doi 101111j1744-313X200700656x

copy 2007 The AuthorsJournal compilation copy 2007 Blackwell Publishing Ltd International Journal of Immunogenetics 34 55ndash63 55

Blackwell Publishing LtdMolecular genetic analysis of porcine mannose-binding lectin genes MBL1 and MBL2 and their association with complement activity

C Phatsara D G J Jennen S Ponsuksilidagger E MuraniDagger D Tesfaye K Schellander amp

K WimmersDagger

Summary

Mannose-binding lectin (MBL) mediates activation of thecomplement system via the lectin pathway Two forms ofMBL MBL-A and MBL-C were characterized in rodentsrabbits bovine and rhesus monkeys whereas only oneform was identified in humans chimpanzees and chickensThe two forms are encoded by two distinct genes namedMBL1 and MBL2 which have been identified in manyspecies including the pig In this report we studied the twoporcine genes MBL1 and MBL2 The porcine MBL geneshad higher identities to bovine rather than primate androdent sequences Both genes were assigned to chromo-some 14 by radiation hybrid panel and linkage mappingBoth MBL genes were highly expressed in liver MBL1was also found to be expressed in the lung testis and brainwhereas low expression of MBL2 was detected in thetestis and kidney New single nucleotide polymorphismsof porcine MBL2 gene were found and genotyped in anexperimental F2 pig population together with a previouslyreported SNP of MBL1 MBL1 genotypes differed in C3cserum concentration ie in vivo complement activityat P lt 01 Correspondingly linkage analysis revealed aquantitative trait locus for C3c serum level close to theposition of the MBL genes The study thus promotes theporcine MBL genes as functional and positional candidategene for complement activity

Introduction

Mannose-binding lectin (MBL) or mannose-bindingprotein (MBP) a member of the collectin family of proteinsis an important factor in the lectin pathway of the com-

plement cascade Binding of MBL leads to opsonizationthrough complement activation and deposition of C3 Itsassociation to MBL-associated serine protease (MASP)initiates the lectin pathway which is homologous to theclassical complement pathway (Matsushita amp Fujita1992) Two types of MBL namely MBL-A and MBL-C were characterized in several species including the ratmouse bovine as well as rhesus monkey (Mizuno et al 1981Drickamer et al 1986 Mogues et al 1996 Kawai et al1997 Hansen et al 2000 Laursen amp Nielsen 2000)Only one MBL was identified in humans chimpanzees(Mogues et al 1996) and chickens (Laursen amp Nielsen2000) In rabbits two forms of MBL (from serum andliver) that are encoded by one gene were characterized(Kawai et al 1998)

Two forms of MBL have been characterized in the pigA full-length porcine liver MBL cDNA of 723-bp with anopen reading frame of 241 amino acids (GenBank acces-sion number NM_214125) was obtained by Agah et al(2001) This MBL shares a higher identity to rat MBL-Cthan MBL-A proteins (567 and 502 respectively)The overall identity between human and porcine MBL is649 at the amino acid level (allowing for eight gaps)Lillie et al (2006) recently reported the isolation ofanother MBL This MBL-A cDNA was also characterizedand proposed to represent the porcine MBL1 gene (Gen-Bank accession number AY771222) which is homologousto the rodent MBL1 gene and MBL1P1 a pseudogene ofhumans and chimpanzees

In this study chromosomal assignment linkage map-ping expression in different tissues and association withcomplement activity of the porcine MBL genes were ana-lysed Furthermore the identity of the porcine MBL genesand the relationship between the two porcine MBL as wellas with those from other species were investigated

Materials and methods

Phylogenetic analysis of MBL

In order to identify the relationships between MBLs indifferent animal species we used deduced amino acidsequences from previously identified genes reported inGenBank for construction of a phylogenetic tree using theneighbour-joining method (NJ) The reliability of internal

Institute of Animal Science Animal Breeding and Husbandry Group University of Bonn Endenicher Allee Bonn Germany dagger Research Institute for the Biology of Farm Animals (FBN) Research Group Functional Genomics Dummerstorf Germany Dagger Research Institute for the Biology of Farm Animals (FBN) Research Unit Molecular Biology Dummerstorf Germany

Received 1 August 2006 revised 30 October 2006 accepted 26 November 2006

Correspondence Klaus Wimmers Research Institute for the Biology of Farm Animals (FBN) Research Unit Molecular Biology Wilhelm-Stahl-Allee 2 18196 Dummerstorf Germany Tel +49 38208 68700 Fax +49 38208 68702 E-mail wimmersfbn-dummerstorfde

56 C Phatsara et al

copy 2007 The AuthorsJournal compilation copy 2007 Blackwell Publishing Ltd International Journal of Immunogenetics 34 55ndash63

branches was assessed by 1000 bootstrap replicates andsites with pairwise deletion in this analysis NJ searcheswere conducted by using the computer program mega3(Kumar et al 2004)

Polymerase chain reaction (PCR) condition

Standard PCR was performed in 20 microL reaction volumeincluding 100 ng of genomic DNA 02 microm of eachprimer 50 microm of dNTP 05 units of T Polymerase(GeneCraft Luumldinghausen Germany) in 1timesPCR buffercontaining 15 mm MgCl2 Thermocycling was per-formed as follows initial denaturation at 95 degC for 5 minfollowed by 35 cycles at 94 degC for 30 s 59 degC for 30 s72 degC for 1 min and final extension at 72 degC for 10 minAdditionally touchdown PCR was also conducted toamplified MBL1 using the following conditions initialdenaturation at 95 degC for 5 min followed by eight cyclesat 94 degC for 30 s 66ndash62 degC for 30 s 72 degC for 1 min(annealing temperature was 05 degC decreased each cycle)then followed by 35 cycles of 94 degC for 30 s 62 degC for30 s 72 degC for 1 min and final extension at 72 degC for 5 minGene-specific primers used are described in Table 1 AllPCR products were resolved by agarose gel electrophoresisand visualized by ethidium bromide staining

Expression study of porcine MBL genes

In order to survey expression of the porcine MBL genes indifferent tissues reverse transcriptase PCR (RT-PCR) wasemployed Total RNA was isolated from the muscleheart spleen tonsil lymph node lung liver kidney testisand brain from adult animals of Duroc and Berlin Minia-ture pig crossbreeds using Tri-Reagent (Sigma TaufkirchenGermany) following manufacturerrsquos instructions All RNAsamples were treated with deoxyribonuclease I (DNase IPromega Mannheim Germany) in the presence of RNaseinhibitor (Promega) for 1 h at 37 degC to remove the resid-ual DNA and DNA-free RNA products were obtainedafter purification by RNeasy Mini kit (Qiagen HildenGermany) The RNA was visualized on 15 formalde-hyde containing agarose gels to check the integrity and theconcentration was measured by spectrometry (Ultraspectphotospectrometer Amersham Biosciences Freiberg

Germany) First-strand cDNA was synthesized from 1 microgof total RNA using random primers and oligo (dT)12 Nin the presence of reverse transcriptase (Superscript IIInvitrogen Karlsruhe Germany) Standard protocols wereapplied to detect MBL1 and MBL2 transcripts as describedpreviously using MBL1-a and MBL2-a primer sets forMBL1 and MBL2 respectively (Table 1) In order to controlfor possible variation of initial RNA input the expressionof the 18S rRNA gene was used as an internal reference

Radiation hybrid (RH) and genetic mapping

The RH mapping of MBL genes was performed usingthe INRA-Minnesota 7000 rad radiation hybrid panel(IMpRH) (Yerle et al 1998) which consisted of 118hamster-porcine hybrid cell lines (Hawken et al 1999)Standard PCR was performed as described previouslyusing MBL1-b and MBL2-b primer sets derived from bothporcine MBL genes sequences (GenBank accession num-bers AF208528 and NM_214125) The entire RH panelwas scored by PCR using INRA protocols (httpwwwtoulouseinrafrlgclgchtm) Data analysis was performedusing software available at IMpRH database (httpimprhtoulouseinrafr) for chromosome assignmentGenetic mapping by two-point linkage analysis was doneusing crimap version 24 (Green et al 1990)

