Journal of Reproductive Immunology 82 (2009) 48–56

Classical and non-classical Major Histocompatibility Complexclass I gene expression in in vitro derived bovine embryos

Julie Doyle a, Shirley A. Ellis b, Grace M. O’Gorman a, Ines Maria Aparicio Donoso a,Patrick Lonergan a, Trudee Fair a,∗

a Agriculture, Food Science & Veterinary Medicine, University College Dublin, Belfield, Dublin 4, Irelandb Bovine Molecular Immunology Group, Institute for Animal Health, Compton RG20 7NN, UK

Received 13 October 2008; received in revised form 12 June 2009; accepted 17 June 2009

Abstract

The role of the Major Histocompatibility Complex class I (MHC-I) genes in human and mouse preimplantation embryo develop-ment has received considerable attention. In contrast, information concerning the role of these genes in bovine embryo developmentis limited. The objective of the current study was to characterize the expression pattern of MHC-I genes during preimplantationembryo development in cattle. To this end, bovine oocytes were harvested from slaughterhouse ovaries, matured, fertilized andcultured in vitro. Samples were collected at immature and mature oocyte, presumptive zygote, 2–4-cell embryo, 8–16-cell embryo,

morula, blastocyst and hatched blastocyst stages of development. MHC-I expression was detected using quantitative real-time-PCR,cDNA sequencing, whole mount immunocytochemistry and Western blotting. We report classical and non-classical MHC-I mRNAexpression in bovine oocytes and developing embryos. Furthermore, we report that the pattern of MHC-I mRNA expression acrossdevelopment was gene- and stage-specific.© 2009 Elsevier Ireland Ltd. All rights reserved.

MHC-I

Keywords: Oocyte; Preimplantation Embryo; mRNA; Development;

1. Introduction

Studies in human and mouse models have shown thatthe Major Histocompatibility Complex class I (MHC-I)genes are involved at a number of levels in the estab-lishment of pregnancy (Warner et al., 1993; Comiskeyet al., 2003). The data gathered from these models have

been conflicting with regard to the expression of classi-cal and non-classical MHC-I genes in male and femalegametes and early cleavage stage embryos. However,

∗ Corresponding author at: Room 246, Veterinary Sciences Centre,School of Agriculture, Food Science & Veterinary Medicine, Col-lege of Life Sciences, University College Dublin, Belfield, Dublin 4,Ireland. Tel.: +353 1 6012147; fax: +353 1 6288421.

E-mail address: trudee.fair@ucd.ie (T. Fair).

0165-0378/$ – see front matter © 2009 Elsevier Ireland Ltd. All rights reservdoi:10.1016/j.jri.2009.06.125

two non-classical MHC-I genes have been identifiedand associated with embryo development and survival,namely the murine Ped gene, which encodes the Qa-2antigen, and the human HLA-G gene, of which there areseveral isoforms (reviewed by Warner et al., 2002). Stud-ies by that group describe pleiotropic effects in termsof increased growth rates, increased rates of cell cleav-age, increased embryo survival to term and a numberof effects on postnatal physiology in mice expressingthe Qa-2-encoding genes (Warner et al., 1991, 1993;Exley and Warner, 1999; Watkins et al., 2006). The roleof the MHC-I has received considerable interest in the

field of human assisted fertilization as recent studiesreport that the secretion of a soluble form of the non-classical MHC-I molecule HLA-G (sHLA-G) by in vitrofertilized (IVF) embryos is associated with higher preg-

ed.

oductive

nS2eelatbdiid

kIIagas(daliteffiToi2

wetideoewvaptano

J. Doyle et al. / Journal of Repr

ancy rates following embryo transfer (Fuzzi et al., 2002;her et al., 2004, 2005a,b; Noci et al., 2005; Yie et al.,005). The identification of a non-invasive marker ofmbryo developmental potential would allow clinicalmbryologists to select the best embryos for transfer,eading to higher IVF pregnancy rates and potentiallyreduction in high risk multiple pregnancies. However,

he issue of the sensitivity of the sHLA-G ELISAs haseen raised (Sageshima et al., 2007) and explored inetail by Sargent et al. (2007), with the authors conclud-ng that further collaboration between research groupss required to establish a more standardized and repro-ucible approach.

In contrast to the situation in mice and humans,nowledge of the bovine MHC-I is relatively limited.t is known that cattle have 6 or more classical classgenes, expressed in different combinations (Ellis et

l., 1999; Birch et al., 2006), compared with the threeenes consistently expressed in mouse and human. Therere approximately 60 full-length validated class I cDNAequences currently listed in the bovine MHC databasehttp://www.ebi.ac.uk/ipd/mhc/bola). These include 9esignated non-classical sequences that are encodedt 4 distinct loci, NC1–NC4 (Birch et al., 2008). Aimited number of studies have been carried out tonvestigate MHC-I expression during pregnancy in cat-le. Classical MHC-I protein expression is absent (Ellist al., 1998) or moderate (Davies et al., 2000) onetal trophoblast at term, and expression of mRNAor non-classical class I has recently been reportedn third trimester trophoblast (Davies et al., 2006).ranscription of classical class I genes and at leastne non-classical class I gene has been reportedn binucleate trophoblast cells (Bainbridge et al.,001).

