Structure, Chromosomal Location, and Tissue-Specific Expression of the Mouse Opticin Gene

Abstract

purpose. To determine the structure, location, and tissue-specific expression of the mouse opticin gene (Optc) and to compare expression in the eye with that of Prelp, collagen II, and collagen IX.

methods. Expressed sequence tags (ESTs) to mouse opticin were identified and the full-length sequence obtained after PCR reactions using a 15-day-postconception (dpc) whole-mouse embryo cDNA library. The mouse chromosomal localization of Optc was determined by radiation hybrid mapping and its genomic structure determined using an Optc-containing BAC clone. Tissue-specific expression of opticin, PRELP, collagen II, and collagen IX mRNAs was investigated by in situ hybridization and by dot blot hybridization for opticin.

results. The Optc gene was localized to mouse chromosome 1 at 74.3 cM and consisted of seven exons spanning 10 kb. The Optc gene was less than 4 kb from the Prelp gene. In situ hybridization localized opticin mRNA exclusively to the presumptive ciliary body during development and to the nonpigmented ciliary epithelium of the adult mouse eye. Expression of Prelp was also detected in the nonpigmented ciliary epithelium of the adult eye. However, expression of collagen types II and IX was detected largely in the developing mouse eye, with type IX expression confined primarily to the presumptive ciliary body.

conclusions. The Optc, Prelp, and fibromodulin (Fmod) genes form a cluster on mouse chromosome 1. Opticin may represent a marker for ciliary body differentiation. Continued expression of opticin in the adult mouse eye suggests functions other than that of putative regulator of vitreous collagen fibrillogenesis.

Opticin is a member of the family of extracellular matrix small leucine-rich repeat proteins (SLRPs) and was originally discovered associated with the collagen fibrils of bovine vitreous. ¹ Subsequently, opticin cDNAs were demonstrated in libraries from adult human iris, ^{2 3} and opticin cDNA was found to be one of the most common transcripts (0.7%) in a library prepared from this tissue. ³ Studies in human tissue have also shown the presence of opticin in brain by Western blot analysis ³ and in ligament and skin by RT-PCR, ¹ but opticin expression has not been detected in several other tissues that have been investigated. 

The SLRPs are characterized by tandem leucine-rich repeats, which may form a horseshoe-shaped solenoid-like structure. ^{4 5 6} They are divided into three classes according to the number of leucine-rich repeats, the spacing of a four-cysteine cluster located N-terminal to the leucine-rich repeats, and the number of exons that encode the gene. Opticin is a class III member, along with epiphycan and osteoglycin/mimecan. ^{7 8 9} Most of the SLRPs are proteoglycans, but some of them (including opticin and PRELP) are not, and opticin is unique in that it is glycosylated with sialylated O-linked oligosaccharides. ¹ Recently, it was reported that the genes for the SLRPs keratocan, decorin, lumican, and epiphycan are present as a cluster on human chromosome 12q22-23. ¹⁰ The human opticin gene (OPTC) has been localized to 1q31-q32 ^{2 3} and two other SLRP genes, PRELP ^{11 12} and fibromodulin (FMOD), ¹³ have been localized to 1q32 in the order FMOD, PRELP, and OPTC. ⁹  

The biological function of the SLRPs in vivo is still somewhat unclear, although several members of the SLRP family bind to collagen fibrils through their leucine-rich repeat domains and appear to regulate fibril diameter and spacing. ¹⁴ Knockout models of mice without decorin, ¹⁵ fibromodulin, ¹⁶ or lumican ¹⁷ all support this hypothesis, in that the mice show irregular forms of collagen fibrils primarily in the dermis, tendon, and cornea, respectively. The vitreous gel contains a dilute network of thin collagen fibrils (composed of collagen types II, IX, and V/XI) that are essential to maintain its structure. ¹⁸ It therefore appears likely that opticin is involved in fibrillogenesis of collagen molecules to form the vitreous gel and is potentially involved in maintaining the spacing between the collagen fibrils of the tissue. ¹⁹  

In this study, we determined the gene structure of mouse Optc, its chromosomal localization, and its relationship to the Prelp and Fmod genes. We also investigated mRNA expression in developing and adult mouse eye and other tissues and compared expression with PRELP, collagen type II, and all three chains of collagen type IX. 

