August 2011
Volume 52, Issue 9
Free
Retina  |   August 2011
Full-Length Transcriptome Analysis of Human Retina-Derived Cell Lines ARPE-19 and Y79 Using the Vector-Capping Method
Author Affiliations & Notes
  • Mio Oshikawa
    From the Department of Rehabilitation Engineering, Research Institute, National Rehabilitation Center for Persons with Disabilities, Tokorozawa, Japan;
  • Chihiro Tsutsui
    From the Department of Rehabilitation Engineering, Research Institute, National Rehabilitation Center for Persons with Disabilities, Tokorozawa, Japan;
    the Department of Applied Chemistry, Graduate School of Engineering, Toyo University, Kawagoe, Japan; and
  • Tomoko Ikegami
    From the Department of Rehabilitation Engineering, Research Institute, National Rehabilitation Center for Persons with Disabilities, Tokorozawa, Japan;
    the Department of Applied Chemistry, Graduate School of Engineering, Toyo University, Kawagoe, Japan; and
  • Yuki Fuchida
    From the Department of Rehabilitation Engineering, Research Institute, National Rehabilitation Center for Persons with Disabilities, Tokorozawa, Japan;
    the Department of Applied Chemistry, Graduate School of Engineering, Toyo University, Kawagoe, Japan; and
  • Maki Matsubara
    From the Department of Rehabilitation Engineering, Research Institute, National Rehabilitation Center for Persons with Disabilities, Tokorozawa, Japan;
  • Shigeru Toyama
    From the Department of Rehabilitation Engineering, Research Institute, National Rehabilitation Center for Persons with Disabilities, Tokorozawa, Japan;
  • Ron Usami
    the Department of Applied Chemistry, Graduate School of Engineering, Toyo University, Kawagoe, Japan; and
  • Kuniyo Ohtoko
    Hitachi High-Technologies Corporation, Hitachinaka, Japan.
  • Seishi Kato
    From the Department of Rehabilitation Engineering, Research Institute, National Rehabilitation Center for Persons with Disabilities, Tokorozawa, Japan;
  • Corresponding author: Seishi Kato, Research Institute, National Rehabilitation Center for Persons with Disabilities, Namiki 4-1, Tokorozawa, Saitama 359-8555, Japan; [email protected]
Investigative Ophthalmology & Visual Science August 2011, Vol.52, 6662-6670. doi:https://doi.org/10.1167/iovs.11-7479
  • Views
  • PDF
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Mio Oshikawa, Chihiro Tsutsui, Tomoko Ikegami, Yuki Fuchida, Maki Matsubara, Shigeru Toyama, Ron Usami, Kuniyo Ohtoko, Seishi Kato; Full-Length Transcriptome Analysis of Human Retina-Derived Cell Lines ARPE-19 and Y79 Using the Vector-Capping Method. Invest. Ophthalmol. Vis. Sci. 2011;52(9):6662-6670. https://doi.org/10.1167/iovs.11-7479.

      Download citation file:


      © ARVO (1962-2015); The Authors (2016-present)

      ×
  • Supplements
Abstract

Purpose.: To collect an entire set of full-length cDNA clones derived from human retina-derived cell lines and to identify full-length transcripts for retinal preferentially expressed genes.

Methods.: The full-length cDNA libraries were constructed from a retinoblastoma cell line, Y79, and a retinal pigment epithelium cell line, ARPE-19, using the vector-capping method, which generates a genuine full-length cDNA. By single-pass sequencing of the 5′-end of cDNA clones and subsequent mapping to the human genome, the authors determined their transcriptional start sites and annotated the cDNA clones.

Results.: Of the 23,616 clones isolated from Y79-derived cDNA libraries, 19,229 full-length cDNA clones were identified and classified into 4808 genes, including genes of >10 kbp. Of the 7067 genes obtained from the Y79 and ARPE-19 libraries, the authors selected 72 genes that were preferentially expressed in the eye, of which 131 clones corresponding to 57 genes were fully sequenced. As a result, we discovered many variants that were produced by different transcriptional start sites, alternative splicing, and alternative polyadenylation.

Conclusions.: The bias-free, full-length cDNA libraries constructed using the vector-capping method were shown to be useful for collecting an entire set of full-length cDNA clones for these retinal cell lines. Full-length transcriptome analysis of these cDNA libraries revealed that there were, unexpectedly, many transcript variants for each gene, indicating that obtaining the full-length cDNA for each variant is indispensable for analyzing its function. The full-length cDNA clones (approximately 80,000 clones each for ARPE-19 and Y79) will be useful as a resource for investigating the human retina.

The retina is composed of a neural cell layer and a retinal pigment epithelial cell layer. The two types of photoreceptor cells, rods and cones, in the neural cell layer convert light signals to changes in membrane potential, organized through complex layers of the neural cells and transmitted to the brain through the fibers of the optic nerve. 1,2 The retinal pigment epithelium (RPE) plays important roles in supporting the function of the photoreceptor cells and serves as a blood-retina barrier. 3,4 Photoreceptor cells and the RPE are not only physiologically interesting, they are pathologically important in relation to retinal degeneration. 5 8 Furthermore, photoreceptor cells serve as a model for the investigation of the development and differentiation of neural cells. 9  
One experimental approach to understand the physiological function of photoreceptors and the RPE at a molecular level is to perform a transcriptome analysis of the genes expressed in these cells. 10 Previously, transcriptome analyses of the human retina were performed using expression profile analysis of expressed sequence tags (EST), 11 microarray, 12 or serial analysis of gene expression (SAGE) 13 ; however, there are problems with each of these methods. First, the tissues used for these analyses contained various cell types; thus, the results did not reflect the expression profile of a single cell type. Second, every analysis was based on the partial sequence of transcripts, and recent studies revealed that many variants are produced from a single gene locus, suggesting that the diversity of variants is due to differences in transcriptional start sites (TSS), 14 alternative splicing, 15 and alternative use of a polyadenylation signal. 16 Furthermore, it is difficult to obtain the expression profile of these variants by partial sequencing analysis. Third, the conventional cDNA libraries used for these expression profile analyses did not reflect the expression levels of all mRNAs in the cell because of a preparatory bias toward abundant, shorter cDNAs, thus excluding rare or longer cDNA genes. Analyzing size-unbiased, full-length cDNA libraries constructed from a single type of cell can overcome these pitfalls. 
The mammalian retina contains approximately 55 distinct cell types. 17 To elucidate an intracellular transcriptional network in retinal cells, it is necessary to analyze the transcriptome derived from specific retinal cell types. One solution for this requirement is to use single cells isolated from the retinal tissue. 18 Another solution is to use a cell line derived from the human retina. The established cell line, ARPE-19, 19 exhibits properties similar to those of the endogenous RPE 20 and is used as a model for analyzing the expression profile of the RPE. 21 24 Because some retinoblastoma cell lines, such as Y79, show cone-specific gene expression 25 and are derived from cone progenitor cells, 26 we used this Y79 cell line in our study for the identity of full-length cDNA clones. 
Recently, we developed an effective method for constructing a full-length cDNA library, named a vector-capping method, 27 which has the following characteristics: it consists of only four steps; it contains neither a PCR step nor a restriction enzyme treatment; several micrograms of total RNA is sufficient to construct a library containing 106 independent clones; the full-length cDNAs can be confirmed by the presence of dG added to the 5′ end of the cDNA in a cap-dependent manner; the full-length content is >95%; the library contains cDNA clones for rare genes and long genes of >10 kbp; and the cDNAs for an antisense gene can be assured. We demonstrated that the cDNA library constructed from the ARPE-19 cell line using the vector-capping method contains size-unbiased, full-length cDNAs. 28  
In this study, we constructed full-length cDNA libraries from Y79 cells using the vector-capping method and analyzed the full-length transcriptome by partial or full-length sequencing of the clones. By comparing the expression profiles from the ARPE-19 and Y79 cell lines, we identified novel transcript variants for each cell type, which will be useful for functional analysis of these genes and will provide new insights into the complex processes of retinal function. 
Methods
Cell Culture and RNA Preparation
The human retinal pigment epithelium cell line ARPE-19 and the human retinoblastoma cell line Y79 were obtained from the American Type Culture Collection (Manassas, VA). ARPE-19 cells were cultured in a 1:1 mixture of Dulbecco's modified Eagle's medium and Ham's F-12 medium (Invitrogen, Carlsbad, CA) containing 10% fetal bovine serum (FBS), whereas Y79 cells were cultured in RPMI 1640 medium (Invitrogen) containing 20% FBS. Both cells were cultured at 37°C in a humidified atmosphere of 5% CO2 and 95% air. For total RNA isolation, cells were harvested by trypsinization, and RNA was isolated using reagent (Isogen; Nippon Gene, Tokyo, Japan). 
Construction of cDNA Library
The construction of our cDNA libraries was carried out using the vector-capping method, as described previously. 27 Briefly, 5 to 10 μg total RNA and 0.15 to 0.3 μg vector primer were mixed and annealed. The first-strand cDNA was synthesized using reverse transcriptase (SuperScript III; Invitrogen). After digestion with or without EcoRI, self-ligation was carried out with T4 RNA ligase. After second-strand cDNA synthesis, the resultant product was used for the transformation of Escherichia coli cells, DH12S, by electroporation. Transformants were plated on LB agar containing 50 μg/mL ampicillin without amplification. Resultant colonies were picked and suspended in LB solution containing 50 μg/mL ampicillin in 96-well or 384-well plates. After incubation, glycerol stocks were prepared by adding 50% glycerol and were stored at −80°C. 
Plasmid Isolation and Sequencing
The isolated plasmid DNA or DNA amplified using a DNA amplification kit (illustra TempliPhi; GE Healthcare, Uppsala, Sweden) was used as a template for sequencing. DNA sequencing from the 5′ end of the cDNA insert was carried out with a capillary DNA sequencer (Applied Biosystems Inc., Foster City, CA) using a reaction kit (BigDye Terminator Cycle sequencing FS Ready Reaction Kit; Applied Biosystems). The full sequence of the cDNA insert was determined using a primer walking method. The determined full sequences were deposited in GenBank/EMBL/DDBJ under accession numbers AB593010-AB593186 and AB621803-AB621830
Genome Mapping and Annotation
The 5′-end sequences were used to query our custom database 29 to identify the abundant genes, such as ribosomal genes and mitochondria-derived sequences. Sequences not registered in our custom database were used to query the Human Genome reference sequence Feb. 2009 (GRCh37/hg19) with BLAT search on the UCSC Genome Browser. If the query sequence was mapped to a known gene locus, the sequence was assigned to that gene. Through the Web sites linked in the browser, including Entrez Gene and UniGene, we retrieved the following information: gene name, gene symbol, gene ID, chromosomal location, and RefSeq accession number. 
Estimation of the Total Number of Genes Composing Libraries
The total number of genes constituting the library was estimated according to non–sampling-based extrapolation, which is used for species richness estimation. 30 The calculation was performed by curve fitting to a gene-accumulation curve using a hyperbolic model: Dt = St α/(β + t α), where Dt denotes the cumulative number of genes for the accumulated number, t, of sequenced clones, S is an asymptotic value, and α and β represent the parameters to be estimated from the data. Finally, the curve fitting was carried out using the software, KaleidaGraph (Synergy Software, Reading, PA). 
Results
Full-Length cDNAs from ARPE-19 cDNA Libraries
The results of in-depth analyses of ARPE-19 libraries (ARe, ARf, ARi, and ARiS) were described in a previous study. 28 In this study, the results of the other ARPE-19 libraries (ARa and ARh) were added to the previous data. By removing unreadable clones, cDNA insert-free clones, mitochondria- and rRNA-coded clones, and incomplete clones, 20,317 full-length cDNA clones were identified. Annotation by mapping their 5′-end sequence to the human genome revealed that the clones were classified into 4854 transcriptional units, or genes. A list of the genes is shown in Supplementary Table S1
Full-Length cDNAs from Y79 cDNA Libraries
Three libraries (RBb, RBd, and RBdS) were constructed using isolated Y79 mRNA and the vector-capping method. The clones (6816 from RBb, 9504 from RBd, and 76,800 from RBdS) were picked from each library and stored as glycerol stocks. All the RBb and RBd clones and part of the RBdS (7296 clones), a total of 23,616 clones, were single-pass sequenced from their 5′ ends. The contents of the RBd/RBdS libraries were compared with those of ARi/ARiS,as shown in Table 1. Both libraries were constructed using the same vector primer and the same methodology. Both the cDNA insert content of the isolated clones and the full-length content of the cDNA inserts were >95%. From three different Y79 libraries, 19,229 clones were identified as full-length cDNAs and were classified into 4808 genes. All identified genes were combined with those of the ARPE-19 libraries and listed in Supplementary Table S1. The total number of genes identified from the ARPE-19 and Y79 libraries was 7067, of which 6733 (95.3%) were included in Entrez Gene. 
Table 1.
 
