Little ROCK is a ROCK1 pseudogene expressed in human smooth muscle cells
© Montefusco et al. 2010
Received: 15 December 2009
Accepted: 14 April 2010
Published: 14 April 2010
Skip to main content
© Montefusco et al. 2010
Received: 15 December 2009
Accepted: 14 April 2010
Published: 14 April 2010
Sequencing of the human genome has identified numerous chromosome copy number additions and subtractions that include stable partial gene duplications and pseudogenes that when not properly annotated can interfere with genetic analysis. As an example of this problem, an evolutionary chromosome event in the primate ancestral chromosome 18 produced a partial duplication and inversion of rho-associated protein kinase 1 (ROCK1 -18q11.1, 33 exons) in the subtelomeric region of the p arm of chromosome 18 detectable only in humans. ROCK1 and the partial gene copy, which the gene databases also currently call ROCK1, include non-unique single nucleotide polymorphisms (SNPs).
Here, we characterize this partial gene copy of the human ROCK1, termed Little ROCK, located at 18p11.32. Little ROCK includes five exons, four of which share 99% identity with the terminal four exons of ROCK1 and one of which is unique to Little ROCK. In human while ROCK1 is expressed in many organs, Little ROCK expression is restricted to vascular smooth muscle cell (VSMC) lines and organs rich in smooth muscle. The single nucleotide polymorphism database (dbSNP) lists multiple variants contained in the region shared by ROCK1 and Little ROCK. Using gene and cDNA sequence analysis we clarified the origins of two non-synonymous SNPs annotated in the genome to actually be fixed differences between the ROCK1 and the Little ROCK gene sequences. Two additional coding SNPs were valid polymorphisms selectively within Little ROCK. Little ROCK-Green Fluorescent fusion proteins were highly unstable and degraded by the ubiquitin-proteasome system in vitro.
In this report we have characterized Little ROCK (ROCK1P1), a human expressed pseudogene derived from partial duplication of ROCK1. The large number of pseudogenes in the human genome creates significant genetic diversity. Our findings emphasize the importance of taking into consideration pseudogenes in all candidate gene and genome-wide association studies, as well as the need for complete annotation of human pseudogenome.
The ROCK1 and 2 serine/threonine kinases regulate many cellular responses such as cell growth, proliferation, and apoptosis through their effects on the cytoskeleton and microtubule network organization [1, 2]. The ROCK1 and ROCK2 proteins share a similar structure characterized by an amino terminal coiled-coil domain containing the kinase activity, a Rho binding site, and a carboxy-terminal pleckstrin homology (PH) domain . Activation by GTP-bound Rho-A involves displacement of the PH domain and exposure of the kinase domain to substrate [4–8]. In vascular smooth muscle cells (VSMC) ROCK1 and 2 activity promotes cellular contraction by direct phosphorylation of the myosin binding subunit (MBS) leading to inhibition of myosin light chain phosphatase activity [9, 10]. Activated Rho kinases can also trigger phosphorylation of MBS through the Zip-like kinase [11, 12] or by phosphorylating the CPI-17 protein, which physically binds and inhibits the actions of PP1M, the catalytic subunit of MLCP [13, 14]. VSMC contraction triggered by activation of the ROCK1 and ROCK2 pathway causes blood vessels to constrict, which increases blood pressure . Inhibitors of ROCK1 and 2 block VSMC contraction and lower blood pressure (BP) in humans , block acetylcholine-induced arterial vasoconstriction , and improve exercise-induced myocardial ischemia .
