Linkage Mapping of Ovine Microphthalmia to Chromosome 23, the Sheep Orthologue of Human Chromosome 18

Abstract

purpose. To characterize the phenotype and map the locus responsible for autosomal recessive inherited ovine microphthalmia (OMO) in sheep.

methods. Microphthalmia-affected lambs and their available relatives were collected in a field, and experimental matings were performed to obtain affected and normal lambs for detailed necropsy and histologic examinations. The matings resulted in 18 sheep families with 48 cases of microphthalmia. A comparative candidate gene approach was used to map the disease locus within the sheep genome. Initially, 27 loci responsible for the microphthalmia–anophthalmia phenotypes in humans or mice were selected to test for comparative linkage. Fifty flanking markers that were predicted from comparative genomic analysis to be closely linked to these genes were tested for linkage to the disease locus. After observation of statistical evidence for linkage, a confirmatory fine mapping strategy was applied by further genotyping of 43 microsatellites.

results. The clinical and pathologic examinations showed slightly variable expressivity of isolated bilateral microphthalmia. The anterior eye chamber was small or absent, and a white mass admixed with cystic spaces extended from the papilla to the anterior eye chamber, while no recognizable vitreous body or lens was found within the affected eyes. Significant linkage to a single candidate region was identified at sheep chromosome 23. Fine mapping and haplotype analysis assigned the candidate region to a critical interval of 12.4 cM. This ovine chromosome segment encompasses an ancestral chromosomal breakpoint corresponding to two orthologue segments of human chromosomes 18, short and long arms. For the examined animals, we excluded the complete coding region and adjacent intronic regions of ovine TGIF1 to harbor disease-causing mutations.

conclusions. This is the first genetic localization for hereditary ovine isolated microphthalmia. It seems unlikely that a mutation in the TGIF1 gene is responsible for this disorder. The studied sheep represent a valuable large animal model for similar human ocular phenotypes.

Congenital microphthalmia is characterized by an abnormally small eye in newborns that can be associated with other ocular abnormalities and represents a heterogeneous disorder occurring in various mammalian species. In humans, microphthalmia is highly variable, with anophthalmia in the most severe cases. Several genes harboring causative mutations for autosomal recessive and dominant syndromic and isolated microphthalmia have been found. ¹ However, although clinical studies in these human defects revealed variable phenotypical expression, the establishment of precise genotype–phenotype correlations in the microphthalmia–anophthalmia spectrum may be difficult, and the molecular genetic basis of numerous human cases is still unknown. ¹ Many inherited malformations of domestic animals are analogous to human hereditary anomalies and have been valuable animal models for the investigation of the pathogenesis of rare human phenotypes with identical molecular basis. ² Naturally occurring large domestic animal models are particularly valuable because the eye size and structure in the animals is larger and more human-like than that of rodents, and their longer life expectancy allows for investigations or treatments over a longer time. In general, the clinical progression in larger animals more closely resembles that in humans. Domestic production animals have the additional advantages of being economic to maintain and of having been bred for easy management. Moreover, a high level of expertise in reproductive technology and veterinary care is available for them. 

In sheep, isolated hereditary bilateral microphthalmia is well known worldwide as a monogenic autosomal recessive inherited disorder with complete penetrance associated with the Texel breed. ^{3 4 5 6 7 8 9} Typically, animals affected by microphthalmia are entirely blind. ¹⁰ Macroscopically, the globes and optic nerves appear hypoplastic and histologically all embryonic components of the eye are present, but lens, ciliary body, iris, and retina are dysplastic. ¹⁰ An abnormal development of the lens vesicle was shown to be the primary event in hereditary ovine microphthalmia by comparative investigation of normal and affected eyes of sheep embryos at different gestational stages. ⁸  

The availability of sheep genome resources has expanded considerably in the past 2 years. The current ovine genetic map includes more than 1400 informative markers (see http://rubens.its.unimelb.edu.au/∼jillm/jill.htm). This powerful tool has been effectively used for linkage analyses in sheep, to identify genome regions associated with Mendelian disorders and economically important production traits. ¹¹ Furthermore, public domain resources now include more than 375,000 genomic BAC end sequences (BES) representing one twelfth of the whole genome sequence (see http://www.ncbi.nlm.nih.gov/). Recently, an ovine radiation hybrid panel has been constructed to build integrated detailed physical maps of entire sheep chromosomes ¹² and a genome-wide, sheep-specific single-nucleotide polymorphism (SNP) chip is under development (see http://www.sheepgenomics.com). In addition, the sequenced cattle genome (see http://www.ncbi.nlm.nih.gov/mapview/map_search.cgi?taxid=9913) provides valuable data, because of the high interspecific sequence and genome similarity between ruminants ¹¹ and the existence of high-resolution comparative maps between the bovine, human, and murine genomes. ^{13 14} Finally, these provide excellent tools for positional cloning of candidate genes or the identification of causal mutations in sheep. ¹⁵  

In this study we report the linkage mapping of recessive inherited ovine microphthalmia (OMO). Initially, several loci responsible for different microphthalmia–anophthalmia phenotypes in humans or mice were selected to test for comparative linkage in sheep with hereditary microphthalmia. Significant linkage to a single candidate region was identified at sheep chromosome (OAR) 23. The linked TGIF1 gene was excluded as a potential candidate for ovine microphthalmia. Fine mapping and haplotype analysis defined the critical interval of 12.4 cM at OAR23, the ovine orthologue of human chromosome (HSA) 18. 

