Food & Beverages

Molecular and immunohistochemical studies on epidermal responses in Atlantic salmon Salmo salar L. induced by Gyrodactylus salaris Malmberg, 1957

Description
Molecular and immunohistochemical studies on epidermal responses in Atlantic salmon Salmo salar L. induced by Gyrodactylus salaris Malmberg, 1957
Published
of 7
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
Share
Transcript
  Molecular and immunohistochemicalstudies on epidermal responses in Atlanticsalmon  Salmo salar   L. induced by Gyrodactylus salaris  Malmberg, 1957 P. W. Kania 1  , O. Evensen 2  , T. B. Larsen 1 and K. Buchmann 1 * 1 Department of Veterinary Disease Biology, Faculty of Life Sciences,University of Copenhagen, Frederiksberg C, Denmark:  2 NorwegianSchool of Veterinary Science, Oslo, Norway (Accepted 6 July 2009; First Published Online 4 September 2009) Abstract Various strains of Atlantic salmon exhibit different levels of susceptibility toinfections with the ectoparasitic monogenean  Gyrodactylus salaris . The basicmechanisms involved in this differential ability to respond to this monogeneanwere elucidated using controlled and duplicated challenge experiments. Highlysusceptible East Atlantic salmon allowed parasite populations to reach up to3000 parasites per host within 6 weeks, whereas less susceptible Baltic salmonnever reached larger parasite burdens than 122 parasites per host during thesame period. The present study, comprising immunohistochemistry and geneexpression analyses, showed that highly susceptible salmon erected a responsemainly associated with an increased expression of interleukin-1 b  (IL-1 b ),interferon- g (IFN- g ), IL-10 and infiltration of CD3-positive cells in the epidermisof infected fins. Less susceptible salmon showed no initial response in fins but3–6 weeks post-infection a number of other genes (encoding the immune-regulating cytokine IL-10, cell marker MHC II and the pathogen-binding proteinserum amyloid A) were found to be up-regulated. No proliferation of epithelialcells was seen in the skin of less susceptible salmon, and IL-10 may play a role inthis regard. It can be hypothesized that resistant salmon regulate the parasitepopulation by restricting nutrients (sloughed epithelial cells and associatedmaterial) and thereby starve the parasites. In association with this ‘scorched-earth strategy’, the production of pathogen-binding effector molecules such asserum amyloid A (SAA) (or others still not detected) may contribute to theresistance status of the fish during the later infection phases. Introduction Gyrodactylus salaris  Malmberg, 1957 infections onAtlantic salmon  Salmo salar  in Norwegian rivers havecaused dramatic ecological and economic losses for thepast four decades (Johnsen, 1978; Johnsen & Jensen, 1986,1991; Heggberget  et al. , 1993; Mo, 1994; Bakke  et al. , 2002).This monogenean parasite was probably introducedduring the 1970s with imported salmon parr fromSwedish hatcheries located in the Baltic drainage areawhere relatively resistant salmon occur (Malmberg, 1993,2004). The Norwegian salmon, in contrast, demonstrateda high susceptibility to the parasite, which resulted indevastating infections within a few years. However,laboratory experiments have shown that susceptiblesalmon strains also occur in other parts of Europe *Fax:  þ 45 35332711E-mail: kub@life.ku.dk  Journal of Helminthology  (2010)  84,  166–172 doi:10.1017/S0022149X09990460 q Cambridge University Press 2009  including Ireland, Scotland and Denmark (Bakke &Mackenzie, 1993; Dalgaard  et al. , 2003, 2004; Heinecke et al. , 2007), which indicates the importance of imple-menting protective measures in order to prevent parasitespread to these areas. Questions associated with thegenetic and physiological background for resistance andsusceptibility are highly relevant, and answers mayeventually lead to control of this parasitosis. A number of studies have elucidated elements in the resistance patternof salmon with regard to  G. salaris  infections (Harris  et al. ,1998; Lindenstrøm  et al. , 2006; Matejusova  et al. , 2006;Collins  et al. , 2007a, b; Kania  et