in chickens also reported adecrease in circulating CD3+, CD4+, CD8a+, TCRab+ or TCRgd+ T-lymphocytes [36–38]. Bucy et al. [49] found CsA to adversely effect

the maturation of chicken T-lymphocytes in the thymus and themigration of mature T-cells from thymus to spleen. This mayprovide an explanation for the overall reduction of peripheral T-lymphocytes in our experiments and other studies.

The effect of CsA-treatment on T-lymphocyte reactivity wasevaluated by ex vivo mitogen stimulation of PBMC with ConA andPHA. Numerous studies have reported a significant inhibitoryeffect of CsA-treatment on ex vivo T-lymphocyte mitogenresponses in different avian species, including turkeys[34,36,37,39,41–44,51]. In these studies lymphocyte activationwas exclusively detected by bulk assays, such as [3H]-thymidine-incorporation assay [34,37,39,41–44,51], glucose consumptiontest [36] or nitric oxide inducing factor (NOIF) bioassay [51]. Bulkassays measure total responses of all cells in the well, but fail togive informations on the phenotype of the actually respondingcells. Thus these methods do not allow discrimination betweenreduced activity of potentially responding cells and reducednumbers of potential responders included in the cell suspension.Flowcytometry-based methods, such as the CFSE proliferationassay [59,60] adapted to the turkey in this study, are able toovercome these disadvantages by simultaneous detection ofactivation parameters and phenotypic characteristics on a singlecell level. In agreement with previous results, we observedsignificantly reduced percentages of proliferated CD4+ andCD8a+ T-lymphocytes within mitogen-stimulated PBMC isolatedfrom CsA-treated turkeys, as compared to untreated groups(P < 0.05). However, this reduction was mainly attributable to asignificant reduction of CD4+ and CD8a+ cell proportions in thewells, which was in consistence with the results of directphenotyping of PBMC. In contrast, the percentages of proliferatedcells within the CD4+ and CD8a+ T-lymphocyte subpopulationswere not affected by in vivo CsA-treatment in this study.

Relative proportions and absolute numbers of B-lymphocytes inblood were determined by discrimination of MHC-II+ lymphocytesfrom monocytes by their FS vs. SS characteristics. MHC-II+

lymphocytes were negative for the T-lymphocyte markers CD4and CD8a when triple staining was applied (Fig. 3J and K),confirming that they were actually B-lymphocytes. In bothexperiments MHC-II+ B-lymphocytes were significantly reducedin the CsA-treated groups compared to untreated groups. Thisreduction occurred later than the decrease of T-lymphocytes inboth experiments. CsA has been reported to not directly affect theactivity of B-lymphocytes in chickens, as shown by unalteredproliferative response to B-lymphocyte mitogens [34,39].

Treatment of turkeys with CsA did not cause clinical symptoms,but resulted in slightly, but significantly reduced body weights, asreported after CsA-treatment of chickens [39,40]. Interestingly,CsA-treated turkey poults developed a prominent increase offeather growth at head and neck, especially at the forehead and atthe wattles. To our knowledge this has not yet been reported foravian species. In humans, hypertrichosis is a well described sideeffect of CsA-treatment, but the underlying mechanisms are onlypoorly understood [62].

CsA-induced suppression had no impact on the early phase ofinfection after inoculation with a virulent aMPV subtype A strain.Virus detection, clinical scores and histopathological lesions weresimilar in treated and untreated groups until day 7 PI. Thereafterdelayed recovery from disease and prolonged presence of the viruswere observed in the CsA-treated groups in both experiments. Theseresults indicate that T-lymphocytes play an important role in aMPV-clearance and control of aMPV-induced clinical disease in turkeys.The design of our study, using CsA, does not allow a cleardiscrimination between the functions of different T-lymphocytessubsets. The unaffected antibody response in CsA-treated groupssuggests that differences in the pathogenesis may have been mainlydue to compromised cell-mediated T-lymphocyte functions in these

Fig. 5. CsA-treatment of turkeys induces enhanced feather growth at the head. (A and B) Untreated turkey, 6 weeks old. (C and D) Six weeks old turkey, which was treated with

CsA (100 mg/kg body weight i.m.) since 4 days of age at intervals of 3–4 days. White arrows mark locations of prominent feather growth at forehead and wattle.

