Research Article | Open Access
An Exon-Based Comparative Variant Analysis Pipeline to Study the Scale and Role of Frameshift and Nonsense Mutation in the Human-Chimpanzee Divergence
Chimpanzees and humans are closely related but differ in many deadly human diseases and other characteristics in physiology, anatomy, and pathology. In spite of decades of extensive research, crucial questions about the molecular mechanisms behind the differences are yet to be understood. Here I report ExonVar, a novel computational pipeline for Exon-based human-chimpanzee comparative Variant analysis. The objective is to comparatively analyze mutations specifically those that caused the frameshift and nonsense mutations and to assess their scale and potential impacts on human-chimpanzee divergence. Genomewide analysis of human and chimpanzee exons with ExonVar identified a number of species-specific, exon-disrupting mutations in chimpanzees but much fewer in humans. Many were found on genes involved in important biological processes such as T cell lineage development, the pathogenesis of inflammatory diseases, and antigen induced cell death. A “less-is-more” model was previously established to illustrate the role of the gene inactivation and disruptions during human evolution. Here this analysis suggested a different model where the chimpanzee-specific exon-disrupting mutations may act as additional evolutionary force that drove the human-chimpanzee divergence. Finally, the analysis revealed a number of sequencing errors in the chimpanzee and human genome sequences and further illustrated that they could be corrected without resequencing.
Chimpanzees (Pan troglodytes) and humans (Homosapiens) are each other's closest living relatives and yet the two primates differ enormously in many characteristics in physiology, anatomy, and pathology . Among the differences, those in the frequency and severity to deadly human diseases such as human immunodeficiency disease (HIV/AIDS), Alzheimer's disease, and Plasmodium falciparum malaria are especially intriguing. In the case of HIV/AIDS, the infections of HIV-1 rarely cause AIDS-like diseases in chimpanzees. Protective strategies appear to have evolved to pacify SIV/HIV . Similarly, the great apes can be infected with viral hepatitis B and C but do not progress to chronic active hepatitis . While the epithelial neoplasm such as carcinomas of the breast, ovary, lung, stomach, colon, pancreas, and prostate cause more than 20% of the deaths in modern human populations, the occurrence rates among the great apes are only 2%–4% or lower [4, 5]. Chimpanzees also appear to be immune to P. falciparum, the most aggressive and acutely life-threatening malaria for humans [6–8]. Finally, the great apes show no sign of the complete pathological lesions of Alzheimer's disease .
The benefit of the comparative analysis of the closely related species is obvious. Through such analyses, it is possible to detect key differences that propelled the chimpanzee-human differentiation. The analysis thus could led to an improved understanding of the molecular mechanisms behind these differences and eventually to new therapies . For example, the HIV-infected chimpanzee differs from HIV-infected human individuals in the following several aspects: (1) a low level of T cell activation and bystander apoptosis in SIV/HIV-infected chimpanzees with no evidence of immunodeficiency ; (2) a rapid establishment of an anti-inflammatory environment, which may prevent the host from developing the aberrant chronic T cell hyperactivation, a hallmark of progression to AIDS during HIV-1 infection ; and (3) An absence of overt CD4+ T cell loss. The nonpathogenic characteristics hence highlight a protective role of down-regulated T cell activation and the establishment of anti-inflammatory profiles early on in immune responses. It is possible that such protection allows the natural host to accept the virus that hence can survive in a symbiotic state. Knowledge about the molecular mechanisms of such protection would allow the development of drugs as well as treatment plans that can imitate biological processes that occur in the HIV-infected chimpanzees. The consequence is to bring the disease under control . The strategy could apply equally well to other differentially evolved human diseases, for example, viral hepatitis B and hepatitis C, should it prove successful. Chimpanzees display a similar response to viral hepatitis B and C as they do to the HIV/AIDS  where infected chimpanzees do not progress to the chronic active hepatitis .