Summary of Single-Pass Sequencing Analysis of Libraries
Table 1.
 
Summary of Single-Pass Sequencing Analysis of Libraries
Library Name ARi/ARiS RBd/RBdS
Cell line ARPE-19 Y79
Total 13824 16800
    Unreadable 928 1229
    Readable 12896 (93.3%) 15571 (92.7%)
        Insert-free vector 465 336
        dT tail 19 6
        Mitochondria 134 118
        rRNA 6 1
        cDNA insert 12272 (95.2%) 15110 (97.0%)
            Full-length 11676 (95.1%) 14577 (96.5%)
            Truncated 514 446
            Poly (A) 82 87
Expression Profile Comparison of the ARPE-19 and Y79 cDNA Libraries
The frequency of abundant genes with ≥0.05% content and the number of low-redundant genes with <0.05% content are shown in Supplementary Figure S1. The numbers of abundant genes from the ARPE-19 library were 311 and 322 for the Y79 library. The number of singletons was 2547 for ARPE-19 and 2434 for Y79. Of the 7067 genes present in the collection, 2595 (36.7%) of the genes were shared between the ARPE-19 and Y79 cell lines. The total number of genes in the ARi/ARiS library and the RBd/RBdS library were estimated from the cumulative curves to be 11,563 and 15,141, respectively (Supplementary Fig. S2). 
Comparison of Abundant Genes
The 455 abundant genes with ≥0.05% content in either the ARPE-19 or the Y79 libraries are listed in Supplementary Table S2. The numbers of clones isolated from the ARPE-19 and Y79 libraries were plotted gene by gene in Figure 1. ARPE-19-specific abundant genes were CRYAB (128 clones, 0.63% in content) highly expressed in lens, an antioxidant MT2A (155 clones, 0.78%), and an intermediate filament VIM (195 clones, 0.98%). On the other hand, Y79 cells expressed an extraordinarily high level of ATP5A1 (328 clones, 1.7%). The overexpression of ATP5A1 in retinoblastoma cells is explained by the gene amplification of the ATP5A1 loci. 31,32 In addition, the high levels of HAUS1 expression (79 clones, 0.41%) can be explained by gene amplification because the TSS of HAUS1 is located approximately 100 bp upstream of the TSS for ATP5A1, and HAUS1 is transcribed in the opposite direction. PCR analyses for the two genes showed that these genes were amplified twice as much as in Y79 cells as they were in ARPE-19 cells (data not shown). The overexpression of DDX1 and its neighboring gene, MYCN, has been reported to result from gene amplification in retinoblastoma cell lines, such as Y79 and RB522A. 33,34 This coamplification was also observed in our data, and we confirmed this result by PCR analysis (data not shown). As a result of the functional breakdown of the most abundant genes, the major functional groups consisted of translation-related proteins, such as ribosomal proteins and elongation factors. The expression levels of 84 ribosomal protein genes showed correlation between ARPE-19 and Y79 (correlation coefficient, 0.78), suggesting that these genes are regulated by a common mechanism in both cell types. 
Figure 1.
 
Comparison of frequencies of abundant genes. The numbers of full-length cDNA clones isolated from ARPE-19 and Y79 were compared gene by gene. The genes with >0.05% content were plotted.
Figure 1.
 
Comparison of frequencies of abundant genes. The numbers of full-length cDNA clones isolated from ARPE-19 and Y79 were compared gene by gene. The genes with >0.05% content were plotted.
Cell Type-Specific Abundant Genes
As shown in Supplementary Table S2, 52 genes were obtained from only the ARPE-19 cell line; of those, the top 20 genes are shown in Table 2. The lists contain the number of clones obtained from the ARPE-19 and Y79 libraries, the number of total ESTs (Nt), and the number of eye-derived ESTs (Ne) retrieved from the UniGene database. The ratio, Ne/Nt, is calculated and listed as an indicator of eye-specific expression. These ARPE-19-specific genes included cytoskeleton genes (CALD1, KRT7, KRT8, KRT18, KRT33B, MYL9, TAGLN, TPM2, and TUBA1C) and extracellular matrix-related genes (COL4A1, CTGF, EFEMP1, EFEMP2, FSTL1, IGFBP3, IGFBP7, SPARC, and TGFBI). The abundant genes from the AREP-19 cDNA library included IL18 (21 clones, Nt = 92, Ne = 2), KRT33B (14 clones, Nt = 5, Ne = 0), and NPPB (40 clones, Nt = 48, Ne = 0), but these were not abundant in the EST database. In particular, NPPB was preferentially expressed in the heart, and its EST was not obtained from eye. On the other hand, 18 genes were obtained from only the Y79 library, as shown in Table 3. NNAT (38 clones, Ne/Nt = 0.12) is a pituitary-enriched gene, 35,36 and its EST is abundant in retinoblastoma. AIPL1 (18 clones, Ne/Nt = 0.85) and GNB3 (25 clones, Ne/Nt = 0.75) are retina-specific genes. SFRP2 (11 clones) has been reported to be involved in Wnt signaling pathway, 37 which is related to antiapoptotic activity. 38 These results reflect some differences between the epithelium-derived ARPE-19 cell line and the neuron-derived Y79 cell line. 
Table 2.
 
Abundant Genes Obtained Only from ARPE-19 Libraries
Table 2.
 
Abundant Genes Obtained Only from ARPE-19 Libraries
No. HP ID Symbol Gene ID Protein ARPE-19 Y79 EST
Nt Ne Ne/Nt
1 HP00436 MT2A 4502 Metallothionein 2A 155 0 702 31 0.04
2 HP04275 CRYAB 1410 Crystallin, alpha B 128 0 1151 272 0.24
3 HP00102 ANXA2 302 Annexin A2 96 0 3874 81 0.02
4 HP00494 S100A6 6277 S100 calcium binding protein A6 85 0 1214 27 0.02
5 HP01162 KRT7 3855 Keratin 7 81 0 1617 9 0.01
6 HP00571 IGFBP7 3490 Insulin-like growth factor binding protein 7 74 0 947 61 0.06
7 HP00021 CYR61 3491 Cysteine-rich, angiogenic inducer, 61 55 0 1118 21 0.02
8 HP00756 KRT18 3875 Keratin 18 44 0 3941 97 0.02
9 HP03186 TAGLN 6876 Transgelin 44 0 1778 79 0.04
10 HP04373 NPPB 4879 Natriuretic peptide precursor B 40 0 48 0 0.00
11 HP01024 MT1E 4493 Metallothionein 1E 34 0 306 6 0.02
12 HP00555 CTGF 1490 Connective tissue growth factor 31 0 1800 56 0.03
13 HP00159 SERPINE1 5054 Serpin peptidase inhibitor, clade E (nexin, plasminogen activator inhibitor type 1), member 1 28 0 1041 16 0.02
14 HP01599 TGFBI 7045 Transforming growth factor, beta-induced, 68 kDa 28 0 4603 174 0.04
15 HP00684 ITGB1 3688 Integrin, beta 1 (fibronectin receptor, beta polypeptide, antigen CD29 includes MDF2, MSK12) 26 0 1509 21 0.01
16 HP00293 MYL9 10398 Myosin, light polypeptide 9, regulatory 22 0 1036 67 0.06
17 HP02548 IL18 3606 Interleukin 18 (interferon-gamma-inducing factor) 21 0 92 2 0.02
18 HP10433 RARRES2 5919 Retinoic acid receptor responder (tazarotene induced) 2 21 0 218 16 0.07
19 HP00514 TM4SF1 4071 Transmembrane 4 L six family member 1 21 0 1176 17 0.01
20 HP00445 KRT8 3856 Keratin 8 20 0 2990 41 0.01
Table 3.
 