Fig. 6. Development of clinical signs following aMPV-inoculation of CsA-treated and vaccinated turkeys (Exp. 1 and 2). Turkeys were individually examined for clinical signs

on a daily base. Results are presented as mean clinical scores; symptoms were nasal and ocular discharge and swollen infraorbital sinus. CA-1 and CA-2: untreated turkeys

inoculated with virulent aMPV subtype A; TA-1 and TA-2: CsA-treated turkeys inoculated with virulent aMPV subtype A. Exp. 1: n = 8; Exp. 2: n = 8–16; Asterisks indicate

significantly different clinical scores between groups (Wilcoxon rank sum test, P < 0.05). Groups not inoculated with aMPV did not express clinical signs throughout either

experiment.

Table 2Histopathological lesions of respiratory mucosa after aMPV-infection (Exp. 1 and 2).

Group CsA aMPV Number of animals with histological lesionsa/total (% of positive animals)

Experiment 1 Experiment 2

Day 14 PI Day 7 PI Day 14 PI

NTb Trc NT Tr NT Tr

CA �d +e 5/7 (71) 0/7A (0) 7/8 (88) 5/8 (63) 3/8A (38) 1/8A (13)

TA +e +e 8/8 (100) 6/8B (75) 7/8 (88) 8/8 (100) 8/8B (100) 8/8B (100)

No mucosal lesions were detected in groups not inoculated with aMPV in either experiment.

Upper case superscript letters indicate significant differences between groups (Fisher’s exact test, P<0.05).a Observed histopathological lesions were lymphoid infiltration of the mucosa and deciliation and desquamation of respiratory epithelium. Sections were analysed as

blinded samples.b Nasal turbinates.c Trachea.d Negativee Positive.


Fig. 7. Detection of aMPV subtype A by RT-PCR from choanal swabs following aMPV-inoculation of CsA-treated and untreated turkeys (Exp. 1 and 2). CA-1 and CA-2: untreated

turkeys inoculated with virulent aMPV subtype A; TA-1 and TA-2: CsA-treated turkeys inoculated with virulent aMPV subtype A. Exp. 1: n = 7–8; Exp. 2: n = 8. Groups not

inoculated with aMPV remained free of aMPV subtype A throughout both experiments.


groups. However, this hypothesis requires further confirmation. Ourresults are in agreement with experiences made with otherPneumovirinae in mammals. Experimental depletion of T-lympho-cytes caused prolonged virus persistence in mice infected withhuman Metapneumovirus (hMPV) [63,64] or human respiratorysyncytial virus (hRSV) [65] as well as in calves infected with thebovine respiratory syncytial virus (BRSV) [66,67]. These mammalianmodels demonstrated a contribution of both CD4+ and CD8+ T-lymphocytes to control and clearance of the infections. Prolonged

Fig. 8. Detection of aMPV-specific antibodies after inoculation of CsA-treated and untreat

B) Virus neutralizing (VN) antibodies in sera. (C and D) ELISA-IgG antibodies in sera, 500-fo

CC-2: untreated, aMPV-free turkeys; TC-1 and TC-2: CsA-treated, aMPV-free turkeys; CA

and aMPV-inoculated turkeys. Values marked with different superscript letters at the sam

Tukey’s comparison of means, P < 0.05); Exp. 1: n = 5–8; Exp. 2: n = 7–8.

virus persistence and exacerbated disease has also been publishedfor infection of CsA-treated chickens with different viral agents, suchas IBV [34], AIV subtype H9N2 [38] or infectious bursal disease virus(IBDV) [35].