The availability of completely sequenced human genomes [15, 16] and that of chimpanzees  would have an immense impact on the comparative human-chimpanzee analyses. For the first time, nucleotide-by-nucleotide comparisons can be performed at whole genome levels. Already such analyses have resulted in some novel observations and in-depth understandings. First, the two closely related species differ significantly in chromosome structures. Cheng et al. (2005) discovered difference in the contents of segmental duplications . About 33% of the human duplications, including some human disease-causing duplication, are not duplicated in the chimpanzees. Harris et al. (2007) reconstructed ancestral states and the structural evolution of the genomes and identified 130 human-specific breakpoints due to rearrangements at an intermediate scale (10 kilobases to 4 megabases) . Recently, Kehrer-Sawatzki and Cooper (2008) revealed a strong spatial association between primate-specific breakpoints and segmental duplications (SDs) .
Chimpanzees and humans were also found to differ in adaptive gene evolution. Clark et al. (2003) observed nonneutral evolution from human-chimpanzee-mouse orthologous gene trios where significantly different patterns of substitution were detected in the human lineage where accelerated evolution was founded in several functional classes, including olfaction and nuclear transport . Marques-Bonet et al. found an association between chromosomal rearrangements and genic evolution in human and chimpanzee [20–22]. Genes located in the rearranged chromosomes that differentiate the genomes of humans and chimpanzees presented lower divergence than genes elsewhere in the genome. These observations were further supported by the chimpanzee sequencing project and the subsequent human-chimpanzee comparative analysis . The lineage-specific adaptive evolution was detected where affected genes differed in the magnitude, regional variation, and the strength of positive and negative selection. In spite of the extensive research, many questions remain, especially impacts of genetic mutations from recent human and chimpanzee evolution. Here I reported a novel computational platform named ExonVar for human-chimpanzee comparative analysis with a new perspective: comparatively analyzing mutations and their impacts on exon structures, specifically those that cause frameshift and nonsense mutations. The objective is to assess the scale and potential impacts of these mutations on human-chimpanzee divergence.
The work was started because of the lack of appropriate computational procedures for such analysis although a few of comparative human-chimpanzee genome analysis methods have been developed. Hahn and Lee are the first group to develop a procedure that can be applied for a large scale analysis of first frameshift  and then nonsense mutations . Their approach, relying on mRNA-genome sequence alignments, has its limitation mainly because of incomplete genome coverage of existing mRNA data especially from chimpanzees. Puente et al. (2006) developed the similar procedure but with a specific focus on cancer genes . Recently, Wetterbom et al. (2009) devised a procedure for a genomewide analysis of chimpanzee genes with premature termination codons . In this approach, exons downloaded from Ensembl and then translated. Exons were subsequently concatenated and scanned for premature termination codons. A main concern for this approach is that the accuracies in the exon boundary prediction will likely have a negative impact on the analysis. Despite numerous developments of useful tools, predicting exons and their precise exon–intron boundaries are still a challenging task [27–29]. An incorrect prediction would introduce incorrect open reading frames, thus false positive results.
It was motivated by our early success in the development of GenVar and by the observations that gene functional modification or inactivation plays a crucial role in human genome evolution. GenVar is a computational pipeline to comparatively analyze closely related bacterial genomes for variant-pathogenesis association studies . The pipeline is unique in that it can analyze sequence variations such as those that cause frameshift, nonsense and indel mutations at the genome scale and within the context of closely related bacterial species. From the analyses, mutations derived from recent evolution can thus be identified, promising a better understanding of the molecular basis underlying differentially evolved phenotypes .
It was also motivated by the fact that genetic changes can result in the acquisition of novel phenotypic traits, manifested by small jaws and teeth, weakened jaw muscles, decreased smell sensitivity, and reduced body hair. This is what so called a “less-is-more” model [31, 32]. For example, the human-specific 2-bp deletion in the coding region of MYH16, a sarcomeric myosin gene, resulted in a frameshift and was linked to the reduction of jaw muscles that allowed humans to have bigger brains . The single base-pair substitution introduced a premature TGA termination codon in the human type I hair keratin gene . The resulting gene inactivation has been suggested to evolve smooth, hairless skin in humans for enhanced thermoregulation  and/or for reduction of parasite loads . Recently, multiple human-specific nonsense mutations have been identified where abolished or modified functions were predicted in the affected genes .