Methods

Animals

The animals belonged to 17 different nonherdbook sheep farms with uncontrolled crossing of Texel or Texel/Whiteheaded mutton crossbred rams to Texel or Texel/Whiteheaded mutton crossbred ewes from a distinct region of northern Germany. For this study, we included only those litters of a single breeding ram in which affected progeny were produced. In general, each ram served at least 50 ewes within a single breeding season. Phenotypes were evaluated by visual inspection. Full blood samples were taken from the vena jugularis and collected in blood sampling tubes prepared with EDTA (Monovette; Sarstedt, Nümbrecht, Germany). 

A breeding experiment designed to augment the number of affected lambs for genetic and pathologic examination purposes was performed in 2004 and 2005 at the University of Veterinary Medicine Hannover, Germany. By mating of an affected ram to known disease carrier ewes, an additional nine affected and five normal lambs were obtained and integrated into the experimental pedigrees (Fig. 1 , families 16–18). Newborn offspring from experimental pedigrees were examined clinically and subsequently euthanatized within their second week of life. All procedures involving animals were performed in compliance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research and were approved by the Lower Saxony (Germany) governmental animal rights protection authorities (Ref. No. 33-42502-04/851). 

In total, the study included 48 lambs with microphthalmia (28 male, 20 female) representing 56 relative affected pairs (9 full-sib pairs, 38 half-sib pairs, and 9 parent–offspring pairs) and 84 informative meioses for OMO. The 18 unrelated two-generation–spanning sheep families segregating for OMO consisted of 119 individuals with an average family size of 6.6 members, ranging from 4 to 18 animals per family (Fig. 1) . 

The entire animals were macroscopically examined, and the eyes were fixed in a solution of 4% paraformaldehyde, 2.5% glutaraldehyde, and 1% sodium cacodylate and sectioned in a median sagittal plane. Tissue samples from various other organs (brain, spinal cord, heart, lung, spleen, lymph nodes, thymus, thyroid, adrenal, pituitary gland, skeletal muscle, liver, kidney, urinary bladder, intestine, rumen, reticulum, omasum, abomasum, and reproductive system) were fixed in 10% formalin. All tissue samples were processed routinely for histopathology and stained with hematoxylin and eosin (HE). 

Linkage Mapping

Genomic DNA was extracted from blood (NucleoSpin 96 Blood Quick Pure Kit; Macherey-Nagel, Düren, Germany). In total, 50 microsatellite markers that were predicted from comparison of the ovine linkage map version 4.4 to human genome build 35.1 (see http://rubens.its.unimelb.edu.au/∼jillm/jill.htm) to be closely linked to the orthologues of putative candidate genes for microphthalmia represented 17 linkage groups (Table 1) . 

Genotyping of microsatellite markers was performed by PCR amplification and subsequent polyacrylamide gel electrophoresis. The PCR reaction with a final volume of 12 μL was performed using 20 ng of genomic ovine DNA, 100 μM dNTPs, 10 picomoles of each primer, and 0.5 U Taq polymerase in the reaction buffer supplied by the manufacturer (MP Biochemicals, Heidelberg, Germany). The thermocycler profile was 94°C for 4 minutes; 32 cycles of 94°C for 30 seconds; marker-specific annealing for 30 seconds and 72°C for 30 seconds; and a final cooling step at 4°C for 10 minutes. Infrared dye (IRD)-labeled PCR products were separated on 6% denaturing polyacrylamide gels by automated sequencer (Gene ReadIR 4200/4300 automated sequencer; LI-COR; Bad Homburg, Germany). Scoring and size determination of alleles was performed independently by two investigators by comparison to a size standard marker (LI-COR) and retyped in case of discrepancy. 

Genotypic data were initially analyzed by using CRI-MAP software, version 2.4. ¹⁶ Pair-wise linkage analysis was performed with the twopoint option to calculate the recombination fractions between OMO and marker loci and corresponding LOD scores. Subsequently, multipoint nonparametric linkage (NPL) and parametric linkage analyses were performed with MERLIN software, version 1.0.1. ¹⁷ We performed multipoint NPL analyses based on the degree of identity by descent allele-sharing among affected members, by using the pair-scoring function. In addition, we calculated multipoint LOD scores under both homogeneity and heterogeneity under the assumption that OMO segregates as a biallelic autosomal recessive trait, with complete penetrance. The frequency of the defect allele in the considered population is unknown, and there are no data available that would make it possible to estimate the frequency in a reliable manner. For the calculations, a frequency of 0.001 for the mutated allele was assumed. Under the assumption of locus heterogeneity, HLOD scores were maximized for varying fractions of linked families (alpha). ¹⁸  

Fine mapping was performed by additional genotyping of 43 microsatellites (Table 2)spanning the recent OAR23 linkage map. ¹² Moreover, we estimated haplotypes by finding the most likely path of gene flow. To reconstruct likely haplotypes and identify recombinations, we applied the best option of MERLIN. 

Candidate Gene Analysis

An ovine BAC clone (CH243-13H3) assumed to contain TGIF1 was identified by BLASTN searches of ovine BES to the HSA18 sequence. ¹² The BAC DNA was prepared with a plasmid kit (Midi; Qiagen, Hilden, Germany) according to the modified protocol for BAC clones (Qiagen). The human reference TGIF1 isoform-c mRNA (GenBank: NM_173208; http://www.ncbi.nlm.nih.gov/Genbank; provided in the public domain by the National Center for Biotechnology Information, Bethesda, MD) was used as query in cross-species BLAST searches identifying five corresponding ovine EST entries (GenBank: CD288212, EE799622, EE830605, EE822729, and EE826815). A partial insert sequence of 11.8 kb of the CH243-13H3 BAC clone was completely determined by a primer-walking strategy until both strands were completely sequenced with dye termination chemistry (BigDye Terminator Sequencing Kit 3.1 on a 3730 capillary sequencer; Applied Biosystems [ABI] Rotkreuz, Switzerland). Initial sequencing primers were designed from the identified EST sequences to sequence the entire ovine TGIF1 gene. Sequence data were then analyzed (Sequencher 4.6; GeneCodes, Ann Arbor, MI). The exact ovine genomic structure was determined by alignment of ovine TGIF1 EST and human and murine TGIF1 mRNA to the determined ovine genomic sequence using Spidey (http://www.ncbi.nlm.nih.gov/IEB/Research/Ostell/Spidey/index.html). Ovine TGIF1 gene sequence was submitted to EMBL (European Molecular Biology Laboratory, Heidelberg, Germany) under accession AM493688. 