CsA-treatment did not reduce clinical scores and histopatho-logical lesions in aMPV-infected turkeys. Thus, T-lymphocytescannot be considered to mediate immunopathological effectsduring aMPV-induced disease in turkeys. In contrast, a contribu-tion of T-lymphocytes to virus-induced immunopathology has

ed turkeys with virulent aMPV subtype A in Exp. 1 (A, C, E) and Exp. 2 (B, D, F). (A and

ld dilution. (E and F) ELISA-IgG antibodies in lacrimal fluid, 6-fold dilution; CC-1 and

-1 and CA-2: untreated and aMPV-inoculated turkeys; TA-1 and TA-2: CsA-treated

e experimental day are significantly different from each other (one-way ANOVA and


been reported for IBDV-infection in chickens [68] and hemorrhagicenteritis virus (HEV) in turkeys [42], as demonstrated by reducedlesions in T-cell-compromised birds. Immunopathological effectsof T-lymphocytes have also been described for hRSV and hMPVinfection in rodents [64,65,69].

Despite the reduction of circulating B-lymphocytes, aMPV-infection of CsA-treated turkeys resulted in higher levels of VN andELISA-IgG antibodies in sera and ELISA-IgG in lacrimal fluids, ascompared to untreated turkeys. Previous results on the effect ofCsA-treatment on humoral immunity in poultry are unequivocal.In some studies antibody development following infection withseveral pathogens was found to be unaffected by CsA [37,39,40],whereas other publications even reported significantly increasedVN antibody levels in CsA-treated birds [34,35]. The prolongedpersistence of aMPV may be responsible for the enhanced humoralresponse of CsA-treated turkeys to aMPV-infection observed in thisstudy. Detection of high ELISA-IgG levels in sera and lacrimal fluidof CsA-treated turkeys suggests that an isotype switch occurreddespite the partial depletion of CD4+ T-lymphocytes, which mayalso have resulted in suppression of T-helper functions. Cihak et al.[33] observed severe reduction of IgA secretion in chickensdepleted of TCRab+ T-cells. Reduction of TCRab+ T-lymphocyteswas also observed in this study, but levels of IgA were notmeasured to evaluate its effect on aMPV-infection. Previous workhas shown that serum antibodies do not provide protection againstaMPV-infection in turkeys [26,27]. Whether the increased anti-body response observed here may have contributed to the aMPV-clearance in T-cell-suppressed turkeys remains unknown.

In summary our work shows, that CsA has suppressive effectson turkey T-lymphocytes in vitro and in vivo. Humoral immunity toaMPV-infection was not negatively affected by CsA-treatmentdespite reduced numbers of B-lymphocytes. CsA-induced T-lymphocyte suppression resulted in delayed recovery from clinicalsigns and histological lesion in aMPV-infected turkeys. Also thepresence of viral RNA was prolonged. These findings suggest thatT-lymphocytes play an important role in the control of primaryaMPV-infection and virus-clearance. The contribution of differentT-lymphocyte subsets to this role remains to be subject of furtherresearch.

Acknowledgements

The authors like to thank Nicole Buhr, Victoria Lebed andChristine Haase for their valuable technical assistance and SonjaBernhardt and Martina Koschorrek for the help with the animalexperiments. Sandimmun was kindly provided by JohannGerschitz (Novartis). The project is funded by the German ResearchSociety (DFG, RA 767/3-2).

References

[1] Gough RE. Avian pneumoviruses. In: Saif YM, editor. Diseases of poultry. IowaState Press; 2003. p. 92–9.

[2] Pringle CR. Virus taxonomy 1999. The universal system of virus taxonomy,updated to include the new proposals ratified by the International Committeeon Taxonomy of Viruses during 1998. Archives of Virology 1999;144(2):421–9.

[3] Bayon-Auboyer MH, Arnauld C, Toquin D, Eterradossi N. Nucleotide sequencesof the F, L and G protein genes of two non-A/non-B avian pneumoviruses (APV)reveal a novel APV subgroup. Journal of General Virology 2000;81:2723–33.

[4] Juhasz K, Easton AJ. Extensive sequence variation in the attachment (G)-protein gene of avian pneumovirus—evidence for two distinct subgroups.Journal of General Virology 1994;75:2873–80.