Phenotype analysis

F2 animals (n = 457) of a three-generation F2 populationbased on the reciprocal cross of Berlin Miniature pig andDuroc (DUMI resource population) were kept andperformance tested at our research farm F2 animals wereborn from mating of 11 F1 sows with 3 F1 boars with amaximum of five parities per sow The animals wereimmunized with Mycoplasma (Mh) (Stellamune Myco-plasma Pfizer Karlsruhe Germany) Aujeszkyrsquos (ADV)(Porcilis Begonia Diluvac Intervet Toumlnisvorst Ger-many) and PRRS (Ingelvac PRRS MLV Boehringer Ingel-heim Germany) vaccines at 6 14 and 16 weeks of agerespectively Blood samples were taken immediately priorto immunization (day 0 time 1 4 and 7) on day 4 (time2 and 5) and day 10 (time 3 and 6) after Mh and ADVvaccination and day 10 (time 8) after PRRS vaccinationBlood samples were cooled immediately and sera andplasma were collected and stored at minus80 degC for furtheruse The concentration of C3c a stable soluble fragmentcleaved from C3b during complement activation wasmeasured by immunonephelometry using a Behring-Nephelometer system and antihuman-C3c-antiserum(Dade Behring GmbH Marburg Germany) for samplestaken before and after the Mh ADV and PRRSV vaccina-tions Results are given as concentration of C3c (g Lminus1)(Wimmers et al 1999 2003) Also in vivo haemolyticcomplement activity was measured in the classical and thealternative pathway For classical complement activity(CH50) sensitized sheep red blood cells were used tomeasure the haemolytic complement activity but substi-tuted by rabbit red blood cells for alternative complement

Table 1 Gene-specific primers (5primendash3prime) used for porcine MBL genes amplification in this study

Primer set SequenceAnnealing temp (degC)

Product size (bp)

MBL1-a CCCCAATATTTCCTGGAGGT 59 222TCCTCCTTCTGTGTGTGGTG

MBL2-a GGGAGAAAAGGGAGAACCAG 59 278CACACAGAGCCTTCACTCCA

MBL1-b AAGGGAGAACCAGGTATAGG 62ndash66 702TGAACCCTGGCCCTGTTG

MBL2-b CTTCGCTCAGGGAAAACAAG 59 319GTCATTCCACTTGCCATCCT

56 C Phatsara et al

copy 2007 The AuthorsJournal compilation copy 2007 Blackwell Publishing Ltd International Journal of Immunogenetics 34 55ndash63

branches was assessed by 1000 bootstrap replicates andsites with pairwise deletion in this analysis NJ searcheswere conducted by using the computer program mega3(Kumar et al 2004)

Polymerase chain reaction (PCR) condition

Standard PCR was performed in 20 microL reaction volumeincluding 100 ng of genomic DNA 02 microm of eachprimer 50 microm of dNTP 05 units of T Polymerase(GeneCraft Luumldinghausen Germany) in 1timesPCR buffercontaining 15 mm MgCl2 Thermocycling was per-formed as follows initial denaturation at 95 degC for 5 minfollowed by 35 cycles at 94 degC for 30 s 59 degC for 30 s72 degC for 1 min and final extension at 72 degC for 10 minAdditionally touchdown PCR was also conducted toamplified MBL1 using the following conditions initialdenaturation at 95 degC for 5 min followed by eight cyclesat 94 degC for 30 s 66ndash62 degC for 30 s 72 degC for 1 min(annealing temperature was 05 degC decreased each cycle)then followed by 35 cycles of 94 degC for 30 s 62 degC for30 s 72 degC for 1 min and final extension at 72 degC for 5 minGene-specific primers used are described in Table 1 AllPCR products were resolved by agarose gel electrophoresisand visualized by ethidium bromide staining

Expression study of porcine MBL genes

In order to survey expression of the porcine MBL genes indifferent tissues reverse transcriptase PCR (RT-PCR) wasemployed Total RNA was isolated from the muscleheart spleen tonsil lymph node lung liver kidney testisand brain from adult animals of Duroc and Berlin Minia-ture pig crossbreeds using Tri-Reagent (Sigma TaufkirchenGermany) following manufacturerrsquos instructions All RNAsamples were treated with deoxyribonuclease I (DNase IPromega Mannheim Germany) in the presence of RNaseinhibitor (Promega) for 1 h at 37 degC to remove the resid-ual DNA and DNA-free RNA products were obtainedafter purification by RNeasy Mini kit (Qiagen HildenGermany) The RNA was visualized on 15 formalde-hyde containing agarose gels to check the integrity and theconcentration was measured by spectrometry (Ultraspectphotospectrometer Amersham Biosciences Freiberg

Germany) First-strand cDNA was synthesized from 1 microgof total RNA using random primers and oligo (dT)12 Nin the presence of reverse transcriptase (Superscript IIInvitrogen Karlsruhe Germany) Standard protocols wereapplied to detect MBL1 and MBL2 transcripts as describedpreviously using MBL1-a and MBL2-a primer sets forMBL1 and MBL2 respectively (Table 1) In order to controlfor possible variation of initial RNA input the expressionof the 18S rRNA gene was used as an internal reference

Radiation hybrid (RH) and genetic mapping

The RH mapping of MBL genes was performed usingthe INRA-Minnesota 7000 rad radiation hybrid panel(IMpRH) (Yerle et al 1998) which consisted of 118hamster-porcine hybrid cell lines (Hawken et al 1999)Standard PCR was performed as described previouslyusing MBL1-b and MBL2-b primer sets derived from bothporcine MBL genes sequences (GenBank accession num-bers AF208528 and NM_214125) The entire RH panelwas scored by PCR using INRA protocols (httpwwwtoulouseinrafrlgclgchtm) Data analysis was performedusing software available at IMpRH database (httpimprhtoulouseinrafr) for chromosome assignmentGenetic mapping by two-point linkage analysis was doneusing crimap version 24 (Green et al 1990)

Phenotype analysis

F2 animals (n = 457) of a three-generation F2 populationbased on the reciprocal cross of Berlin Miniature pig andDuroc (DUMI resource population) were kept andperformance tested at our research farm F2 animals wereborn from mating of 11 F1 sows with 3 F1 boars with amaximum of five parities per sow The animals wereimmunized with Mycoplasma (Mh) (Stellamune Myco-plasma Pfizer Karlsruhe Germany) Aujeszkyrsquos (ADV)(Porcilis Begonia Diluvac Intervet Toumlnisvorst Ger-many) and PRRS (Ingelvac PRRS MLV Boehringer Ingel-heim Germany) vaccines at 6 14 and 16 weeks of agerespectively Blood samples were taken immediately priorto immunization (day 0 time 1 4 and 7) on day 4 (time2 and 5) and day 10 (time 3 and 6) after Mh and ADVvaccination and day 10 (time 8) after PRRS vaccinationBlood samples were cooled immediately and sera andplasma were collected and stored at minus80 degC for furtheruse The concentration of C3c a stable soluble fragmentcleaved from C3b during complement activation wasmeasured by immunonephelometry using a Behring-Nephelometer system and antihuman-C3c-antiserum(Dade Behring GmbH Marburg Germany) for samplestaken before and after the Mh ADV and PRRSV vaccina-tions Results are given as concentration of C3c (g Lminus1)(Wimmers et al 1999 2003) Also in vivo haemolyticcomplement activity was measured in the classical and thealternative pathway For classical complement activity(CH50) sensitized sheep red blood cells were used tomeasure the haemolytic complement activity but substi-tuted by rabbit red blood cells for alternative complement

Table 1 Gene-specific primers (5primendash3prime) used for porcine MBL genes amplification in this study

Primer set SequenceAnnealing temp (degC)

Product size (bp)

MBL1-a CCCCAATATTTCCTGGAGGT 59 222TCCTCCTTCTGTGTGTGGTG

MBL2-a GGGAGAAAAGGGAGAACCAG 59 278CACACAGAGCCTTCACTCCA

MBL1-b AAGGGAGAACCAGGTATAGG 62ndash66 702TGAACCCTGGCCCTGTTG

MBL2-b CTTCGCTCAGGGAAAACAAG 59 319GTCATTCCACTTGCCATCCT

Molecular genetic analysis of MBL1 and MBL2 57

copy 2007 The AuthorsJournal compilation copy 2007 Blackwell Publishing Ltd International Journal of Immunogenetics 34 55ndash63

activity (AH50) Haemolytic activity was expressed as thetitre that lysed 50 of the erythrocytes (CH50 U mLminus1AH50 U mLminus1) (Wimmers et al 1999 2003)

SNP detection in porcine MBL2 gene

Partial porcine MBL2 gene was amplified from 15 pigs ofthe F2 DUMI resource population by using standard PCRwith MBL2-b primer set (Table 1) Primers were designedbased on the conserved region between exon 4 of thehuman and porcine MBL2 genes (GenBank accessionnumbers NP_000233 and NM_214125) The targetproducts were sequenced using CEQtrade 8000 GeneticAnalysis System (Beckman Coulter Krefeld Germany)Individual sequences were aligned and compared usingthe web-based program multalin (httpribosometoulouseinrafrmultalinmultalinhtml) for identificationof sequence variation