With regard to early preimplantation bovine embryos,e have previously reported the detection of MHC-I

xpression at the mRNA level (Fair et al., 2004). Fur-hermore, our study showed an association betweenncreased MHC-I mRNA abundance, faster embryoevelopment and higher embryo quality in cattle (Fairt al., 2004). To our knowledge, this was the first reportf MHC-I mRNA expression in bovine pre-attachmentmbryos, although the specific MHC-I genes expressedere not determined. These findings suggest that the initro produced bovine embryo could provide a compar-tive model for a standardized large-scale analysis ofutative markers of embryo quality. Thus, the objec-

ive of the current study was to characterize the mRNAnd protein expression profile of MHC-I classical andon-classical genes during early preimplantation devel-pment in cattle.

Immunology 82 (2009) 48–56 49

2. Materials and methods

2.1. Oocyte recovery and in vitro embryoproduction (IVP)

Cumulus oocyte complexes (COCs) were obtained byaspiration of surface visible follicles from the ovaries ofslaughtered beef-cross heifers. Oocytes were matured,fertilized and cultured in vitro using standard previouslydescribed techniques (Rizos et al., 2002). Samples werecollected at the following stages of development: imma-ture oocytes (germinal vesicle stage), mature oocytes(metaphase II stage), presumptive zygotes (20 hourspost-insemination, hpi), 2–4-cell (48 hpi), 8–16-cell(Day 3 pi), morula (Day 5 pi), blastocyst (Day 7 pi) andhatched blastocyst (Day 8 pi). The immature and matureoocytes were carefully denuded, snap frozen in pools of20 and their cumulus cells were frozen and stored sep-arately. All other stages were collected in pools of 20,snap frozen, and stored at −80 ◦C until use. A minimumof three replicate pools of each stage of development wascollected (replicate = day of ovary collection).

2.2. RNA extraction and reverse transcription

Total RNA was extracted from each pool of 20embryos and digested with DNase I, using the Abso-lutely RNA Nanoprep Kit (Stratagene, La Jolla CA,#400753) according to the manufacturer’s instructions.Reverse transcription was performed using SuperscriptII reverse transcriptase (Invitrogen Life Technologies,Carlsbad, CA, #18064-014) supplemented with 200 ngof random primers (Invitrogen, #48190-011) accordingto the manufacturer’s instructions.

2.3. MHC-I primers for PCR

Six sets of primers were used: a generic MHC-I primerset, Bov 7 and 11 (Birch et al., 2006), which ampli-fies most classical MHC-I sequences and non-classicalalleles from the NC1 locus and specific primer sets forNC1, NC2, NC3 and NC4 (Birch et al., 2008) and forthe endogenous house keeping control gene bovine his-tone 2A.Z (H2A.Z). Details of annealing temperatures,oligonucleotide sequences and accession numbers arelisted in Table 1.

2.4. Sequencing of MHC-I mRNA products

In order to detect identifiable MHC sequencesimmature oocytes were recovered from the ovaries ofHolstein-Friesian cows (n = 200) at slaughter. Oocytes

50 J. Doyle et al. / Journal of Reproductive Immunology 82 (2009) 48–56

Table 1Summary of primers used for Q-RT-PCR and cDNA sequencing.

Primer name Primer sequence (5′–3′) Location Annealing temp. (◦C) Fragment size (BP)