Abundant Genes Obtained Only from Y79 Libraries
Table 3.
 
Abundant Genes Obtained Only from Y79 Libraries
No. HP ID Symbol Gene ID Protein ARPE-19 Y79 EST
Nt Ne Ne/Nt
1 HP10162 HAUS1 115106 HAUS augmin-like complex, subunit 1 0 79 261 65 0.25
2 HP05252 NNAT 4826 Neuronatin 0 38 291 36 0.12
3 HP00291 CKB 1152 Creatine kinase, brain 0 32 1978 165 0.08
4 HP05165 MYCN 4613 v-myc myelocytomatosis viral-related oncogene, neuroblastoma derived (avian) 0 28 255 32 0.13
5 HP06683 GNB3 2784 Guanine nucleotide binding protein (G protein), beta polypeptide 3 0 25 144 108 0.75
6 HP02108 MARCKSL1 65108 MARCKS-like 1 0 20 1010 42 0.04
7 HP05221 AIPL1 23746 Aryl hydrocarbon receptor interacting protein-like 1 0 18 157 133 0.85
8 HP00777 ASS1 445 Argininosuccinate synthetase 1 0 16 953 48 0.05
9 HP05523 PRAME 23532 Preferentially expressed antigen in melanoma 0 16 478 24 0.05
10 HP05290 XRCC4 7518 X-ray repair complementing defective repair in Chinese hamster cells 4 0 15 152 13 0.09
11 HP02086 PCBP4 57060 Poly(rC) binding protein 4 0 13 680 96 0.14
12 HP05324 C18orf24 220134 Chromosome 18 open reading frame 24 0 12 116 2 0.02
13 HP05871 PRR6 201161 Proline-rich 6 0 12 139 8 0.06
14 HP05169 ACLY 47 ATP citrate lyase 0 11 867 40 0.05
15 HP05231 HPCA 3208 Hippocalcin 0 11 130 17 0.13
16 HP05237 SFRP2 6423 Secreted frizzled-related protein 2 0 11 565 40 0.07
17 HP05238 PSMG1 8624 Proteasome (prosome, macropain) assembly chaperone 1 0 10 236 15 0.06
18 HP01582 PPP2CA 5515 Protein phosphatase 2 (formerly 2A), catalytic subunit, alpha isoform 0 10 1342 30 0.02
Transcriptional Start Sites
Given that the vector-capping method provides for the genuine full-length cDNA by starting at the cap site, mapping the 5′-terminal sequence of the cDNA to the human genome sequence enabled us to determine the precise TSS. The distributions of TSSs for GAPDH and ACTG1 in ARPE-19, Y79, and DBTSS 39 were compared and are shown in Figure 2. The most frequent TSS was identical and the pattern was similar among the three distributions, suggesting that TSS distribution is not tissue specific for these genes. 
Figure 2.
 
Comparison among distributions of transcriptional start sites. Gray bars: ARPE-19 library; black bars: Y79 library; white bars: DBTSS. Position 1 is defined as a major TSS.
Figure 2.
 
Comparison among distributions of transcriptional start sites. Gray bars: ARPE-19 library; black bars: Y79 library; white bars: DBTSS. Position 1 is defined as a major TSS.
Selection of Eye Preferentially Expressed Genes
Eye preferentially expressed genes were selected from our collection based on the value of Ne/Nt. Our list contained 234 genes with Ne/Nt ≥0.1. The UniGene EST profile regards the 5′-end and 3′-end EST sequences derived from the same EST clone as two different clones. To avoid this redundancy, we recounted the true number, Nt* and Ne*. Table 4 shows the list of 72 genes with Ne*/Nt* ≥ 0.2. As shown below, many eye-characteristic known genes were contained in this list. The number of tissues or organs expressing each gene, Nc, is also listed here. The tissue origin of eye-derived ESTs is also shown, along with the number of ESTs in Supplementary Table S3
Table 4.
 
Eye Preferentially Expressed Genes
Table 4.
 