It is thus hypothesized that functional modification or inactivation likely assumes greater roles in human-chimpanzee differentiation. GenVar is a powerful tool to test the hypothesis since it can assess the genome variations at a global scale. GenVar, however, is a CPU-demanding pipeline that presents a significant challenge in the human-chimpanzee genome study. First, there was an impractically high computational cost and running time due to the complexities of the human, chimpanzee and other eukaryotic genomes. The genomes are massive; human genome, for example, consists of over three billion base pairs (bps) with over 30,000 genes and 61,318 transcripts whereas average bacterial genomes are about 3 million bps with around 3,000 genes/proteins. The genes from these genomes are further more complicated than their bacterial counterparts. They can reach up to millions of base pairs but only tiny portions are actual coding sequences that are further imbedded in vast arrays of non-coding DNA. Many genes are alternatively spliced, forming multiple protein products [37, 38]; moreover, duplication is a ubiquitous phenomenon in humans and many other eukaryotic genomes and can reach a level with that bacterial genomes can barely match [39, 40].
The complexities consequently result in genomic DNA inputs with gigantic sizes and database inputs with heterogeneous protein components, thus slowing the GeneWise-based analysis process and complicating result interpretations. This can be manifested by our earlier experiments. A small-sized human chromosome (e.g., chromosome 21) needs over three months of computational time in a Linux node with a dual Intel processor. By contrast, bacteria, with the same number of genes, need less than 24 hours . Second, the result from the analysis was unreliable and inaccurate when GeneWise was applied to align the interrupted, mega-sized gene sequences (intron-exon structures) with their homologous proteins. The unreliability was obvious: approximately 24.48% of disrupted genes were revealed in human chromosomes 21 when compared to homologous proteins from humans, chimpanzees, and macaques but a majority of the disruption mutations turned out to be false positive when the alignments were manually examined.
ExonVar followed the same strategy as GenVar with critical modifications to overcome the challenge. First, all the analyses were performed on coding exons instead of genes. The coding exons are much smaller, even smaller than bacterial genes. With the divide and conquer strategy, an immediate result is an improved performance. With the same human chromosomes 21, the analysis can be finished within 24 hours at the same Linux node. The modifications, furthermore, come with an increased accuracy and simplified interpretation. All exon-disrupting mutations detected were found in their corresponding trace sequences. Finally, the human-chimpanzee analysis could be extended to include the genomes of macaques and other mammals to define lineage-specific sequence variants. With the computational pipeline, human and chimpanzee exons were analyzed. This paper presents a sample of total results to illustrate the analysis procedure and to assess the scale and biological impacts of the species-specific, exon-disrupting mutations.
2. Materials and Methods
2.1. Genome Data
Genome sequences of eight mammalian species were downloaded from Ensembl database in (2007) (ftp://ftp.ensembl.org/pub/). The downloaded data are release-46 (Table 1). Each genome is comprised of proteins, assembled chromosomes, and multiple MySQL files in the database. All genomes have completely assembled chromosomes with some variations at the time when the data were downloaded. For example, sequence data for humans consist of 22 autosomal chromosomes, 1 mitochondrion, 2 sex chromosomes, and three other unassembled DNA fragments. The MySQL files include several species-specific database tables including gene.txt.table, gene_stable_id.txt.table, exon.txt.table, exon_transcript.txt.table, transcript.txt.table, translation_stable_id.txt.table, and translation.txt.table. The purpose is to extract genomic information at the levels of genes, exons, and proteins, and to define the exon-gene and exon-protein relationships.
2.2. The ExonVar Implementation
ExonVar consists of three steps: first detect sequence mutations specifically those that cause frameshift and nonsense in the coding sequences are identified (i). Once identified, the occurrence patterns of the mutations were defined among the genomes of humans, chimpanzees, macaque and five other mammals (ii). Those that are specific to humans or chimpanzees were subsequently validated (In Silico) to tag those that are caused by sequencing errors (iii). Following sections give a detailed description about the procedures.