Bov 7 GGCTACGTGGACGACACG Exon 2 55 410Bov 11 CCCTCCAGGTAGTTCCT Exon 3

NC1F GGTCCCCCGAGACACAC Exon 1 60 430NC1R TCAGGGCGAGGTAATCTT Exon 3

NC2F1 CCGAGGATAGAGCCTACA Exon 2 55 310NC2R AGCATTGTGCTTGGTGACT Exon 3

NC3&4F GGCTCCCACTCCCTGAGC Exon 2 60 350NC3R1 TCTGCCGTCGTAGGCCTT Exon 3

NC3&4F GGCTCCCACTCCCTGAGC Exon 2 60 350NC4R1 CGCACCGTCATAGGCGTT Exon 3

ExonExon

H2A.Z GGTAAGGCTGGGAAGGACTCGATGCATTTCCTGCCAATTC

were pooled and following in vitro maturation (IVM),oocytes were inseminated in vitro with frozen thawedsemen from one of two MHC-homozygous Holstein-Friesian bulls, expressing either the haplotype A18(N*01301 class I allele) or the haplotype A31 (N*02101and N*02201 class I alleles). Following 24 h of co-culture with spermatozoa the presumptive zygoteswere washed and further cultured in synthetic oviductfluid medium as described by Rizos et al. (2002).cDNA was prepared from pools of immature oocytes(n = 10 oocytes × 5 replicates), presumptive zygotes(n = 20 zygotes × 1 replicate per bull) and Day 7blastocysts (n = 10 blastocysts × 5 replicates per bull).PCR amplification was carried out in a 50 �l vol-ume reaction using Platinum® Taq DNA Polymerase(Invitrogen, #10966-018) according to the manufac-turer’s instructions. Each reaction contained 4 �l oftemplate DNA. All samples were amplified with eachof the MHC-I primer sets listed in Table 1. Followingverification by agarose gel electrophoresis, individ-ual PCR products were purified, cloned using theTOPO TA Cloning® Kit (Invitrogen, #K4600-01) andsequenced (Macrogen Inc., Seoul, Korea). Approxi-mately eight colony inserts were sequenced per PCRreaction. All sequenced transcripts were compared to theNational Centre for Biotechnology Information (NCBI)Genbank database using the Blastn search algorithm(http://blast.ncbi.nlm.nih.gov/Blast.cgi).

2.5. Quantitative real-time-PCR (qRT-PCR)

qRT-PCR was carried out using the MXP3000Real-Time-PCR System (Stratagene) and SYBR GreenMaster Mix (Brilliant® Sybr® Green QPCR Master Mix,

2 60 2084

Stratagene) according to the manufacturer’s instructions.All PCR reactions were carried out in duplicate for eachstage of development for a minimum of three replicates.At the completion of cycling, melting curve analysis wascarried out to establish the specificity of the ampliconsproduced.

The comparative CT method was used for quantifica-tion of mRNA expression levels (Livak and Schmittgen,2001). The quantification was normalised to the endoge-nous control H2A.Z. The �CT value was determined bysubtracting the H2A.Z CT value for each sample fromthe MHC primer set CT value of the sample. Calcula-tion of the ��CT involved using the highest sample CTvalue (i.e. the sample with the lower target expression)as an arbitrary constant to subtract all other �CT samplevalues. Fold changes in the relative mRNA expressionof target was determined by using the formula 2−��CT.Where product was not detected after 50 cycles of PCRan arbitrary CT value of 50 was assigned to allow thesedata to be included in the analysis. Quantitative MHC-ImRNA expression data were collected for the genericprimer set and for two of the non-classical genes, NC1and NC2.

2.6. Whole mount immunolabelling of oocytes

Whole mount immunolocalization of MHC-I proteinwas carried out on immature and in vitro maturedoocytes and in vitro produced 2–4-cell and blastocystembryos produced as described above. Oocyte sam-

ples were processed for immunocytochemistry eitherwith or without their cumulus investments. In thecase of oocytes that were denuded, cumulus invest-ments were removed by manual pipetting. Following

oductive

dpSspw5B4aaNppiba#awF#c(4(cTcws

2

Zspagunc5

2

migc

J. Doyle et al. / Journal of Repr

enudation, the zona pellucida was removed using arewarmed (37 ◦C) pronase solution (0.5% Protease,igma–Aldrich, Poole, Dorset, UK #81748). Theamples were prepared for immunocytochemistry asreviously described (Fair et al., 2001). Briefly, samplesere blocked for 2 h in PBS containing 1% Triton-X,% rabbit serum (Sigma–Aldrich, #R4505) and 0.05%SA (blocking buffer) and incubated overnight at◦C with the primary antibody, mouse monoclonalntibody ILA88 (Toye et al., 1990) which recognizesll free cattle MHC-I heavy chain molecules includingC1, but not NC2, at a concentration of 1 �l antibodyer 100 �l blocking buffer. The following day, excessrimary antibodies were removed by extensive washingn PBS. The signal was amplified by incubation withiotinylated secondary antibodies (biotinylated rabbitnti-mouse immunoglobulins; Dako A/S, Denmark,E0464) diluted with blocking buffer 1:500 for 4 ht 4 ◦C and 1 h at room temperature. The signalas labelled by incubating with streptavidin Alexaluor® 488 conjugate (Molecular Probes, Invitrogen,S32354). The processed oocytes were mounted onlean glass slides in fluorescent mounting mediumDako A/S, #002627) supplemented with 1 �l/10 ml′,6-diamidino-2-phenylindole*dihydrochloride hydraDAPI) (Sigma–Aldrich, #32670) and covered with alean coverslip that was fixed to the slide with nail polish.he slides were stored at 4 ◦C in the dark. As a negativeontrol, the primary antibody was omitted or substitutedith a matched isotype, during the processing of several

amples (n = 5) at each stage of development.

.7. Microscopy and image processing

Immunolabeled oocytes were analysed using the Carleiss LSM 510 UVMETA Axiovert 200 M confocalystem. The fluorochrome was excited using the appro-riate combination of excitation and barrier filters andrgon/krypton laser lines for Alexa Fluor® 488 conju-ate (498 nm) and DAPI (650 nm). Images were obtainedsing a Plan-Neofluar 40×/1.3 oil DIC. Two filter chan-els were used to view the presence of Alexa Fluor® 488onjugate and DAPI (Ch2-1: BP 385-470 and Ch3-2: BP05-530).