Materials and Methods

cDNA Cloning of Mouse Opticin

Human opticin cDNA (AJ133790) was used to search GenBank (provided by the National Institutes of Health and available in the public domain at http://www.ncbi.nlm.nih.gov) for mouse ESTs to opticin. Based on the sequence of several EST clones, specific primers were designed to mouse opticin both in sense and antisense directions. PCR was performed using a 15-day-postconception (dpc) mouse embryo cDNA library (Clontech, Palo Alto, CA) with vector primers. PCR products were subcloned into a vector (pGEM-T Easy; Promega, Madison, WI) and sequenced manually by using a kit (Thermosequenase; Amersham Pharmacia Biotech, Piscataway, NJ). To establish the potential full-length cDNA of mouse opticin, including 5′- and 3′-untranslated regions (UTRs), the cDNA sequences determined by PCR reactions were compared with the mouse EST database. Several EST clones including 5′- or 3′-UTRs were found, and the full-length cDNA sequence of mouse opticin was assembled from the sequences of PCR products and EST clones. The putative transcriptional start site was designated from the longest available 5′ EST sequence (AA223016). 

Annotation of a BAC Clone Containing Fmod, PRELP, and Optc

A mouse BAC clone (control number 24885, clone address 164[K13]) containing the Optc gene was obtained after library screening by Incyte Genomics (St. Louis, MO) using the following mouse opticin primers: OP17, 5′-CTG GAT TCC ATC CCT GGG CCT-3′, and OP29, 5′-ACA GGG AGC CGT GGC AGG CA-3′. To analyze the genomic structure of mouse Optc, primers designed from the cDNA sequence were used for intronic bridging PCR reactions. PCR products were subcloned into a vector (pGEM-T Easy; Promega) and sequenced on an automated sequencer (model 377; Perkin Elmer-Applied Biosystems, Inc., Foster City, CA) using vector primers. A mouse BAC clone that contains the Fmod, Prelp, and Optc genes is present in GenBank (AC026760), but the annotation is presently not complete. Similarly, human BAC clones containing the PRELP and OPTC genes (AC022000, AL391817) were identified together with a third human BAC clone that contains FMOD and PRELP (AL359837). The human BAC clones were further annotated by using cDNA sequences of PRELP and opticin, and the remaining gaps were filled in by performing bridging PCR reactions on BAC DNAs. In addition, PCR reactions were performed to determine the distance between the mouse Prelp and Optc genes by using AC026760 with the following primers: PR1, 5′-CGT CCC TTC GTG AGG ATG GT-3′, and OP14, 5′-GGG GAC AGC TAG ACA CCC CCA-3′. The PCR product was ligated to the vector (pGEM-T Easy; Promega) and sequenced. 

In Situ Hybridization

Probes used for in situ hybridization were as described in Table 1 . All probes were amplified by PCR from a 15-dpc embryo cDNA preparation (Marathon-Ready; Clontech). PCR products were subcloned into a vector (pBluescript-SK; Stratagene, La Jolla, CA). Antisense and sense riboprobes were transcribed by T3 and T7 RNA polymerase by a commercial system (In Vitro Transcription System; Promega) in the presence of ³³P-UTP (1000 Ci/mmol). 

Albino BALB/c or 129SV mice of agouti color were used for in situ hybridization. In this manner, the detailed morphology of ocular structures could be determined from the presence or absence of pigmented cells in the uveal tract. All animals were treated humanely in accordance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. Coitus was confirmed by the presence of a vaginal plug the next morning and designated as 0.5 dpc. Whole selected embryos (12.5, 15.5, and 17.5 dpc) together with whole eyes dissected from adult mice were embedded in optimal cutting temperature (OCT) compound and frozen in dry ice-acetone immediately. Thick sections (10 μm) were cut in a cryotome at a parasagittal angle that passed through the eye. The sections were fixed in 4% paraformaldehyde and digested with 0.2 μg/ml proteinase K for 5 minutes at room temperature. After acetylation with 0.25% acetic anhydride, sections were dehydrated with a graded series of ethanol. Hybridization was performed using a ³³P-UTP-labeled riboprobe overnight at 50°C. Sections were washed in 50% formamide in 2× SSC for 30 minutes at 55°C, followed by digestion with RNase A. Sections were washed again under highly stringent conditions (0.1× SSC for 60 minutes at 60°C) and then dehydrated with ethanol. Sections were exposed in emulsion (Hypercoat LM-1; Amersham Pharmacia Biotech) for 3 to 6 days at 4°C and developed (D-19 developer; Eastman Kodak, Rochester, NY). Sections were visualized, and photomicrographs were taken under a dark-field microscope (model DC200; Leica, Deerfield, IL) and processed using image management software (Photoshop; Adobe, San Jose, CA). 