The three ovine TGIF1 exons with flanking sequences were PCR amplified from genomic DNA (AmpliTaqGold; ABI) at standard conditions at 60°C annealing. Three sets of primer pairs designated I, II, and III, representing each individual TGIF1 exon were used: I, 5′-CCTCGCTCACTCTGACAGC-3′ and 5′-CTTTCAGACAAAGCGCCAAG-3′; II, 5′-GGAAGAACAGCGCTTGAAAC-3′ and 5′-TGACTGTCAAGTGTTTCTGTTTCTC-3′; and III, 5′-ACTGGGAGGAGTGGCTCTTT-3′ and 5′-GCTCTGAGTTTATAGTTCTTGGAATG-3′. For the resequencing, we used eight sheep: four unrelated affected lambs, two unrelated putative carrier sheep, and two unrelated rams from the Texel breed, respectively. The subsequent sequencing of the PCR products in both directions was performed after shrimp alkaline phosphatase (Roche, Basel, Switzerland) and exonuclease I (NEB; Axonlab, Baden, Switzerland) treatment by using the PCR primers with dye termination chemistry (BigDye Terminator Sequencing Kit 3.1 on the 3730 capillary sequencer; ABI). 

Results

Animals

Seventeen unrelated rams with at least two affected offspring were identified and, in total, 39 affected offspring and seven normal littermates were collected (Fig. 1) . Affected newborn lambs showed abnormally small eye balls (Fig. 2A) , ranging from 0.7 to 1.1 cm in diameter (Fig. 2B)compared with an average diameter of 2.25 cm in control animals. The anterior eye chamber was small or absent, and there was no recognizable vitreous body or lens (Fig. 2C) . The observed pathologic features are summarized in Table 3 . No lesions were found in the other investigated organs. The experimental breeding showed normal reproductive capacity of the affected ram at mating to unrelated carrier ewes producing nine affected among 14 offspring (Fig. 1) . The clinical examination of the nine affected animals from the breeding experiment showed that they were entirely blind, but did not show any further abnormality. 

Linkage Analysis

Linkage mapping was used to determine whether 27 known candidate loci influencing congenital isolated or syndromic microphthalmia–anophthalmia phenotypes in humans or mice were involved in the ovine microphthalmia phenotype in the examined sheep families. At least two polymorphic flanking microsatellite markers for each candidate locus were genotyped in all samples (Table 1) . As some of the candidate genes map to the same chromosome region, 17 linkage groups on 16 sheep chromosomes were studied (Table 1) . The twopoint and multipoint linkage results between OMO and ovine microsatellite markers are compiled in Table 1 . The twopoint linkage analysis yielded a maximum LOD score of 3.52 at θ = 0.03 with microsatellite marker ADCY1AP on OAR23, which was confirmed by nonparametric multipoint linkage analysis (Table 1 ; Fig. 3 ). Several positive linkage LOD or NPL scores on different chromosomes were mutually contradictory and therefore could be regarded as suggestive linkage results (Table 1) . The subsequent fine mapping and haplotype analysis firmly assigns OMO to a 12.4 cM interval of OAR23 between the markers DU274690 and DU259631 (Table 4) . Therefore, all 24 recombinant chromosomes were taken together to describe the most likely region of OMO on the chromosome map (Table 4) . 

This region of the OAR23 linkage map is compared with the corresponding sequence map for human and cattle (Fig. 4) . Thereby, the definitive marker order in sheep was confirmed by sequence alignments of the used ovine microsatellite markers to the corresponding bovine chromosome 24 sequence (Fig. 4) . 

Candidate Gene Analysis

The OMO linked region of OAR23 corresponds to segments of HSA18, based on the recent comparative sheep–human map. ¹² At least one intrachromosomal rearrangement has occurred between sheep and human as the OMO linked segment of OAR23 is orthologue to the inverted centromeric region of HSA18q and the telomeric end of HSA18p, respectively (Fig. 4) . Besides this rearrangement, there is a perfect colinearity between the sheep and human maps. Thus, the human homologue for ovine microphthalmia is expected to be found in the two indicated HSA18 intervals (Fig. 4) . The candidate gene TGIF1 located on HSA18p11.3 appeared to be within the linked region. ¹² We identified an ovine BAC clone containing the entire TGIF1 orthologue and experimentally confirmed the chromosome location by fluorescent in situ hybridization assignment of CH243-13H3 to OAR23q24 (data not shown). Genomic sequencing of the TGIF1 gene and mRNA to genomic sequence alignment using sheep TGIF1 EST sequences revealed that the ovine TGIF1 gene spans 7.7 kb and consists of three exons. All splice donor–splice acceptor sites conform to the GT/AG rule. The ovine TGIF1 transcript contains an open reading frame of 819 bp encoding a deduced protein of 272 amino acids which shows 99%, 91%, 90%, and 86% identity to the orthologous bovine, canine, human (isoform c), and murine TGIF1 protein, respectively. 