Eye Preferentially Expressed Genes
No. HP ID Symbol Gene ID Protein ARPE-19 Y79 EST
Nc Nt* Ne* Ne*/Nt*
1 HP05997 ARL4D 379 ADP-ribosylation factor-like 4D 0 4 27 198 44 0.22
2 HP10845 AMTN 401138 Amelotin 1 0 2 2 1 0.50
3 HP04607 ANGPTL7 10218 Angiopoietin-like 7 3 0 17 110 54 0.49
4 HP09211 ANO2 57101 Anoctamin 2 0 1 8 49 25 0.51
5 HP05221 AIPL1 23746 Aryl hydrocarbon receptor interacting protein-like 1 0 18 6 139 123 0.88
6 HP07380 AANAT 15 Arylalkylamine N-acetyltransterase 0 3 5 10 2 0.20
7 HP06949 ATOH7 220202 Atonal homolog 7 (Drosophila) 0 4 3 5 3 0.60
8 HP09128 CDHR1 92211 Cadherin-related family member 1 0 2 17 130 43 0.33
9 HP08811 CABP2 51475 Calcium-binding protein 2 0 1 1 2 2 1.00
10 HP06370 CRABP1 1381 Cellular retinoic acid-binding protein 1 1 0 21 84 27 0.32
11 HP03154 CRABP2 1382 Cellular retinoic acid-binding protein 2 0 8 33 503 108 0.21
12 HP05082 CCL26 10344 Chemokine (C-C motif) ligand 26 2 0 10 15 4 0.27
13 HP08659 CHRND 1144 Cholinergic receptor, nicotinic, delta 0 1 7 16 4 0.25
14 HP09189 CPLX4 339302 Complexin 4 0 1 4 8 5 0.63
15 HP05329 CRX 1406 Cone-rod homeobox (CRX) 0 2 3 65 60 0.92
16 HP09172 DDX51 317781 DEAD (Asp-Glu-Ala-Asp) box polypeptide 51 0 1 27 153 51 0.33
17 HP06349 FAM57B 83723 Family with sequence similarity 57, member B 1 1 7 99 49 0.49
18 HP08732 FAM78B 149297 Family with sequence similarity 78, member B 0 1 8 10 4 0.40
19 HP03371 GSG1 83445 Germ cell-associated 1 0 7 8 148 68 0.46
20 HP01066 GPX3 2878 Glutathone peroxidase 3 (plasma) 3 0 38 1381 324 0.23
21 HP08116 GDF5 8200 Growth differentiation factor 5 1 0 13 30 8 0.27
22 HP06683 GN83 2784 Guanine nucleotide-binding protein (G protein), beta polypeptide 3 0 25 10 124 98 0.79
23 HP07212 GNGT1 2792 Guanine nucleotide-binding protein (G protein), gamma-transducing activity polypeptide 1 0 7 8 64 28 0.44
24 HP09011 GUCA1C 9626 Guanylate cyclase activator 1C 0 1 6 20 11 0.55
25 HP08466 HMX1 3166 H6 family homeobox 1 0 1 5 26 7 0.27
26 HP10162 HAUS1 115106 HAUS augmin-like complex, subunit 1 0 79 33 203 63 0.31
27 HP06472 HHIPL1 84439 HHIP-like 1 2 0 5 8 2 0.25
28 HP05343 LOC284260 284260 Hypothetical gene supported by BC011527; BC021928 0 5 3 12 10 0.83
29 HP05388 LOC389023 389023 Hypothetical gene supported by BC032913; BC048425 0 4 4 13 3 0.23
30 HP09087 LOC645967 645967 Hypothetical LOC645967 0 1 6 8 3 0.38
31 HP09273 LOC645249 645249 Hypothetical protein LOC645249 0 1 5 9 2 0.22
32 HP09025 LOC651900 651900 Hypothetical protein LOC651900 0 1 2 3 1 0.33
33 HP07725 LOC729421 729421 Hypothetical protein LOC729421 1 0 6 13 3 0.23
34 HP08405 IMPG2 50939 Interphotoreceptor matrix proteoglycan 2 0 1 6 24 8 0.33
35 HP08667 ISL2 64843 ISL LIM homeobox 2 0 1 9 19 6 0.32
36 HP07245 KIRREL2 84063 Kin of IRRE-like 2 (Drosophila) 0 1 7 23 7 0.30
37 HP08656 LRRC38 126755 Leucine-rich repeat containing 38 0 1 7 10 5 0.50
38 HP05297 LHX3 8022 LIM homeobox 3 0 9 3 11 8 0.73
39 HP09033 LYPD2 137797 LY6/PLAUR domain containing 2 0 1 3 8 3 0.38
40 HP03430 MAB21L1 4081 Mab-21-like 1 (C. elegans) 1 1 10 65 14 0.22
41 HP09007 MAK 4117 Male germ cell-associated kinase 0 1 12 52 15 0.29
42 HP06805 MATK 4145 Megakaryocyte-associated tyrosine kinase 0 2 17 126 29 0.23
43 HP06819 NRL 4901 Neural retina leucine zipper 0 7 8 60 50 0.83
44 HP05315 NEUROD1 4760 Neurogenic differentiation 1 0 9 8 97 31 0.32
45 HP06849 NEUROG1 4762 Neurogenin 1 0 4 3 5 2 0.40
46 HP05226 OTX2 5015 Orthodenticle homeobox 2 0 3 5 67 45 0.67
47 HP09106 OTOP3 347741 Otopetrin 3 0 1 2 2 1 0.50
48 HP05219 OTX2OS1 100309464 Otx2 opposite strand transcript 1 0 5 1 3 3 1.00
49 HP06376 PAX6 5080 Paired box 6 3 0 18 207 58 0.28
50 HP08866 PDE6A 5145 Phosphodiesterase 6A, cGMP-specific, rod. alpha 0 1 7 83 57 0.69
51 HP06842 PDE6B 5158 Phosphodiesterase 6B, cGMP-specific, rod, beta 0 1 17 64 13 0.20
52 HP08566 PLCH2 9651 Phospholipase C, eta 2 0 2 12 50 11 0.22
53 HP04634 PRH1 5554 Proline-rich protein HaeIII subfamily 1 1 2 30 866 323 0.37
54 HP05707 PPP1R3F 89801 Protein phosphatase 1, regulatory (inhibitor) subunit 3F 0 1 12 88 24 0.27
55 HP09147 RCVRN 5957 Recoverin 0 1 8 135 116 0.86
56 HP08658 RCOR2 283248 REST corepressor 2 0 2 12 61 14 0.23
57 HP07905 RTBDN 83546 Retbindin 0 1 9 62 32 0.52
58 HP03156 RAX 30062 Retina and anterior neural fold homeobox 0 1 3 20 18 0.90
59 HP06193 RAX2 84839 Retina and anterior neural fold homeobox 2 0 5 1 35 35 1.00
60 HP03153 RXRG 6258 Retinoid X receptor, gamma 0 3 16 75 15 0.20
61 HP07871 RBP3 5949 Retinol-binding protein 3, interstitial 0 3 7 67 55 0.82
62 HP06965 ARHGDIG 398 Rho GDP dissociation inhibitor (GDI) gamma 0 1 3 11 3 0.27
63 HP09269 RHBDL3 162494 Rhomboid, veinlet-like 3 (Drosophila) 0 1 8 40 9 0.23
64 HP01198 SERPINF1 5176 Serpin peptidase inhibitor, clade F, member 1 0 8 40 744 149 0.20
65 HP08216 SIX3 6496 SIX homeobox 3 1 0 7 29 18 0.62
66 HP05260 SLC1A7 6512 Solute carrier family 1 (glutamate transporter), member 7 0 3 9 34 22 0.65
67 HP05823 SAMD11 148398 Sterile alpha motif domain containing 11 0 7 22 82 16 0.20
68 HP06208 TLX2 3196 T-cell leukemia homeobox 2 0 1 5 19 6 0.32
69 HP05543 TSPAN10 83882 Tetraspanin 10 2 0 7 40 10 0.25
70 HP07915 TULP1 7287 Tubby-like protein 1 0 1 10 46 35 0.76
71 HP02611 UNC119 9094 Unc-119 homolog (C. elegans) 0 2 35 222 54 0.24
72 HP08996 VAX2 25806 Ventral anterior homeobox 2 0 1 6 10 4 0.40
Three genes—FAM57B (Ne*/Nt* = 0.49), PRH1 (Ne*/Nt* = 0.37), and MAB21L1 (Ne*/Nt* = 0.22)—were found in both the ARPE-19 and the Y79 libraries. FAM57B encodes a membrane protein with unknown function, and its gene was expressed preferentially in the retina, brain, and testis. PRH1 is a salivary proline-rich secreted protein, 40 and its gene was highly expressed in the lacrimal gland. MAB21L1 is a homolog of mab-21, which is a cell fate-determining gene found in Caenorhabditis elegans that may be involved in the development of the eye and the cerebellum. 41  
ARPE-19–Specific Genes
The following 11 genes were only cloned from the ARPE-19 cells: SIX3 (Ne*/Nt* = 0.62, 1 clone), AMTN (Ne*/Nt* = 0.50, 1 clone), ANGPTL7 (Ne*/Nt* = 0.49, 3 clones), CRABP1 (Ne*/Nt* = 0.32, 1 clone), PAX6 (Ne*/Nt* = 0.28, 3 clones), GDF5 (Ne*/Nt* = 0.27, 1 clone), CCL26 (Ne*/Nt* = 0.27, 2 clones), TSPAN10 (Ne*/Nt* = 0.25, 2 clones), HHIPL1 (Ne*/Nt* = 0.25, 2 clones), GPX3 (Ne*/Nt* = 0.23, 3 clones), LOC729421 (Ne*/Nt* = 0.23, 1 clone). ANGPTL7 and LOC729421 were not RPE-specific genes because they were also in retinoblastoma-derived EST sequences. ANGPTL7 is known as a morphogen of the cornea. 42 SIX3 and PAX6 are transcription factors involved in eye development. 43,44 Interestingly, AMTN is an ameloblast-specific protein involved in the formation of dental enamel. 45 CRABP1 is a retinoic acid-binding protein involved in development and differentiation. CCL26 is Cys-Cys cytokine that has chemotactic activity for eosinophils. 46 TSPAN10 is a tetraspanin cloned from RPE/choroid cDNA libraries. 47  
Y79-Specific Genes
Fifty-eight genes were cloned only from Y79 cells, and the major functional group consisted of 15 transcription factors: RAX2 (Ne*/Nt* = 1.0, 5 clones), CRX (Ne*/Nt* = 0.92, 2 clones), RAX (Ne*/Nt* = 0.90, 1 clone), NRL (Ne*/Nt* = 0.83, 7 clones), LHX3 (Ne*/Nt* = 0.73, 9 clones), OTX2 (Ne*/Nt* = 0.67, 3 clones), ATOH7 (Ne*/Nt* = 0.60, 4 clones), VAX2 (Ne*/Nt* = 0.40, 1 clone), NEUROG1 (Ne*/Nt* = 0.40, 4 clones), NEUROD1 (Ne*/Nt* = 0.32, 9 clones), TLX2 (Ne*/Nt* = 0.32, 1 clone), ISL2 (Ne*/Nt* = 0.32, 1 clone), HMX1 (Ne*/Nt* = 0.27, 1 clone), RCOR2 (Ne*/Nt* = 0.23, 2 clones), and RXRG (Ne*/Nt* = 0.20, 3 clones). The homeodomain-type transcription factors (OTX2, CRX, RAX, RAX2), the bZIP-type transcription factor (NRL), and the bHLH-type transcription factors (ATOH7, NEUROG1, NEUROD1) are known to be involved in the transcription of photoreceptor-specific genes. 9 In addition, LHX3, VAX2, TLX2, ISL2, and HMX1 are classified as homeodomain transcription factors. Finally, RXRG is a negative regulator of S-cones. 48  
The second most abundant group consisted of the following 13 signal transduction-related genes: CABP2 (Ne*/Nt* = 1.0, 1 clone), RCVRN (Ne*/Nt* = 0.86, 1 clone), GNB3 (Ne*/Nt* = 0.79, 25 clones), TULP1 (Ne*/Nt* = 0.76, 1 clone), PDE6A (Ne*/Nt* = 0.69, 1 clone), GUCA1C (Ne*/Nt* = 0.55, 1 clone), GNGT1 (Ne*/Nt* = 0.44, 7 clones), MAK (Ne*/Nt* = 0.29, 1 clone), ARHGDIG (Ne*/Nt* = 0.27, 1 clone), PPP1R3F (Ne*/Nt* = 0.27, 1 clone), MATK (Ne*/Nt* = 0.23, 2 clones), PLCH2 (Ne*/Nt* = 0.22, 2 clones), and PDE6B (Ne*/Nt* = 0.20, 1 clone). RCVRN, PDE6A, GUCA1C, GNGT1, and PDE6B play a primary role in the phototransduction process. 2 GNB3 was the most abundant signal transduction-related gene, and it is a cone-specific G protein. 49 TULP1 is associated with retinitis pigmentosa. 50,51  
The present collection contains 12 genes encoding membrane-bound and secreted proteins: RBP3 (Ne*/Nt* = 0.82, 3 clones), SLC1A7 (Ne*/Nt* = 0.65, 3 clones), RTBDN (Ne*/Nt* = 0.52, 1 clone), ANO2 (Ne*/Nt* = 0.51, 1 clone), OTOP3 (Ne*/Nt* = 0.5, 1 clone), GSG1 (Ne*/Nt* = 0.46, 7 clones), LYPD2 (Ne*/Nt* = 0.38, 1 clone), IMPG2 (Ne*/Nt* = 0.33, 1 clone), CDHR1 (Ne*/Nt* = 0.33, 2 clones), KIRREL2 (Ne*/Nt* = 0.3, 1 clone), CHRND (Ne*/Nt* = 0.25, 1 clone), and RHBDL3 (Ne*/Nt* = 0.23, 1 clone). RBP3 plays a role in transporting retinol from the photoreceptor to the RPE. 52 SLC1A7 is a chloride conductance-coupled glutamate transporter. 53 RTBDN has high sequence similarity to a riboflavin binding protein. 54 ANO2 is a Ca2+-dependent chloride channel in photoreceptor synapses. 55 Interestingly, OTOP3 is a paralog of a multi-transmembrane domain protein, OTOP1, which exists at the otoconia of the inner ear. 56 IMPG2 is a proteoglycan located between the photoreceptor and the RPE layers. 57 CDHR1 is a photoreceptor-specific cadherin. 58 CHRND is an acetylcholine receptor at synapses. 59  
In addition, the list contains three genes annotated functionally. AIPL1 (Ne*/Nt* = 0.88, 18 clones) functions as a chaperone for the biosynthesis of rod PDE. 60 CPLX4 (Ne*/Nt* = 0.63, 1 clone) is related to the secretion of neurotransmitter from retinal ribbon synapse. 61 AANAT (Ne*/Nt* = 0.2, 3 clones) is an enzyme involved in melatonin biosynthesis. 62 Furthermore, the list contains noncoding genes and rare predicted genes. OTX2OS1 (Ne*/Nt* = 1.0, 5 clones) is a noncoding gene that is located upstream of the eye-specific transcription factor, OTX2, and is transcribed in the opposite direction. 
Of the Y79-specific genes, 10 did not exist in the retinoblastoma-derived EST database. Most genes are classified into a rare gene (Nt* < 10), but an EST sequence for DDX51 and UNC119 did not exist in the retinoblastoma-derived EST database in spite of Nt* > 100. In addition, it should be noted that 19 genes were included in pineal gland-derived EST sequences. 
Full-Sequence Analysis of Eye Preferentially Expressed cDNA Clones