Genotype analysis

For genotyping of the DUMI F2 animals (n = 457) atMBL1 and MBL2 polymorphic sites PCR restrictionfragment length polymorphism (PCR-RFLP) assays wereperformed using HinfI and AdeI restriction enzymesFor the MBL1 gene the animals were genotyped at apreviously identified C to T substitution within intron 1 atposition 328 of the sequence (AF208528) reported byMarklund et al (2000) For the MBL2 gene the animalswere genotyped at the G to A transition at position 645of sequence NM_214125 as found in this study The frag-ments covering polymorphic site of both genes wereamplified using specific primer sets (MBL1-b MBL2-bTable 1) as given previously with touchdown and standardconditions for MBL1 and MBL2 respectively Digestionof the products was carried out in 10 microL of 1times restrictionbuffer and incubated at 37 degC overnight to ensure completedigestion Digested PCR products were visualized on 25agarose gel to determine genotype distribution Data werechecked for any genotyping errors by using the programpedcheck version 11 (OrsquoConnell amp Weeks 1998)

Association study

In order to investigate the effects of different MBL1 andMBL2 genotypes on AH50 CH50 and C3c concentrationanalysis of variance was performed using the procedurelsquomixedrsquo and the lsquorepeated measurersquo statement of the sassoftware package (SAS System for Windows Release 802)A model was fitted in order to identify other significantenvironmental and genetic effects apart from the MBL1and MBL2 genotypes and its interaction by stepwise elim-ination of non-significant effects Animal effect was thesubject specified in the repeated statement For eachimmunological trait the least square means between MBLgenotype classes at each time of blood sampling werecompared The repeated measures (first-order autoregressiveR-matrix) mixed model for analysing haemolytic complementactivity traits was as follows

Y = micro + SIREi + DAMj + PARITYk + TREATMENTl + GENOTYPEm + TIMEn + SEXo + animalijklmno + (GENOTYPETIME)mn + εijklmno

where Yijklmno is the phenotypic data (CH50 AH50 andC3c level) micro is overall mean sirei is the fixed effect of sire(i = 1ndash3) damj is the fixed effect of dam (j = 1ndash11) parityk

is the fixed effect of parity (k = 1ndash5) treatmentl is the fixedeffect of treatment (vaccinated and non-vaccinated l = 1ndash2)genotypem is the fixed effect of genotype (m = 1ndash3)timen is the fixed effect of time point of measurement priorand after vaccinations (n = 1ndash8) sexo is the fixed effect ofsex (o = 1ndash2) animalijklmno is the random effect of animal(genotypetime)mn is the interaction between MBL2 geno-type and time point and εijklmno is the residual error

Linkage quantitative trait locus (QTL) study

The association analysis was complemented by linkageQTL analysis which enables to provide evidence foreffects of the MBL genes in the absence of linkage disequi-librium between the analysed silent SNPs and otherpotentially existing causal mutations Linkage QTL anal-ysis was carried out by least square regression using theprogram qtl express (httpqtlcapedacuk) (Seatonet al 2002) This model assumes a biallelic QTL fixed foralternative alleles in each parental line For each F2 ani-mal genotypes of five markers including MBL1 andMBL2 on chromosome 14 (SSC14) were used to estimatethe probability of having none one or two alleles of theputative QTL of the respective founder line (grandparentgeneration) in 1 centimorgan (cm) intervals The proba-bilities are used to calculate additive and dominancecoefficients for a putative QTL at each position and thetrait values are then regressed onto these coefficients ForQTL analysis phenotypes CH50 AH50 and C3c adjustedfor systematic effects were used ie residuals of repeatedmeasures analyses using the model detailed previouslybut without the fixed effect of genotype of MBL1 andMBL2 Chromosome-wide 5 significance threshold wasdetermined empirically by permutation (10 000 itera-tions) and transformed to genome-wide 5 significancethreshold by subsequent Bonferroni correction for thenumber of autosomes (Churchill amp Doerge 1994)

Results and discussion

Phylogenetic analysis of MBL

The predicted amino acid sequences of MBL-A and MBL-C across species (porcine human rhesus monkey mouserat and chicken) were aligned Unrooted NJ tree showingphylogenetic relationships of MBL based on the NJmethod reconstructed by mega3 (Fig 1) indicates theexistence of three distinct forms of MBL among all speciesinvestigated (MBL-A and MBL-C in mammals and MBLin chicken) The MBL-A and MBL-C branches each com-prise three different subbranches ie rodent (mouse and rat)primate (rhesus andor human) and non-primate artiodactyl

58 C Phatsara et al

copy 2007 The AuthorsJournal compilation copy 2007 Blackwell Publishing Ltd International Journal of Immunogenetics 34 55ndash63

(porcine and cattle) The artiodactyl sub-branch indicateshigher identity of porcine MBLs to the bovine homolo-gous loci (76 and 98 bootstrap values for MBL-A andMBL-C respectively) than to the loci of the other speciesThe result also indicates a closer relationship of non-primate to primate than to the rodent group These observa-tions correspond to results of Lillie et al (2006) whichindicate three separate branches of MBL-A for artiodac-tyls primates and rodents with the first being closer toprimate than to rodents

Comparison of porcine MBL genes

The amino acid sequences of the two porcine MBL iso-forms are homologous to each other as shown in Fig 2

The identity of these two lectins is about 566 allowingfor nine gaps Both porcine MBLs have three cysteineresidues at the N-terminal domain that are involved inoligomerization A collagen domain can be recognized fromthe amino acid sequence with its characteristic Gly-X-Yrepetitive pattern where X and Y can be any amino acid(Haringkansson amp Reid 2000) In animal MBL the collagenregions comprise 19 Gly-X-Y triplets (Holmskov et al2003) Both porcine MBL-A and MBL-C have the samerepetitive pattern as the MBL of other species including asingle interruption at the middle of the collagen domain asfound in rat MBLs The single interruption in the Gly-X-Y repeat pattern of this protein falls in the same part ofcollagen domain suggesting a significance of interactionbetween this MBL with a common effector protein

Figure 1 Unrooted NJ tree showing phylogenetic relationships of MBL In total 12 predicted amino acid sequences were used for constructing the tree Amino acid sequences are from GenBank (accession no in parentheses) human MBL (NP_000233) chicken MBL (NP_989680) mouse MBL-A and MBL-C (NP_034905 and NP_034906) rat MBL-A and MBL-C (AAH88159 and NP_073195) rhesus monkey MBL-A and MBL-C (AAX84952 and AAX84958) bovine MBL-A and MBL-C (NP_001010994 and NP_776532) porcine MBL-A and MBL-C (NP_001007195 and NP_999290) The numbers at the nodes are the bootstrap scores (percentage of 1000 replicates)

Figure 2 Alignment of two full-length porcine MBL cDNA-deduced amino acid sequences reported by Agah et al (2001) and Lillie et al (2006) Shaded boxes represent identical amino acid residues between both MBLs Dash (ndash) in a sequence indicates a gap Locations of putative structural domains are indicated based on porcine MBL-A (Lillie et al 2006) Cys depicts cysteine residue position GEKGEP indicates C1q receptor interaction site PGKMGP indicates MASP-binding motif and EPN indicates mannose sugar specificity site

Molecular genetic analysis of MBL1 and MBL2 59

copy 2007 The AuthorsJournal compilation copy 2007 Blackwell Publishing Ltd International Journal of Immunogenetics 34 55ndash63

(Drickamer et al 1986) Amino acid consensus sequenceGEKGEP which is involved in C1q receptor (C1qRp)interaction (Arora et al 2001) is present in both porcineMBL This suggests that porcine MBL has the same abilityas C1q to stimulate phagocytosis in the complementsystem (Holmskov et al 2003) The PGKXGP sequencerepresenting part of a putative MASP-binding motif isfound in the porcine MBL-A suggesting its potential toactivate the lectin-complement pathway (Lillie et al 2006)Although this motif is altered in MBL-C (PGMVGP) itsfunction is retained which is confirmed by the result ofporcine MBL-C to functionally activate the lectin path-way in MBL-deficient human sera (Agah et al 2001)The most variable domain in MBL which contains hydro-phobic amino acids necessary for forming the triple helicalcoil (Kawai et al 1998) is also present in the porcineMBLs It is interesting to note that porcine MBL-C missesnine amino acids at this neck region compared to MBL-A This difference between porcine MBL-A and MBL-Cmay be correlated with their α-helix coil formation abilityCarbohydrate recognition domain (CRD) of both porcineMBLs shows a high homology to each other Furthermoreporcine MBL-A and MBL-C contain a mannose-bindingEPN motif (Glu-Pro-Asn) in the CRD This indicates theability of porcine MBLs to recognize the mannose sugar(Stahl amp Ezekowitz 1998)