[5] Toquin D, de Boisseson C, Beven V, Senne DA, Eterradossi N. Subgroup C avianmetapneumovirus (MPV) and the recently isolated human MPV exhibit acommon organization but have extensive sequence divergence in their puta-tive SH and G genes. Journal of General Virology 2003;84:2169–78.

[6] Majo N, Allan GM, O’Loan CJ, Pages A, Ramis AJ. A sequential histopathologicand immunocytochemical study of chickens, turkey poults, and broiler bree-

ders experimentally infected with turkey rhinotracheitis virus. Avian Diseases1995;39(4):887–96.

[7] Cook JK, Kinloch S, Ellis MM. In vitro and in vivo studies in chickens and turkeyson strains of turkey rhinotracheitis virus isolated from the two species. AvianPathology 1993;22(1):157–70.

[8] Liman M, Rautenschlein S. Induction of local and systemic immune reactionsfollowing infection of turkeys with avian Metapneumovirus (aMPV) subtypesA and B. Veterinary Immunology and Immunopathology 2007;115(3–4):273–85.

[9] Chary P, Rautenschlein S, Njenga MK, Sharma JM. Pathogenic and immuno-suppressive effects of avian pneumovirus in turkeys. Avian Diseases2002;46(1):153–61.

[10] Chary P, Rautenschlein S, Sharma JM. Reduced efficacy of hemorrhagic enteri-tis virus vaccine in turkeys exposed to avian pneumovirus. Avian Diseases2002;46(2):353–9.

[11] Marien M, Decostere A, Martel A, Chiers K, Froyman R, Nauwynck H. Synergybetween avian pneumovirus and Ornithobacterium rhinotracheale in turkeys.Avian Pathology 2005;34(3):204–11.

[12] Van de Zande S, Nauwynck H, Pensaert M. The clinical, pathological andmicrobiological outcome of an Escherichia coli O2: K1 infection in avianpneumovirus infected turkeys. Veterinary Microbiology 2001;81(4):353–65.

[13] Rubbenstroth D, Ryll M, Behr KP, Rautenschlein S. Pathogenesis of Riemerellaanatipestifer in turkeys after experimental mono-infection via respiratoryroutes or dual infection together with the avian metapneumovirus. AvianPathology 2009;38(6):497–507.

[14] Jones RC. Avian pneumovirus infection: questions still unanswered. AvianPathology 1996;25(4):639–48.

[15] Catelli E, Cecchinato M, Savage CE, Jones RC, Naylor CJ. Demonstration of lossof attenuation and extended field persistence of a live avian metapneumovirusvaccine. Vaccine 2006;24(42–43):6476–82.

[16] Cook JKA, Ellis MM, Dolby CA, Holmes HC, Finney PM, Huggins MB. A liveattenuated turkey rhinotracheitis virus-vaccine. 1. Stability of the attenuatedstrain. Avian Pathology 1989;18(3):511–22.

[17] Naylor CJ, Jones RC. Demonstration of a virulent subpopulation in a prototypelive attenuated turkey rhinotracheitis vaccine. Vaccine 1994;12(13):1225–30.

[18] Sharma JM, Chary P, Gerbyshak-Szudy H. Immunopathogenesis of avianpneumovirus of turkeys. In: Hafez HM, editor. Proceedings of the 4thInternational Symposium on Turkey Diseases. DVG—German Veterinary Med-ical Society; 2002. p. 145–8.

[19] Sharma JM, Chary P, Khatri M, Cha R, Palmquist JM. Pathogenesis and control ofavian pneumovirus. In: Heffels-Redmann U, Kaleta EF, editors. IV. Symposiumon Avian Corona and Pneumovirus Infections. VVB Laufersweiler Verlag; 2004.p. 318–21.

[20] Chary P, Njenga MK, Sharma JM. Protection by recombinant viral proteinsagainst a respiratory challenge with virulent avian metapneumovirus. Veteri-nary Immunology and Immunopathology 2005;108(3–4):427–32.