(i) Develop the exon-based, variant discovering procedure. The development followed the analysis procedure described in GenVar but with a critical modification . Instead of comparatively analyzing entire genes, which are hard to handle because of massive size and complicated structures, this procedure focuses solely on exons, the simplest and smallest coding units in eukaryotic genomes. Briefly, two required inputs are first established; one is Exon-based genomic DNA input and another is the Exon-based Peptide DataBase input (ExPepDB). The genomic DNA input was defined as an extended area of the predicted exons where 100 base pairs were added at both and ends. The ExPepDB, established for each genomic DNA input, consists of homologous peptide exons from all included genomes. The peptide exons are the amino acid sequences of the exons, translated by an in-house developed, Perl-based program. The program basically performed BLAST analyses against species-specific protein databases and subsequent protein-exon sequence mapping. Note that data downloaded from Ensembl have explicit specifications about protein-exon relationships. The program takes the relationship as a constraint. From that, amino acid sequences of any possible overlapped coding exons can be unequivocally determined. Once the two required inputs are established, they are comparatively analyzed using GeneWise [42–44] to identify sequence variations.
(ii) Identify lineage-specific exon-disrupting mutations. To identify human- or chimpanzee-specific mutations, a tree-based evolutionary scheme was devised based on a previously published mammalian species tree . In this scheme, macaques, cows, dogs, opossums, mice and rats, which are closely related, but phylogenetically outside of the chimpanzee and human clade, are used as out-groups (Figure 1). Mutations that follow the species tree were considered to be lineage-specific (panels I to V). Among them, those that are specific to chimpanzees or humans (in panel I and II) were defined as chimpanzee- or human-specific respectively.
(iii) Insilico validate the lineage-specific exon-disrupting mutations. The first step is to establish species-specific trace databases for humans and chimpanzees. The trace data were downloaded from NCBI Trace archive (ftp://ftp.ncbi.nih.gov/pub/TraceDB/). The human-specific trace database covered 188, 150, 586, 226 base pairs (bp) of trace sequences from the diploid genome sequences of Dr. Craig J. Venter, Dr. James D. Watson, and other human sequencing projects, and has a total coverage of 58X based on the human genome size of 3, 253, 037, 807 bps. The chimpanzee-trace database is much smaller, which includes 43, 176, 085, 998 bp trace sequences from Pan troglodytes, a coverage of merely 15X. The database was extended to include 321, 835, 223 bp trace sequences from pongo_pygmaeus (sequencing project) and 15, 420, 096, 318 bp trace sequences from pongo_pygmaeus_abelii (sequencing project). The extension led to a 5.4X increase in the genome coverage (see results and discussion for reasoning to include Pongo sequences).
The second step is to select query sequences. The objective is to validate the frame-disrupting mutations in target exons. The query sequences are the orthologous exons from the closest related genome, for example, exons in humans to those in chimpanzees or vice verse. The queries are then used to search the species-specific trace DNA database. Trace sequences with at least a 95% sequence identity are extracted. They were named candidate trace sequences to represent those that were sequenced directly from the chromosomal regions where the disrupted exons are detected. The candidate trace sequences, along with query-specific ExPepDB, are comparatively analyzed with GeneWise. The GeneWise alignments are evaluated with a set of predetermined rules. The assumption underlying the rules is that mutations derived from evolution (accepted mutations) will lead to the consistent occurrence of the variations among the candidate trace sequences whereas sequencing errors to randomness if ever occurred. Specifically, the rules say that for a given query,(i)there will be at least n candidate trace sequences that have a coverage on the regions where the exon-disrupting mutations occur where and(ii)among the n candidate trace sequences at least m display exactly identical mutations in types and positions where ;(iii)none of the candidate trace sequences show a complete alignment with their orthologous sequences where no disruptions were displayed.
If all statements described above are evaluated as true, the status of the mutations will be considered as true and recognized as accepted mutations. Otherwise, the status of the mutations will be recognized as either false (statement I is true, but both II and III are false) or undetermined (statement I is false or both statement I and II are true but III is false).
2.3. Program Implementation
All implementations were based on the PERL script language. Wise2 was downloaded from EMBL-EBI (http://www.sanger.ac.uk/Software/Wise2/). ExonVar is available to noncommercial users upon request.