Expression of porcine MBL genes

RT-PCR of 10 tissues from adult pigs indicated differentialexpression of porcine MBL genes as shown in Fig 3 BothMBL genes were highly expressed in liver Low MBL1expression was also found in the lung testis and brainwhereas low expression of MBL2 was detected in thetestis and kidney Porcine MBL1 expression pattern inthis study is similar to the results reported by Lillie et al(2006) showing high expression of MBL1 in liver aswell Differential expression of MBL1 and MBL2 inmurine tissues was reported by Wagner et al (2003)Real-time RT-PCR study revealed that the liver is themajor site of expression for both MBL genes Also lowexpression was found in the kidney brain spleen andmuscle but only murine MBL1 is expressed in the testis(Wagner et al 2003)

Polymorphisms of porcine MBL genes and genotypes

Two SNPs were found at the positions 579 (G to A) and645 (G to A) of porcine MBL2 cDNA (GenBank acces-sion number NM_214125) in F2 DUMI pigs This twoSNPs were located at codons 193 (AAG) and 215 (GTG)of predicted amino acid sequence (GenBank accessionnumber AAD45377) but did not affect amino acid com-position in the translated protein (Lys and Val at codon193 and 215 respectively) The SNP (at codon 215)affecting an AdeI restriction site was found to be segregat-ing in the F2 DUMI resource population The AdeI PCR-RFLP generates fragments of 319 bp (allele A) 286 bpand 33 bp (allele G) (Fig 4) For MBL2 genotypingshowed allele G and allele A frequencies of 041 and 059respectively The distributions of gene frequencies were021 040 and 039 for GG GA and AA genotypesrespectively Genotyping F2 DUMI pigs at the SNP withinintron 1 of MBL1 (position 328 of the sequenceAF208528) reported by Marklund et al (2000) revealedfrequencies of alleles C and T of 067 and 033 respec-tively Frequencies of genotypes CC CT and TT were048 038 and 014 respectively

Radiation hybrid (RH) and genetic mapping

As shown in Fig 5 the results of RH mapping assignedboth porcine MBL genes to SSC14 with retention fre-quencies of 16 for both genes The most significantlylinked markers (two-point analysis) for porcine MBL1and MBL2 were SW210 (89 cR LOD = 332) and SW1552(35 cR LOD = 1066) respectively The closest markeron the linkage map was S0007 with recombinationfrequencies and two-point LOD scores of 032 334 and023 826 for porcine MBL1 and MBL2 respectivelyMBL2 chromosomal location established by RH mappingin our study is confirmed by gene assignment of porcineMBL2 to position 32260 cR of SSC14 with nearest geneand markers DKK1 and SW1552 respectively as reportedby Meyers et al (2005) and Yasue et al (2006) It hasbeen shown in previous comparative genomic analysesthat a large portion of SSC14 is homologous to humanchromosome 10 (HSA10) Rearrangement of the geneorder on SSC14 involves three regions (46ndash51 74ndash81 and82ndash88 Mb) of HSA10 (Nonneman amp Rohrer 2004) Inaddition Yasue et al (2006) found the HSA10q arm to be

Figure 3 Tissue-specific expression patterns of MBL1 and MBL2 genes assayed by RT-PCR RT-PCR of ribosomal 18S was performed to provide an internal reference

Figure 4 Mendelian inheritance of the G gt A SNP at position 645 of porcine MBL2 (NM_214125) in the F2 DUMI resource population (The 33-bp fragment was not visible in this figure)

60 C Phatsara et al

copy 2007 The AuthorsJournal compilation copy 2007 Blackwell Publishing Ltd International Journal of Immunogenetics 34 55ndash63

corresponding to SSC14q24-qter and their study alsodemonstrated the occurrence of intrachromosomalrearrangements Collectin genes including SFPTA SFPTDand MBL1 were found to be located at a collectin clusterin humans (Guo et al 1998) mouse (Akiyama et al 1999)cattle (Gjerstorff et al 2004) and also in pigs (van Eijket al 2000) The MBL2 gene in human is located onthe same chromosome as the collectin cluster whereas inmouse and bovine it is located on another chromosome(Gjerstorff et al 2004) The comparative data indicatethat the porcine MBL1 might be located between SFPTAand SFPTD genes on SSC14 (Fig 5)

Association and linkage QTL analysis

The association analysis using the MBL1 and MBL2 andtheir interactions with time point revealed no effect onhaemolytic complement activity in classical (CH50) andalternative (AH50) pathways of the SNPs that were ana-lysed (Table 2) C3c protein level which reflects in vivocomplement activity tended to be higher in MBL1 hetero-zygous genotypes (CT) than in the homozygousgenotypes (CC and TT) (P = 0067) There was a highlysignificant effect of time of measurement (P lt 0001)Interactions of time and MBL genotypes in the repeated

Figure 5 Comparative mapping of the 463ndash60 cM region (flanking markers SW210 and S0007) of SSC14 with the human and mouse genome maps Gene positions on SSC14 were taken from Meyers et al (2005) Yasue et al (2006) and van Eijk et al (2000) The position used for human and mouse genes are from NCBI database (Build 362 and Build 361 respectively httpwwwncbinlmnihgovmapview) Dash lines (- - -) indicate chromosomal rearrangement Human MBL1 is the pseudogene MBL1P1

Molecular genetic analysis of MBL1 and MBL2 59

copy 2007 The AuthorsJournal compilation copy 2007 Blackwell Publishing Ltd International Journal of Immunogenetics 34 55ndash63

(Drickamer et al 1986) Amino acid consensus sequenceGEKGEP which is involved in C1q receptor (C1qRp)interaction (Arora et al 2001) is present in both porcineMBL This suggests that porcine MBL has the same abilityas C1q to stimulate phagocytosis in the complementsystem (Holmskov et al 2003) The PGKXGP sequencerepresenting part of a putative MASP-binding motif isfound in the porcine MBL-A suggesting its potential toactivate the lectin-complement pathway (Lillie et al 2006)Although this motif is altered in MBL-C (PGMVGP) itsfunction is retained which is confirmed by the result ofporcine MBL-C to functionally activate the lectin path-way in MBL-deficient human sera (Agah et al 2001)The most variable domain in MBL which contains hydro-phobic amino acids necessary for forming the triple helicalcoil (Kawai et al 1998) is also present in the porcineMBLs It is interesting to note that porcine MBL-C missesnine amino acids at this neck region compared to MBL-A This difference between porcine MBL-A and MBL-Cmay be correlated with their α-helix coil formation abilityCarbohydrate recognition domain (CRD) of both porcineMBLs shows a high homology to each other Furthermoreporcine MBL-A and MBL-C contain a mannose-bindingEPN motif (Glu-Pro-Asn) in the CRD This indicates theability of porcine MBLs to recognize the mannose sugar(Stahl amp Ezekowitz 1998)

Expression of porcine MBL genes

RT-PCR of 10 tissues from adult pigs indicated differentialexpression of porcine MBL genes as shown in Fig 3 BothMBL genes were highly expressed in liver Low MBL1expression was also found in the lung testis and brainwhereas low expression of MBL2 was detected in thetestis and kidney Porcine MBL1 expression pattern inthis study is similar to the results reported by Lillie et al(2006) showing high expression of MBL1 in liver aswell Differential expression of MBL1 and MBL2 inmurine tissues was reported by Wagner et al (2003)Real-time RT-PCR study revealed that the liver is themajor site of expression for both MBL genes Also lowexpression was found in the kidney brain spleen andmuscle but only murine MBL1 is expressed in the testis(Wagner et al 2003)

Polymorphisms of porcine MBL genes and genotypes

Two SNPs were found at the positions 579 (G to A) and645 (G to A) of porcine MBL2 cDNA (GenBank acces-sion number NM_214125) in F2 DUMI pigs This twoSNPs were located at codons 193 (AAG) and 215 (GTG)of predicted amino acid sequence (GenBank accessionnumber AAD45377) but did not affect amino acid com-position in the translated protein (Lys and Val at codon193 and 215 respectively) The SNP (at codon 215)affecting an AdeI restriction site was found to be segregat-ing in the F2 DUMI resource population The AdeI PCR-RFLP generates fragments of 319 bp (allele A) 286 bpand 33 bp (allele G) (Fig 4) For MBL2 genotypingshowed allele G and allele A frequencies of 041 and 059respectively The distributions of gene frequencies were021 040 and 039 for GG GA and AA genotypesrespectively Genotyping F2 DUMI pigs at the SNP withinintron 1 of MBL1 (position 328 of the sequenceAF208528) reported by Marklund et al (2000) revealedfrequencies of alleles C and T of 067 and 033 respec-tively Frequencies of genotypes CC CT and TT were048 038 and 014 respectively