The full-length sequence was determined for 57 (131 clones) of the 72 eye preferentially expressed genes with Ne*/Nt* ≥ 0.2. The sequenced results were shown in Supplementary Table S4. Comparing the exon-intron structures of each clone with those in RefSeq, eight genes contained different first exons, suggesting that these genes may have alternative TSS. As a result, clones for LHX3, RTBDN, and ANO2 exhibited a different N-terminal sequence. On the other hand, the N-terminal sequence of NRL was shortened, whereas that of TSPAN10 was lengthened because of different TSSs. Fifteen genes provided splicing variants, which were due to the lack of an exon (PDE6A, RAX, OTX2, OTX2OS1), insertion of an exon (RXRG, MAK, PAX6, OTX2OS1), shifting of a splicing site (ISL2, OTOP3, CABP2, SLC1A7, AIPL1, NRL), and non-splicing (LHX3). All three clones for SLC1A7 and all three clones for OTX2OS1 were different splicing isoforms. 
Eighteen genes showed alternative polyadenylation. In most cases, the alternative polyadenylation occurred at different sites of the last exon, thus generating different sizes of 3′UTRs. In addition, we found polyadenylation occurring at the 3′-extended region of the middle exons in AIPL1, TULP1, and ANO2 and at different last exons in OTX2OS1, HMX1, and FAM78B
The exon-intron structures of 15 clones for AIPL1 were compared (Supplementary Fig. S3). Although the TSSs and exon 1 were almost identical, the clones were classified into seven variants because of the different 3′ splicing site of exon 1, the skipping of exon 3, a different 5′ splice site of exon 4, a different 3′ splice site, or alternative polyadenylation at the last exon. Interestingly, this gene contained five single nucleotide polymorphisms in its transcript. Thus, these clones were classified into two haplotypes, and their allelic origin was identified. Ultimately, there appears not to be a haplotype-specific variant. 
Eye Preferentially Expressed Unannotated Genes
There were 16 eye preferentially expressed genes with Ne*/Nt* ≥ 0.2, which are currently unannotated in Entrez Gene. The full-length sequences of 22 clones for these genes were determined and are listed in Supplementary Table S4. Clones for six genes were obtained only from ARPE-19 cells. Every clone was unique and <1 kbp in length. HP11113 (Ne*/Nt* = 1) was 746 bp long and contained seven exons and no open-reading frame (ORF) coding for a protein of >50 amino acids. HP10990 (Ne*/Nt* = 0.67) was 577 bp long and contained three exons, which encoded a protein of 94 amino acids. Three ESTs for HP10990 were derived from fetal eyes, lens, and blood vessel. 
Clones for 10 genes were obtained only from Y79 cells, of which five genes with Ne*/Nt* = 1 had an exon-intron structure. Most ESTs that matched to these sequences were derived only from retinoblastomas; thus, these clones were termed retinoblastoma-specific gene X (RBSGX, X = 1∼5). 
Antisense Genes
Because our cDNA library is designed to guarantee unidirectional insertion into the vector, an antisense gene can be easily identified. Antisense genes derived from ARPE-19 cells have been previously reported. 28 Our Y79-derived antisense genes are listed in Supplementary Table S5. Every sequence is an antisense transcript against the first exon of the known gene. 
Very-Long-Sized Clones
cDNA libraries previously constructed from ARPE-19 cells were shown to contain many long cDNA clones (≥7 kbp).27,28 Our Y79 libraries also contained 33 genes with inserts ≥7 kbp, as shown in Supplementary Table S6. Five clones were fully sequenced, and the results are shown in Supplementary Table S4. The longest cDNA (12,786 bp) encoded Dmx-like 1 (DMXL1), which contains a WD repeat domain at its C-terminal.63 The number of exons and polyadenylation signal sites in this clone were identical to those registered in RefSeq (NM_005509.4, 11,173 bp, 3027 amino acids, 43 exons). Our clone extended 302 bp at the 5′ end. Furthermore, the 3′ splicing site at exon 2 of our clone shifted 1312 bp downstream so that the first ORF was shortened to encode a 98-amino acid protein. The largest ORF started at exon 6 and encoded for a 2854-amino acid protein, but it is unclear which ORF was actually translated. 
The second largest clone of 11,089 bp encoded a 3184-amino acid, golgin B1 (GOLGB1). The GOLGB1 clone obtained from ARPE-19 cells was 11,198 bp and encoded a 3269-amino acid protein. Compared with RefSeq (NM_004487.3, 11,185 bp, 3259 amino acids, 22 exons), three sequences had similar TSSs and the same polyadenylation signal site. Our clone lacked exon 7, resulting in the deletion of 36 amino acid residues. Furthermore, there was a 28-bp extension at the 3′ end of exon 2 and an initiation codon shift in exon 3 to cause a shortening of an N-terminal sequence by 39 amino acids. 
Recently, a long gene responsible for retinitis pigmentosa, EYS, was identified. 64,65 EYS (NM_001142800.1) spans 2 Mbp on the genome, contains 43 exons, and is transcribed into a 10,589-nucleotide mRNA encoding a 3144-amino acid protein. Our Y79 library contained three clones for EYS, one of which was 7989 bp. This clone ended early at exon 11 and encoded a 594-amino acid protein that was identical to that of 1 of 3 RefSeqs (NM_198283.1), but their TSSs and first exon differed. 
Discussion
Preparation of cDNA libraries using the vector-capping method enabled us to analyze a full-length transcriptome in which the clones contained sequences from their cap site to their poly(A) tail. The expression profile analyses previously performed used sequencing-based methods, such as EST and SAGE, or microarray-based methods, both of which are based only on partial transcript sequence analyses. The advantage of analyzing full-length transcripts is that accurate information regarding a gene's TSS, polyadenylation sites, and splicing isoforms can be identified using one technique. For example, three different variants of AIPL1 and a difference in haplotype were identified using this technique. This method lacks a PCR step and an amplification step, which means that gene expression levels and transcript size are no longer caveats in the making of cDNA libraries. As a result, we were able to obtain a precise expression profile of genes within these cells; these ranged in size from <100 bp to >10,000 bp. 
Comparing the expression profiles of bias-free, full-length transcripts from the retinoblastoma-derived cell line Y79 and the RPE-derived cell line ARPE19 allowed for easy identification of cell-specific genes. In the case of abundant genes, those isolated from the ARPE-19 libraries encoded epithelium-specific proteins, such as cytoskeleton and secreted and membrane proteins, reflecting the characteristics of the epithelium. On the other hand, the genes isolated from Y79 libraries contained only two photoreceptor-specific genes, GNB3 and AIPL1, suggesting that the expression levels of photoreceptor-specific genes may be low in Y79 cells. Analyzing the results of our cell line comparisons and the known EST data helped to identify photoreceptor-specific genes expressed at low levels in Y79 cells. Based on the value of Ne/Nt calculated from EST data, we identified many photoreceptor-specific transcription factors from Y79 cDNA libraries involved in the development and differentiation of the photoreceptor and the retinoblastoma-specific unannotated genes. 
Many transcriptional variants of abundant genes have been previously reported. 14 16 Using our method, we were able to identify variants of rare genes. Each identified variant altered the amino acid sequence of the encoded protein. These amino acid changes could affect important protein-protein interactions or could alter biological activity. Further investigations should be conducted to better understand the functional differences between these identified gene variants. This is especially significant with regard to genes responsible for retinal degeneration. In this study, we carried out full sequencing of only cell-specific genes. Our collection contains multiple clones for functionally interesting genes. The full sequencing of these clones will provide information on novel variants, including novel splicing isoforms. These variants will provide greater insight into the protein network within specific cell types. 
In conclusion, our cDNA libraries are powerful tools for the analyses of full-length transcriptomes from retina-derived cells. This method of constructing cDNA libraries is especially useful to obtain and compare transcript variants. Conventional methods, such as EST analyses and RNA-Seq, enable us to expect some variants, but they do not provide the sequence information for full-length transcript variants. Comprehensive analysis of structure and function of expressed genes requires obtaining and expressing full-length cDNA clones. The clones in our collection can be expressed in vitro and in mammalian cells because the vector used to create our libraries carries a T7 promoter and an SV40 promoter, thus providing the translated protein for use in functional analyses. 29 All cDNA clones from our libraries were deposited in the RIKEN BRC DNA bank (http://dna.brc.riken.jp/index.html) and are available to any researcher. We hope these full-length cDNA collections will be widely used for better understanding the protein network in retinal cells. 
Supplementary Materials
Figure sf01, PDF - Figure sf01, PDF 
Figure sf02, PDF - Figure sf02, PDF 
Figure sf03, PDF - Figure sf03, PDF 
Table st1, XLS - Table st1, XLS 
Table st2, XLS - Table st2, XLS 
Table st3, XLS - Table st3, XLS 
Table st4, XLS - Table st4, XLS 
Table st5, XLS - Table st5, XLS 
Table st6, XLS - Table st6, XLS 
Footnotes
 Supported by a grant for the Genome Network Project from the Ministry of Education, Culture, Sports, Science and Technology, Japan.
Footnotes
 Disclosure: M. Oshikawa, None; C. Tsutsui, None; T. Ikegami, None; Y. Fuchida, None; M. Matsubara, None; S. Toyama, None; R. Usami, None; K. Ohtoko, None; S. Kato, None
The authors thank Keiko Sugai for technical assistance and IOVS Volunteer Editor Stephen Gee (National Institutes of Health) for editing this manuscript. 
References
Dowling JE . The Retina: An Approachable Part of the Brain. Cambridge, MA: Belknap Press; 1987.
Luo DG Xue T Yau KW . How vision begins: an odyssey. Proc Natl Acad Sci U S A. 2008;105:9855–9862. [CrossRef] [PubMed]
Marmor MF Wolfensberger TJ . The Retinal Pigment Epithelium. Oxford, UK: Oxford University Press; 1998.
Strauss O . The retinal pigment epithelium in visual function. Physiol Rev. 2005;85:845–881. [CrossRef] [PubMed]
Rattner A Sun H Nathans J . Molecular genetics of human retinal disease. Annu Rev Genet. 1999;33:89–131. [CrossRef] [PubMed]
Phelan JK Bok D . A brief review of retinitis pigmentosa and the identified retinitis pigmentosa genes. Mol Vis. 2000;6:116–124. [PubMed]
Pacione LR Szego MJ Ikeda S Nishina PM McInnes RR . Progress toward understanding the genetic and biochemical mechanisms of inherited photoreceptor degenerations. Annu Rev Neurosci. 2003;26:657–700. [CrossRef] [PubMed]