The resequencing of the three coding exons and flanking intron regions of TGIF1 using four affected as well as four healthy sheep revealed four SNPs which are located within TGIF1 intron 1 (c.16+356A→G and c.17-186T→G), exon 3 (c.420A→G), and the 3′UTR (*16T→C), respectively. The codon in exon 3 with the A/G transition codes for proline in both SNP variants and none of the three intronic polymorphisms affected splice sites in the TGIF1 gene. Two SNP (c.17-186T→G and c.420A→G) were observed only in comparisons to the established genomic DNA sequence derived from the ovine BAC clone. The eight sequenced sheep were all homozygous for the respective G allele at both positions. Assuming linkage disequilibrium between the two remaining polymorphic TGIF1 sites (c.16+356A→G and *16T→C) two different haplotypes (A-T and G-C) were constructed for the SNP genotypes in the examined sheep. Both haplotypes occurred in affected and unaffected sheep, respectively, in the homozygous and heterozygous states. 

Discussion

The observed segregation pattern of microphthalmia within the experimental breeding pedigrees can be explained by autosomal recessive inheritance. Both sexes were affected, and 9 of 14 offspring from the mating of an affected ram to known carrier females were reported to have the condition. Under the assumption of a biallelic recessive locus this does not fit perfectly to the theoretical expectation of 50% affected and 50% normal offspring, respectively, but it may be explained by the small number of animals used. In general, the total number of 39 recorded affected offspring from mating of 17 normal Texel or Texel/Whiteheaded mutton crossbred rams to approximately 850 normal Texel or Texel/Whiteheaded mutton crossbred ewes supports the assumption of a recessive deleterious allele. Taken together, these observations confirm the previously reported autosomal recessive mode of inheritance for this hereditary disorder in Texel sheep. ^{3 4 5 6 7 8 9} Because the ewes were collected from several unrelated sheep flocks, allelism of disease among the different familial origins was confirmed by test breeding. 

The clinical and pathologic examinations indicate a slightly variable expressivity of the fully penetrant phenotype, as previously reported. ^{8 10} The described pathology is consistent with prior reports. ^{8 10} Neither an anterior eye chamber nor a lens was found in any of the examined eyes, but several structures could be observed (e.g., lacrimal glands) that are of ectodermal origin, probably deriving from the lens vesicle. This supports the hypothesis of van der Linde-Sipmann et al. ⁸ assuming an abnormal development of the lens vesicle as that primary event leading to disintegration of the lens and subsequent secondary defects in other eye structures due to the overgrowth of mesenchymal tissue. This leads to regression of the embryonic globe that had reached at least the stage of the optic cup, which closely resembles autosomal recessive lens aplasia in mice. ¹⁹ Hereditary microphthalmia in sheep characterized by smaller eyes and lens dysplasia may be affected by the lens degeneration due to developmental arrest that causes collapse of the surrounding ocular tissue (reduction in eye size) and abnormal patterning of the anterior segment structures (cornea defects). Similar associations between lens defects, small eye, and malformal anterior chamber have been reported. ¹  

The presented large animal model for isolated congenital microphthalmia was mapped to the sheep genome for the first time. We have successfully used a comparative candidate linkage mapping approach where we have applied information on established comparative maps between ruminants and human genomes. As the 12.4 cM interval identified on OAR23 shows conserved synteny with regions on HSA18 OMO is likely to represent a mutation in an orthologue gene. ¹² The initially selected candidate gene TGIF1 encoding TGFB-induced factor, a homeodomain protein that acts as a transcription factor during early brain development, is located within the critical interval. Mutations within human TGIF1 causes holoprosencephaly. ²⁰ Labs ¹¹ concluded from her study of newborn lambs with microphthalmia that there was a primary anomaly of the lateral prosencephalon and the primary optic vesicle and that this would result in an abnormal interaction between the primary optic vesicle and the ectoderm, with the absence of lens formation. 

Direct-sequence screening analysis excluded the complete coding region of ovine TGIF1 to harbor disease-causing mutations in the examined sheep. Due to the assumption of a single recessive founder mutation within the Texel breed, we expect homozygosity in affected individuals. The presented data indicate that the detected SNP of ovine TGIF1 can be excluded as disease-harboring mutations in the examined sheep. Therefore, it seems to be likely that the entire TGIF1 gene can be excluded as a candidate for microphthalmia in this breed. 

The presented sequence-alignment–based sheep–human comparison helps to define the exact candidate gene regions on HSA18. The corresponding centromeric region of HSA18q between 0.00 and 6.95 Mb and the telomeric end of HSA18p between 16.78 and 22.75 Mb contains 51 and 49 annotated genes, respectively (human genome build 36.2). A comparison with cattle genome sequence suggests that the critical region in ruminants is smaller than 10 Mb (Fig. 4) . Up to the present, we could not identify any obvious candidate genes in the defined region assumed to harbor the human orthologue of OMO. Therefore, we expect the microphthalmia phenotype to be caused by a novel gene that has not previously been associated with ocular disorders or eye development in any species. It is therefore of considerable interest to reveal the nature of the microphthalmia mutation in sheep, because this could provide novel insights into the molecular mechanisms leading to similar phenotypes in human and mammalian eye development. Recently, a genetic linkage was reported to bovine chromosome 18 for cattle showing an inherited ocular defect characterized by microphthalmia, torsion of the eyes, and complete blindness. ²¹ The genomic evolutionary orthologue, HSA16, harbors several potential candidate genes for the bovine disorder ²² indicating another example for a possible valuable large animal model in this field. 