Radiation hybrid (RH) and genetic mapping

As shown in Fig 5 the results of RH mapping assignedboth porcine MBL genes to SSC14 with retention fre-quencies of 16 for both genes The most significantlylinked markers (two-point analysis) for porcine MBL1and MBL2 were SW210 (89 cR LOD = 332) and SW1552(35 cR LOD = 1066) respectively The closest markeron the linkage map was S0007 with recombinationfrequencies and two-point LOD scores of 032 334 and023 826 for porcine MBL1 and MBL2 respectivelyMBL2 chromosomal location established by RH mappingin our study is confirmed by gene assignment of porcineMBL2 to position 32260 cR of SSC14 with nearest geneand markers DKK1 and SW1552 respectively as reportedby Meyers et al (2005) and Yasue et al (2006) It hasbeen shown in previous comparative genomic analysesthat a large portion of SSC14 is homologous to humanchromosome 10 (HSA10) Rearrangement of the geneorder on SSC14 involves three regions (46ndash51 74ndash81 and82ndash88 Mb) of HSA10 (Nonneman amp Rohrer 2004) Inaddition Yasue et al (2006) found the HSA10q arm to be

Figure 3 Tissue-specific expression patterns of MBL1 and MBL2 genes assayed by RT-PCR RT-PCR of ribosomal 18S was performed to provide an internal reference

Figure 4 Mendelian inheritance of the G gt A SNP at position 645 of porcine MBL2 (NM_214125) in the F2 DUMI resource population (The 33-bp fragment was not visible in this figure)

60 C Phatsara et al

copy 2007 The AuthorsJournal compilation copy 2007 Blackwell Publishing Ltd International Journal of Immunogenetics 34 55ndash63

corresponding to SSC14q24-qter and their study alsodemonstrated the occurrence of intrachromosomalrearrangements Collectin genes including SFPTA SFPTDand MBL1 were found to be located at a collectin clusterin humans (Guo et al 1998) mouse (Akiyama et al 1999)cattle (Gjerstorff et al 2004) and also in pigs (van Eijket al 2000) The MBL2 gene in human is located onthe same chromosome as the collectin cluster whereas inmouse and bovine it is located on another chromosome(Gjerstorff et al 2004) The comparative data indicatethat the porcine MBL1 might be located between SFPTAand SFPTD genes on SSC14 (Fig 5)

Association and linkage QTL analysis

The association analysis using the MBL1 and MBL2 andtheir interactions with time point revealed no effect onhaemolytic complement activity in classical (CH50) andalternative (AH50) pathways of the SNPs that were ana-lysed (Table 2) C3c protein level which reflects in vivocomplement activity tended to be higher in MBL1 hetero-zygous genotypes (CT) than in the homozygousgenotypes (CC and TT) (P = 0067) There was a highlysignificant effect of time of measurement (P lt 0001)Interactions of time and MBL genotypes in the repeated

Figure 5 Comparative mapping of the 463ndash60 cM region (flanking markers SW210 and S0007) of SSC14 with the human and mouse genome maps Gene positions on SSC14 were taken from Meyers et al (2005) Yasue et al (2006) and van Eijk et al (2000) The position used for human and mouse genes are from NCBI database (Build 362 and Build 361 respectively httpwwwncbinlmnihgovmapview) Dash lines (- - -) indicate chromosomal rearrangement Human MBL1 is the pseudogene MBL1P1

60 C Phatsara et al

copy 2007 The AuthorsJournal compilation copy 2007 Blackwell Publishing Ltd International Journal of Immunogenetics 34 55ndash63

corresponding to SSC14q24-qter and their study alsodemonstrated the occurrence of intrachromosomalrearrangements Collectin genes including SFPTA SFPTDand MBL1 were found to be located at a collectin clusterin humans (Guo et al 1998) mouse (Akiyama et al 1999)cattle (Gjerstorff et al 2004) and also in pigs (van Eijket al 2000) The MBL2 gene in human is located onthe same chromosome as the collectin cluster whereas inmouse and bovine it is located on another chromosome(Gjerstorff et al 2004) The comparative data indicatethat the porcine MBL1 might be located between SFPTAand SFPTD genes on SSC14 (Fig 5)

Association and linkage QTL analysis

The association analysis using the MBL1 and MBL2 andtheir interactions with time point revealed no effect onhaemolytic complement activity in classical (CH50) andalternative (AH50) pathways of the SNPs that were ana-lysed (Table 2) C3c protein level which reflects in vivocomplement activity tended to be higher in MBL1 hetero-zygous genotypes (CT) than in the homozygousgenotypes (CC and TT) (P = 0067) There was a highlysignificant effect of time of measurement (P lt 0001)Interactions of time and MBL genotypes in the repeated

Figure 5 Comparative mapping of the 463ndash60 cM region (flanking markers SW210 and S0007) of SSC14 with the human and mouse genome maps Gene positions on SSC14 were taken from Meyers et al (2005) Yasue et al (2006) and van Eijk et al (2000) The position used for human and mouse genes are from NCBI database (Build 362 and Build 361 respectively httpwwwncbinlmnihgovmapview) Dash lines (- - -) indicate chromosomal rearrangement Human MBL1 is the pseudogene MBL1P1

Molecular genetic analysis of MBL1 and MBL2 61

copy 2007 The AuthorsJournal compilation copy 2007 Blackwell Publishing Ltd International Journal of Immunogenetics 34 55ndash63

measures models reflect the dependency of the profile ofcomplement activity along the experiment from the geno-type (Fig 6) For in vivo complement activity a slightMBL1 genotype-dependent deviation of the profiles ofC3c concentration over time was found (P = 0056) Thisdeviation is most prominent late after Mh vaccination(times 3 and 4) (Fig 6) No significant effect of interactionbetween genotypes and time point on haemolytic com-plement activity assayed in the alternative and classical

pathway was found The profiles of the haemolytic com-plement activities AH50 and CH50 between differentgenotypes of porcine MBL1 and MBL2 were similar overtime as shown in Fig 6 Both parameters of in vivo haemo-lytic complement activity CH50 and AH50 depend onsequences of the complement cascade that do not directlyinvolve MBL1 or MBL2 In contrast C3c serum con-centration reflects in vivo complement activation after thevaccination that may act on the lectin pathway controlledby MBL Through the activation of complement systemMBL has been found to trigger complement activityresulting in C3 activation and release of C3 fragments (Pre-sanis et al 2003) Interestingly it has recently been shownthat human MBL binds several Mycoplasma strains(Hamvas et al 2005) The lack of association of the SNPsin MBL1 and MBL2 with haemolytic complement activ-ity emphasizes the specificity of the results obtained forC3c serum concentration Moreover linkage analysisrevealed a genome-wide significant QTL for C3c serumconcentration late after Mh vaccination (time 4) on SSC14in the interval of MBL1 and MBL2 (Fig 7) Significantdominance effects were found at the QTL site corre-sponding to the finding of differences between hetero-zygous and homozygous MBL1 genotypes The plot of theF ratio from least square interval mapping for evidence ofQTL for C3c serum concentration represents at each posi-tion a sum-up of effects depending on the flanking markersFor instance in mouse there is an increasing number ofQTL studies where large QTL when fine mapped turnedout to be the result of multiple linked loci Thus the posi-tion of the QTL might be a symptom of effects of the two

Table 2 Least square means of haemolytic complement activity traits (AH50 CH50) and C3c serum concentration for the effect of MBL1 and MBL2 genotypes in DUMI resource population

Genotype AH50 CH50 C3c

MBL1CC 5629 plusmn 247 6812 plusmn 331 0190 plusmn 0004CT 5813 plusmn 260 6525 plusmn 355 0198 plusmn 0005TT 5427 plusmn 340 6290 plusmn 431 0192 plusmn 0005

Effect (P )MBL1 0355 0313 0067Time lt 0001 lt 0001 lt 0001MBL1time 0690 0479 0056

MBL2GG 5824 plusmn 314 6759 plusmn 436 0201 plusmn 0006GA 6058 plusmn 260 6889 plusmn 350 0201 plusmn 0005AA 5669 plusmn 263 6554 plusmn 361 0194 plusmn 0005

Effect (P )MBL2 0141 0499 0136Time lt 0001 lt 0001 lt 0001MBL2time 0664 0723 0967

Figure 6 Plots of least square means of haemolytic complement activity in the alternative (AH50 U mLminus1) and classical pathway (CH50 U mLminus1) and C3c protein level (g Lminus1) for the interaction time points and porcine MBL1 and MBL2 genotypes