Dot Blot Hybridization

A mouse multiple-tissue expression array (Clontech) was used for dot blot hybridization. An 870-bp PCR product (forward primer: 5′-CAA CCA CTC GTC CTC TCC CT-3′; reverse primer: 5′-CAG CCG ATT CAG GCG GAC GT-3′) obtained from a 15-dpc embryo cDNA (Marathon-Ready; Clontech) was used as a probe. The probe was labeled with ³²P-dCTP (3000 Ci/mmol; Random Primed DNA Labeling Kit; Roche Molecular Biochemicals, Indianapolis, IN). Hybridization was performed at 65°C overnight in hybridization solution (ExpressHyb; Clontech). The membrane was washed with 2× SSC and 1% SDS at 65°C and 0.1× SSC and 0.1% SDS at 50°C. The membrane was exposed to x-ray film (XAR-5; Kodak) with intensifying screens for 2 days at −80°C before the film was developed. 

Radiation Hybrid Mapping of the Mouse Optc Gene

The mouse Optc gene was mapped by screening a mouse/hamster RH panel (Research Genetics, Huntsville, AL) using the following primers: I-1, 5′-ATA GGT CTC GCA TGG GCC AA-3′, and I-4, 5′-TGG CTC TTT CTA GGA CCT CT-3′. The data were analyzed by computer (Auto-RHMAPPER, provided by Whitehead Institute/Massachusetts Institute of Technology Center for Genome Research radiation hybrid map of the mouse genome and available in the public domain at http://www.genome.wi.mit.edu/cgi-bin/mouse_rh/rhmap-auto/rhmapper.cgi). The mouse Optc gene was mapped at 74.3 cM on mouse chromosome 1 and placed 5.23 cR from D1Mit348 (lod >3.0). 

Results

cDNA Sequence of Mouse Opticin

The full-length sequence of mouse opticin cDNA was 1901 bp. Mouse opticin cDNA had a 290-bp UTR at the 5′ end and a 618-bp UTR at the 3′ end, both of which were longer than reported for human opticin cDNA (GenBank AF333980 for the mouse was submitted with this manuscript, AJ133790 for humans). The open reading frame (ORF) of human and mouse sequences was almost the same length, 999 bp and 986 bp, respectively. Alignment of the human and mouse cDNA sequences showed that the sequence of opticin was well conserved (Fig. 1) . Identity of the ORF was 77% at the nucleotide level and 73% at the amino acid level. The seven leucine-rich repeat domains and the position of all cysteine residues were conserved in both species. The signal peptide cleavage site was determined with sequence analysis software (SignalP ver. 1.1; provided in the public domain by the Center for Biological Sequence Analysis, Technical University of Denmark, Lyngby, Denmark, and available at http://www.cbs.dtu.dk/services/SignalP/). The alignment was performed on computer (ClustalW ver.1.81; SGI, Mountain View, CA). 

Structure of the Optc Gene

The genomic structure of mouse Optc was elucidated by exon-bridging PCR reactions using a BAC clone (AC026760) that was found to contain the whole Optc gene but was not annotated. PCR products were cloned and subsequently sequenced to complete the annotation of the Optc gene (Fig. 2A) . The mouse Optc gene had seven exons and the total size of the gene was approximately 10 kb. (The complete sequence of the mouse Optc gene was submitted to GenBank AF333981). The human OPTC gene was shown to have eight exons (Fig. 2A) . Exons 1 to 6 of the mouse Optc gene correspond to the human OPTC gene, but the last exon in the 3′-UTR is different. In human OPTC, exon 8 contains the polyadenylation signal and is located 5 kb downstream of exon 7 (Fig. 2A) . In mouse Optc, exon 8 does not exist, and a longer exon 7 contains the polyadenylation signal (Fig. 1) . Exon 7 of mouse Optc includes a 3′-UTR of 618 bp. The consensus sequences for each exon–intron boundary of both human OPTC and mouse Optc are shown in Figure 2B . 