In conclusion, a further refinement of the candidate region may be necessary before a systematic mutation analysis of the genes located in the region can be undertaken. As there appears to be a fairly low recombination rate in the identified candidate region, it would be hard to achieve a high-resolution localization by linkage analysis. Therefore, we are now adopting a linkage disequilibrium strategy using SNP markers as the next step toward positional cloning of the OMO gene. Identifying the minimum shared haplotype associated with OMO could take advantage of the recombination events that have accumulated during the transmission of OMO from a putative common founder. 

 Supported by research grants from the Swiss National Foundation and the H. W. Schaumann Stiftung, Hamburg, Germany.

 Submitted for publication January 15, 2007; revised March 2, 2007; accepted May 16, 2007.

 Disclosure: J. Tetens, None; M. Ganter, None; G. Müller, None; C. Drögemüller, None

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

 

Figure 1.

 

Experimental sheep pedigrees segregating for OMO. (*) The affected ram used for the experimental mating (families 16–18). The matings for each family represent matings of a single male to multiple unrelated carrier females.

Table 1.

 

List and Human Map Position of Candidate Loci and Map Information for Corresponding Ovine Microsatellite Markers Used in the Initial Linkage Analysis

Region	Candidate Gene	OMIM	Human Chrom.	Sheep Chrom.	cM (SheepMap ver. 4.5)	Microsatellite	No. of Alleles	PIC	CRI-MAP Twopoint Linkage			MERLIN Nonparametric Multipoint Linkage
																		Rec. Frac.	LOD	NPL Score	P
1	EYA3	601655	1p36	1	16.8	INRA197	6	0.70	0.50	−∞	−0.17	0.6
	FOXE3	107250	1p32	1	23.8	EPCDV21	7	0.41	0.50	−∞	−0.50	0.7
				1	40.3	BMS835	11	0.78	0.50	−∞	−0.46	0.7
2	SIX3	603714	2p16-p21	3	66.3	FCB129	7	0.71	0.50	−∞	0.70	0.2
	OTX1	600036	2p13	3	80.6	BMS710	9	0.79	0.50	−∞	0.90	0.2
				3	100.2	ILSTS49	5	0.46	0.50	−∞	0.13	0.4
3	MITF	156845	3p14.2-p14.1	19	31.7	MILVET8	7	0.65	0.50	−∞	−0.79	0.8
				19	64.8	FCB304	6	0.55	0.50	−∞	−0.68	0.8
4	SOX2	206900	3q26.3-q27	1	229.5	BM6506	7	0.38	0.50	−∞	−0.61	0.7
				1	235.3	LS6	11	0.80	0.50	−∞	0.60	0.7
				1	250.3	MCM130	9	0.72	0.50	−∞	0.35	0.6
5	PITX2	601542	4q25-q27	6	16.0	INRA133	8	0.26	0.50	−∞	1.20	0.12
				6	29.7	MCM53	6	0.74	0.16	0.18	0.71	0.2
				6	45.0	MCMA14	6	0.65	0.50	−∞	0.35	0.4
6	FOXC1	601090	6p25	20	30.8	BM1258	6	0.47	0.50	−∞	−0.07	0.5
				20	66.2	BM1818	5	0.65	0.50	−∞	0.24	0.6
				20	91.4	MCMA23	4	0.42	0.21	0.12	0.59	0.3
7	GJA1	164200	6q21-q23.2	8	61.4	BMS434	9	0.60	0.50	−∞	−0.78	0.8
				8	88.6	BMS1724	7	0.46	0.50	−∞	−1.09	0.9
				8	108.0	BMS1967	9	0.61	0.24	0.13	−0.06	0.5
8	SHH	120200	7q36	4	120.0	BMS648	6	0.65	0.50	−∞	−0.06	0.5
				4	139.5	HH64	8	0.44	0.50	−∞	−0.29	0.6
9	EYA1	601653	8q13.3	9	60.1	ILSTS11	5	0.68	0.50	−∞	−0.26	0.6
				9	78.4	MCM42	5	0.56	0.50	−∞	−0.49	0.7
10	PITX3	107250	10q25	22	34.5	BM1314	7	0.51	0.50	−∞	0.50	0.3
				22	57.2	MAF36	7	0.59	0.50	−∞	0.31	0.4
11	PAX6	106210	11p13	15	46.5	MAF65	4	0.55	0.50	−∞	0.27	0.4
	NNO1	600165	11p	15	69.4	HBB2	7	0.72	0.50	−∞	0.18	0.4
				15	79.2	BMS2812	5	0.67	0.50	−∞	0.05	0.5
				15	92.5	SHP4	8	0.75	0.50	−∞	−0.03	0.5
				15	96.0	BMS1660	5	0.69	0.50	−∞	0.11	0.5
12	DACH1	603803	13q22	10	53.9	BMS975	7	0.74	0.50	−∞	−1.15	0.9
				10	71.1	INRA5	6	0.65	0.50	−∞	−0.85	0.8
13	MFRP	605738	15q12-q15	7	75.0	INRA107	8	0.67	0.50	−∞	−0.56	0.7
	BMP4	112262	14q22-q23	7	92.1	INRA71	3	0.53	0.50	−∞	−0.60	0.7
	SIX6	606326	14q22.3-q23	7	105.5	BMS1620	7	0.37	0.50	−∞	−0.53	0.7
	SIX4	606342	14q23	7	117.1	MCM149	11	0.78	0.50	−∞	−0.12	0.5
	CHX10	142993	14q24	7	126.2	BMS2721	6	0.53	0.21	0.06	0.87	0.2
				7	136.7	BMS2614	5	0.59	0.36	0.01	0.60	0.3
14	MCOP	251600	14q32	18	77.0	HH47	10	0.83	0.50	−∞	−1.14	0.9
				18	88.7	TGLA122	11	0.82	0.50	−∞	−0.93	0.8
15	MAF	177075	16q22-q23	14	13.0	TGLA357	7	0.64	0.50	−∞	0.21	0.4
				14	26.6	CSRD47	7	0.56	0.50	−∞	−0.14	0.6
16	TGIF1	602630	18p11.3	23	23.5	BMS2270	6	0.66	0.11	0.84	3.20	0.0007
	RAX	601881	18q21	23	43.1	ADCY1AP	8	0.76	0.03	3.52	4.18	0.00001
				23	67.9	DIK2727	10	0.83	0.22	0.27	1.78	0.04
17	VSX1	122000	20p11.23- p11.22	13	65.0	HUJ616	7	0.56	0.50	−∞	0.31	0.4
	EYA2	601654	20q13.1	13	98.2	CSTBJ12	5	0.68	0.17	0.21	1.04	0.15
	BMP7	112267	20q13	13	115.4	MMP9	7	0.70	0.50	−∞	−0.23	0.6
				13	125.9	BMS995	6	0.66	0.50	−∞	0.34	0.4