.8. Western blot analysis

To separate the proteins according to their apparent

olecular masses, SDS-PAGE was performed accord-

ng to Laemmli (1970). Immature (n = 100 per lane onel) and mature oocytes (n = 100 per lane on gel), 2–4-ell (n = 120 per lane on gel) and Day 8 blastocyst (n = 80

Immunology 82 (2009) 48–56 51

per lane on gel)-stage embryos were produced by routineIVP as described above. In addition, Day 14 elongatedblastocysts (n = 1 per lane on gel) from superovulatedheifers were also analysed. The oocytes were denudedof their cumulus investments by gentle pipetting andrepeated washing in PBS and the cumulus cells wererecovered for analysis by centrifugation. All sampleswere resuspended in lysis buffer (M-PER MammalianProtein Extraction Reagent from Thermo Scientific),frozen and thawed three times and centrifuged for 15 minat 10,000 × g at 4 ◦C. The isolated protein was dena-tured by boiling for 5 min at 95 ◦C in loading buffer(Laemmli Sample Buffer, Bio-Rad Laboratories, Inc.,Hercules, CA, USA, #161-0737) supplemented with 5%mercaptoethanol and subsequently loaded on to a 10%polyacrylamide gel and separated by SDS-PAGE. Onceseparated, proteins were transferred onto a nitrocellulosemembrane and blocked in 5% (w/v) nonfat dry milk for1 h at room temperature. Immunoblotting was performedby incubating the membrane overnight at 4 ◦C with theprimary antibody ILA88 (1 �g/ml), and following wash-ing, incubated for 90 min at room temperature with thesecondary antibody anti mouse (Cell Signaling Tech-nology, Danvers, MA, #7076) at 1:1000 dilution. Afterthree washes for 10 min, the signal was visualized usingSuperSignal West Pico Chemiluminescent Substrate Kit(Pierce Biotechnology, Rockford, IL). The protein con-tent of the cumulus cells and Day 14 blastocysts wascalculated using the Bradford Assay (Bradford, 1976)and for the oocyte and embryo samples by extrapolat-ing from previously published data (Grealy et al., 1996;Thompson et al., 1998). Five micrograms of protein fromcumulus cells; 12.5 �g from oocytes; 15.5 �g from 2–4-cell embryos; 29 �g from Day 8 blastocysts and 25 �gfrom a single Day 14 blastocyst were loaded on to eachgel (n = 3).

2.9. Statistical analysis

The mRNA expression data were analysed using oneway repeated measures ANOVA (followed by Tukey’sMultiple Comparison Test). Differences of P < 0.05 wereconsidered significant.

3. Results

3.1. Identification of MHC-I mRNA transcripts

The primer pair Bov 7 and 11 generated a number

of discrete sequences (n = 50); of the alleles identifiedin post-fertilization samples, some corresponded to thealleles expressed by the homozygous bulls (N*01301,N*02101, N*02201), while the majority corresponded

52 J. Doyle et al. / Journal of Reproductive Immunology 82 (2009) 48–56

Table 2CT values of MHC-I mRNA during preimplantation development.

Stage of development mRNA H2A.ZAverage CT

aGeneric MHC-I NC1 NC2

Average CTa Average �CT

b Average CTa Average �CT

b Average CTa Average �CT

b

DGV 28.8 ± 0.8 30.7 ± 1.2 1.9 ± 1.5a 37.3 ± 0.5 9.4 ± 0.8ac 31.1 ± 1 2.7 ± 1.2a

CUMULUS 19.9 ± 0.3 21.4 ± 1.5 1.5 ± 1.3a 27.6 ± 0.6 8.5 ± 0.1ac 21.1 ± 0.1 1.25 ± 1.1a

DMII 29.7 ± 0.6 32.7 ± 0.4 3.0 ± 0.6a 35.1 ± 0.7 6.3 ± 0.9a 35.0 ± 0.7 5.3 ± 0.3ac

MII GC 19.6 ± 0.3 20.0 ± 0.7 0.5 ± 0.45a 25.5 ± 0.3 6.4 ± 0.4a 23.0 ± 0.1 3.3 ± 0.4a

P ZYG 28.2 ± 1.8 29.2 ± 1.95 0.9 ± 0.45a 33.0 ± 1.9 6.1 ± 0.15a 31.4 ± 2.9 3.8 ± 1.1a

2–4 C 28.6 ± 0.6 33.4 ± 0.7 4.8 ± 1.0ab Not detected 21.4 ± 0.9bde 33.9 ± 0.25 6.4 ± 0.7ab

8–16 C 24.2 ± 0.7 30.5 ± 1.9 6.2 ± 1.4b Not detected 25.7 ± 1.5be 33.7 ± 1.9 9.4 ± 1.1bc

MORULA 25.6 ± 0.9 31.8 ± 1.1 6.2 ± 0.9b 40.1 ± 1.6 13.6 ± 1.3ad 35.7 ± 1.1 10.1 ± 0.3b

BL 25.4 ± 0.6 31.5 ± 1.0 6.1 ± 0.9b 42.0 ± 3.0 16.3 ± 2.4bc 34.0 ± 1.0 8.2 ± 0.65bc

HBL 24.6 ± 0.9 32.15 ± 1.3 7.6 ± 1.2b 38.7 ± 0.6 14.7 ± 1.5bc 33.4 ± 1.1 9.2 ± 1.0b

the thremalisatiP < 0.05

3.3. MHC-I protein expression

MHC-I proteins were localized in oocytes andembryos at all of the stages examined (Table 3, Fig. 4).

