We performed transcriptome sequencing of canine retinal tissue by 454 GS-FLX and Ion Torrent PGM platforms. RNA-Seq analysis by CLC Genomics Workbench mapped expression of 10,360 genes. Gene ontology analysis of retinal transcriptome revealed abundance of transcripts known to be involved in vision associated processes. The
The retina is composed of a neural cell layer and a retinal pigment epithelial cell layer. 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 [
Massively parallel, high-throughput sequencing platforms have provided possibility for genome-wide observations of the transcriptional makeup of retina genes. The transcriptome is the complete set of transcripts in a cell at a specific stage or under given physiological condition [
The retinal tissues of both eyes were taken from a female nondescript dog, approximately 4-5 years old, that had an automobile accident and succumbed during the treatment at the Department of Veterinary Surgery & Radiology, AAU, Anand, Gujarat, India. Tissues were washed with sterile phosphate buffer saline solution, transferred immediately in the “RNA later,” and stored in liquid nitrogen for downstream processing.
Total RNA was extracted from 100 mg of both tissues using TRIzol (Invitrogen Life Technologies, CA) reagent as per the manufacturer’s instructions. DNase treatment was given to remove the DNA contamination. The quantity and quality of RNA were evaluated using NanoDrop1000 spectrophotometer (Thermo Fisher Scientific) as well as Bioanalyzer 2100 (Agilent Technologies, CA). Total mRNA was isolated from total RNA sample using mRNA isolation kit (Roche Diagnostics, Switzerland) as described in the manufacturer’s protocols. Total isolated mRNA was again quality checked on Bioanalyzer 2100 using RNA 6000 nano Chip kit (Agilent Technologies, CA). cDNA and library preparations were carried out using kits of 454 GS-FLX sequencing and Ion Torrent mRNA library preparation. Both platforms based cDNA libraries were sequenced on 454 GS-FLX and Ion Torrent PGM sequencers as per the manufacturer’s instructions. The brief steps for sequencing are mRNA fragmentation, adapter ligation, cDNA preparation, emulsion PCR based library amplification, and library enrichment.
The generated reads of both datasets were pooled and subjected to quality screening using PRINSEQ. Reads with less than 60 bases of read length, with mean read quality of less than 20, and with duplicate reads were removed [
The genes that were expressed with RPKM (reads per kilobase of exon model per million mapped reads) value of ≥0.5 were taken for functional annotation. Genes which were expressed with RPKM < 0.5 were excluded from the functional annotation. The genes (gene ontology) were annotated using Database for Annotation, Visualization and Integrated Discovery (DAVID) Version 6.7 [
The downstream analysis was carried out from total data obtained from 454 GS-FLX and Ion Torrent. The duplicate reads, chimeric reads, minimum length (<30 bp), quality mean read length (<20), and end trimming were performed using PRINSEQ tool [
The annotations of contigs were carried out by BLASTx searches of all contigs with reference to the NCBI nonredundant (nr) database using Blast2GO [
All assembled transcripts (using CAP3) were submitted to in-house local BLASTx against protein sequence database of
The assembled contigs were uploaded to online tool ORFPredictor [
The SSRs were identified from assembled sequences using SSR Locator [