 The observed marker features and calculated twopoint LOD scores in regard to OMO are reported. Chrom., chromosome; Rec. Frac., recombinant fraction.

Table 2.

 

OAR23 Microsatellite Markers Used in the Fine Mapping Analysis

Microsatellite	Sheep Chrom. 23 cM (SheepMap v4.7)	Alleles (n)	PIC	Informative Meioses (n)	CRI-MAP Twopoint Linkage			MERLIN Parametric Multipoint Linkage					MERLIN Nonparametric Multipoint Linkage
										Rec. Frac.	LOD	LOD	α	HLOD	NPL Score	P
UW72A	0.0	2	0.23	13	0.47	−∞	−∞	0.36	0.82	1.54	0.06
BL6	2.4	11	0.77	43	0.47	−∞	−∞	0.35	0.87	1.73	0.04
MCM172	2.9	7	0.66	35	0.46	−∞	−∞	0.35	0.86	1.74	0.04
BM226	7.6	8	0.70	19	0.48	−∞	−∞	0.34	0.79	1.77	0.04
BMS2526	8.3	6	0.59	33	0.49	−∞	−∞	0.34	0.78	1.77	0.04
DU280225	18.0	12	0.83	45	0.24	0.32	−∞	0.31	0.92	2.16	0.02
DU330122	18.0	8	0.54	42	0.25	0.25	−∞	0.32	0.98	2.93	0.002
CSRD148	18.6	10	0.83	26	0.39	−∞	−∞	0.32	0.98	2.93	0.002
MCMA1	21.9	8	0.72	32	0.26	0.13	−∞	0.42	1.49	3.45	0.0003
CABB12	22.6	6	0.70	24	0.23	0.23	−∞	0.42	1.47	3.49	0.0002
BMS2270	23.5	6	0.66	32	0.11	0.84	−∞	0.47	1.66	3.63	0.00014
CSSM31	30.3	10	0.51	16	0.07	1.13	−∞	0.56	2.40	3.66	0.00012
ILSTS65	30.9	2	0.34	14	0.10	0.67	−∞	0.55	2.40	3.68	0.00012
DIK4464	34.0	6	0.57	11	0.27	0.04	−∞	0.47	2.11	3.55	0.0002
AGLA269	34.0	14	0.84	17	0.30	0.03	−∞	0.47	2.11	3.55	0.0002
DU268178		9	0.80	48	0.18	1.12	−∞	0.54	2.76	3.65	0.00013
DU216028	34.6	10	0.81	50	0.22	0.72	−∞	0.57	3.00	4.38	0.00001
DU340520	35.1	6	0.67	39	0.25	0.20	−∞	0.58	3.14	4.43	<0.00001
DU252884	35.4	2	0.37	19	0.17	0.51	−∞	0.68	3.82	4.50	<0.00001
DU274690	35.7	4	0.42	19	0.37	0.01	−∞	0.71	4.15	4.53	<0.00001
DU520666	37.7	4	0.27	20	0.00	0.60	6.51	0.91	6.85	4.78	<0.00001
DU416699	41.7	3	0.36	14	0.00	2.11	7.95	0.94	8.09	5.35	<0.00001
DU288059	41.7	4	0.55	16	0.00	0.60	8.32	0.95	8.40	5.47	<0.00001
CL638456		4	0.39	42	0.00	4.21	8.14	0.95	8.25	5.42	<0.00001
DU189183		7	0.60	42	0.04	2.67	8.79	1.00	8.79	5.48	<0.00001
DU377056	41.7	5	0.29	23	0.00	2.11	8.41	0.96	8.45	5.48	<0.00001
DU373930	41.7	8	0.63	19	0.06	1.64	8.35	0.96	8.42	5.47	<0.00001
ADCY1AP	42.7	8	0.76	43	0.03	3.52	9.62	1.00	9.62	5.49	<0.00001
DU239182	43.2	17	0.84	38	0.05	2.17	10.15	1.00	10.15	5.49	<0.00001
DU316110	44.7	6	0.69	33	0.05	1.92	10.70	1.00	10.70	5.50	<0.00001
DU441780	44.7	14	0.68	53	0.03	3.79	10.70	1.00	10.70	5.50	<0.00001
MAF35	44.7	5	0.52	43	0.09	1.53	10.70	1.00	10.70	5.50	<0.00001
DU291927	45.6	5	0.55	48	0.03	3.22	10.59	1.00	10.59	5.41	<0.00001
DU259631	48.1	6	0.59	43	0.04	2.70	−∞	0.85	6.68	5.05	<0.00001
UW69A	48.1	6	0.42	41	0.08	1.46	−∞	0.83	6.60	5.05	<0.00001
DU408420	48.1	3	0.64	16	0.00	2.11	−∞	0.83	6.61	5.05	<0.00001
CSRD172	50.0	3	0.42	25	0.00	2.10	−∞	0.80	5.43	4.70	<0.00001
DU171881	50.0	5	0.62	32	0.12	0.59	−∞	0.79	5.32	4.69	<0.00001
BMS1332	50.0	6	0.56	21	0.00	1.51	−∞	0.74	5.00	4.67	<0.00001
HEL6	51.7	3	0.37	2	0.00	0.30	−∞	0.73	4.24	4.26	0.00001
MCM136	52.8	10	0.80	42	0.09	1.18	−∞	0.63	3.14	4.06	0.00002
DIK5068	62.1	5	0.63	26	0.48	−∞	−∞	0.30	0.79	2.36	0.009
CGBP	63.2	3	0.12	0	0.00	−∞	−7.98	0.32	0.83	2.34	0.01
DIK2727	67.5	10	0.83	17	0.22	0.27	−∞	0.35	0.88	2.27	0.012
DU264615	74.5	5	0.55	34	0.16	0.47	−∞	0.23	0.45	1.51	0.07
URB031	83.9	3	0.23	4	0.00	0.30	−0.51	0.41	0.54	1.17	0.12