Given the importance of ROCK1 and ROCK2 to BP and by extension cardiovascular diseases we sought to understand whether genetic differences in these genes contribute to the normal variation of blood pressure that exists in the general population. The ROCK1 and ROCK2 proteins are products of separate genes located on chromosomes 18 and 2, respectively. A ROCK2 gene polymorphism located adjacent to the coiled-coiled domain (ROCK2-T432N) has been associated with BP . At the start of our study computational analysis of ROCK1 gene revealed that the single nucleotide polymorphism database (dbSNP) lists several ROCK1 coding region variants, assigned to two different loci on chromosome 18. Reported studies designed to determine the genomic differences that distinguish the human chromosome 18 from its homolog in great apes (chimpanzee, orangutan, and gorilla) identified a chromosome 18 pericentric break causing an inversion and transposition event that included part of ROCK1 as well as USP14 and THOC1 [20, 21]. The result of this chromosomal event, which occurred at some point before humans evolutionarily separated from great apes, was the placement of USP14, THOC1 and a partial duplication of ROCK1 in the sub-telomeric region of the p arm of chromosome 18 [20, 21]. Full-length ROCK1 remained in the peri-centromeric region of 18q. This partial duplication corresponds to the region of ROCK1 (the last for exons and introns) that included numerous non-uniquely annotated coding SNPs.
Partial gene duplications commonly produce pseudogenes, and we considered whether the partial duplication of ROCK1 at 18p11.32 represented a ROCK1 pseudogene . Approximately half of all mammalian protein families include pseudogenes http://pseudofam.pseudogene.org, with the greatest representation found in housekeeping and ribosomal families of genes . While pseudogenes are commonly considered to be genetic "fossils" that have no biological function, there are examples of functional pseudogenes. Expressed pseudogene transcripts can contribute to the synthesis of small interfering RNA species that regulate parent transcripts [24, 25], and disease-related pseudogenes have also been reported . A pseudogene can be found for approximately twenty percent of kinase genes [27, 28]; however a ROCK1 or -2 pseudogene has not been described. The microtubule-affinity regulating kinase family has the largest number of pseudogenes followed by p70S6 kinase . Kinase pseudogenes that produce mRNA transcripts have been identified, but as yet there is no documented function for these expressed pseudogenes.
Here we report the characterization of the gene produced by partial duplication of human ROCK1, which we named Little ROCK (ROCK1P1). We demonstrate expression of Little ROCK transcript in human vascular smooth muscle cells. Despite the high level of nucleotide identity, we define sequences specific to Little ROCK and resolve the location of non-synonymous coding polymorphisms that were previously reported to be located non-uniquely within both ROCK1 and Little ROCK.
C526G (rs2847092) and T865C (rs1045144) are located in the exon 3 and exon 4 of Little ROCK respectively corresponding to exon 31 and 32 of ROCK1. The DNA analysis of 90 unique human Caucasian DNA samples using a TaqMan assay for T865C and by direct sequencing of G526C showed heterozygote genotypes for all samples analyzed (Figure 3A). Sequence analysis of cDNA synthesized from human VSMC RNA revealed that G526 and T865 belong to Little ROCK (Figure 3B). From these results we conclude that both these nucleotide "variants" are actually fixed sequence differences between the ROCK1 and the Little ROCK genes.
A662G (rs1045142) and C667T (rs2663698) nucleotide variants are both within the exon 3 of Little ROCK corresponding to exon 31 of ROCK1 gene. The direct sequence analysis of 90 unique human Caucasian DNA samples showed polymorphic results where both homozygous GT and heterozygous GT and AC individuals were represented within the population. Sequence analysis of the VSMC ROCK1 cDNA sequence showed exclusively GT at the two positions (Figure 3B), which indicates that ROCK1 is not the source of these polymorphisms. By comparison, the Little ROCK cDNA sequence showed GT, AC or both. Sequence analysis further demonstrated that both polymorphisms are in complete linkage disequilibrium, forming two haplotypes: A662-C667 (Haplotype AC) and G662-T667 (Haplotype GT) (Figure 3C). To confirm the results and to determine the haplotype frequencies we genotyped the Heart SCORE cohort using a custom designed TaqMan assay. The frequency of the Haplotype AC was similar in Heart SCORE African American and Caucasian participants (minor allele frequency 0.33 versus 0.31, respectively). The two haplotypes were in Hardy-Weinberg equilibrium (p > 0.05) in both racial groups.