3. Results and Discussion
3.1. Comparative Analysis of the Exon-Disrupting Mutations
Following the procedure of ExonVar, all 504,862 predicted human and chimpanzee exons were analyzed. The analysis revealed significant differences between chimpanzees and humans. The first is the number of the disrupted exons: there are 1,931 disrupted exons in humans while that number is 3,742, approximately doubled in chimpanzees. The second is the occurrence patterns of the exon-disrupting mutations, which describes how the mutations are distributed among the genomes. In the human genome, only a small portion of the mutations are involved in cross-species variation between humans and chimpanzees. In chromosome X, for example, the number is 7.3% (eight out of 109 of the disrupted exons). Examples include HUMAN_1_851437 (solute carrier family 26, member 9 isoform A), HUMAN_1_837766 (calcium activated chloride channel 3 precursor) (Table 2). The exon-disrupting mutations are human-specific where they were detected in humans only. A majority of the disrupting mutations are, however, not species-specific. For instance, the exon-disrupting mutations in HUMAN_X_817906 (cancer/testis antigen 2) were detected by another human exon; that in HUMAN_X_806072 (intestinal protein OCI-5) by an exon from the dog genome, and that in HUMAN_X_815259 (green-sensitive opsin) by an exon from the macaque genome.
|Note: Numbers under the columns of frameshifts and premature stop codons are the chromosomal positions of the sequence mutations. “” presents no annotation available from Ensemble database.|
On the contrary, a majority of the exon-disrupting mutations in chimpanzees were involved in human homologues, for example, 59.7% (114 out of 193 disrupted exons) in chromosome X. Among the exons, CHIMP_X_104156 (ADP-ribosylation factor-like protein 13A), CHIMP_X_99308 (Cylicin-1), CHIMP_X_146199 (AF4/FMR2 family member 2), CHIMP_X_144941 and CHIMP_X_144913 (melanoma-associated antigen C3) have disrupted reading frames when compared to their human and macaque orthologs, thus specific to chimpanzees. A similar phenomenon was found in other chromosomes (data not shown). Note that the exon names were designed for clarity and consistency where exon identification numbers from Ensembl database, for example, 142174 in CHIMP_X_142174, were used as bases with the prefixes of species abbreviations (CHIMP) and chromosome numbers (X).
3.2. Validation of the Species-Specific Exon-Disrupting Mutations
The exon-disrupting mutations can come from either sequence artifacts or accepted mutations . With chimpanzee-specific trace databases, a total of 189 chimpanzee-specific frame-disrupted exons were examined with the predetermined rules. Case studies are represented here to illustrate the rules and their applications.
Case 1 1. One chimpanzee-specific frameshift mutation was detected in CHIMP_X_139912 (cancer/testis antigen CT45). A total of 62 candidate trace sequences were extracted from the chimpanzee-specific trace database. An identical frameshift was revealed in all 62 sequences (Figure 2). Rules applied here are the following: (1) multiple candidate trace sequences were identified; (2) the exact same mutations in at least two seed sequences were observed; (3) no candidate trace sequences showed complete (undisrupted) open reading frame. The conclusion is that the frameshift mutation can be accepted as an adapted mutation derived from the recent human-chimpanzee evolution.
Case 2 2. Four chimpanzee-specific frameshift mutations were detected in CHIMP_X_142174 (HIV Tat-specific factor (1) eight trace sequences were identified from the chimpanzee-specific trace database and only one was revealed to have the frameshift mutations. Some of the sequences had perfect alignments with its human and macaque orthologs (Figure 3). Rules applied here are the following: (1) Identified multiple candidate trace sequences; (2) observed the frameshift mutation in one candidate trace sequence only, which I believed to be the seed sequence that was used for the original genome assembly; and (3) detected undisrupted open reading frames from multiple candidate trace sequences including one from Pongo pygmaeus abelii. The conclusion is that the frameshift mutations were rejected and the disrupted exon was predicted to be due to sequencing artifacts.
Exon-disrupting mutations were also validated in the genes or homologues of cancer/testis antigen CT45-3, melanoma-associated antigen C3, the inhibitor of growth family, member 1, protocadherin-11 X-linked precursor, potassium channel tetramerisation domain containing 9-like, testis-expressed sequence 13A protein, tumor necrosis factor receptor superfamily member 18 precursor, heat-shock protein beta-7, eyes absent homolog 3, (EC 184.108.40.206), neuroblastoma breakpoint family, member 11, tropomyosin alpha-3 chain (Tropomyosin-3), voltage-dependent R-type calcium channel subunit alpha-1E, ADP-ribosylation factor-like protein 13A, SOX-13 protein (Type 1 diabetes autoantigen ICA12) (Islet cell antigen 12) (Table 3) and 36 other chimpanzee genes (data not shown).