Fig. 1. Generic MHC-I mRNA expression across preimplantation

a CT, cycle threshold: cycle number in which amplification crossesb �CT, subtracted gene product of interest CT, H2a mRNA CT: nor

with different subscripts within the same column differ significantly (

to known classical alleles listed on the MHC database(www.ebi.ac.uk/ipd/mhc/bola). Only one NC sequencewas generated using these generic primers, and this wasidentical to NC1*00301. PCR amplification using thenon-classical class I gene-specific primers for NC1, NC2,NC3 and NC4, demonstrated transcription of these genesin bovine oocytes, presumptive zygotes and blastocysts.Although these primers have previously been shown tobe gene-specific (Birch et al., 2008) a number of cloneswere sequenced and confirmed that this was the case.A selection of images of ethidium bromide gels show-ing PCR amplified product of NC1, NC2, NC3 andNC4 primer sets are presented as supplementary data(Supplementary Fig. 6).

3.2. Quantitative analysis of MHC-I mRNAexpression

MHC-I mRNA transcripts were detected at all inves-tigated stages of development (Table 2); however,expression appeared to be gene-specific. Transcriptsamplified by the generic primers Bov 7 and 11 (Fig. 1)were detected in abundance in immature oocytes andtheir cumulus cells; there was a steep decrease inabundance following oocyte maturation followed bya transient increase at the presumptive zygote stage.Interestingly, mRNA expression remained high in thecumulus cells during maturation. Levels decreased by2-fold at the 2–4-cell stage and remained low through tothe morula and blastocyst stages. The mRNA expression

pattern of the non-classical gene NC1 differed from theglobal MHC-I mRNA profile in that mRNA abundancewas 12.48-fold lower in the immature oocytes and 7-foldlower in their cumulus cells compared to the matured

shold set in the geometric portion of amplification curve.on of PCR cycles for subtracted gene target with H2a mRNA. Values).

oocytes and their cumulus cells, respectively (Table 2,Fig. 2). Subsequent to the zygote stage, NC1 mRNAwas not detected until the morula, blastocyst and hatchedblastocyst stages when trace levels were observed. ThemRNA expression profile of the non-classical gene NC2was quite similar to that of the global MHC-I profileoutlined above (Table 2, Fig. 3). In particular, a tran-sient rise in mRNA abundance was detected at the zygotestage. Thereafter, mRNA expression levels remained lowthrough to the morula and blastocyst stages.

development. DGV, denuded germinal vesicle stage oocyte; DMII,denuded metaphase II stage oocyte; ZYG, presumptive zygote (17 hpi);2–4 C, 2–4-cell embryo; 8–16 C, 8–16-cell embryo; MORULA, 32–64-cell embryo; BL, blastocyst stage embryo; HBL, hatched blastocyst.Data shown are mean fold difference ± SEM.

J. Doyle et al. / Journal of Reproductive

Fig. 2. NC1 mRNA expression across preimplantation develop-ment. Different subscripts (a and b) differ significantly (P < 0.05),Figs. 1–3: DGV, denuded germinal vesicle stage oocyte; DMII,denuded metaphase II stage oocyte; ZYG, presumptive zygote (17 hpi);2–4 C, 2–4-cell embryo; 8–16 C, 8–16-cell embryo, MORULA, 32–64-cell embryo; BL, blastocyst stage embryo; HBL, hatched blastocyst.Data shown are mean fold difference ± SEM.

Fig. 3. NC2 mRNA expression across preimplantation develop-ment. Different subscripts (a and b) differ significantly (P < 0.05),Figs. 1–3: DGV, denuded germinal vesicle stage oocyte; DMII,denuded metaphase II stage oocyte; ZYG, presumptive zygote (17 hpi);2–4 C, 2–4-cell embryo; 8–16 C, 8–16-cell embryo; MORULA, 32–64-cell embryo; BL, blastocyst stage embryo; HBL, hatched blastocyst.Data shown are mean fold difference ± SEM.

Table 3Summary of immunocytochemistry data.

Stage of development N Antibody Location Pattern of labelling

COC (GV) 13 ILA88 Cytoplasm DiffuseMII 5 ILA88 Cytoplasm Diffuse/peripheral2–4 C 12 ILA88 Cytoplasm PeripheralBlastocyst 8 ILA88 Cytoplasm Peripheral

Immunology 82 (2009) 48–56 53

The immature COCs were characterized by bright dif-fuse labelling of the surrounding cumulus cells, theoocyte nucleus and the cytoplasm (Fig. 4A and B).The majority of matured oocytes presented a similarlabelling pattern, however, a small proportion of theseoocytes showed protein localization to the periphery ofthe oocyte (Fig. 4G and H). Western blot analysis con-firmed the detection of protein by the ILA88 antibody inbovine immature and mature oocytes and their surround-ing cumulus cells, 2–4-cell embryos, Day 8 blastocystsand Day 14 elongated blastocysts. The signal from thecumulus cells was greater than that of the oocytes andappeared to be increased during maturation (Fig. 5).