The custom Little ROCK/ROCK1 TaqMan assay can discriminate the T865C alleles in both genomic DNA and cDNA. Quantitative RT-PCR analysis of four unique human VSMC lines demonstrated that ROCK1 transcript had a lower threshold cycle for detection compared with Little ROCK (Ct mean ± SD 19.26 ± 0.68 versus 22.39 ± 1.29, p < 0.05), which is consistent with higher relative transcript abundance. By comparison, the ROCK1 threshold cycles were not lower than Little ROCK in analysis of genomic DNA from the same cell lines. Therefore, we conclude that while we confirmed expression of Little ROCK by 5'RACE and RT-PCR, its expression in smooth muscle cells is reduced compared with ROCK1.
Completion of the human genome sequence has identified numerous chromosome copy number additions and subtractions that include partial gene duplication. Here we characterize Little ROCK, created by partial duplication and translocation of a portion of chromosome 18 immediately following the separation of humans from great apes [20, 21, 30]. Pseudogenes are one product of these duplication events, and several kinase pseudogenes have been described. We conclude that Little ROCK is a ROCK1 pseudogene for several reasons. First, despite a high degree of sequence identity with ROCK1, Little ROCK includes a disproportionate number of non-synonymous changes in the coding sequence. An excess of non-synonymous coding changes is characteristic of pseudogenes, perhaps reflecting a lower level of purifying selection as the parent gene . Second, pseudogenes lack regulatory CpG islands in their promoters , and unlike ROCK1 and ROCK2 the first Little ROCK exon is not preceded by a predicted CpG island. We detected the Little ROCK transcript in cultured human VSMC, which is unusual because many pseudogenes are not expressed . The expression pattern of Little ROCK transcript was different than ROCK1, a finding that likely reflected that the ROCK1 upstream gene regulatory promoter region was not included in the partial duplication event that created Little ROCK, and because Little ROCK is located near a telomere [20, 21]. Finally, we found the Little ROCK protein to be highly unstable and capable of rendering a stable protein, EGFP, subject to degradation by the ubiquitin-proteasome system. Therefore, despite evidence of transcript expression, the highly unstable Little ROCK peptide is unlikely to accumulate to sufficient quantities to have a direct functional impact. Therefore, based upon the unique presence of a partial duplication of ROCK1, and the fact that a corresponding ROCK2 duplication was not found, our findings are consistent with Little ROCK being the sole pseudogene of the ROCK family of kinases. We have reported our findings to the HUGO Gene Nomenclature Committee and Little ROCK has been assigned the symbol: ROCK1P1.
Following completion of the human genome sequence and analysis of human gene variants it has been estimated that 12% of the human genome is affected by chromosomal gains and losses . Indeed, an emerging problem is the incomplete annotation of stable chromosomal duplications and the pseudogenes often contained within the duplicated regions. Determining the location and sequence differences associated with chromosomal duplication is challenging because of the high level of sequence identity between pseudogenes and their parent genes. To illustrate this problem we have clarified the chromosomal location of four variants listed in SNP databases that at the time of preparing this manuscript were assigned both to ROCK1 and to Little ROCK. Two of these reported polymorphisms (rs2847092 and rs1045144) were in fact fixed nucleotide differences that define Little ROCK compared with ROCK1. By comparison we demonstrate two polymorphisms in complete linkage disequilibrium located exclusively in Little ROCK (rs1045142 and rs2663698). The polymorphic differences detected in Heart SCORE participants were found in similar allele frequencies in Caucasians and African Americans suggesting that the variants were created sometime after the time of the chromosome 18 event that created Little ROCK and before divergence of Homo sapiens. These examples illustrate the need to identify sequence differences of chromosomal duplications in the ongoing 1,000 Genomes Project.
Pseudogenes are commonly felt to be "junk" DNA, yet there are examples of pseudogenes that have functional regulatory effects [33, 34]. As an expressed pseudogene, Little ROCK may also have a direct biological effect, perhaps affecting VSMC and blood vessel function. The instability of the Little ROCK fusion protein would suggest that any functional role may be unlikely to be explained on a peptide level. Future studies will explore the cross-talk between the Little ROCK and ROCK1 transcripts.