Including trace sequences from two Pongo sequencing projects increased the genome coverage and enhanced the capability in the validation of the exon-disrupting mutations in the chimpanzee genome. For example, two frameshift mutations were found in CHIMP_1_235104, an exon of SOX-13 protein gene but only one chimpanzee candidate trace sequence was extracted from the chimpanzee-specific trace database. Fortunately two other candidate trace sequences were identified from the Pongo pygmaeus abelii trace database. All three candidates bear the same frameshift mutation, leading us to consider the frameshift as an accepted mutation. In CHIMP_1_178077, an exon for heat-shock protein beta-7, one homologous trace sequence from Pongo pygmaeus abelii was identified and found to share a nonsense mutation with three other chimpanzee trace sequences. A similar phenomenon was observed in many other exon-disrupting mutations in chimpanzee exons such as CHIMP_1_229134, an exon for tropomyosin alpha-3 chain (Tropomyosin-3), and CHIMP_X_91363, an exon for inhibitor of growth protein 1 homolog. On the other hand, differences were also observed between chimpanzees and pongos. For instance, a frameshift mutation was detected in 14 chimpanzee candidate trace sequences but not in any of the five pongo candidates. The observation is not surprising considering an approximate 13.04 Mya to 17.74 Mya differentiation time between Pan troglodytes and Pongo pygmaeus, a more distant relationship than between Pan troglodytes and Homo sapiens (a range of 5.65 Mya to 8.27 Mya) (http://www.timetree.org/) .
In summary, the analysis found an apparent asymmetry where more species-specific, exon-disrupting mutations were detected in chimpanzees than those in humans. One of the possible assumptions is that there is a bias towards detecting exons that are intact in humans because of more careful/complete annotations. Indeed, errors in annotations were detected in CHIMP_X_142174 and many other chimpanzee exons. We expected, however, such annotation errors would have limited impacts on the asymmetry because of our experimental design where all exons were extended by 100 base pairs at both end of the exons and then comparatively analyzed. The results were further Insilico validated.
3.3. The Structural Impacts
The lineage-specific exon-disrupting mutations were detected on many coding exons from the chimpanzee genome. Several examples were presented here to illustrate their potential structural and biological impacts, first through structure analysis of the affected genes and their resulting proteins, and then via the analyses of biological functions of the genes. Indeed, a combining effect of nonsense, frameshift, and exon missing mutations (data not shown) had completely reshaped the structure of the chimpanzee melanoma-associated gene C3 (Figure 4). Its exons were fractured where two human exons were split into a total of five chimpanzee ones (Figure 4(I)). Furthermore, the resulting protein sequence lost one of two functional MAGE domains (Figure 4(II)). Similar impacts were observed on melanoma-associated gene C2, SOX-13 protein (Type 1 diabetes autoantigen ICA12), tumor necrosis factor receptor superfamily member 18 precursor and many other genes where the overall structures of genes and their protein products were affected.
3.4. Functional Impacts
Genes with the species-specific exon-disrupting mutations appear to be involved in some critical physiological and biological processes. In chimpanzees, for example, affected genes included those encoding a series of the melanoma-associated antigens, for example, MAGE-C1 antigen, MAGE-3, MAGE-C2 and MAGE-1 antigen in addition to MAGE-C3 described above. Additional affected genes include those for the inhibitor of the growth family, member 1 (ING1), the testis-expressed sequence 13A protein, neuroblastoma breakpoint family, member 11, the tumor necrosis factor receptor superfamily member 18 precursor (TNFSF18), SOX-13 protein, heat-shock protein beta-7 (HSPB7), Eya3 and protocadherin-11 X-linked precursor (PCDH11X). These are genes/homologues with crucial roles in cell differentiations and developments.