In humans, the OPTC gene is clearly present in a gene cluster with FMOD and PRELP on chromosome 1q32 (GenBank NT_004523) with FMOD considered to be located 128 kb upstream of PRELP (GenBank Chromosome 1 Contig Map). However, some gaps still existed in the available sequence. We, therefore, performed PCR reactions to fill the gap between the PRELP and OPTC genes, together with a gap within the OPTC gene, by using human BAC clone (AC022000). The distance from the 3′ end of the longest transcript of PRELP to the 5′ end of OPTC was found to be 4.5 kb in humans and was completely resequenced. 

In the case of the mouse, we attempted to annotate a mouse BAC clone (AC026760) to elucidate the relationship of each gene at this locus. Although this BAC clone clearly contained all three genes, it was not possible to annotate the orientation of Fmod and its distance from Prelp and Optc. The published cDNA sequence of mouse PRELP did not contain the full length of the 3′-UTR (GenBank AF261886), and the downstream sequence from mouse BAC clone AC026760 was therefore used to survey the EST database. Several ESTs were identified that overlapped with the previously published sequence of exon 3 of mouse Prelp. The longest EST identified that contained a polyadenylation signal at the correct location was determined to be 2.5 kb by comparison with BAC clone AC026760. The longest mRNA of mouse Prelp was therefore predicted to be 4.5 kb, and this prediction was confirmed by Northern blot analysis (Henry SP, unpublished observations, 2000). PCR reactions were performed to bridge from the 3′ end of mouse Prelp to the Optc gene, assuming the same relationship between these genes as occurs in humans. PCR products were cloned and sequenced. The results show that the mouse Prelp gene is located approximately 3.5 kb upstream of the Optc gene and that they are arranged on the same DNA strand in a head-to-tail manner. The Prelp and Optc genes therefore form a very similar gene cluster in both the human and the mouse. We were unable to annotate the exact location and orientation of the mouse Fmod gene, but preliminary results suggest that it, as in the human, lies at a considerable distance upstream of Prelp. There are no apparent intervening genes, in either the human or mouse, between FMOD and PRELP. 

Expression of mRNA for Opticin, PRELP, and Type IX Collagen during Eye Development in the Mouse

The eye from a 12.5-dpc embryo did not show any signal of opticin by in situ hybridization (Figs. 3A 3B) . In the 15.5-dpc embryo, opticin expression was observed specifically in the region of the anterior tip of the developing retina (Figs. 3C 3D) . No significant signal for opticin was detected in any other structure in the eye, including the lens and cornea. In the 17.5-dpc embryonic eye, a strong signal was observed only in the region of the developing ciliary body (Figs. 3E 3F) . During development, at 12.5, 15.5, and 17.5 dpc no clear signal for opticin expression was detected in any other tissue of the whole embryo (data not shown). In the adult eye, the signal for opticin was detected only in the nonpigmented epithelium of the ciliary body in both the BALB/c albino mouse (Figs. 4A 4B) and the agouti mouse (Fig. 4C) . No other structures showed a positive signal for opticin mRNA. No signal was observed on the slide with the use of a sense control in the agouti mouse, in which the pigment is clearly observed (Fig. 4D) . 

The expression of PRELP mRNA could not be detected during development of the eye at 12.5, 15.5, or 17.5 dpc (data not shown), but PRELP expression was seen in other tissues during development, including cartilaginous components of the vertebrae and ribs, the cochlea, and the neurohypophysis (Figs. 5A 5B 5C 5D) . However, in the adult eye, PRELP mRNA was detected in the nonpigmented epithelium of the ciliary body in a pattern of expression that resembled that of opticin mRNA (Figs. 4E 4F) although the signal was much weaker. 