In this report we have characterized Little ROCK, an expressed pseudogene derived from partial duplication of ROCK1. The large number of pseudogenes in the human genome creates significant genetic diversity that can have physiological importance. The finding of genetic variants distinct to Little ROCK emphasizes the importance of taking into consideration pseudogenes in all candidate gene and genome-wide association studies, as well as the need for complete annotation of human pseudogenome.
Human ROCK1 and Little ROCK genomic, mRNA, and protein sequences were obtained from UCSC genome browser http://genome.ucsc.edu/cgi-bin/hgGateway and Ensembl genome browser http://www.ensembl.org/. In Ensembl genome browser, the identification numbers for Little ROCK and ROCK1 genes are ENSG00000215585 and ENSG00000067900, respectively.
PCR, sequencing, site-directed mutagenesis primer and TaqMan probe Sequences
PCR or Sequencing
The Tufts Medical Center Institutional Review Board approved these studies. 90 human DNA samples collected from Caucasian patients with cardiovascular disease were used for gene sequencing studies. Heart Strategies Concentrating on Risk Evaluation (Heart SCORE) is a single-site prospective community-based cohort study investigating the mechanisms underlying population disparities in cardiovascular disease [35, 36]. Our sample included 1,191 individuals (425 African Americans and 766 Caucasians) who provided consent and a DNA sample. Observed genotype frequencies were compared with those expected under Hardy-Weinberg equilibrium (HWE) using a χ2 test.
Custom TaqMan genotyping assays created for rs1045144 and the haplotype rs1045142/rs2663698 (See Table 1 for primer and probe sequences) were purchased from Applied Biosystems (Assays on Demand). The assays were performed following manufacturing instructions on a 7900HT real-time PCR system. The reaction volume was 5 μl and included 10 ng of DNA, 2.5 μl of Universal PCR master mix (2×) and 0.1 μl of 40× probes. The reaction conditions were: one step at 95° for 10 minutes followed by 40 cycles of 15 seconds at 92°C and 1 minute at 60°C. Real time PCR results and genotype calls were using the SDS 2.3 program (Applied Biosystems).
Immortalized human VSMC were provided by Dr. Mendelsohn. Details about explants, isolation, and immortalization of VSMC are reported in Pace MC et. al. . VSMC cultures were maintained at 37°C in 5% CO2 humid atmosphere in a growth medium containing high glucose DMEM, Fetal Bovine Serum (10%), and Penicillin/Streptomycin (1×). Total RNA was extracted from VSMC using Trizol solution (Invitrogen, Carlsbad, CA) following the manufacturer's instructions. 5' RACE experiments were carried out on human VSMC RNA with a custom oligonucleotide primer that recognizes both ROCK1 and Little ROCK transcripts (5RACELT, Table 1) according to the GeneRacer kit instructions (Invitrogen). Due to the sequence similarity between ROCK1 and Little ROCK, one single gene-specific primer was designed in order to select both Little ROCK and ROCK1 mRNAs. RT experiments were carried out using SuperScriptIII enzyme (Invitrogen) on 5 μg of total RNA extracted from human VSMC. 1/10 of the RT reaction volume was used for the following PCR. A set of eight human organ cDNAs were purchased from Clontech (Clontech, Mountain View, CA). We carried out PCRs for Little ROCK, ROCK1 and GADPH cDNAs on all cDNA samples. GAPDH was use as a cDNA loading control. The Little ROCK primers were: cDNALTF and cDNACMR, resulting in a 614 bp PCR product; ROCK1 primers were: ROCKF and ROCKR resulting in a 230 bp PCR product; and GAPDH primers were: GAPDHF and GAPDHR, producing a 146 bp PCR product. Primer sequences are listed in Table 1. The PCR reaction was carried out in a final volume of 30 μl including 3 μl of cDNA, 1× polymerase reaction buffer (1.5 mM MgCl2), 0.1 mM dNTPs, 0.1 μM each primer, and 2 units of AmpliTaq DNA Polymerase (Applied Biosystems). Standard PCR conditions have been used: an initial step at 95° for 10 minutes followed by 35 cycles of three steps at 95° for 30 sec, 60° for 20 sec and 72° for 50 sec. A final step at 72° for 1 min was added to complete the elongation reactions.