[21] Kapczynski DR, Sellers HS. Immunization of turkeys with a DNA vaccineexpressing either the F or N gene of avian metapneumovirus. Avian Diseases2003;47(4):1376–83.

[22] Liman M, Peiser L, Zimmer G, Prosting M, Naim HY, Rautenschlein S. Agenetically engineered prime-boost vaccination strategy for oculonasal deliv-ery with poly(D,L-lactic-co-glycolic acid) microparticles against infection ofturkeys with avian Metapneumovirus. Vaccine 2007;25(46):7914–26.

[23] Qingzhong Y, Barrett T, Brown TDK, Cook JKA, Green P, Skinner MA, et al.Protection against turkey rhinotracheitis pneumovirus (TRTV) induced by afowlpox virus recombinant expressing the TRTV fusion glycoprotein (F).Vaccine 1994;12(6):569–73.

[24] Cook JKA, Holmes HC, Finney PM, Dolby CA, Ellis MM, Huggins MB. A liveattenuated turkey rhinotracheitis virus-vaccine. 2. The use of the attenuatedstrain as an experimental vaccine. Avian Pathology 1989;18(3):523–34.

[25] Kapczynski DR, Perkins LL, Sellers HS. Mucosal vaccination with formalin-inactivated avian metapneumovirus subtype C does not protect turkeysfollowing intranasal challenge. Avian Diseases 2008;52(1):28–33.

[26] Naylor CJ, Worthington KJ, Jones RC. Failure of maternal antibodies to protectyoung turkey poults against challenge with turkey rhinotracheitis virus. AvianDiseases 1997;41(4):968–71.

[27] Rubbenstroth D, Rautenschlein S. Investigations on the protective role ofpassively transferred antibodies against avian metapneumovirus infectionin turkeys. Avian Pathology 2009;38(6):427–36.

[28] Jones RC, Naylor CJ, Alafaleq A, Worthington KJ, Jones R. Effect of Cyclophos-phamide immunosuppression on the immunity of turkeys to viral rhinotra-cheitis. Research in Veterinary Science 1992;53(1):38–41.

[29] Seo SH, Pei J, Briles WE, Dzielawa J, Collisson EW. Adoptive transfer ofinfectious bronchitis virus primed alphabeta T cells bearing CD8 antigenprotects chicks from acute infection. Virology 2000;269(1):183–9.

[30] Seo SH, Webster RG. Cross-reactive, cell-mediated immunity and protection ofchickens from lethal H5N1 influenza virus infection in Hong Kong poultrymarkets. Journal of Virology 2001;75(6):2516–25.

[31] Pei J, Briles WE, Collisson EW. Memory T cells protect chicks from acuteinfectious bronchitis virus infection. Virology 2003;306(2):376–84.

[32] Cihak J, Hoffmann-Fezer G, Wasl M, Merkle H, Kaspers B, Vainio O, et al.Inhibition of the development of spontaneous autoimmune thyroiditis in theobese strain (OS) chickens by in vivo treatment with anti-CD4 or anti-CD8antibodies. Journal of Autoimmunity 1998;11(2):119–26.

[33] Cihak J, Hoffmann-Fezer G, Zieglerheibrock HWL, Stein H, Kaspers B, Chen CH,et al. T-cells expressing the V-beta-1 T-cell receptor are required for IgA


production in the chicken. Proceedings of the National Academy of Sciences ofthe United States of America 1991;88(23):10951–5.

[34] Dhinakar Raj G, Jones RC. Effect of T-cell suppression by cyclosporin onprimary and persistent infections of infectious bronchitis virus in chickens.Avian Pathology 1997;26(2):257–76.

[35] Poonia B, Charan S. T-cell suppression by Cyclosporin A enhances infectiousbursal disease virus infection in experimentally infected chickens. AvianPathology 2001;30(4):311–9.

[36] Isobe T, Shimizu SY, Yoshihara S, Yokomizo Y. Cyclosporin A, but not bursect-omy, abolishes the protective immunity of chickens against Leucocytozooncaulleryi. Developmental and Comparative Immunology 2000;24(4):433–41.