 The observed marker features and calculated twopoint and multipoint LOD scores in regard to OMO are reported. Chrom., chromosome; Rec. Frac., recombinant fraction.

Figure 2.

 

Gross appearance of microphthalmic eyes from an affected lamb (left) in comparison with an age-matched control globe (right). In vivo (A), isolated after necropsy (B), and sectioned in a median sagittal plane (C).

Table 3.

 

Pathology Features of Ovine Microphthalmia

Eye Structure	Phenotype Description • Microscopic Details
Eye lids, conjunctiva, nictitating membrane	Regular
Cornea	Diffuse clouded showing distinct vascularization, pigmentation and a granulated surface
	• Variable lymphohistiocytic infiltration
	• Absence of the Descemet’s membrane and posterior epithelium
Anterior eye chamber	Not visible
Lens, vitreous body	Not recognizable
Iris and ciliary body	• Incorporated within or restricted to the periphery of the tissue mass
Posterior eye chamber, retina	White mass composed of connective tissue with blood vessels, smooth muscle, cartilage and fat tissue
	• Anterior: islets of lacrimal glands and multiple cystic structures frequently lined by squamous epithelium and filled with keratin
	• Posterior: partly covered by detached retinal structures sometimes forming rosettes
Optic nerve	Reduced diameter
	• Highly cellular

Figure 3.

 

Results of the parametric (LOD) and nonparametric (NPL) multipoint linkage analysis across OAR23 with 18 half- and full-sib families with the MERLIN software. The maximum parametric LOD score is 10.70 for three markers with relative positions at 44.7 cM. The markers ranging from 42.7 to 45.6 cM are significantly linked to OMO in all 18 families (α = 1). The highest NPL score is 5.50 for three markers with relative positions at 44.7 cM. The calculated nonparametric linkage is highly significant in the region between 18.0 and 63.2 cM with an error probability of P ≤ 0.01.

Table 4.

 

List of Recombinants Present in the Whole Set of Genotyped Sheep Families, Arrayed According to the Position of the Recombination Event^*