4. Discussion

The findings of the present study reveal that tran-scripts for both classical and non-classical MHC-I genesare present during preimplantation embryo developmentin cattle from the immature oocyte until at least thehatched blastocyst stage. Furthermore, our findings sug-gest that the pattern of mRNA expression is regulated ona gene-specific basis.

Although MHC-I expression has been reported inin vitro derived and nuclear transfer bovine embryos,generic oligonucleotide primers (Fair et al., 2004) or panmonoclonal antibodies were used (Pfister-Genskow etal., 2005). Thus, there is no information available con-cerning the precise pattern of classical and non-classicalMHC-I expression during early embryo developmentin cattle. In contrast, the subject has been studiedextensively in both mouse and human species and theconsensus is that expression of classical MHC-I isdown-regulated in mature oocytes, mature sperm cellsand developing embryos, while non-classical MHC-Iexpression can be detected on sperm cell precursorsand fertilized oocytes and is up-regulated during theearly cleavage stages (summarized by Fernandez et al.,1999).

In the present study, we report the detection of bothclassical and non-classical MHC-I mRNA in bovineoocytes and in their surrounding cumulus cells. Follow-ing the first cleavage and subsequent divisions, globalMHC-I mRNA expression decreased to trace levels.In contrast, using whole mount immunocytochem-istry, MHC-I protein was observed across development,including the early cleavage and blastocyst stages. A sim-ilar variance between pan HLA-G protein and mRNA

expression during human embryo development has beenreported (Yao et al., 2005). The authors postulated thatthe explanation for the variance between protein andmRNA expression could be that the protein detected is

54 J. Doyle et al. / Journal of Reproductive Immunology 82 (2009) 48–56

ture (Ambryos

lowing a cDNA microarray comparison of bovine Day7 and Day 14 embryos, the relative quantity of MHC-ImRNA was higher in Day 14 embryos.

Fig. 5. Western blot analysis of MHC-I antigen expression in bovine

Fig. 4. Immunocytochemical localization of MHC-I protein on imma2-cell (E), 4-cell (F), Day 7 blastocyst (G) and Day 9 blastocyst (H) e

of maternal origin until the embryonic MHC-I transcrip-tome is fully activated and regulated. In order to confirmthe validity of the whole mount results, further analysiswas undertaken using Western blotting. A band of appro-priate molecular weight (47 kDa) was observed in allinvestigated samples. When the intensity of the bands isconsidered in relation to the amount of protein loaded perstage of development, our findings suggest that MHC-I protein content increased dramatically during meioticmaturation in both the oocyte and its surrounding cumu-lus cells, subsequently decreased following fertilizationand development to the 2–4-cell stage, decreased furtherby the blastocyst stage and thereafter had increased bythe Day 14 elongating blastocyst stage. These findingsare comparable to the quantitative mRNA data; in partic-

ular the mRNA expression profile of NC1 is very similarduring oocyte maturation. The detection of MHC-I pro-tein at the 2–4-cell stage is in contrast to the low mRNAlevels detected with the three primer sets and as refer-

and B) and in vitro matured (C and D) oocytes and in vitro derivedand an example of a negative control (I). Scale bar = 40 �m.

enced above, suggests that the protein is of maternalorigin. The increase in MHC-I protein content in theDay 14 compared to Day 7 blastocyst is reflective of themRNA findings of Ushizawa et al. (2004), where fol-

cumulus cells from germinal vesicle (lane 1) and metaphase II (lane 2)-stage oocytes, germinal vesicle (lane 3) and metaphase II (lane 4)-stageoocytes, 2–4-cell embryos (lane 5), blastocyst stage embryos (lane 6)and one Day 14 embryo (lane 7), using the mouse monoclonal ILA88antibody.

oductive

ertct2aMfm

4soteiaaicccatbabd

tIpso

A

F(

A

b2

R

B

J. Doyle et al. / Journal of Repr

These results are very interesting when one consid-rs that this period of development is associated withapid growth and cellular differentiation and again raiseshe question of an association between the kinetics ofell division and MHC-I gene expression, particularly inhe light of the findings in mice (Warner and Brenner,001), humans (Juriscova et al., 1996) and cattle (Fair etl., 2004). Thus, it appears that up-regulation of specificHC-I genes is associated with, and perhaps important

or, early embryo development in a number of mam-alian species.A total of 17 discrete classical MHC-I alleles andnon-classical alleles were identified in the current

tudy. Because the mRNA was recovered from pools ofocytes or pools of embryos it is impossible to determinehe number of classical and non-classical MHC-I genesxpressed by each oocyte or each blastocyst. However, its interesting to note that using the generic primers whichre compatible with classical MHC-I and NC1 alleles,ll of the sequences generated from oocyte cDNA weredentified as classical MHC-I in contrast to the blastocystDNA sequences, of which 10% were identified as non-lassical NC1. Although caution must be exercised whenonsidering the abundance of one gene compared tonother in pooled samples, these data suggest a change inhe ratio of classical to non-classical MHC-I expressionetween the two developmental stages. Further analysesre ongoing and as new technologies emerge it is wille possible to build an MHC-I profile of each stage ofevelopment from single oocyte and embryo samples.