Little ROCK coding region was cloned into pEGFP-C2 expression vector in order to obtain EGFP-Little ROCK fusion protein constructs with Little ROCK in frame with the C-terminus of EGFP. We amplified using cDNALTF and cDNACMR primers and cloned a fragment of Little ROCK cDNA, including the entire coding sequence and part of both the 5' and the 3' UTRs into the pCR4-TOPO vector (Invitrogen). A clone of Little ROCK was obtained and we amplified a fragment by using the primers LTGFPF (carrying the EcoRI consensus sequence at 5' end) and M13F that is part of pCR4-TOPO vector sequence and downstream the EcoRI restriction site. The PCR product was digested with EcoRI and ligated into the pEGFP-C2 expression vector at EcoRI cloning sites. Sequencing reactions confirmed that the nucleotide sequence of Little ROCK was in frame with EGFP. We isolated the Little ROCK clone showing the A662/C667 nucleotides and we generated the Little ROCK clone carrying G662/T667 nucleotides by using the Stratagene (Stratagene, Cedar Creek, TX) site-directed mutagenesis quick change mutagenesis kit and primers MU2LTF and MU2LTR (Table 1). The two clones were respectively named pEGFP-LRAC and pEGFP-LRGT.
HeLa cells were cultivated in growth medium (10% FBS, 1× Penicillin/streptomycin, DMEM) until 60-80% confluent. Transfection reactions were carried out with PolyFect reagent (Qiagen, Valencia, CA) using a DNA/PolyFect ratio of 1 (μg)/10 (μl) following the manufacturer's instructions. The cells were washed twice with 1× PBS 48 hours after transfection and fresh medium containing 10 μM of MG132 or DMSO (same solvent used to dissolve MG132) as a control was added. The cells were incubated at 37° for 4 hours. Protein lysates were prepared in lysis buffer (50 mM Tris pH 7.5, 150 mM NaCl, 5% Glycerol, 1% Triton, 10 mM MgCl2, 1 mM EGTA, 1 mM DTT, 25 mM NaF, 20 mM b-Glycerophosphate, 1 mM Na3VO4, 2 mM PMSF, 1× Protein Inhibitors cocktail). After protein quantification by BCA protein assay (Pierce, Rockford, IL), 50 μg of protein lysates were loaded onto a 12% SDS page. Protein was then transferred onto a nitrocellulose membrane and then treated with blocking solution (1× TBST, 5% skim milk). The mouse anti-GFP antibody (Covance Inc, Princeton NJ, USA), was diluted 1:2000 in 1× TBST and 2% skim milk. The secondary Goat anti-mouse HRP conjugated antibody (Santa Cruz Biotechnology, Santa Cruz, CA) was diluted 1:10000 in 1× TBST and 2% skim milk. ECL plus western blotting detection system (GE Healthcare, Bukinghamshire, UK) was used for protein detection. The blotted membranes were then analyzed on a Typhoon scanner and the protein band intensity was measured by ImageQuant TL software (GE Healthcare). Differences in protein abundance were compared by t-test.
We would like to thank Eric Wooten, Sarah Greytak, Alyson K. Hedgepeth, David Housman and Michelle Arya for their helpful comments and suggestions. This project is funded, in part, under a grant with the Pennsylvania Department of Health (S.E.R.). (Contract ME-02-384). The Department specifically disclaims responsibility for any analyses, interpretations, or conclusions. This project was supported by the following National Institutes of Health grants HL077378 (M.E.M.), HL069770 (M.E.M).
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.