[37] Kim IJ, You SK, Kim H, Yeh HY, Sharma JM. Characteristics of bursal Tlymphocytes induced by infectious bursal disease virus. Journal of Virology2000;74(19):8884–92.

[38] Kwon JS, Lee HJ, Lee DH, Lee YJ, Mo IP, Nahm SS, et al. Immune responses andpathogenesis in immunocompromised chickens in response to infection withthe H9N2 low pathogenic avian influenza virus. Virus Research 2008;133(2):187–94.

[39] Ganapathy K, Bradbury JM. Effects of Cyclosporin A on the immune responsesand pathogenesis of a virulent strain of Mycoplasma gallisepticum in chickens.Avian Pathology 2003;32(5):495–502.

[40] Hatkin J, Giambrone JJ, Blagburn BL. Correlation of circulating antibody andcellular-immunity with resistance against Cryptosporidium baileyi in broiler-chickens. Avian Diseases 1993;37(3):800–4.

[41] Khehra RS, Jones RC. Investigation into avian pneumovirus persistence inpoults and chicks using Cyclosporin A immunosuppression. Research inVeterinary Science 1999;66(2):161–3.

[42] Suresh M, Sharma JM. Hemorrhagic enteritis virus-induced changes in thelymphocyte subpopulations in turkeys and the effect of experimental immu-nodeficiency on viral pathogenesis. Veterinary Immunology and Immunopa-thology 1995;45(1–2):139–50.

[43] Loa CC, Lin TL, Wu CC, Bryan T, Hooper T, Schrader D. The effect of immuno-suppression on protective immunity of turkey poults against infection withturkey coronavirus. Comparative Immunology Microbiology and InfectiousDiseases 2002;25(2):127–38.

[44] Fitzgerald SD, Reed WM, Furukawa AM, Zimels E, Fung L. Effect of T-lympho-cyte depletion on the pathogenesis of marble spleen disease virus-infection inring-necked pheasants. Avian Diseases 1995;39(1):68–73.

[45] Morris PJ. Cyclosporin A. Transplantation 1981;32(5):349–54.[46] Schreiber SL, Crabtree GR. The mechanism of action of Cyclosporine-A and

FK506. Immunology Today 1992;13(4):136–42.[47] Resch K, Szamel M. Molecular mechanisms of the immunosuppressive action

of Cyclosporin A. International Journal of Immunopharmacology 1997;19(9–10):579–85.

[48] Zenke G, Baumann G, Wenger R, Hiestand P, Quesniaux V, Andersen E, et al.Molecular mechanisms of immunosuppression by Cyclosporins. Annals of theNew York Academy of Sciences 1993;685:330–5.

[49] Bucy RP, Li JM, Xu XY, Char D, Chen CLH. Effect of Cyclosporine A on theontogeny of different T-cell sublineages in chickens. Journal of Immunology1990;144(9):3257–65.

[50] Bumstead JM, Howes K, Payne LN. Effect of Cyclosporin A on normal, mitogen-stimulated and Mareks disease virus-exposed and transformed chicken lym-phoid-cells. Avian Pathology 1985;14(1):99–113.

[51] Karaca K, Kim IJ, Reddy SK, Sharma JM. Nitric oxide inducing factor as ameasure of antigen and mitogen-specific T cell responses in chickens. Journalof Immunological Methods 1996;192(1–2):97–103.

[52] Schnetzler M, Oommen A, Nowak JS, Franklin RM. Characterization of chickenT-cell growth-factor. European Journal of Immunology 1983;13(7):560–6.

[53] Wilding GP, Baxter-Jones C, Grant M. Ciliostatic agent found in rhinotracheitis.Veterinary Record 1986;118(26):735–1735.

[54] Cook JKA, Darbyshire JH, Peters RW. Use of chicken tracheal organ-cultures forisolation and assay of avian infectious bronchitis virus. Archives of Virology1976;50(1–2):109–18.