In conclusion, the current study presents novel datahat demonstrate both classical and non-classical MHC-mRNA expression in bovine oocytes and developingreimplantation embryos. Furthermore, our findingsuggest that the expression pattern is both gene- and stagef development-specific.

cknowledgements

Supported by the University College Dublin Seedunding Award (SF015) and Science Foundation Ireland05/PICA/B813)

ppendix A. Supplementary data

Supplementary data associated with this article cane found, in the online version, at doi:10.1016/j.jri.009.06.125

eferences

ainbridge, D.R., Sargent, I.L., Ellis, S.A., 2001. Increased expressionof major histocompatibility complex (MHC) class in transplan-

Immunology 82 (2009) 48–56 55

tation antigens in bovine trophoblast cells before fusion withmaternal cells. Reproduction 122, 907–913.

Birch, J., Codner, G., Guzman, E., Ellis, S.A., 2008. Genomic locationand characterisation of non-classical MHC class I genes in cattle.Immunogenetics 60, 267–273.

Birch, J., Murphy, L., Machugh, N.D., Ellis, S.A., 2006. Generationand maintenance of diversity in the cattle MHC class I region.Immunogenetics 58, 670–679.

Bradford, M.M., 1976. A rapid and sensitive method for the quanti-tation of microgram quantities of protein utilizing the principle ofprotein-dye binding. Anal. Biochem. 72, 248–254.

Comiskey, M., Goldstein, C.Y., De Fazio, S., Mammolenti, M., New-mark, J.A., Warner, C.A., 2003. Evidence that HLA-G is thefunctional homolog of mouse Qa-2, the Ped gene product. Hum.Immunol. 64, 999–1004.

Davies, C.J., Eldridge, J.A., Fischer, P.J., Schafer, D.H., 2006. Evi-dence for expression of both classical and non-classical majorhistocompatibility complex class I genes in bovine trophoblastcells. Am. J. Reprod. Immunol. 55, 188–200.

Davies, C.J., Fisher, P.J., Schlafer, D.H., 2000. Temporal and regionalregulation of major histocompatibility complex class I expressionat the bovine uterine/placental interface. Placenta 21, 194–202.

Ellis, S.A., Sargent, I.L., Charleston, B., Bainbridge, D.R., 1998. Reg-ulation of MHC class I gene expression is at transcriptional andpost-transcriptional level in bovine placenta. J. Reprod. Immunol.37, 103–115.

Ellis, S.A., Holmes, E.C., Staines, K.A., Smith, K.B., Stear, M.J., McK-eever, D.J., MacHugh, N.D., Morrison, W.I., 1999. Variation inthe number of expressed MHC genes in different cattle class Ihaplotypes. Immunogenetics 50, 19–28.

Exley, G.E., Warner, C.M., 1999. Selection in favor of the Ped fast hap-lotype occurs between mid-gestation and birth. Immunogenetics49, 653–659.

Fair, T., Gutierrez-Adan, A., Murphy, M., Rizos, D., Martin, F.,Boland, M.P., Lonergan, P., 2004. Search for the bovine homologof the murine Ped gene and characterization of its messengerRNA expression during bovine preimplantation development. Biol.Reprod. 70, 488–494.

Fair, T., Hyttel, P., Lonergan, P., Boland, M.P., 2001. Immunolocal-ization of nucleolar proteins during bovine oocyte growth, meioticmaturation and fertilization. Biol. Reprod. 64, 1516–1525.

Fernandez, N., Cooper, J., Spinks, M., Abd-Elrahman, M., Fiszer, D.,Kurpisz, M., Dealtry, G.A., 1999. Critical review of the role of themajor histocompatibility complex in fertilization, preimplantationdevelopment and feto-maternal interactions. Hum. Reprod. Update5, 234–248.

Fuzzi, B., Rizzo, R., Criscuoli, L., Noci, I., Melchiorri, L., Scarselli, B.,et al., 2002. HLA-G expression in early embryos is a fundamentalprerequisite for the obtainment of pregnancy. Eur. J. Immunol. 32,311–315.

Grealy, M., Diskin, M.G., Sreenan, J.M., 1996. Protein content of cattleoocytes and embryos from the two-cell to the elongated blastocyststage at day 16. J. Reprod. Fertil. 107 (2), 229–233.

Juriscova, A., Casper, R.F., MacLusky, N.J., Mills, G.B., Librach, C.L.,1996. HLA-G expression during preimplantation embryo develop-ment. PNAS U.S.A. 93, 161–165.

Laemmli, U.K., 1970. Cleavage of structural proteins during the

assembly of the head of bacteriophage T4. Nature 227 (5259),680–685.