Coleman HR Chan CC Ferris FL3rd Chew EY . Age-related macular degeneration. Lancet. 2008;372:1835–1845. [CrossRef] [PubMed]
Hennig AK Peng GH Chen S . Regulation of photoreceptor gene expression by Crx-associated transcription factor network. Brain Res. 2008;1192:114–133. [CrossRef] [PubMed]
Swaroop A Zack DJ . Transcriptome analysis of the retina. Genome Biol. 2002;3:REVIEWS1022.
Bortoluzzi S d'Alessi F Danieli GA . A novel resource for the study of genes expressed in the adult human retina. Invest Ophthalmol Vis Sci. 2000;41:3305–3308. [PubMed]
Chowers I Gunatilaka TL Farkas RH . Identification of novel genes preferentially expressed in the retina using a custom human retina cDNA microarray. Invest Ophthalmol Vis Sci. 2003;44:3732–3741. [CrossRef] [PubMed]
Sharon D Blackshaw S Cepko CL Dryja TP . Profile of the genes expressed in the human peripheral retina, macula, and retinal pigment epithelium determined through serial analysis of gene expression (SAGE). Proc Natl Acad Sci U S A. 2002;99:315–320. [CrossRef] [PubMed]
Kimura K Wakamatsu A Suzuki Y . Diversification of transcriptional modulation: large-scale identification and characterization of putative alternative promoters of human genes. Genome Res. 2006;16:55–65. [CrossRef] [PubMed]
Modrek B Resch A Grasso C Lee C . Genome-wide detection of alternative splicing in expressed sequences of human genes. Nucleic Acids Res. 2001;29:2850–2859. [CrossRef] [PubMed]
Beaudoing E Freier S Wyatt JR Claverie JM Gautheret D . Patterns of variant polyadenylation signal usage in human genes. Genome Res. 2000;10:1001–1010. [CrossRef] [PubMed]
Masland RH . The fundamental plan of the retina. Nat Neurosci. 2001;4:877–886. [CrossRef] [PubMed]
Wahlin KJ Lim L Grice EA Campochiaro PA Zack DJ Adler R . A method for analysis of gene expression in isolated mouse photoreceptor and Müller cells. Mol Vis. 2004;10:366–375. [PubMed]
Dunn KC Aotaki-Keen AE Putkey FR Hjelmeland LM . ARPE-19, a human retinal pigment epithelial cell line with differentiated properties. Exp Eye Res. 1996;62:155–169. [CrossRef] [PubMed]
Dunn KC Marmorstein AD Bonilha VL Rodriguez-Boulan E Giordano F Hjelmeland LM . Use of the ARPE-19 cell line as a model of RPE polarity: basolateral secretion of FGF5. Invest Ophthalmol Vis Sci. 1998;39:2744–2749. [PubMed]
Tian J Ishibashi K Handa JT . The expression of native and cultured RPE grown on different matrices. Physiol Genomics. 2004;17:170–182. [CrossRef] [PubMed]
Chowers I Kim Y Farkas RH . Changes in retinal pigment epithelial gene expression induced by rod outer segment uptake. Invest Ophthalmol Vis Sci. 2004;45:2098–2106. [CrossRef] [PubMed]
Cai H Del Priore LV . Gene expression profile of cultured adult compared to immortalized human RPE. Mol Vis. 2006;12:1–14. [PubMed]
Tian J Ishibashi K Honda S Boylan SA Hjelmeland LM Handa JT . The expression of native and cultured human retinal pigment epithelial cells grown in different culture conditions. Br J Ophthalmol. 2005;89:1510–1517. [CrossRef] [PubMed]
Bogenmann E Lochrie MA Simon MI . Cone cell-specific genes expressed in retinoblastoma. Science. 1988;240:76–78. [CrossRef] [PubMed]
Xu XL Fang Y Lee TC . Retinoblastoma has properties of a cone precursor tumor and depends upon cone-specific MDM2 signaling. Cell. 2009;137:1018–1031. [CrossRef] [PubMed]
Kato S Ohtoko K Ohtake H Kimura T . Vector-capping: a simple method for preparing a high-quality full-length cDNA library. DNA Res. 2005;12:53–62. [CrossRef] [PubMed]
Oshikawa M Sugai Y Usami R Ohtoko K Toyama S Kato S . Fine expression profiling of full-length transcripts using a size-unbiased cDNA library prepared with the vector-capping method. DNA Res. 2008;15:123–136. [CrossRef] [PubMed]
Kato S Sekine S Oh SW . Construction of a human full-length cDNA bank. Gene. 1994;150:243–250. [CrossRef] [PubMed]
Chao A . Species richness estimation. In: Balakrishnan N Read CB Vidakovic B eds. Species Estimation and Applications: Encyclopedia of Statistical Sciences. New York: Wiley; 2005:7907–7916.
Godbout R Bisgrove DA Honoré LH Day RS3rd . Amplification of the gene encoding the alpha-subunit of the mitochondrial ATP synthase complex in a human retinoblastoma cell line. Gene. 1993;123:195–201. [CrossRef] [PubMed]
Godbout R Pandita A Beatty B Bie W Squire JA . Comparative genomic hybridization analysis of Y79 and FISH mapping indicate the amplified human mitochondrial ATP synthase alpha-subunit gene (ATP5A) maps to chromosome 18q12→q21. Cytogenet Cell Genet. 1997;77:253–256. [CrossRef] [PubMed]
Godbout R Squire J . Amplification of a DEAD box protein gene in retinoblastoma cell lines. Proc Natl Acad Sci U S A. 1993;90:7578–7582. [CrossRef] [PubMed]
Manohar CF Salwen HR Brodeur GM Cohn SL . Co-amplification and concomitant high levels of expression of a DEAD box gene with MYCN in human neuroblastoma. Genes Chromosomes Cancer. 1995;14:196–203. [CrossRef] [PubMed]
Dou D Joseph R . Structure and organization of the human neuronatin gene. Genomics. 1996;33:292–297. [CrossRef] [PubMed]
Usui H Morii K Tanaka R . cDNA cloning and mRNA expression analysis of the human neuronatin: high level expression in human pituitary gland and pituitary adenomas. J Mol Neurosci. 1997;9:55–60. [CrossRef] [PubMed]
Bovolenta P Esteve P Ruiz JM Cisneros E Lopez-Rios J . Beyond Wnt inhibition: new functions of secreted Frizzled-related proteins in development and disease. J Cell Sci. 2008;121:737–746. [CrossRef] [PubMed]
Melkonyan HS Chang WC Shapiro JP . SARPs: a family of secreted apoptosis-related proteins. Proc Natl Acad Sci U S A. 1997;94:13636–13641. [CrossRef] [PubMed]
Wakaguri H Yamashita R Suzuki Y Sugano S Nakai K . DBTSS: database of transcription start sites, progress report 2008. Nucleic Acids Res. 2008;36:D97–D101. [CrossRef] [PubMed]
Maeda N Kim HS Azen EA Smithies O . Differential RNA splicing and post-translational cleavages in the human salivary proline-rich protein gene system. J Biol Chem. 1985;260:11123–11130. [PubMed]
Margolis RL Stine OC McInnis MG . cDNA cloning of a human homologue of the Caenorhabditis elegans cell fate-determining gene mab-21: expression, chromosomal localization and analysis of a highly polymorphic (CAG)n trinucleotide repeat. Hum Mol Genet. 1996;5:607–616. [CrossRef] [PubMed]
Peek R Kammerer RA Frank S Otte-Höller I Westphal JR . The angiopoietin-like factor cornea-derived transcript 6 is a putative morphogen for human cornea. J Biol Chem. 2002;277:686–693. [CrossRef] [PubMed]
Oliver G Mailhos A Wehr R Copeland NG Jenkins NA Gruss P . Six3, a murine homologue of the sine oculis gene, demarcates the most anterior border of the developing neural plate and is expressed during eye development. Development. 1995;121:4045–4055. [PubMed]
Ton CC Hirvonen H Miwa H . Positional cloning and characterization of a paired box- and homeobox-containing gene from the aniridia region. Cell. 1991;67:1059–1074. [CrossRef] [PubMed]
Iwasaki K Bajenova E Somogyi-Ganss E . Amelotin—novel secreted, ameloblast-specific protein. J Dent Res. 2005;84:1127–1132. [CrossRef] [PubMed]
Shinkai A Yoshisue H Koike M . A novel human CC chemokine, eotaxin-3, which is expressed in IL-4-stimulated vascular endothelial cells, exhibits potent activity toward eosinophils. J Immunol. 1999;163:1602–1610. [PubMed]
Wistow G Bernstein SL Wyatt MK . Expressed sequence tag analysis of human RPE/choroid for the NEIBank Project: over 6000 non-redundant transcripts, novel genes and splice variants. Mol Vis. 2002;8:205–220. [PubMed]
Roberts MR Hendrickson A McGuire CR Reh TA . Retinoid X receptor (gamma) is necessary to establish the S-opsin gradient in cone photoreceptors of the developing mouse retina. Invest Ophthalmol Vis Sci. 2005;46:2897–2904. [CrossRef] [PubMed]
Lee RH Lieberman BS Yamane HK Bok D Fung BK . A third form of the G protein beta subunit, 1: immunochemical identification and localization to cone photoreceptors. J Biol Chem. 1992;267:24776–24781. [PubMed]
Hagstrom SA North MA Nishina PL Berson EL Dryja TP . Recessive mutations in the gene encoding the tubby-like protein TULP1 in patients with retinitis pigmentosa. Nat Genet. 1998;18:174–176. [CrossRef] [PubMed]
Banerjee P Kleyn PW Knowles JA . TULP1 mutation in two extended Dominican kindreds with autosomal recessive retinitis pigmentosa. Nat Genet. 1998;18:177–179. [CrossRef] [PubMed]
Gonzalez-Fernandez F . Evolution of the visual cycle: the role of retinoid-binding proteins. J Endocrinol. 2002;175:75–88. [CrossRef] [PubMed]
Arriza JL Eliasof S Kavanaugh MP Amara SG . Excitatory amino acid transporter 5, a retinal glutamate transporter coupled to a chloride conductance. Proc Natl Acad Sci U S A. 1997;94:4155–4160. [CrossRef] [PubMed]
Wistow G Bernstein SL Wyatt MK . Expressed sequence tag analysis of human retina for the NEIBank Project: retbindin, an abundant, novel retinal cDNA and alternative splicing of other retina-preferred gene transcripts. Mol Vis. 2002;8:196–204. [PubMed]
Stöhr H Heisig JB Benz PM . TMEM16B, a novel protein with calcium-dependent chloride channel activity, associates with a presynaptic protein complex in photoreceptor terminals. J Neurosci. 2009;29:6809–6818. [CrossRef] [PubMed]
Hurle B Ignatova E Massironi SM . Non-syndromic vestibular disorder with otoconial agenesis in tilted/mergulhador mice caused by mutations in otopetrin 1. Hum Mol Genet. 2003;12:777–789. [CrossRef] [PubMed]
Acharya S Foletta VC Lee JW . SPACRCAN, a novel human interphotoreceptor matrix hyaluronan-binding proteoglycan synthesized by photoreceptors and pinealocytes. J Biol Chem. 2000;275:6945–6955. [CrossRef] [PubMed]
Rattner A Smallwood PM Williams J . A photoreceptor-specific cadherin is essential for the structural integrity of the outer segment and for photoreceptor survival. Neuron. 2001;32:775–786. [CrossRef] [PubMed]
Hutchins JB . Acetylcholine as a neurotransmitter in the vertebrate retina. Exp Eye Res. 1987;45:1–38. [CrossRef] [PubMed]
Liu X Bulgakov OV Wen XH . AIPL1, the protein that is defective in Leber congenital amaurosis, is essential for the biosynthesis of retinal rod cGMP phosphodiesterase. Proc Natl Acad Sci U S A. 2004;101:13903–13908. [CrossRef] [PubMed]
Reim K Wegmeyer H Brandstätter JH . Structurally and functionally unique complexins at retinal ribbon synapses. J Cell Biol. 2005;169:669–680. [CrossRef] [PubMed]
Klein DC . Arylalkylamine N-acetyltransferase: the ‘Timezyme.’ J Biol Chem. 2007;282:4233–4237. [CrossRef] [PubMed]
Kraemer C Enklaar T Zabel B Schmidt ER . Mapping and structure of DMXL1, a human homologue of the DmX gene from Drosophila melanogaster coding for a WD repeat protein. Genomics. 2000;64:97–101. [CrossRef] [PubMed]
Abd El-Aziz MM Barragan I O'Driscoll CA . EYS, encoding an ortholog of Drosophila spacemaker, is mutated in autosomal recessive retinitis pigmentosa. Nat Genet. 2008;40:1285–1287. [CrossRef] [PubMed]
Collin RW Littink KW Klevering BJ . Identification of a 2 Mb human ortholog of Drosophila eyes shut/spacemaker that is mutated in patients with retinitis pigmentosa. Am J Hum Genet. 2008;83:594–603. [CrossRef] [PubMed]
Figure 1.
 