62 C Phatsara et al

copy 2007 The AuthorsJournal compilation copy 2007 Blackwell Publishing Ltd International Journal of Immunogenetics 34 55ndash63

MBL genes Association analyses of the intronic and synony-mous SNPs of MBL1 and MBL2 failed to consistently revealsignificant effects probably as a result of the fact that theyare not in linkage disequilibrium with a causative poly-morphism A causative polymorphism may either exist inthe regulatory regions of the genes affecting its expressionor at coding sites where non-synonymous nucleotide pol-ymorphisms may affect the protein function In humansthree SNPs have been identified in exon 1 at codons 52(Arg rarr Cys) 54 (Gly rarr Asp) and 57 (Gly rarr Glu) (Sumiyaet al 1991 Lipscombe et al 1993 Madsen et al 1994)showing association with many innate immunological factors(Holmskov et al 2003) Furthermore the point mutationsin exon 1 of the human MBL gene are frequently describedas being associated with MBL plasma concentrationreduced ligand-binding capacity and failure to activatecomplement (Larsen et al 2004) Also polymorphismsin the MBL promoter have been shown to be associatedwith prevalence of infectious diseases (Mullighanet al 2002 Hamvas et al 2005 Zhang et al 2005)

In summary least square means of C3c serum concen-tration ie in vivo complement activity estimated forMBL1 genotypes differed at (P lt 01) Correspondinglylinkage analysis revealed a QTL for C3c serum level closeto the position of the MBL genes The study thus pro-motes the porcine MBL genes as functional and positionalcandidate gene for in vivo complement activity mediatedvia the lectin pathway

Acknowledgements

The experiments described in this study were performedin accordance with all appropriate regulations Animalused in this study were kept at the Research Station Franken-forst of the Institute of Animal Science University ofBonn Germany

References

Agah A Montalto MC Young K amp Stahl GL (2001) Isolation cloning and functional characterization of porcine mannose-binding lectin Immunology 102 338

Akiyama J Volik SV Plajzer-Frick I Prince A Sago H Weier HUG Vanderbilt JN Hawgood S amp Poulain FR (1999) Characterization of the mouse collectin gene locus American Journal of Respiratory Cell and Molecular Biology 21 193

Arora M Munoz E amp Tenner AJ (2001) Identification of a site on mannan-binding lectin critical for enhancement of phagocytosis Journal of Biological Chemistry 276 43087

Churchill GA amp Doerge RW (1994) Empirical threshold values for quantitative trait mapping Genetics 138 963

Drickamer K Dordal MS amp Reynolds L (1986) Mannose-binding proteins isolated from rat liver contain carbohydrate recognition domains linked to collagenous tails mdash complete primary structures and homology with pulmonary surfactant apoprotein Journal of Biological Chemistry 261 6878

van Eijk M Haagsman HP Skinner T Archibald A Reid KBM amp Lawson PR (2000) Porcine lung surfactant protein D complementary DNA cloning chromosomal localization and tissue distribution Journal of Immunology 164 1442

Gjerstorff M Hansen S Jensen B Dueholm B Horn P Bendixen C amp Holmskov U (2004) The genes encoding bovine SP-A SP-D MBL-A conglutinin CL-43 and CL-46 form a distinct collectin locus on Bos taurus chromosome 28 (BTA28) at position q189 Animal Genetics 35 333

Green P Falls K amp Crooks S (1990) Documentation for crimap version 24 Washington University School of Medicine St Louis MO

Guo N Mogues T Weremowicz S Morton CC amp Sastry KN (1998) The human ortholog of rhesus mannose-binding protein A gene is an expressed pseudogene that localizes to chromosome 10 Mammalian Genome 9 246

Haringkansson K amp Reid KBM (2000) Collectin structure a review Protein Science 9 1607

Hamvas RMJ Johnson M Vlieger AM Ling C Sherriff A Wade A Klein NJ Turner MW amp Webster ADB (2005) Role for mannose-binding lectin in the prevention of mycoplasma infection Infection and Immunity 73 5238

Hansen S Thiel S Willis A Holmskov U amp Jensenius JC (2000) Purification and characterization of two mannan-binding lectins from mouse serum Journal of Immunology 164 2610

Hawken RJ Murtaugh J Flickinger GH Yerle M Robic A Milan D Gellin J Beattie CW Schook LB amp Alexander LJ (1999) A first-generation porcine whole-genome radiation hybrid map Mammalian Genome 10 824

Holmskov U Thiel S amp Jensenius JC (2003) Collectins and ficolins humoral lectins of the innate immune defense Annual Review of Immunology 21 547

Kawai T Suzuki Y Eda S Ohtani K Kase T Fujinaga Y Sakamoto T Kurimura T amp Wakamiya N (1997) Cloning and characterization of a cDNA encoding bovine mannan-binding protein Gene 186 161

Kawai T Suzuki Y Eda S Ohtani K Kase T Sakamoto T Uemura H amp Wakamiya N (1998) Molecular and biological characterization of rabbit mannan-binding protein (MBP) Glycobiology 8 237

Kumar S Tamura K amp Nei M (2004) mega3 integrated software for molecular evolutionary genetics analysis and sequence alignment Briefings in Bioinformatics 5 150

Larsen F Madsen HO Sim RB Koch C amp Garred P (2004) Disease-associated mutations in human mannose-binding lectin compromise oligomerization and activity of the final protein Journal of Biological Chemistry 279 21302

Figure 7 Plot of the F-ratio from least square interval mapping for evidence of QTL for C3c serum concentration after Mh and prior to ADV vaccination (time 4) on SSC14 The x-axis indicates the relative position on the linkage map Arrows on the x-axis indicate the position of markers The y-axis represents the F-value Lines indicate the 5 genome-wide and 5 chromosome-wide significance thresholdsQTL at 24 cM with F-value = 82 LOD = 35 additive genetic effect = 0004 plusmn 0006 dominance effect = 0051 plusmn 0012 fraction of phenotypic variance in the F2 explained by the QTL = 4

Molecular genetic analysis of MBL1 and MBL2 63

copy 2007 The AuthorsJournal compilation copy 2007 Blackwell Publishing Ltd International Journal of Immunogenetics 34 55ndash63

Laursen SB amp Nielsen OL (2000) Mannan-binding lectin (MBL) in chickens molecular and functional aspects Developmental and Comparative Immunology 24 85

Lillie BN Hammermueller JD MacInnes JI Jacques M amp Hayes MA (2006) Porcine mannan-binding lectin A binds to Actinobacillus suis and Haemophilus parasuis Developmental and Comparative Immunology 30 954

Lipscombe RJ Sumiya M Hill AVS Lau YL Levinsky RJ Summerfield JA amp Turner MW (1993) High frequencies in African and non-African populations of independent mutations in the mannose binding-protein gene Human Molecular Genetics 2 342

Madsen HO Garred P Kurtzhals JAL Lamm LU Ryder LP Thiel S amp Svejgaard A (1994) A new frequent allele is the missing link in the structural polymorphism of the human mannan-binding protein Immunogenetics 40 37

Marklund L Shi X amp Tuggle CK (2000) Rapid communication mapping of the mannose-binding lectin 2 (MBL2) gene to pig chromosome 14 Journal of Animal Science 78 2992

Matsushita M amp Fujita T (1992) Activation of the classical complement pathway by mannose-binding protein in association with a novel C1s-like serine protease Journal of Experimental Medicine 176 1497

Meyers SN Rogatcheva MB Larkin DM Yerle M Milan D Hawken RJ Schook LB amp Beever JE (2005) Piggy-BACing the human genome II A high-resolution physically anchored comparative map of the porcine autosomes Genomics 86 739

Mizuno Y Kozutsumi Y Kawasaki T amp Yamashina I (1981) Isolation and characterization of a mannan-binding protein from rat liver Journal of Biological Chemistry 256 4247

Mogues T Ota T Tauber AI amp Sastry KN (1996) Characterization of two mannose-binding protein cDNAs from rhesus monkey (Macaca mulatta) structure and evolutionary implications Glycobiology 6 543

Mullighan CG Heatley S Doherty K Szabo F Grigg A Hughes TP et al (2002) Mannose-binding lectin gene polymorphisms are associated with major infection following allogeneic hemopoietic stem cell transplantation Blood 99 3524

Nonneman D amp Rohrer GA (2004) Comparative mapping of

human chromosome 10 to pig chromosomes 10 and 14 Animal Genetics 35 338

OrsquoConnell JR amp Weeks DE (1998) pedcheck a program for identification of genotype incompatibilities in linkage analysis American Journal of Human Genetics 63 259

Presanis JS Kojima M amp Sim RB (2003) Biochemistry and genetics of mannan-binding lectin (MBL) Biochemical Society Transactions 31 748