In the 17.5-dpc embryonic eye, mRNA of COL9A1 was expressed in the retina and developing ciliary body (Figs. 6A 6B) . A weak signal was also detected in the cornea, but not in the lens or sclera, confirming earlier results. ^{20 21} COL9A2 mRNA was detected only in the developing ciliary body, similar to opticin mRNA (Figs. 6C 6D) . No positive signal was observed in the retina, cornea, lens, iris, or sclera. COL9A3 mRNA was weakly detected in the ciliary body (Fig. 6E and arrowhead in Fig. 6F ), and, unexpectedly, was also present in the choroid. In the adult eye, mRNA of COL9A1 was detected in the ciliary body and inner nuclear layer of the retina (Figs. 4G 4H) . Expression of COL9A2 and COL9A3 was also detected in the ciliary body, but the signal was extremely weak (data not shown). The expression of COL2A1 was detected in the developing cornea, retina, and sclera of the embryonic eye at 17.5 dpc (Figs. 6G 6H) . In the adult eye, COL2A1 expression was detected in the lens and cornea as well as in the retina, sclera, and ciliary body. These expression patterns for COL2A1 were similar to those previously observed by Savontaus et al. ^{22 23} In all experiments, sense control probes for each gene were also processed and no signal was detected above background. 

Dot blot hybridization was performed to survey the expression of opticin in adult tissues. The probe used for these experiments was longer than that used for in situ hybridization but included the same sequence. A strong signal was detected for adult eye mRNA after dot blot hybridization (Fig. 7) , and a weak signal was also detected for heart, brain, testis, thyroid, and epididymis. It is uncertain whether the weak signals represent a low level of opticin expression or cross-reactivity with the mRNA of other molecules, especially other members of the SLRP family. No other tissue-specific RNA present on the blot showed a positive signal. 

Discussion

In this study, we present the gene structure of mouse Optc and investigate mRNA expression in the developing and adult mouse eye. Expression of opticin was first detected in the 15.5-dpc embryonic eye at the anterior tip of the developing neuroretina. However, it was impossible to distinguish morphologically a particular cell type expressing opticin mRNA in the hematoxylin-eosin–stained sections at this stage. In the 17.5-dpc embryonic eye, opticin mRNA was positive in cells developing between the iris and neuroretina. Because expression of opticin mRNA was specifically restricted to the nonpigmented epithelium of the ciliary body in the adult, the positive cells in the 15.5- and 17.5-dpc embryonic eyes probably are committed to become this cell type. A previous study showed that the developing ciliary body could be distinguished morphologically in the mouse embryonic eye at approximately 16 dpc. ²⁴ Our in situ hybridization data indicate that functional differentiation of the ciliary body had started by 15.5 dpc and that opticin mRNA may represent a marker for ciliary body differentiation. 

In humans, cDNA transcripts for opticin have been observed at high levels in adult iris libraries. ³ In addition, most of the human opticin ESTs that can be identified in GenBank are reportedly derived from retinal cDNA. By contrast, in the current study opticin expression was observed by in situ hybridization only in the ciliary body. It is possible that the human tissues used to make these cDNA libraries were contaminated by ciliary body or that the pattern of opticin expression is different in humans. We are currently undertaking further studies to resolve this issue. 

Although opticin expression was observed only by in situ hybridization in the ciliary body and dot blot analysis revealed a strong opticin mRNA signal only from the eye, several mouse opticin ESTs were found in GenBank from nonocular tissues, including brain, kidney, mammary gland, uterus, urinary bladder, pineal gland, and cerebellum. Taken together, these data suggest that opticin expression occurs only at high levels in the eye, but low-level expression may be present in many other tissues, such as parts of the brain. 

The functions of opticin in the eye remain uncertain. The nonpigmented epithelium of the ciliary body is assumed to be the source of opticin in the vitreous cavity. In humans, opticin has been shown to be closely associated with the collagen fibrils of vitreous ¹ suggesting a role in vitreous collagen fibrillogenesis. In mice, we have confirmed that the ciliary body expresses type IX collagen ^{20 21 22} as well as opticin, whereas the type II collagen of the vitreous appears to be partially or totally derived from other eye tissues both in the mouse ^{20 22 23} and chicken. ²⁵ Why type IX collagen and opticin should be expressed only in the nonpigmented layer of the presumptive ciliary body is not clearly understood, nor is it understood whether they are coordinately regulated. In previous work, expression of α1(IX) was detected in the optic cup as early as 10.5 dpc and began to be concentrated in the anterior part of the retina as early as 13.5 dpc. ²⁰ However, type IX collagen expression appeared to be decreased in the adult mouse (Fig. 4H) , whereas opticin expression appeared to be maintained (Fig. 4B) . This suggests that opticin may perform other functions and is being continuously released into the vitreous cavity of the adult mouse. However, the adult mouse has very little if any vitreous gel, and it therefore appears unlikely that opticin is necessary to maintain the spacing between the collagen fibrils within the gel. 