al. , 2007), but no finalconclusion of decisive factors has been reached. Thepresentstudypresentsmolecularandimmunohistochem-icalstudies,whichcanexplain,atleastpartly,thedifferentvulnerability of East Atlantic and Baltic salmon strains. Materials and methods Fish East Atlantic salmon eggs from the Danish River Skjernwere hatched and fry reared to the age of 4 months at theDanish Center for Wild Salmon, Jutland, and sub-sequently brought to the experimental university facility,University of Copenhagen, Frederiksberg. The averageweightandbodylengthwere6.7 ^ 4.2gand9.5 ^ 1.8cm,respectively. Baltic salmon eggs from the River Ume A¨lvin Sweden were directly imported from the Swedishhatchery and hatched at the Copenhagen Universityfacility, and fry were kept for 4 months until experimen-tation. The average weight and length were 6.5 ^ 2.7gand 9.5 ^ 1.3cm, respectively. All fish were acclimatizedinfour120ltanks for2weeksbeforeexperimentation andfed (1% of biomass daily) the same pelleted dry feed(1.5mm, 47% protein, 20% lipids and 14% carbohydrates;Ecostart 17, Biomar, Brande, Denmark). Parasites A laboratory stock of the Norwegian form of  Gyrodactylus salaris  from the River Lærdalselva wassrcinally imported from Oslo (Norway) and propagatedin our laboratory on Scottish salmon for 5 years beforeexperimentation (Dalgaard  et al. , 2003). Water and fish tanks The water used was a 50:50 mixture of deionized waterand municipal water (Frederiksberg County), which wasdechlorinated by continuous aeration for 1 week beforeuse. The 120l tanks were kept aerated and with internal biofilters (Eheim, Germany) in a thermostat-controlledroom at 12–13 8 C with a 12h light:12h dark cycle. Experimental design A total of 180 fish were used and allocated randomly toindividual groups. For each strain of salmon, a totalof 90 salmon were divided into three groups (twoexperimental replicates and one control group). Thus, 120salmon (60 fish from River Ume A¨lv and 60 fish fromRiverSkjern)wereexposedtoparasites.Thiswasdonebyexposing (for 24h) two groups each of 30 River Ume A¨lvfish to 600 live parasites and two groups each of 30 RiverSkjern fish to 600 live parasites. The infection wasperformed by placing the fish with infected salmon finscarrying parasites in 120l aquaria with lowered waterlevels. Another 60 salmon (2  £  30 control fish) were keptand handled similarly, but without inducing infection(sham infection with uninfected fins). Sampling for molecular analysis At days 0, 16, 22, 29, 36 and 43, a total of fourindividuals from each tank were randomly sampled formolecular analysis. Then sampling for parasite countingwas performed (five other fish were removed for thispurpose) to prevent any effect on gene expression dueto handling. Fish were anaesthetized and killed inMS-222 (200mg/l). Fin tissue samples from the indivi-dual fish were conserved in RNAlater e (Sigma-Aldrich,Copenhagen, Denmark). Parasite counting Subsamples (five specimens) of fish in all six groupswere examined (days 0, 16, 22, 29, 36 and 43). FollowinganaesthesiaoffishinMS-222(50mg/l),theparasiteswerecounted on all body regions (Buchmann & Uldal, 1997)under a dissection microscope (7–40  £  magnification).The infection level was expressed as mean intensity(mean number of parasites per infected fish) (Bush et al. , 1997). Isolation of RNA and cDNA synthesis RNA was isolated from sonicated fin and skin tissuewith GenElute e Total RNA kit (Sigma-Aldrich). Removalof genomic DNAwas conducted with deoxyribonucleaseI (Sigma-Aldrich). Finally, the quality was checked with1% agarose electrophoresis of 2 m l of the RNA and bymeasuring the ratio of absorptions at 260 and 280nm. Anamount of 400ng RNAwas used in each 20 m l reaction forcDNA synthesis with TaqMan w Reverse TranscriptionReagents (Applied Biosystems, Naerum, Denmark). ThecDNA was diluted ten times in RNase-free water(Invitrogen, Taastrup, Denmark). Real-time quantitative PCR (Q-PCR) A total of 2.5 m l of the diluted cDNA were used astemplate in a 12.5 m l real-time quantitative polymerasechainreactionusingJumpStart e TaqReadyMix e (Sigma-Aldrich) in the Mx3000P e  real-time PCR