Comparison of frequencies of abundant genes. The numbers of full-length cDNA clones isolated from ARPE-19 and Y79 were compared gene by gene. The genes with >0.05% content were plotted.
Figure 1.
 
Comparison of frequencies of abundant genes. The numbers of full-length cDNA clones isolated from ARPE-19 and Y79 were compared gene by gene. The genes with >0.05% content were plotted.
Figure 2.
 
Comparison among distributions of transcriptional start sites. Gray bars: ARPE-19 library; black bars: Y79 library; white bars: DBTSS. Position 1 is defined as a major TSS.
Figure 2.
 
Comparison among distributions of transcriptional start sites. Gray bars: ARPE-19 library; black bars: Y79 library; white bars: DBTSS. Position 1 is defined as a major TSS.
Table 1.
 
Summary of Single-Pass Sequencing Analysis of Libraries
Table 1.
 
Summary of Single-Pass Sequencing Analysis of Libraries
Library Name ARi/ARiS RBd/RBdS
Cell line ARPE-19 Y79
Total 13824 16800
    Unreadable 928 1229
    Readable 12896 (93.3%) 15571 (92.7%)
        Insert-free vector 465 336
        dT tail 19 6
        Mitochondria 134 118
        rRNA 6 1
        cDNA insert 12272 (95.2%) 15110 (97.0%)
            Full-length 11676 (95.1%) 14577 (96.5%)
            Truncated 514 446
            Poly (A) 82 87
Table 2.
 
Abundant Genes Obtained Only from ARPE-19 Libraries
Table 2.
 
Abundant Genes Obtained Only from ARPE-19 Libraries
No. HP ID Symbol Gene ID Protein ARPE-19 Y79 EST
Nt Ne Ne/Nt
1 HP00436 MT2A 4502 Metallothionein 2A 155 0 702 31 0.04
2 HP04275 CRYAB 1410 Crystallin, alpha B 128 0 1151 272 0.24
3 HP00102 ANXA2 302 Annexin A2 96 0 3874 81 0.02
4 HP00494 S100A6 6277 S100 calcium binding protein A6 85 0 1214 27 0.02
5 HP01162 KRT7 3855 Keratin 7 81 0 1617 9 0.01
6 HP00571 IGFBP7 3490 Insulin-like growth factor binding protein 7 74 0 947 61 0.06
7 HP00021 CYR61 3491 Cysteine-rich, angiogenic inducer, 61 55 0 1118 21 0.02
8 HP00756 KRT18 3875 Keratin 18 44 0 3941 97 0.02
9 HP03186 TAGLN 6876 Transgelin 44 0 1778 79 0.04
10 HP04373 NPPB 4879 Natriuretic peptide precursor B 40 0 48 0 0.00
11 HP01024 MT1E 4493 Metallothionein 1E 34 0 306 6 0.02
12 HP00555 CTGF 1490 Connective tissue growth factor 31 0 1800 56 0.03
13 HP00159 SERPINE1 5054 Serpin peptidase inhibitor, clade E (nexin, plasminogen activator inhibitor type 1), member 1 28 0 1041 16 0.02
14 HP01599 TGFBI 7045 Transforming growth factor, beta-induced, 68 kDa 28 0 4603 174 0.04
15 HP00684 ITGB1 3688 Integrin, beta 1 (fibronectin receptor, beta polypeptide, antigen CD29 includes MDF2, MSK12) 26 0 1509 21 0.01
16 HP00293 MYL9 10398 Myosin, light polypeptide 9, regulatory 22 0 1036 67 0.06
17 HP02548 IL18 3606 Interleukin 18 (interferon-gamma-inducing factor) 21 0 92 2 0.02
18 HP10433 RARRES2 5919 Retinoic acid receptor responder (tazarotene induced) 2 21 0 218 16 0.07
19 HP00514 TM4SF1 4071 Transmembrane 4 L six family member 1 21 0 1176 17 0.01
20 HP00445 KRT8 3856 Keratin 8 20 0 2990 41 0.01
Table 3.
 
Abundant Genes Obtained Only from Y79 Libraries
Table 3.
 