RNA sequencing carried out on Ion Torrent yielded a total of 569,066 quality reads with mean read length of 145.79 bp and on 454 GS-FLX yielded a total of 231,088 quality reads with mean read length of 373.19 bp. Out of 226,684 counted fragments, 222,296 reads mapped uniquely whereas 4,388 reads mapped nonspecifically with reference genome (Table
Number and proportion of reads mapped to reference genome. Mapping statistics were carried out using CLC Genomics with
Canine retina (expressed genes) | |
---|---|
Total reads | 800,154 |
Counted fragments | 226,684 |
Uniquely mapped | 222,296 |
Mapped unspecifically | 4,388 |
Uncounted fragments | 573,470 |
Number of genes | 10,360 |
Number of genes in reference | 28,455 |
Among the reads mapping uniquely to protein coding genes, 34.05% located within exon reads and 14.63% on exon-exon reads. Nearly 48.00% located within the introns and 2.75% in the exon-intron reads (Table
Distribution of mapped reads to different transcript types and gene regions with reference
Uniquely mapped | Nonspecifically mapped | Mapped reads | ||
---|---|---|---|---|
Number of reads | Number of reads | Number of reads | % | |
Total exon reads | 75,883 | 1,305 | 77,188 | 34.05 |
Exon-exon reads | 32,385 | 768 | 33,153 | 14.63 |
Exon-intron reads | 6,210 | 35 | 6,245 | 2.75 |
Total intron reads | 107,818 | 2,280 | 110,098 | 48.57 |
Total gene reads | 222,296 | 4,388 | 226,684 | 100.00 |
In our transcriptome study, 10,360 genes were expressed with RPKM values ≥ 0.50. In order to categorize the genes with different level of expression, genes were categorized based on RPKM values into three groups, that is, high (≥200 RPKM), medium (≥10–200 RPKM), and low (≥0.4–10 RPKM) expressed genes. There were 36 (0.78%) highly expressed genes, 1,850 (40.14%) moderately expressed genes, and 2,723 (59.08%) low expressed genes in the retina.
The functional annotations of genes expressed in retinal tissue were performed using DAVID 6.7 web based annotation tool [
The enrichment of expressed genes in biological process was observed for 316 genes in 28 GO terms which ranged between 25 and 6 genes. The assignment of GO terms (
Functional annotation of expressed genes of retinal tissue in GO term: biological process.
The enrichment of expressed genes in cellular component was observed for 293 genes which ranged between 102 and 5 genes in 34 GO terms (
Functional annotation of expressed genes of retinal tissue in GO term: cellular component.
The enrichment of expressed genes in cellular component was observed for 293 genes which ranged between 75 and 6 genes in 27 GO terms, namely, nucleotide binding (75 genes), purine nucleotide binding (73 genes), purine ribonucleotide binding, ribonucleotide binding (70 genes), purine nucleoside binding (51 genes), nucleoside binding (51 genes), adenyl nucleotide binding (47 genes), adenyl ribonucleotide binding (44 genes), ATP binding (44 genes), zinc ion binding (33 genes), guanyl nucleotide binding (27 genes), guanyl ribonucleotide binding (27 genes), GTP binding (24 genes), protein kinase activity (23 genes), protein tyrosine kinase activity (11 genes), protein serine/threonine kinase activity (11 genes), and protein domain specific binding (9 genes). In addition, symporter activity (13 genes), solute : cation symporter activity (12 genes), phosphatase activity (9 genes), and nucleoside-triphosphatase regulator activity (9 genes) were enriched with
Functional annotation of expressed genes of retinal tissue in GO term: molecular function.
In an attempt to determine the retina associated expression of candidate genes, expression of 160 genes out of 221 genes (based on RetNet (
GO terms biological process enrichment of expressed genes in retinal tissue.