Seaton G Haley CS Knott SA Kearsey M amp Visscher PM (2002) QTL express mapping quantitative trait loci in of simple and complex pedigrees Bioinformatics 18 339

Stahl PD amp Ezekowitz RAB (1998) The mannose receptor is a pattern recognition receptor involved in host defense Current Opinion in Immunology 10 50

Sumiya M Super M Tabona P Levinsky RJ Arai T Turner MW amp Summerfield JA (1991) Molecular basis of opsonic defect in immunodeficient children Lancet 337 1569

Wagner S Lynch NJ Walter W Schwaeble WJ amp Loos M (2003) Differential expression of the murine mannose-binding lectins A and C in lymphoid and nonlymphoid organs and tissues Journal of Immunology 170 1462

Wimmers K Lipperheide C Ponsuksili S Schmoll F Hardge T Petersen B amp Schellander K (1999) Haemolytic complement activity and C3c serum concentration in pigs Archiv fuumlr Tierzucht 42 93

Wimmers K Mekchay S Schellander K amp Ponsuksili S (2003) Molecular characterization of the pig C3 gene and its association with complement activity Immunogenetics 54 714

Yasue H Kiuchi S Hiraiwa H Ozawa A amp Hayashi T (2006) Assignment of 101 genes localized in HSA10 to a swine RH (IMpRH) map to generate a dense humanndashswine comparative map Cytogenetic and Genome Research 112 121

Yerle M Pinton P Robic A Alfonso A Palvadeau Y Delcros C et al (1998) Construction of a whole genome radiation hybrid panel for high-resolution gene mapping in pigs Cytogenetics and Cell Genetics 82 182

Zhang HX Zhou GQ Zhi LT Yang H Zhai Y Dong XJ Zhang XM Gao X Zhu YP amp He FC (2005) Association between mannose-binding lectin gene polymorphisms and susceptibility to severe acute respiratory syndrome coronavirus infection Journal of Infectious Diseases 192 1355

62 C Phatsara et al

copy 2007 The AuthorsJournal compilation copy 2007 Blackwell Publishing Ltd International Journal of Immunogenetics 34 55ndash63

MBL genes Association analyses of the intronic and synony-mous SNPs of MBL1 and MBL2 failed to consistently revealsignificant effects probably as a result of the fact that theyare not in linkage disequilibrium with a causative poly-morphism A causative polymorphism may either exist inthe regulatory regions of the genes affecting its expressionor at coding sites where non-synonymous nucleotide pol-ymorphisms may affect the protein function In humansthree SNPs have been identified in exon 1 at codons 52(Arg rarr Cys) 54 (Gly rarr Asp) and 57 (Gly rarr Glu) (Sumiyaet al 1991 Lipscombe et al 1993 Madsen et al 1994)showing association with many innate immunological factors(Holmskov et al 2003) Furthermore the point mutationsin exon 1 of the human MBL gene are frequently describedas being associated with MBL plasma concentrationreduced ligand-binding capacity and failure to activatecomplement (Larsen et al 2004) Also polymorphismsin the MBL promoter have been shown to be associatedwith prevalence of infectious diseases (Mullighanet al 2002 Hamvas et al 2005 Zhang et al 2005)

In summary least square means of C3c serum concen-tration ie in vivo complement activity estimated forMBL1 genotypes differed at (P lt 01) Correspondinglylinkage analysis revealed a QTL for C3c serum level closeto the position of the MBL genes The study thus pro-motes the porcine MBL genes as functional and positionalcandidate gene for in vivo complement activity mediatedvia the lectin pathway

Acknowledgements

The experiments described in this study were performedin accordance with all appropriate regulations Animalused in this study were kept at the Research Station Franken-forst of the Institute of Animal Science University ofBonn Germany

References

Agah A Montalto MC Young K amp Stahl GL (2001) Isolation cloning and functional characterization of porcine mannose-binding lectin Immunology 102 338

Akiyama J Volik SV Plajzer-Frick I Prince A Sago H Weier HUG Vanderbilt JN Hawgood S amp Poulain FR (1999) Characterization of the mouse collectin gene locus American Journal of Respiratory Cell and Molecular Biology 21 193

Arora M Munoz E amp Tenner AJ (2001) Identification of a site on mannan-binding lectin critical for enhancement of phagocytosis Journal of Biological Chemistry 276 43087

Churchill GA amp Doerge RW (1994) Empirical threshold values for quantitative trait mapping Genetics 138 963

Drickamer K Dordal MS amp Reynolds L (1986) Mannose-binding proteins isolated from rat liver contain carbohydrate recognition domains linked to collagenous tails mdash complete primary structures and homology with pulmonary surfactant apoprotein Journal of Biological Chemistry 261 6878

van Eijk M Haagsman HP Skinner T Archibald A Reid KBM amp Lawson PR (2000) Porcine lung surfactant protein D complementary DNA cloning chromosomal localization and tissue distribution Journal of Immunology 164 1442

Gjerstorff M Hansen S Jensen B Dueholm B Horn P Bendixen C amp Holmskov U (2004) The genes encoding bovine SP-A SP-D MBL-A conglutinin CL-43 and CL-46 form a distinct collectin locus on Bos taurus chromosome 28 (BTA28) at position q189 Animal Genetics 35 333

Green P Falls K amp Crooks S (1990) Documentation for crimap version 24 Washington University School of Medicine St Louis MO

Guo N Mogues T Weremowicz S Morton CC amp Sastry KN (1998) The human ortholog of rhesus mannose-binding protein A gene is an expressed pseudogene that localizes to chromosome 10 Mammalian Genome 9 246

Haringkansson K amp Reid KBM (2000) Collectin structure a review Protein Science 9 1607

Hamvas RMJ Johnson M Vlieger AM Ling C Sherriff A Wade A Klein NJ Turner MW amp Webster ADB (2005) Role for mannose-binding lectin in the prevention of mycoplasma infection Infection and Immunity 73 5238

Hansen S Thiel S Willis A Holmskov U amp Jensenius JC (2000) Purification and characterization of two mannan-binding lectins from mouse serum Journal of Immunology 164 2610

Hawken RJ Murtaugh J Flickinger GH Yerle M Robic A Milan D Gellin J Beattie CW Schook LB amp Alexander LJ (1999) A first-generation porcine whole-genome radiation hybrid map Mammalian Genome 10 824

Holmskov U Thiel S amp Jensenius JC (2003) Collectins and ficolins humoral lectins of the innate immune defense Annual Review of Immunology 21 547

Kawai T Suzuki Y Eda S Ohtani K Kase T Fujinaga Y Sakamoto T Kurimura T amp Wakamiya N (1997) Cloning and characterization of a cDNA encoding bovine mannan-binding protein Gene 186 161

Kawai T Suzuki Y Eda S Ohtani K Kase T Sakamoto T Uemura H amp Wakamiya N (1998) Molecular and biological characterization of rabbit mannan-binding protein (MBP) Glycobiology 8 237

Kumar S Tamura K amp Nei M (2004) mega3 integrated software for molecular evolutionary genetics analysis and sequence alignment Briefings in Bioinformatics 5 150

Larsen F Madsen HO Sim RB Koch C amp Garred P (2004) Disease-associated mutations in human mannose-binding lectin compromise oligomerization and activity of the final protein Journal of Biological Chemistry 279 21302

Figure 7 Plot of the F-ratio from least square interval mapping for evidence of QTL for C3c serum concentration after Mh and prior to ADV vaccination (time 4) on SSC14 The x-axis indicates the relative position on the linkage map Arrows on the x-axis indicate the position of markers The y-axis represents the F-value Lines indicate the 5 genome-wide and 5 chromosome-wide significance thresholdsQTL at 24 cM with F-value = 82 LOD = 35 additive genetic effect = 0004 plusmn 0006 dominance effect = 0051 plusmn 0012 fraction of phenotypic variance in the F2 explained by the QTL = 4

Molecular genetic analysis of MBL1 and MBL2 63

copy 2007 The AuthorsJournal compilation copy 2007 Blackwell Publishing Ltd International Journal of Immunogenetics 34 55ndash63

Laursen SB amp Nielsen OL (2000) Mannan-binding lectin (MBL) in chickens molecular and functional aspects Developmental and Comparative Immunology 24 85

Lillie BN Hammermueller JD MacInnes JI Jacques M amp Hayes MA (2006) Porcine mannan-binding lectin A binds to Actinobacillus suis and Haemophilus parasuis Developmental and Comparative Immunology 30 954

Lipscombe RJ Sumiya M Hill AVS Lau YL Levinsky RJ Summerfield JA amp Turner MW (1993) High frequencies in African and non-African populations of independent mutations in the mannose binding-protein gene Human Molecular Genetics 2 342