[55] Reed LJ, Muench HJ. A simple method for estimating fifty percent endpoints.American Journal of Hygiene 1938;27(3):493–7.

[56] Li Z, Nestor KE, Saif YM, Fan Z, Luhtala M, Vainio O. Cross-reactive anti-chickenCD4 and CD8 monoclonal antibodies suggest polymorphism of the turkey CD8alpha molecule. Poultry Science 1999;78(11):1526–31.

[57] Lawson S, Rothwell L, Lambrecht B, Howes K, Venugopal K, Kaiser P. Turkeyand chicken interferon-gamma, which share high sequence identity, arebiologically cross-reactive. Developmental and Comparative Immunology2001;25(1):69–82.

[58] Burgess SC, Davison TF. Counting absolute numbers of specific leukocytesubpopulations in avian whole blood using a single-step flow cytometrictechnique: comparison of two inbred lines of chickens. Journal of Immuno-logical Methods 1999;227(1–2):169–76.

[59] Dalgaard TS, Norup LR, Rubbenstroth D, Wattrang E, Juul-Madsen HR. Establish-ment of a carboxyfluorescein succinimidyl ester (CFSE) based method for flowcytometric analysis of antigen-specific proliferation of chicken T cells. in press.

[60] Lyons AB, Parish CR. Determination of lymphocyte division by flowcytometry.Journal of Immunological Methods 1994;171(1):131–7.

[61] Cavanagh D, Mawditt K, Britton P, Naylor CJ. Longitudinal field studies ofinfectious bronchitis virus and avian pneumovirus in broilers using type-specific polymerase chain reactions. Avian Pathology 1999;28(6):593–605.

[62] Akgul S, Balci YI, Unal S, Alikasifoglu A, Gurgey A. Hypertrichosis: the possibleside effect of cyclosporin in an infant with hemophagocytic lymphohistiocy-tosis receiving HLH-2004 chemotherapy protocol. Turkish Journal of Hema-tology 2009;26(3):154–6.

[63] Alvarez R, Harrod KS, Shieh WJ, Zaki S, Tripp RA. Human metapneumoviruspersists in BALB/c mice despite the presence of neutralizing antibodies.Journal of Virology 2004;78(24):14003–11.

[64] Kolli D, Bataki EL, Spetch L, Guerrero-Plata A, Jewell AM, Piedra PA, et al. Tlymphocytes contribute to antiviral immunity and pathogenesis in experi-mental human metapneumovirus infection. Journal of Virology 2008;82(17):8560–9.

[65] Graham BS, Bunton LA, Wright PF, Karzon DT. Role of T-lymphocyte subsets inthe pathogenesis of primary infection and rechallenge with respiratory syn-cytial virus in mice. Journal of Clinical Investigation 1991;88(3):1026–33.

[66] Taylor G, Thomas LH, Wyld SG, Furze J, Sopp P, Howard CJ. Role of T-lymphocyte subsets in recovery from respiratory syncytial virus-infectionin calves. Journal of Virology 1995;69(11):6658–64.

[67] Thomas LH, Cook RS, Howard CJ, Gaddum RM, Taylor G. Influence of selectiveT-lymphocyte depletion on the lung pathology of gnotobiotic calves and thedistribution of different T-lymphocyte subsets following challenge with bo-vine respiratory syncytial virus. Research in Veterinary Science 1996;61(1):38–44.

[68] Rautenschlein S, Yeh HY, Njenga MK, Sharma JM. Role of intrabursal T cells ininfectious bursal disease virus (IBDV) infection: T cells promote viral clearancebut delay follicular recovery. Archives of Virology 2002;147(2):285–304.

[69] Cannon MJ, Openshaw PJM, Askonas BA. Cytotoxic T-cells clear virus butaugment lung pathology in mice infected with respiratory syncytial virus.Journal of Experimental Medicine 1988;168(3):1163–8.

Documents

Effects of Cyclosporin A induced T-lymphocyte depletion on the course of avian Metapneumovirus (aMPV) infection in turkeys