GO ID | GO terms | Count |
|
Genes |
---|---|---|---|---|
GO:0050953 | Sensory perception of light stimulus | 17 | 0.006 | GNAT1, RP1, RPGR, RPE65, RCVRN, PRPH2, PDE6G, CNGA1, GUCY2D, SAG, PDE6A, PDE6B, PDE6D, PRCD, PDC, CLN5, and RHO |
|
||||
GO:0007601 | Visual perception | 17 | 0.006 | GNAT1, RP1, RPGR, RPE65, RCVRN, PRPH2, PDE6G, CNGA1, GUCY2D, SAG, PDE6A, PDE6B, PDE6D, PRCD, PDC, CLN5, and RHO |
|
||||
GO:0009314 | Response to radiation | 11 | 0.011 | GNAT1, GNGT1, SLC1A2, SLC1A3, UACA, CASP9, BAX, RCVRN, BCL2L1, SNAI2, and RHO |
|
||||
GO:0009416 | Response to light stimulus | 9 | 0.017 | GNAT1, GNGT1, SLC1A2, SLC1A3, UACA, CASP9, BAX, RCVRN, and RHO |
The reads were assembled using CAP3 which generated a total of 29,683 contigs with N50 of 619 and mean length of 560.9 bases (Table
Summary statistics of
Features | Values |
---|---|
Clean reads for assembly | 800,154 |
N50 length | 619 |
N75 length | 485 |
N90 length | 384 |
Contig number | 29,683 |
Contig bases | 16,649,067 |
Maximum contig length | 6,416 |
Minimum contig length | 42 |
Mean length of cleaned reads | 560.9 |
Mode length of cleaned reads | 510 |
Median length of cleaned reads | 501 |
Number of reads per contig | 27.95 |
Distribution of contigs length in canine transcriptome assembly.
Full-length cDNAs are important resources for many genetic and genomic researchers and useful to predict protein sequences [
Open reading frame identification in RNA-Seq is important in gene prediction and identification of candidate protein coding regions [
Out of 29,683 contigs, a total of 2,470 SSRs/microsatellites were identified in 2,298 contigs, including dinucleotide (1,582) (68.84%), trinucleotide (592) (25.76%), tetranucleotide (216) (9.40%), pentanucleotide (58) (2.52%), and hexanucleotide (22) (0.96%) (Figure
In order to identify the active biological pathways in canine tissue, the assembled contigs were used to obtain the enzyme commission (EC) against the Kyoto Encyclopaedia of Genes and Genomes (KEGG) database. A total of 3,782 contigs were assigned to 3,570 enzyme commission (EC) numbers (Table
KEGG biochemical mappings for
KEGG pathway | EC count |
---|---|
Metabolism |
|
Overview | 135 |
Carbohydrate metabolism | 238 |
Energy metabolism | 96 |
Lipid metabolism | 176 |
Nucleotide metabolism | 79 |
Amino acid metabolism | 164 |
Metabolism of other amino acids | 53 |
Glycan biosynthesis and metabolism | 115 |
Metabolism of cofactors and vitamins | 67 |
Metabolism of terpenoids and polyketides | 19 |
Biosynthesis of other secondary metabolites | 17 |
Xenobiotics biodegradation and metabolism | 54 |
Genetic Information Processing |
|
Transcription | 21 |
Translation | 54 |
Folding, sorting, and degradation | 97 |
Replication and repair | 32 |
Environmental Information Processing |
|
Signal transduction | 330 |
Signaling molecules and interaction | 10 |
Cellular Processes |
|
Transport and catabolism | 115 |
Cell motility | 18 |
Cell growth and death | 70 |
Cell communication | 73 |
Organismal Systems |
|
Immune system | 151 |
Endocrine system | 196 |
Circulatory system | 26 |
Digestive system | 67 |
Excretory system | 37 |
Nervous system | 163 |
Sensory system | 15 |
Development | 30 |
Environmental adaptation | 20 |
Human Diseases |
|
Cancers | 320 |
Immune diseases | 23 |
Neurodegenerative diseases | 114 |
Substance dependence | 42 |
Cardiovascular diseases | 12 |
Endocrine and metabolic diseases | 24 |
Infectious diseases | 297 |
Total |
|
The transcriptome of canine tissue was primarily examined to identify a wide range of candidate genes that might be functionally associated with melanogenesis, phototransduction, and retinol metabolism (Additional File 2, Table S1 a). The present study indicated that 33 contigs were associated with melanogenesis pathways including adenylate cyclase with the EC number EC:4.6.1.1 encoded by five contigs, while protein kinase A (EC:2.7.11.11), CREB-binding protein (EC:2.3.1.48), RAF protooncogene serine/threonine-protein kinase (EC:2.7.11.1), and mitogen-activated protein kinase kinase 1 (EC:2.7.12.2) were encoded by two contigs. However, mitogen-activated protein