system(Stratagene, La Jolla, California, USA). Cycle conditionsfor all reactions were one cycle of pre-denaturation at94 8 C for 2min and 45 cycles with denaturation at 94 8 Cfor 30s. Annealing and elongation were in one step at60 8 C for 1min with end point measurement. Expressionanalyses were performed for the immune-relevant genesencoding interferon- g  (IFN- g ) (GenBank accession num- ber AY795563), MX isoforms 1, 2 and 3 (GenBankaccession numbers for MX1: U66475, MX2: U66476,MX3: U66477), MHC I (conserved region of ten sequencespresented by Grimholt  et al. , 2002) and IgM (GenBank Molecular and immunohistochemical studies in  S. salar  167  accession number S48658). Primers, probes (designedfrom the above-mentioned sources) and MgCl 2  concen-trations used are listed in table 1. Elongation factor  a 1(ELF-1 a ) (GenBank accession number AF321836) wasusedasthehousekeepinggene,asithasbeenshowntobethe least regulated gene among several others and veryuseful for the study of expression of immune-relevantgenes (Raida & Buchmann, 2008).  Analysis of Q-PCR data Data were analysed according to the 2 2 DD C t method byLivak & Schmittgen (2001). Change in threshold cyclenumber ( D C t ) was calculated by the difference in the  C t  between the target gene and the housekeeping gene foreach individual. The means of the control groups(uninfected groups for each sampling point for eachsalmon strain) were calculated. For each salmon strain,the  DD C t  values were then calculated as the difference between the  D C t  of the infected individual and theaverage of the uninfected group at the sampling point.The  t -test was used to test whether the  DD C t  values of thereplicas were different. If not, the data of a replica weremerged into one group. The values for each group at eachsampling point were averaged and standard error of themean calculated before calculating folds as 2 – DD C t . The95% confidence intervals of the fold value were thencalculated according to Livak & Schmittgen (2001). If thefold value was greater than 1, the fold was denoted as afold increase (positive values); if it was less than 1, thenegative reciprocal was calculated and the result denotedas a fold decrease (negative values); and if the fold valuewas equal to 1, the target gene was consideredunregulated. However, in order to detect substantialregulations, only regulation of more than threefold wasconsidered. The  t -test was used to detect differences between groups using a 5% probability level.  Histology and immunohistochemistry Tail fins infected with  G. salaris  were taken from RiverSkjern salmon and River Ume A¨lv salmon at day 22. Thesamples (from two fish from each fish group; RiversSkjern and Ume) were fixed for 24h in 10% neutralformalin and subsequently transferred to 70% ethanol.Following dehydration in graded ethanol series, the finswere paraffin embedded and sectioned (3 m m) andmounted on pretreated slides. The slides were eitherstaineddirectlywithhaematoxylinandeosinaccordingtostandard methods, or parallel sections were reacted withantibodies raised against human CD3 (polyclonal rabbitanti-human CD3, A0452, DAKO, Glostrup, Denmark)(Bakke-McKellep  et al. , 2007) in order to detectcell proliferation in fins from the two salmon types(Lilleeng  et al. , 2009).After the primary incubation and subsequentwashing, the Envision w system (DAKO) was usedwith the secondary peroxidase-conjugated antibody, Table 1. Genes investigated with data on primers and probes (Kania  et al. , 2007) used for QPCR, together with MgCl 2 concentrations.Gene Probe (5 0 –3 0 ) Primer forward (5 0 –3 0 ) Primer reverse (5 0 –3 0 ) MgCl 2  (m M )ELF1 a  ctgtcggtgtcatcaaggctgttg gctgtgcgtgacatgagg actttgtgaccttgccgc 5.5IL-1 b  ttgctggagagtgctgtggaagaa aggacaaggacctgctcaact ccgactccaactccaacacta 3.5IL-10 acctcgacacggtgttgcccac gggtgtcacgctatggacag tgtttccgatggagtcgatg 5.5IFN- g  ttgatgggctggatgactttagga aagggctgtgatgtgtttctg tgtactgagcggcattactcc 5.5IgM accgacagggacagcatggg actgtccatgcagcaacacc ctccaacgccatacagcaga 3.5TCR- a  cagcgcacacaaggctaattcg acagcttgcctggctacaga tgtcccctttcactctggtg 3.5MHC I tggtgtcctggcagaaagacgg