The data provided by studying the expression of the individual genes encoding the three type IX collagen chains (Col9a1, Col9a2, and Col9a3) in the 17.5-dpc embryonic mouse eye was intriguing. Although, all three genes were preferentially expressed in the ciliary body, Col9a1 was also expressed by the retina. By contrast, the Col9a2 gene was apparently expressed in the choroid, but Col9a1 and Col9a3 were not. The significance of these differential expression patterns remains uncertain. Expression of the three genes would be expected to be in a 1:1:1 ratio to allow stoichiometric assembly of the three chains of type IX collagen. ^{26 27}  

PRELP expression has been demonstrated in tissues including cartilage and lung, ¹² but expression in the eye is a novel finding. By contrast, high-level expression of opticin is found in the eye, but expression levels are low in other tissues and expression could not be demonstrated in cartilage or lung. ¹ It is possible that PRELP is secreted by the ciliary body and has a function in the vitreous humor, although to date we have not been able to demonstrate the presence of any members of the SLRP family apart from opticin in the bovine vitreous humor. Alternatively, it is possible that, because the Prelp and Optc genes are so close (<4 kb), a regulatory element that drives opticin expression also moderately affects PRELP expression. 

In conclusion, OPTC is a well-conserved gene between humans and mice and forms part of a gene cluster with FMOD and PRELP. Expression of opticin in the adult mouse eye is specifically localized to the nonpigmented epithelium of the ciliary body. The high level of opticin expression in the adult eye suggests that it has an important, as yet undetermined role in maintaining homeostasis within the eye. 

 Supported in part by Grant R37 AR30481 from the National Institutes of Health and by a grant from Fight for Sight. PNB is a Wellcome Trust Senior Research Fellow in Clinical Science.

 Submitted for publication February 15, 2001; revised April 27, 2001; accepted May 15, 2001.

 Commercial relationships policy: N.

 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.

 

Table 1.

 

Probes Used for In Situ Hybridization of Mouse Tissues

Gene	Primer	Sequence	Length (bp)	Location
Optc	OP2	5′-AGCTTGGTGCTGCAGAAGGCA-3′	620	Exon 2–4
	OP11	5′-CAGCCGATTCAGGCGGACGT-3′
Prelp	PRELP-S	5′-GACACACGCAGACAGGCACCAACTGGGAGAC-3′	550	Exon 1–2
	PRELP-A	5′-GAATTCGGTTGTCCAGGTTGACCCACCTCAGG-3′
Col2a1	2A1-S	5′-ACACACTGGTAAGTGGGGCA-3′	435	3′-UTR
	2A1-A	5′-TGGGGCTGGGAACAGTCACT-3′
Col9a1	9A1-S	5′-GTGCTCTTGGCTTAAGAGGA-3′	400	3′-UTR
	9A1-A	5′-TGATGTCAGAGGTGAAACCT-3′
Col9a2	9A2-S	5′-AGCAACCAGCCAGGACAGAG-3′	470	3′-UTR
	9A2-A	5′-TACACAAAGGCCAGAGTGGT-3′
Col9a3	9A3-S	5′-CTTCAGTAGGAAATGGCTCC-3′	500	3′-UTR
	9A3-A	5′-CAGATGGTGCAGTGTAGTTC-3′

 S, sense; A, antisense.

Figure 1.

 

The alignment of human and mouse cDNA of opticin and deduced protein sequences. Lowercase letters: 5′- and 3′-UTR sequences; uppercase letters: ORFs; box: translation initiation codon and termination codon; underline: leucine-rich repeat domains; highlight: conserved cysteine residues; double underline: polyadenylation signal; arrow: signal peptide cleavage site.

Figure 2.