Abundant Genes Obtained Only from Y79 Libraries
No. HP ID Symbol Gene ID Protein ARPE-19 Y79 EST
Nt Ne Ne/Nt
1 HP10162 HAUS1 115106 HAUS augmin-like complex, subunit 1 0 79 261 65 0.25
2 HP05252 NNAT 4826 Neuronatin 0 38 291 36 0.12
3 HP00291 CKB 1152 Creatine kinase, brain 0 32 1978 165 0.08
4 HP05165 MYCN 4613 v-myc myelocytomatosis viral-related oncogene, neuroblastoma derived (avian) 0 28 255 32 0.13
5 HP06683 GNB3 2784 Guanine nucleotide binding protein (G protein), beta polypeptide 3 0 25 144 108 0.75
6 HP02108 MARCKSL1 65108 MARCKS-like 1 0 20 1010 42 0.04
7 HP05221 AIPL1 23746 Aryl hydrocarbon receptor interacting protein-like 1 0 18 157 133 0.85
8 HP00777 ASS1 445 Argininosuccinate synthetase 1 0 16 953 48 0.05
9 HP05523 PRAME 23532 Preferentially expressed antigen in melanoma 0 16 478 24 0.05
10 HP05290 XRCC4 7518 X-ray repair complementing defective repair in Chinese hamster cells 4 0 15 152 13 0.09
11 HP02086 PCBP4 57060 Poly(rC) binding protein 4 0 13 680 96 0.14
12 HP05324 C18orf24 220134 Chromosome 18 open reading frame 24 0 12 116 2 0.02
13 HP05871 PRR6 201161 Proline-rich 6 0 12 139 8 0.06
14 HP05169 ACLY 47 ATP citrate lyase 0 11 867 40 0.05
15 HP05231 HPCA 3208 Hippocalcin 0 11 130 17 0.13
16 HP05237 SFRP2 6423 Secreted frizzled-related protein 2 0 11 565 40 0.07
17 HP05238 PSMG1 8624 Proteasome (prosome, macropain) assembly chaperone 1 0 10 236 15 0.06
18 HP01582 PPP2CA 5515 Protein phosphatase 2 (formerly 2A), catalytic subunit, alpha isoform 0 10 1342 30 0.02
Table 4.
 
Eye Preferentially Expressed Genes
Table 4.
 
Eye Preferentially Expressed Genes
No. HP ID Symbol Gene ID Protein ARPE-19 Y79 EST
Nc Nt* Ne* Ne*/Nt*
1 HP05997 ARL4D 379 ADP-ribosylation factor-like 4D 0 4 27 198 44 0.22
2 HP10845 AMTN 401138 Amelotin 1 0 2 2 1 0.50
3 HP04607 ANGPTL7 10218 Angiopoietin-like 7 3 0 17 110 54 0.49
4 HP09211 ANO2 57101 Anoctamin 2 0 1 8 49 25 0.51
5 HP05221 AIPL1 23746 Aryl hydrocarbon receptor interacting protein-like 1 0 18 6 139 123 0.88
6 HP07380 AANAT 15 Arylalkylamine N-acetyltransterase 0 3 5 10 2 0.20
7 HP06949 ATOH7 220202 Atonal homolog 7 (Drosophila) 0 4 3 5 3 0.60
8 HP09128 CDHR1 92211 Cadherin-related family member 1 0 2 17 130 43 0.33
9 HP08811 CABP2 51475 Calcium-binding protein 2 0 1 1 2 2 1.00
10 HP06370 CRABP1 1381 Cellular retinoic acid-binding protein 1 1 0 21 84 27 0.32
11 HP03154 CRABP2 1382 Cellular retinoic acid-binding protein 2 0 8 33 503 108 0.21
12 HP05082 CCL26 10344 Chemokine (C-C motif) ligand 26 2 0 10 15 4 0.27
13 HP08659 CHRND 1144 Cholinergic receptor, nicotinic, delta 0 1 7 16 4 0.25
14 HP09189 CPLX4 339302 Complexin 4 0 1 4 8 5 0.63
15 HP05329 CRX 1406 Cone-rod homeobox (CRX) 0 2 3 65 60 0.92
16 HP09172 DDX51 317781 DEAD (Asp-Glu-Ala-Asp) box polypeptide 51 0 1 27 153 51 0.33
17 HP06349 FAM57B 83723 Family with sequence similarity 57, member B 1 1 7 99 49 0.49
18 HP08732 FAM78B 149297 Family with sequence similarity 78, member B 0 1 8 10 4 0.40
19 HP03371 GSG1 83445 Germ cell-associated 1 0 7 8 148 68 0.46
20 HP01066 GPX3 2878 Glutathone peroxidase 3 (plasma) 3 0 38 1381 324 0.23
21 HP08116 GDF5 8200 Growth differentiation factor 5 1 0 13 30 8 0.27
22 HP06683 GN83 2784 Guanine nucleotide-binding protein (G protein), beta polypeptide 3 0 25 10 124 98 0.79
23 HP07212 GNGT1 2792 Guanine nucleotide-binding protein (G protein), gamma-transducing activity polypeptide 1 0 7 8 64 28 0.44
24 HP09011 GUCA1C 9626 Guanylate cyclase activator 1C 0 1 6 20 11 0.55
25 HP08466 HMX1 3166 H6 family homeobox 1 0 1 5 26 7 0.27
26 HP10162 HAUS1 115106 HAUS augmin-like complex, subunit 1 0 79 33 203 63 0.31
27 HP06472 HHIPL1 84439 HHIP-like 1 2 0 5 8 2 0.25
28 HP05343 LOC284260 284260 Hypothetical gene supported by BC011527; BC021928 0 5 3 12 10 0.83
29 HP05388 LOC389023 389023 Hypothetical gene supported by BC032913; BC048425 0 4 4 13 3 0.23
30 HP09087 LOC645967 645967 Hypothetical LOC645967 0 1 6 8 3 0.38
31 HP09273 LOC645249 645249 Hypothetical protein LOC645249 0 1 5 9 2 0.22
32 HP09025 LOC651900 651900 Hypothetical protein LOC651900 0 1 2 3 1 0.33
33 HP07725 LOC729421 729421 Hypothetical protein LOC729421 1 0 6 13 3 0.23
34 HP08405 IMPG2 50939 Interphotoreceptor matrix proteoglycan 2 0 1 6 24 8 0.33
35 HP08667 ISL2 64843 ISL LIM homeobox 2 0 1 9 19 6 0.32
36 HP07245 KIRREL2 84063 Kin of IRRE-like 2 (Drosophila) 0 1 7 23 7 0.30
37 HP08656 LRRC38 126755 Leucine-rich repeat containing 38 0 1 7 10 5 0.50
38 HP05297 LHX3 8022 LIM homeobox 3 0 9 3 11 8 0.73
39 HP09033 LYPD2 137797 LY6/PLAUR domain containing 2 0 1 3 8 3 0.38
40 HP03430 MAB21L1 4081 Mab-21-like 1 (C. elegans) 1 1 10 65 14 0.22
41 HP09007 MAK 4117 Male germ cell-associated kinase 0 1 12 52 15 0.29
42 HP06805 MATK 4145 Megakaryocyte-associated tyrosine kinase 0 2 17 126 29 0.23
43 HP06819 NRL 4901 Neural retina leucine zipper 0 7 8 60 50 0.83
44 HP05315 NEUROD1 4760 Neurogenic differentiation 1 0 9 8 97 31 0.32
45 HP06849 NEUROG1 4762 Neurogenin 1 0 4 3 5 2 0.40
46 HP05226 OTX2 5015 Orthodenticle homeobox 2 0 3 5 67 45 0.67
47 HP09106 OTOP3 347741 Otopetrin 3 0 1 2 2 1 0.50
48 HP05219 OTX2OS1 100309464 Otx2 opposite strand transcript 1 0 5 1 3 3 1.00
49 HP06376 PAX6 5080 Paired box 6 3 0 18 207 58 0.28
50 HP08866 PDE6A 5145 Phosphodiesterase 6A, cGMP-specific, rod. alpha 0 1 7 83 57 0.69
51 HP06842 PDE6B 5158 Phosphodiesterase 6B, cGMP-specific, rod, beta 0 1 17 64 13 0.20
52 HP08566 PLCH2 9651 Phospholipase C, eta 2 0 2 12 50 11 0.22
53 HP04634 PRH1 5554 Proline-rich protein HaeIII subfamily 1 1 2 30 866 323 0.37
54 HP05707 PPP1R3F 89801 Protein phosphatase 1, regulatory (inhibitor) subunit 3F 0 1 12 88 24 0.27
55 HP09147 RCVRN 5957 Recoverin 0 1 8 135 116 0.86
56 HP08658 RCOR2 283248 REST corepressor 2 0 2 12 61 14 0.23
57 HP07905 RTBDN 83546 Retbindin 0 1 9 62 32 0.52
58 HP03156 RAX 30062 Retina and anterior neural fold homeobox 0 1 3 20 18 0.90
59 HP06193 RAX2 84839 Retina and anterior neural fold homeobox 2 0 5 1 35 35 1.00
60 HP03153 RXRG 6258 Retinoid X receptor, gamma 0 3 16 75 15 0.20
61 HP07871 RBP3 5949 Retinol-binding protein 3, interstitial 0 3 7 67 55 0.82
62 HP06965 ARHGDIG 398 Rho GDP dissociation inhibitor (GDI) gamma 0 1 3 11 3 0.27
63 HP09269 RHBDL3 162494 Rhomboid, veinlet-like 3 (Drosophila) 0 1 8 40 9 0.23
64 HP01198 SERPINF1 5176 Serpin peptidase inhibitor, clade F, member 1 0 8 40 744 149 0.20
65 HP08216 SIX3 6496 SIX homeobox 3 1 0 7 29 18 0.62
66 HP05260 SLC1A7 6512 Solute carrier family 1 (glutamate transporter), member 7 0 3 9 34 22 0.65
67 HP05823 SAMD11 148398 Sterile alpha motif domain containing 11 0 7 22 82 16 0.20
68 HP06208 TLX2 3196 T-cell leukemia homeobox 2 0 1 5 19 6 0.32
69 HP05543 TSPAN10 83882 Tetraspanin 10 2 0 7 40 10 0.25
70 HP07915 TULP1 7287 Tubby-like protein 1 0 1 10 46 35 0.76
71 HP02611 UNC119 9094 Unc-119 homolog (C. elegans) 0 2 35 222 54 0.24
72 HP08996 VAX2 25806 Ventral anterior homeobox 2 0 1 6 10 4 0.40
Figure sf01, PDF
Figure sf02, PDF
Figure sf03, PDF
Table st1, XLS
Table st2, XLS
Table st3, XLS
Table st4, XLS
Table st5, XLS
Table st6, XLS
×
×

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.

×