Marker	cM	1	2	3	4	5	6	7	8	9	10	11	12	13	14	15	16	17	18	19	20	21	22	23	24
UW72A	0.0	135	135	135	135	135	135	135	135	135	135	135	135	135	135	135	135	133	135	135	135	135	135	133	133
BL6	2.4	190	180	190	186	180	190	180	190	194	186	174	180	190	190	180	180	190	190	190	180	180	190	182	190
MCM172	2.9	143	135	143	141	135	135	135	143	143	135	143	135	135	135	135	135	149	135	147	135	135	135	143	149
BM226	7.6	126	126	140	126	126	124	126	126	124	126	126	126	124	126	126	126	132	126	140	126	140	126	126	132
BMS2526	8.3	150	156	158	150	150	152	150	150	156	150	154	150	154	154	150	150	150	154	150	150	150	150	154	150
DU280225	18.0	442	442	424	436	444	444	444	442	426	432	444	444	432	436	424	444	444	436	442	444	436	434	424	444
DU330122	18.0	291	289	289	289	289	289	289	291	277	303	289	289	303	289	289	289	289	289	289	289	293	287	289	289
CSRD148	18.6	391	397	401	375	401	393	401	391	401	391	397	291	375	375	401	401	401	375	397	291	393	377	401	401
MCMA1	21.9	126	136	140	124	136	124	136	126	150	130	126	136	136	136	136	136	136	136	136	136	126	144	140	136
CABB12	22.6	240	230	230	240	230	232	230	240	228	230	240	230	230	234	240	230	230	234	230	230	240	226	230	230
BMS2270	23.5	94	92	90	94	92	92	92	94	96	92	90	92	92	92	92	92	92	92	92	92	92	92	90	92
CSSM31	30.3	161	131	149	165	131	131	131	161	131	161	131	131	131	131	131	131	131	131	131	131	131	163	149	131
ILSTS65	30.9	115	115	115	115	115	117	115	115	117	115	115	115	115	115	115	115	115	115	117	115	117	115	115	115
DIK4464	34.0	226	224	218	224	224	216	224	226	226	226	230	226	226	226	226	226	226	224	216	226	224	226	218	226
AGLA269	34.0	238	266	272	300	270	300	270	238	266	300	266	270	272	266	266	270	270	266	300	270	300	238	272	270
DU268178		106	106	116	110	110	116	110	106	106	112	116	116	114	114	110	123	123	106	116	116	112	106	116	112
DU216028	34.6	221	213	221	217	213	225	213	221	219	193	193	225	217	217	213	213	193	213	225	193	217	221	221	193
DU340520	35.1	225	219	221	225	225	219	225	225	221	225	221	219	223	223	225	225	225	219	219	221	225	221	221	225
DU252884	35.4	324	324	324	324	324	326	324	324	324	326	326	326	324	324	324	324	326	324	326	326	326	324	324	326
DU274690	35.7	188	188	188	188	188	190	188	188	192	192	182	188	188	192	188	188	188	188	190	182	188	192	188	188
DU520666	37.7	226	226	226	226	224	226	224	226	226	226	226	226	226	226	226	226	226	226	226	224	226	226	226	226
DU416699	41.7	158	158	156	158	158	133	158	158	158	158	158	158	156	158	158	158	158	158	133	158	158	158	156	158
DU288059		110	110	120	110	120	115	120	110	115	115	110	115	120	120	110	110	110	110	115	115	110	120	120	110
CL638456	41.7	182	182	182	182	182	171	182	182	182	182	182	171	182	182	182	182	171	182	171	171	171	182	182	171
DU189183	41.7	213	199	213	205	199	199	199	213	213	213	199	198	199	199	199	199	199	199	199	198	215	199	213	199
DU377056	41.7	289	289	289	289	289	302	289	289	289	289	289	302	289	289	289	289	289	289	302	302	289	289	289	289
DU373930		315	315	315	307	307	307	307	315	315	315	315	307	307	307	315	315	315	315	307	307	318	307	315	315
ADCY1AP	42.7	106	108	87	116	87	108	87	106	114	106	106	108	87	87	106	106	106	108	108	108	114	87	87	106
DU239182	43.2	291	292	292	292	294	280	294	291	272	280	292	261	272	294	292	292	292	292	280	261	292	292	292	292
DU441780	44.7	146	146	210	179	161	146	161	146	196	148	210	146	146	161	210	210	210	146	146	146	210	210	210	210
DU316110	44.7	131	129	131	131	129	131	129	131	131	131	129	132	131	131	129	129	129	129	131	132	131	131	131	129
MAF35	44.7	106	104	106	110	104	106	104	106	104	104	106	106	106	104	106	106	106	104	106	106	106	106	106	106
DU291927	45.6	247	243	247	245	243	247	243	247	243	241	247	247	247	243	247	247	247	243	247	247	243	243	247	247
DU259631	48.1	154	154	158	152	152	154	152	154	154	154	154	154	154	152	152	152	152	154	154	154	154	154	158	152
UW69A	48.1	154	156	154	158	158	154	158	154	158	156	158	158	158	158	158	158	158	156	154	158	154	158	154	158
DU408420	48.1	288	288	286	288	288	286	288	288	290	288	296	294	294	288	288	288	288	288	286	294	288	294	286	288
CSRD2172	50.0	188	188	192	192	192	188	192	188	192	192	188	188	192	188	192	192	192	188	188	188	192	188	192	192
DU171881	50.0	164	178	164	178	162	164	162	164	176	164	164	162	164	162	164	164	162	178	164	162	164	162	164	162
BMS1332	50.0	136	138	142	142	142	140	142	138	142	140	142	140	142	140	142	142	140	138	140	140	140	140	142	140
HEL6	51.7	181	181	185	185	185	185	185	185	185	185	181	185	185	185	185	185	185	181	185	185	185	185	185	185
MCM136	52.8	148	142	138	146	164	164	138	148	144	164	142	142	164	142	140	140	140	142	164	142	164	142	146	140
DIK5068	62.1	90	92	92	90	92	92	92	90	98	90	90	90	92	90	90	90	92	90	94	90	94	90	92	92
CGBP	63.2	265	265	265	265	265	265	265	265	265	265	265	265	265	265	265	265	265	265	265	265	265	265	265	265
DIK2727	67.5	208	194	194	194	202	212	202	208	226	208	206	212	208	212	208	208	184	212	212	212	212	212	194	184
DU264615	74.5	118	124	124	118	124	116	124	118	124	118	116	116	124	118	118	124	116	118	116	116	124	116	124	116
URB031	83.9	234	234	234	234	234	234	234	234	234	234	234	234	234	234	234	234	234	234	234	234	234	234	234	234

*  The alleles of markers on the chromosome containing the OMO locus of each individual is shown as a column. The unshaded versus dark-shaded portions show the position of the marker where recombination was first observed. Regions of light shading indicate uninformative markers. Therefore the critical region is delimited by the gray bars. The haplotype analysis suggests the position of OMO between the two bold typed flanking markers.

Figure 4.

 

Comparative chromosome maps of the ovine microphthalmia critical interval. The human-sheep alignment is based on the OAR23 radiation hybrid map ¹² indicating two different corresponding HSA18 blocks of conserved gene order. The sheep–cattle alignment is based by sequence alignments of microsatellite markers used within this study and the current bovine assembly (see http://www.ncbi.nlm.nih.gov/mapview/map_search.cgi?taxid=9913). It indicates an approximate size of 9.5 Mb genomic chromosome segment for the critical interval between the markers DU274690 and DU259631, identified by haplotype analysis. #, two markers that were not mapped on the ovine linkage map. Their position was inferred from comparative sequence analysis between sheep and cattle. *, the single marker that could not be placed in the cattle genome sequence.

The authors thank Hauke Peters for excellent assistance with collecting the blood samples.