The melanoma-associated genes are unique in that they are strictly tumor- and testis-specific and furthermore, their protein products can be recognized by autologous cytolytic T lymphocytes in many human tumors . The inhibitor of the growth family, member 1 (ING1) is a breast cancer suppressor gene . The acute expression of transfected constructs encoding ING1 inhibited cell growth while chronic expression of ING1 antisense constructs promoted cell transformation . The gene encoding the testis-expressed sequence 13A protein is a spermatogonically-expressed, germ-cell-specific genetic factor. Cytogenetic analysis indicated that the gene is located near breakpoints on the X chromosome in the azoospermic patients with X-autosome translocations , suggesting that it may be one of the direct genetic risk factors for azoospermia. Gene of the neuroblastoma breakpoint family, member 11 has been implicated in recent genome duplications, illegitimate recombination, and chromosomal translocation in a neuroblastoma patient .
Tumor necrosis factor receptor superfamily member 18 precursor (TNFSF18) is a member of the tumor necrosis factor (TNF) and TNF receptor (TNFR) gene superfamilies. Data accumulated from experiments indicated that it plays important roles in regulating cell proliferation, differentiation, and survival. This was manifested from Gurney’s experiment  where cotransfection of the tumor factor receptor and its ligand in Jurkat T leukemia cells inhibited antigen-receptor-induced cell death. Recently Kim et al. (2006) showed the role of the gene in mediating the inflammatory activation of macrophages that can destabilize atherosclerotic plaques . Bae et al., (2008) showed that TNFSF18 may be involved in the pathogenesis of inflammatory diseases . The stimulation of this gene induced the expression of pro-inflammatory cytokines and matrix metalloproteinase (MMP)-9 and up-regulated ICAM-1 expression levels.
Gene of SOX-13 protein is a transcription factor of the sex-determining region [SRY]-type high mobility group [HMG] box) family. Among other regulatory functions, Sox13 is the first lineage specific gene identified that modulates T cell lineage development. This gene promotes gamm/adelta T cell development while opposes alpha/beta T cell differentiation . Mice deficient in Sox13 expression exhibited impaired development of gamma/delta T cells but not alphabeta T cells. Previous research suggested a possible role of the heat-shock protein beta-7 (HSPB7) in cardiovascular development. Krief et al. (1999) found that it was selectively expressed in cardiovascular and insulin-sensitive tissues . In obese Zucker rats, its mRNA was increased in skeletal muscle, brown, and white adipose tissues but remained unchanged in the heart. Finally, few data were available for functions of Eya3 except the fact that the gene is expressed in the developing eye including the branchial arches and CNS but not cranial placode .
Among these exon-disrupting mutations, those on protocadherin-11 X-linked precursor (PCDH11X) are especially inspiring. Single frameshift mutations were detected in PCDH11X of both chimpanzee and human genomes when compared to their human Y-linked paralogue (PCDH11Y). The mutations resulted in the loss of the N-terminal 29-residue signal peptide in the X-linked protocadherin-11 proteins (Figure 5), suggesting that PCDHY and PCDHX may function at different cellular locations and differentially regulated. Indeed, Blanco et al. (2000) observed differential regulation in a pluripotential cell line. PCDHX predominated before retinoic acid treatment whereas PCDHY predominated after retinoic acid treatment . In addition, Yang et al. (2005) observed that the PCDHY is selectively expressed in apoptosis- and hormone-resistant human prostate cancer cells . An upgraded expression of PCDHY activated WNT signaling and drove neuroendocrine transdifferentiation. The observations suggested a significant different role of the PCDHY for the process through which prostate cancers progress to hormone resistance in humans. No Y-linked protocadherin-11 was observed in the chimpanzee genome. It will be interesting to investigate the biological impacts of the lack of the Y-linked protocadherin gene, structural changes of the melanoma-associated antigens and many others on physiological properties, especially disease responses of humans and chimpanzees.
So far, a total of 19 human-specific, exon-disrupting mutations were examined and among them, those in HUMAN_1_851772, HUMAN_1_837766 and HUMAN_1_851437 were validated (Table 3). HUMAN_1_851772 is an exon for the gene that encodes complement receptor type 1 precursor (C3b/C4b receptor) (CD35 antigen). The gene was identified as the immune-adherence receptor for the complement fragments C3b and C4b (CR1) . Its deficiency in erythrocytes was associated with systemic lupus erythematosus . HUMAN_1_851437 is an exon of gene for solute carrier family 26, member 9 isoform A (SLC26A9). The analysis of SLC26A9 expression patterns with homeostatic stress suggested a important role in the mediation of the response of the airway to stress . HUMAN_1_837766 is an exon of gene encoding calcium activated chloride channel 3 (CLCA3). As early as 1999, CLCA3 was characterized as a truncated, secreted member of the human family of Ca2+-activated Cl-channels . Compared to its 125-kDa transmembrane paralogous protein, the CLCA3 encodes a 37-kDa glycoprotein that corresponds to the N-terminal extracellular domain, suggesting that it does not act as a channel protein but has distinct, yet unidentified functions.