gcgacaggtttctaccccagt tgtcaggtgggagcttttctg 3.5MHC II atggtgtatcactgggacccgtcc gtggagcacatcagcctcact gacgcaccgatggctatctta 5.5IFN- a  tgcagcacagatgtactgatcatcca cgtcatctgcaaagattgga gggcgtagcttctgaaatga 5.5MX caactggaggaaccagcagtcaaga ttgaggtgatggtgaaagacc gctctgagccagcagtaagaa 5.5SAA tctgaccctcgttgtaggagctcaag ctgcttctagctggacttg tcagtggtaccgcttcc 5.5TNF- a 1 tggaatggagcatcagctggagatactc ggtgggtatcttttgcacc actggcaacgatgcagg 4.0TNF- a 2 caaaatggagccctcaactggagatactcTumour necrosis factor (TNF)- a 1 and 2 were amplified in a duplex reaction with the TNF- a 1 forward and reverse primerstogether with both the TNF- a 1 and 2 probes.Table 2. Course of   Gyrodactylus salaris  infection of salmon from River Skjern and River Ume A¨lv through 43 days (ten fish per sample,duplicates pooled).Salmon strain Day 0 Day 16 Day 22 Day 29 Day 36 Day 43River SkjernMean intensity (SD) 0 99.0 (51.8) 178.6 (68.6) 482.1 (271.1) 1088.0 (863.5) 1362.0 (1099.9)min–max 8–43 77–310 146–1060 410–2500 500–3000River Ume A¨lvMean intensity (SD) 0 48.5 (38.2) 44.0 (30.8) 42.7 (36.2) 44.0 (12.5) 17.5 (11.7)min–max 18–22 10–114 9–105 19–61 6–41 168  P.W. Kania  et al.  and 3-amino-9-ethylcarbazole was used as substrate.Counterstaining with Mayer’s haematoxylin was sub-sequently performed.The enumeration of CD3-positive cells in microscopicsections was carried out on serial sections from the fins of each of the groups using computer-assisted microscopy.The number of pixels in the studied regions wasconverted to areas using the Image-Pro w  Analysisprogram (Media Cybernetics, L. P., Silver Spring,Maryland, USA). Results Challenge infection The duplicate groups of River Skjern and River UmeA¨lv, respectively, did not differ significantly within thestrain with respect to parasite loads and gene expressionprofile. Therefore, the duplicate groups were pooled forfurther calculations and presentation. Investigationsconfirmed that the  G. salaris  population increased rapidlyon Atlantic salmon from River Skjern, but showed amodest performance on skin and fins of Baltic salmonfrom River Ume A¨lv. From day 22 post-infection, themean intensity decreased on River Ume A¨lv salmon(maximum infection of 122 parasites per host at day 15),whereas the parasite burden on River Skjern salmoncontinued to increase to high levels (maximum intensityof 3000 parasites per fish on day 43) (table 2). Gene expression in fins The tested genes all showed a constitutive expressionalthough at a different level (fig. 1). Followingchallenge, River Skjern salmon (highly susceptible) Fig. 1. Basic transcript of investigated genes in fin tissue of twosalmon strains at day 0 (uninfected Control fish).  C t  values andSEMforeachgene.Thresholdvalues( C t )areindicatedbyshaded(River Skjern) and white columns (River Ume A¨lv). Numbersindicate the genes: 1 for ELF-1 a ; 2 for MHC I; 3 for MX 1, 2 and 3;4 for IL-10; 5 for IgM; 6 for MHC II; 7 for IFN- a ; 8 for SAA; 9 forTCR- a ; 10 for IL-1 b ; 11 for IFN- g ; 12 for TNF- a 1 and 2.Table 3. Expression of immune-relevant genes in fins from the two salmon strains duringinfection with  Gyrodactylus salaris  expressed as fold up-regulation.River Skjern River Ume A¨lvGene  DD C t  SEM Fold  DD C t  SEM FoldMHC I – – – – – –MHC IIDay 22 0.23 0.27 0.85  2 2.19 0.59 4.56 *IL-1 b Day 36  2 1.95 0.50 3.86 * 0.49 0.26 0.71Day 43  2 3.06 0.98 8.36 * 1.26 0.50 0.42IL-10Day 22  2 0.85 0.39 1.80  2 3.38 0.80 10.43 *Day 29  2 2.33 0.47 5.04 *  2 1.62 0.64 3.07 *Day 36  2 1.02 0.45 2.03  2 1.02 0.46 2.03 *Day 43  2 1.99 0.57 3.96 * 1.68 0.61 0.31TNF- a 1 and 2 – – – – – –IFN- a  – – – – – –IFN- g Day 22  2 1.38 0.59 2.61  2 3.06 0.94 8.31 *Day 29  2 2.20 0.41 4.61 *  2 1.74 0.58 3.34 *Day 36  2 1.94 0.47 3.82 *  2 1.41 0.48 2.65MX1,2,3 – – – – – –TCR- a Day 22  2 0.54 0.43 1.46  2 2.04 0.53 4.12 *IgMDay 22 0.49 0.26 0.71  2 1.67 0.50 3.18 *SAADay 22 0.31 0.25 0.81  2 1.91 0.75 3.75 *Genes that were regulated are indicated in bold. Only days with significant regulation areshown with detailed  DD C t  values, SEM and fold.