 

(A) Genomic organization of OPTC and PRELP in humans and the mouse. The PRELP and OPTC genes have been mapped on chromosome 1q32 for humans. The cluster of human genes was derived from annotated sequences (GenBank NT_004523). Because sequencing of the cluster is presently not complete, each intron for human OPTC is not represented as an exact size (gap marked by vertical dashed line). The genomic structure of human PRELP was determined by annotation of a BAC clone (AC022000) and with an EST clone to the longest 3′-UTR (AI584095). Although the FMOD gene is clearly upstream of PRELP, it was not included in the diagram, because the exact orientation and location of the gene is still unknown. (B) Exon–intron boundary of the human and mouse OPTC genes. Uppercase letters: exon sequences; lowercase letters: flanking intronic sequences. The sequences for human were derived from BAC clone AC022000 and for mouse were determined by sequencing of PCR products from BAC clone AC026760.

Figure 3.

 

Messenger RNA expression of opticin during development of the mouse eye: (A, B) 12.5-dpc, (C, D) 15.5-dpc, and (E, F) 17.5-dpc mouse embryonic eyes. (A, C, and E) Bright-field views; (B, D, and F) dark-field views. (A, B) No positive signal was observed in the 12.5-dpc embryonic eye. (C, D) The small regions at the tip of the neuroretina showed a signal in the 15.5-dpc embryonic eye (D, arrowhead). (E, F) A strong signal was observed only in the small areas between the retina and iris in the 17.5-dpc embryonic eye (F, arrowhead). Scale bar, (A, B) 150 μm; (C, D) 300 μm; (E, F) 400 μm.

Figure 4.

 

Messenger RNA expression of opticin, PRELP, and COL9A1 in the adult mouse eye. All micrographs show only the region in the vicinity of the ciliary body. (A, B) Bright- and dark-field views of the ciliary body of an albino mouse. Opticin messenger RNA was specifically expressed in the nonpigmented epithelial cells of the ciliary body (B, arrowhead). (C, D) Dark-field, high-magnification views of the ciliary body of a pigmented mouse. Section in (C) was processed with an antisense probe and in (D) with a sense probe. The pigmented cells in the retina, ciliary body, and iris are red-brown. Positive signal was clearly seen in the nonpigmented epithelium of the ciliary body (C, arrowhead). A sense control did not show any signal (D). (E, F) Bright- and dark-field views of expression of PRELP. A relatively weak signal is seen in the nonpigmented epithelium of the ciliary body (F, arrowhead). (G, H) Bright- and dark-field views showing expression of COL9A1. A signal was observed in the nonpigmented epithelium of the ciliary body and inner nuclear layer of the retina (H, arrowhead). c, cornea; cb, ciliary body; gl, ganglion layer; i, iris; in, inner nuclear layer; on, outer nuclear layer, vb, vitreous body. Scale bar, (A, B, E–H) 200 μm; (C, D) 100 μm.

Figure 5.

 

Messenger RNA expression of PRELP in the 15.5-dpc mouse embryo. Signal was seen in the vertebral column (A) and ribs (B). The signal was strong in the peripheral region of the cartilage in both vertebrae and ribs. Signal was also seen in the cartilaginous component of the cochlea (C) and the pituitary gland (D). The neurohypophysis (D, arrow) and a part of the forebrain were positive. CH, cochlea; SB, sphenoid bone; AH, adenohypophysis; N, neurohypophysis; R, rib; S, spinal cord; V, vertebral column. Scale bar, (A, B) 275 μm; (C, D) 400 μm.

Figure 6.

 

Messenger RNA expression of chains of type IX collagen in the 17.5-dpc embryonic eye (A–F) and type II collagen (G, H). (A, C, E, and G) Bright-field views; (B, D, F, and H) dark-field views. (A, B) Expression of COL9A1 mRNA. The ciliary body showed a strong signal. In addition, a positive signal was seen in the cornea and outer nuclear layer of the retina. (C, D) Expression of COL9A2 mRNA. A positive signal was seen only in the presumptive ciliary body. (E, F) Expression of COL9A3 mRNA. The presumptive ciliary body showed a weak signal for COL9A3 mRNA (F, arrowhead). For unknown reasons, a strong signal was seen in the choroid. (G, H) Expression of COL2A1 mRNA. A positive signal was seen in the developing cornea, retina, and sclera. Scale bar, 400 μm.

Figure 7.

 

A mouse multiple tissue expression array was used for dot blot hybridization to investigate expression of opticin mRNA in adult tissues. The eye showed a strong signal for opticin. Weak signals were detected in the heart, brain, testis, thyroid, and epididymis.