Livak, K.J., Schmittgen, T.G., 2001. Analysis of relative gene expres-sion data using real-time quantitative PCR and the 22DDCTmethod. Methods 25, 402–408.

oductiv

8379–8385.

56 J. Doyle et al. / Journal of Repr

Noci, I., Fuzzi, B., Rizzo, R., Melchiorri, L., Criscuoli, L., Dabizzi, S.,et al., 2005. Embryonic soluble HLA-G as a marker of develop-mental potential in embryos. Hum. Reprod. 20, 138–146.

Pfister-Genskow, M., Myers, C., Childs, L.A., Lacson, J.C., Patterson,T., Betthauser, J.M., et al., 2005. Identification of differentiallyexpressed genes in individual bovine preimplantation embryosproduced by nuclear transfer: improper reprogramming of genesrequired for development. Biol. Reprod. 72, 546–555.

Rizos, D., Lonergan, P., Boland, M.P., Arroyo-Garcia, R., Pintado, B.,de la Fuente, J., Gutierrez-Adan, A., 2002. Analysis of differentialmessenger RNA expression between bovine blastocysts producedin different culture systems: implications for blastocyst quality.Biol. Reprod. 66, 589–595.

Sageshima, N., Shobu, T., Awai, K., Hashimoto, H., Yamashita, M.,Takeda, N., et al., 2007. Soluble HLA-G is absent from humanembryo cultures: a reassessment of sHLA-G detection methods. J.Reprod. Immunol. 75, 11–22.

Sargent, I., Swales, A., Ledee, N., Kozma, N., Tabiasco, J., LeBouteiller, P., 2007. sHLA-G production by human IVF embryos:can it be measured reliably? J. Reprod. Immunol. 75, 128–132.

Sher, G., Keskintepe, L., Fisch, J.D., Acacio, B.A., Ahlering, P., Bat-zofin, J., Ginsburg, M., 2005a. Soluble human leukocyte antigenG expression in phase I culture media at 46 h after fertilizationpredicts pregnancy and implantation from day 3 embryo transfer.Fertil. Steril. 83, 1410–1413.

Sher, G., Keskintepe, L., Batzofin, J., Fisch, J., Acacio, B., Ahler-ing, P., Ginsburg, M., 2005b. Influence of early ICSI-derivedembryo sHLA-G expression on pregnancy and implantation rates:a prospective study. Hum. Reprod. 20, 1359–1363.

Sher, G., Keskintepe, L., Nouriani, M., Roussev, R., Batzofin, J., 2004.

Expression of sHLA-G in supernatants of individually cultured46-h embryos: a potentially valuable indicator of ‘embryo compe-tency’ and IVF outcome. Reprod. Biomed. Online 9, 74–78.

Thompson, J.G., Sherman, A.N., Allen, N.W., McGowan, L.T., Ter-vit, H.R., 1998. Total protein content and protein synthesis within

e Immunology 82 (2009) 48–56

pre-elongation stage bovine embryos. Mol. Reprod. Dev. 50,139–145.

Toye, P.G., MacHugh, N.D., Bensaid, A.M., Alberti, S., Teale, A.J.,Morrison, W.I., 1990. Transfection into mouse cells of genesencoding 2 serologically and functionally distinct bovine class IMHC molecules. Immunology 70, 20–26.

Ushizawa, K., Herath, C.B., Kaneyama, K., Shiojima, S., Hirasawa, A.,Takashashi, T., et al., 2004. cDNA microarray analysis of bovineembryo gene expression profiles during the pre-implantationperiod. Reprod. Biol. Endocrinol. 2, 77.

Warner, C.M., Tyas, D.A., Goldstein, C., Comiskey, M., Cohen, J.,Brenner, C.A., 2002. Genotyping: the HLA system and embryodevelopment. Reprod. Biomed. Online 4, 133–139.

Warner, C.M., Brenner, C.A., 2001. Genetic regulation of preimplan-tation embryo survival. Curr. Top. Dev. Biol. 52, 151–192.

Warner, C.M., Panda, P., Almquist, C.D., Xu, Y., 1993. Preferentialsurvival of mice expressing the qa-2 antigen. J. Reprod. Fertil. 99,145–147.

Warner, C.M., Brownell, M.S., Rothschild, M.F., 1991. Analysis oflitter size and weight in mice differing in Ped gene phenotypeand the q region of the h-2 complex. J. Reprod. Immunol. 19,303–313.

Watkins, A., Wilkins, A., Osmond, C., Warner, C.M., Comiskey, M.,Hanson, M., Fleming, T.P., 2006. The influence of mouse Pedgene expression on postnatal development. J. Physiol. 571, 211–220.

Yao, Y.Q., Barlow, D.H., Sargent, I.L., 2005. Differential expres-sion of alternatively spliced transcripts of HLA-G in humanpreimplantation embryos and inner cell masses. J. Immunol. 175,

Yie, S.M., Balakier, H., Motamedi, G., Librach, C.L., 2005. Secretionof human leukocyte antigen-G by human embryos is associatedwith a higher in vitro fertilization pregnancy rate. Fertil. Steril. 83,30–36.