*Significantly different from time point control,  P , 0.05 ( t -test). Molecular and immunohistochemical studies in  S. salar  169  showed significant regulation of a few genes in the fintissue investigated. Thus, a significant up-regulation of the interleukin-1 b  (IL-1 b ) gene was seen on days 36 and43, interferon- g  (IFN- g ) on days 29 and 36 and IL-10 ondays 29 and 43. River Ume A¨lv salmon (less susceptible)showed significant up-regulation of genes encoding IL-10(up to 11-fold) on days 29, 36 and 43; MHC II on day 22;IgM on day 22; serum amyloid A (SAA) on day 22; T-cellreceptor (TCR)- a  on day 22; and finally, IFN- g  on days 22and 29 (table 3).  Histology and immunohistochemistry Histology indicatedinfiltration withinflammatorycellsin the epidermis of both groups, but a trend was seen formore inflammatory cells (polymorphonuclear granulo-cytes) in the epidermis of highly susceptible salmon (notshown). Fin epidermis from highly susceptible salmon(River Skjern) showed a higher occurrence of CD3-positive cells compared with fin epidermis from the lesssusceptible(RiverUmeA¨ lv)salmonthatcontainedfewof these cells (fig. 2a and b). The percentage occurrence of the cell types in the epidermis in the two groups of fishwas compared quantitatively and showed a significantlyhigher percentage of CD3-positive cells in the highlysusceptible fish (2.5% compared with 0.51% in the lesssusceptible fish (fig. 3)). Discussion The present work has confirmed that the River Skjernsalmon is highly susceptible to infections with theNorwegian strain of   G. salaris . This is in accordancewith previous work on this and other strains of EastAtlantic salmon (Bakke & Mackenzie, 1993; Bakke  et al. ,2002; Dalgaard  et al. , 2003, 2004; Heinecke  et al. , 2007;Kania  et al. , 2007). Likewise, the low susceptibility of Baltic salmon strains to infection with this parasite wasconfirmed. However, the basic elements in fish skinresponsible for the observed differences in susceptibilityarestill enigmatic. This is also the case with other parasitesystems such as the ectoparasite  Lepeophtheirus salmonis infecting salmon skin (Fast  et al. , 2006; Jones  et al. , 2008).During a series of previous investigations, a number of immune-relevant molecules have been in focus withoutproviding a clear description of the events leadingto resistance to  G. salaris  infection. Thus, authorshave focused on lysozyme in fin tissue (Buchmann &Uldal, 1997), mucous cell densities and physiologicalproperties (Buchmann & Uldal, 1997; Sterud  et al. , 1998),mucus characteristics (Buchmann, 1998b; Jørndrup &Buchmann, 2005), complement (Buchmann, 1998a; Harris et al. , 1998), FIP2 molecules (Collins  et al. , 2007a), athymidylate kinase (Collins  et al. , 2007b) and differentialexpressionofcytokinegenes(Kania et al. ,2007).Althoughthese factors may be involved to some extent in resistancetowards G. salaris ,theexactmechanismofactionisnotyetproperly described. The present investigation haselucidated the difference in the regulation of a numberof central immune-relevant genes in fins from the twosalmon types following infection. In addition, immuno-histochemical studies have demonstrated clear differ-ences in infiltration of CD3-positive cells in the epidermisof the two salmon strains during infection. The exactnature of the reaction of these salmon cells with thiscommercial antibody raised against the epsilon chain of human CD3 is unclear (Bakke-McKellep  et al. , 2007).Thus, the CD3 nature of these cells has been questioned(Liu  et al. , 2008). However, the antibody seems to identify Fig. 2. Immunohistochemical detection of CD3-positive cells(stained red) in salmon tail fin epidermis at 22 days post-infection. (a) Less susceptible River Ume Aˆ lv salmon; (b) highlysusceptible River Skjern salmon, note the higher density of stained cells compared to River Ume Aˆlv salmon epidermis.Fig. 3. Percentage of CD3-positive cells in tail fin epidermis fromRiver Ume A¨lv (black column) and River Skjern (white).Significantly different densities ( P ¼ 0.004,  t -test). 170  P.W. Kania  et al.
Search
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks
SAVE OUR EARTH

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!

x