Madsen HO Garred P Kurtzhals JAL Lamm LU Ryder LP Thiel S amp Svejgaard A (1994) A new frequent allele is the missing link in the structural polymorphism of the human mannan-binding protein Immunogenetics 40 37

Marklund L Shi X amp Tuggle CK (2000) Rapid communication mapping of the mannose-binding lectin 2 (MBL2) gene to pig chromosome 14 Journal of Animal Science 78 2992

Matsushita M amp Fujita T (1992) Activation of the classical complement pathway by mannose-binding protein in association with a novel C1s-like serine protease Journal of Experimental Medicine 176 1497

Meyers SN Rogatcheva MB Larkin DM Yerle M Milan D Hawken RJ Schook LB amp Beever JE (2005) Piggy-BACing the human genome II A high-resolution physically anchored comparative map of the porcine autosomes Genomics 86 739

Mizuno Y Kozutsumi Y Kawasaki T amp Yamashina I (1981) Isolation and characterization of a mannan-binding protein from rat liver Journal of Biological Chemistry 256 4247

Mogues T Ota T Tauber AI amp Sastry KN (1996) Characterization of two mannose-binding protein cDNAs from rhesus monkey (Macaca mulatta) structure and evolutionary implications Glycobiology 6 543

Mullighan CG Heatley S Doherty K Szabo F Grigg A Hughes TP et al (2002) Mannose-binding lectin gene polymorphisms are associated with major infection following allogeneic hemopoietic stem cell transplantation Blood 99 3524

Nonneman D amp Rohrer GA (2004) Comparative mapping of

human chromosome 10 to pig chromosomes 10 and 14 Animal Genetics 35 338

OrsquoConnell JR amp Weeks DE (1998) pedcheck a program for identification of genotype incompatibilities in linkage analysis American Journal of Human Genetics 63 259

Presanis JS Kojima M amp Sim RB (2003) Biochemistry and genetics of mannan-binding lectin (MBL) Biochemical Society Transactions 31 748

Seaton G Haley CS Knott SA Kearsey M amp Visscher PM (2002) QTL express mapping quantitative trait loci in of simple and complex pedigrees Bioinformatics 18 339

Stahl PD amp Ezekowitz RAB (1998) The mannose receptor is a pattern recognition receptor involved in host defense Current Opinion in Immunology 10 50

Sumiya M Super M Tabona P Levinsky RJ Arai T Turner MW amp Summerfield JA (1991) Molecular basis of opsonic defect in immunodeficient children Lancet 337 1569

Wagner S Lynch NJ Walter W Schwaeble WJ amp Loos M (2003) Differential expression of the murine mannose-binding lectins A and C in lymphoid and nonlymphoid organs and tissues Journal of Immunology 170 1462

Wimmers K Lipperheide C Ponsuksili S Schmoll F Hardge T Petersen B amp Schellander K (1999) Haemolytic complement activity and C3c serum concentration in pigs Archiv fuumlr Tierzucht 42 93

Wimmers K Mekchay S Schellander K amp Ponsuksili S (2003) Molecular characterization of the pig C3 gene and its association with complement activity Immunogenetics 54 714

Yasue H Kiuchi S Hiraiwa H Ozawa A amp Hayashi T (2006) Assignment of 101 genes localized in HSA10 to a swine RH (IMpRH) map to generate a dense humanndashswine comparative map Cytogenetic and Genome Research 112 121

Yerle M Pinton P Robic A Alfonso A Palvadeau Y Delcros C et al (1998) Construction of a whole genome radiation hybrid panel for high-resolution gene mapping in pigs Cytogenetics and Cell Genetics 82 182

Zhang HX Zhou GQ Zhi LT Yang H Zhai Y Dong XJ Zhang XM Gao X Zhu YP amp He FC (2005) Association between mannose-binding lectin gene polymorphisms and susceptibility to severe acute respiratory syndrome coronavirus infection Journal of Infectious Diseases 192 1355

Molecular genetic analysis of MBL1 and MBL2 63

copy 2007 The AuthorsJournal compilation copy 2007 Blackwell Publishing Ltd International Journal of Immunogenetics 34 55ndash63

Laursen SB amp Nielsen OL (2000) Mannan-binding lectin (MBL) in chickens molecular and functional aspects Developmental and Comparative Immunology 24 85

Lillie BN Hammermueller JD MacInnes JI Jacques M amp Hayes MA (2006) Porcine mannan-binding lectin A binds to Actinobacillus suis and Haemophilus parasuis Developmental and Comparative Immunology 30 954

Lipscombe RJ Sumiya M Hill AVS Lau YL Levinsky RJ Summerfield JA amp Turner MW (1993) High frequencies in African and non-African populations of independent mutations in the mannose binding-protein gene Human Molecular Genetics 2 342

Madsen HO Garred P Kurtzhals JAL Lamm LU Ryder LP Thiel S amp Svejgaard A (1994) A new frequent allele is the missing link in the structural polymorphism of the human mannan-binding protein Immunogenetics 40 37

Marklund L Shi X amp Tuggle CK (2000) Rapid communication mapping of the mannose-binding lectin 2 (MBL2) gene to pig chromosome 14 Journal of Animal Science 78 2992

Matsushita M amp Fujita T (1992) Activation of the classical complement pathway by mannose-binding protein in association with a novel C1s-like serine protease Journal of Experimental Medicine 176 1497

Meyers SN Rogatcheva MB Larkin DM Yerle M Milan D Hawken RJ Schook LB amp Beever JE (2005) Piggy-BACing the human genome II A high-resolution physically anchored comparative map of the porcine autosomes Genomics 86 739

Mizuno Y Kozutsumi Y Kawasaki T amp Yamashina I (1981) Isolation and characterization of a mannan-binding protein from rat liver Journal of Biological Chemistry 256 4247

Mogues T Ota T Tauber AI amp Sastry KN (1996) Characterization of two mannose-binding protein cDNAs from rhesus monkey (Macaca mulatta) structure and evolutionary implications Glycobiology 6 543

Mullighan CG Heatley S Doherty K Szabo F Grigg A Hughes TP et al (2002) Mannose-binding lectin gene polymorphisms are associated with major infection following allogeneic hemopoietic stem cell transplantation Blood 99 3524

Nonneman D amp Rohrer GA (2004) Comparative mapping of

human chromosome 10 to pig chromosomes 10 and 14 Animal Genetics 35 338

OrsquoConnell JR amp Weeks DE (1998) pedcheck a program for identification of genotype incompatibilities in linkage analysis American Journal of Human Genetics 63 259

Presanis JS Kojima M amp Sim RB (2003) Biochemistry and genetics of mannan-binding lectin (MBL) Biochemical Society Transactions 31 748

Seaton G Haley CS Knott SA Kearsey M amp Visscher PM (2002) QTL express mapping quantitative trait loci in of simple and complex pedigrees Bioinformatics 18 339

Stahl PD amp Ezekowitz RAB (1998) The mannose receptor is a pattern recognition receptor involved in host defense Current Opinion in Immunology 10 50

Sumiya M Super M Tabona P Levinsky RJ Arai T Turner MW amp Summerfield JA (1991) Molecular basis of opsonic defect in immunodeficient children Lancet 337 1569

Wagner S Lynch NJ Walter W Schwaeble WJ amp Loos M (2003) Differential expression of the murine mannose-binding lectins A and C in lymphoid and nonlymphoid organs and tissues Journal of Immunology 170 1462

Wimmers K Lipperheide C Ponsuksili S Schmoll F Hardge T Petersen B amp Schellander K (1999) Haemolytic complement activity and C3c serum concentration in pigs Archiv fuumlr Tierzucht 42 93

Wimmers K Mekchay S Schellander K amp Ponsuksili S (2003) Molecular characterization of the pig C3 gene and its association with complement activity Immunogenetics 54 714

Yasue H Kiuchi S Hiraiwa H Ozawa A amp Hayashi T (2006) Assignment of 101 genes localized in HSA10 to a swine RH (IMpRH) map to generate a dense humanndashswine comparative map Cytogenetic and Genome Research 112 121

Yerle M Pinton P Robic A Alfonso A Palvadeau Y Delcros C et al (1998) Construction of a whole genome radiation hybrid panel for high-resolution gene mapping in pigs Cytogenetics and Cell Genetics 82 182

Zhang HX Zhou GQ Zhi LT Yang H Zhai Y Dong XJ Zhang XM Gao X Zhu YP amp He FC (2005) Association between mannose-binding lectin gene polymorphisms and susceptibility to severe acute respiratory syndrome coronavirus infection Journal of Infectious Diseases 192 1355