kinase 1/3 (EC:2.7.12.2), dopachrome tautomerase (EC:5.3.3.12), phosphatidylinositol phospholipase C, beta (EC:3.1.4.11), calcium/calmodulin-dependent protein kinase (CaMkinase) II (EC:2.7.11.17), and classical protein kinase C (EC:2.7.11.13) were encoded by one contig (Additional File 2). Contigs associated with phototransduction pathways were 15, with rhodopsin kinase (EC:2.7.11.14, 1 contig) and rod cGMP-specific 3′,5′-cyclic phosphodiesterase (EC:3.1.4.35, 2 contigs) (Additional File 2, Table S1 b). Enzymes involved in retinol metabolism were also encoded by the canine retinal tissue contigs (11 contigs). In the retinol metabolism, main enzymes detected were, namely, alcohol dehydrogenase (EC:1.1.1.1, 1 contig), retinol dehydrogenase (EC:1.1.1, 4 contigs), retinoid isomer hydrolase (EC:3.1.1.64, 3 contigs), and retinal dehydrogenase (EC:1.2.1.36, 1 contig) (Additional File 2, Table S1 c).
We performed RNA-Seq of canine retina using 454 GS-FLX and Ion Torrent PGM. Mapping identified the expression of 10,360 genes out of 28,455 reference genes of canine genome. The expressed genes were utilized for the gene ontology analysis and functional annotation.
In the retinal transcriptome, 316 genes were found to be enriched into the 28 molecular function GO category (
Cellular component refers to the place in the cell where a gene product is active. These terms reflect our understanding of eukaryotic cell structure. In the retinal transcriptome, 293 genes were found to be enriched into the 34 cellular component GO category (
The BLASTx search of contigs revealed 42% of the hits matching
BLASTx hits distribution with
Length distribution of putative predicted full-length cDNA of assembled contigs of
Open reading frame (ORF) length distribution of assembled contigs of
Distribution of simple sequence repeats (SSRs) among different nucleotide types found in the transcriptome of
Full-length cDNAs are the valuable source of information for many genetic and genomic researchers and can be useful to predict the protein sequences [
The transcriptome sequencing provides an excellent source for mining and development of gene-associated markers [
Vision is one of the most fascinating mechanisms of the interactions of a biological system and the process of phototransduction where the electromagnetic radiation is converted into biologically recognizable signals by the retinal photoreceptor cell. The phototransduction cascade of vertebrate serves as a benchmark system in signal transduction for a number of light stimuli including the remarkable ability of rod cells to respond reliably to single photon [
In the present study, we have analyzed transcriptome data and identified >10000 expressed genes in canine retinal tissue. The enrichment of genes in GO terms, namely, sensory perception of light stimulus, visual perception, response to radiation, and response to light stimulus, suggested the abundance of genes specifically present in retinal tissue and involved in vision related processes. Several highly expressed genes encoding proteins involved in melanogenesis, phototransduction, and retinol metabolism were identified in the retina. Moreover, a large number of cDNA SSRs were predicted which can be used for subsequent marker development, genetic linkage, and QTL analysis. Overall, the canine retina transcriptome represents a valuable resource for future functional and comparative genomic studies for effective way to treat vision related problem of this globally vulnerable species.
Retinal pigment epithelium
Inherited retinal degeneration
Reads per kilobase per million mapped reads
Expressed Sequence Tag
Gene ontology
Contig assembly program-3
Open reading frame
Kyoto Encyclopedia of Genes and Genomes
Nonredundant
Simple sequence repeat.
The authors have no conflict of interests in this study.
The present study was carried out by financial support provided by the Department of Biotechnology, Government of India, New Delhi. The authors wish to thank Dr. J. V. Solanki, Scientist Emeritus, College of Veterinary Science and Animal Husbandry, Anand, for critical reading of the paper.