In spite of decades of extensive research, some questions remains about what is exactly behind the differences between two closely-related primates . The availability of completely sequenced genomes of humans and chimpanzees allows nucleotide-by-nucleotide comparisons at whole genome scale. Yet, an immense technique challenge remains due to the complexities of human and chimpanzee genomes. With the divide and conquer strategy, ExonVar is expected to provide new perspective in the field. Instead of comparatively analyzing eukaryotic genes, all the analyses focused exclusively on exons, the simplest and smallest coding units in the genomes.
With all the exons on human and chimpanzee genomes, this analysis demonstrated that the newly developed pipeline could improve performance and increase accuracy, and reveal significant differences between humans and chimpanzees. A large number of species-specific, exon-disrupting mutations were revealed in chimpanzees. The exon-disrupting mutation fractured exons, truncated protein domains, and thus forced the structural shifts of many chimpanzee genes away from that of humans. The differentially evolved exons are further involved in many crucial biological processes such as T cell lineage development, the pathogenesis of inflammatory diseases, antigen-receptor-induced cell death, cancer progression and many other important biological processes. It is thus hypothesized that the frameshift and nonsense mutations could play a great role in the human-chimpanzee divergence. Finally, the analysis discovered a number of sequencing errors but a majority of them can be corrected without resequencing. The pipeline thus will be valuable in improving genome annotations, enhancing understanding of human evolution, and eventually, providing drug candidates and strategies for better disease treatments.
This publication was made possible by NIH Grant #P20 RR016454 from the INBRE Program of the National Center for Research Resources and by supports from Boise State University. The author thankful for the critical reading of the manuscript by his colleague Dr. Jim Smith at Department of Biological Science of Boise State University.
- Chimpanzee Sequencing and Analysis Consortium, “Initial sequence of the chimpanzee genome and comparison with the human genome,” Nature, vol. 437, pp. 69–87, 2005.
- G. Silvestri, D. L. Sodora, R. A. Koup et al., “Nonpathogenic SIV infection of sooty mangabeys is characterized by limited bystander immunopathology despite chronic high-level viremia,” Immunity, vol. 18, no. 3, pp. 441–452, 2003.
- E. Muchmore, H. Popper, D. A. Peterson, M. F. Miller, and H. M. Lieberman, “Non-A, non-B hepatitis-related hepatocellular carcinoma in a chimpanzee,” Journal of Medical Primatology, vol. 17, no. 5, pp. 235–246, 1988.
- H. Schmidt and E. Liss, “Different uptake of two alkylating substances by ascites tumor cells,” Arzneimittelforschung, vol. 28, no. 11, pp. 2150–2152, 1978.
- D. Beniashvili, “Comparative study on the action of some carcinogens inducing transplacental blastomogenesis in rabbits,” Voprosy Onkologii, vol. 24, no. 3, pp. 77–83, 1978.
- S. H. Qari, Y. P. Shi, N. J. Pieniazek, W. E. Collins, and A. A. Lal, “Phylogenetic relationship among the malaria parasites based on small subunit rRNA gene sequences: monophyletic nature of the human malaria parasite, Plasmodium falciparum,” Molecular Phylogenetics and Evolution, vol. 6, no. 1, pp. 157–165, 1996.
- A. A. Escalante, E. Barrio, and F. J. Ayala, “Evolutionary origin of human and primate malarias: evidence from the circumsporozoite protein gene,” Molecular Biology and Evolution, vol. 12, no. 4, pp. 616–626, 1995.
- A. A. Escalante and F. J. Ayala, “Phylogeny of the malarial genus Plasmodium, derived from rRNA gene sequences,” Proceedings of the National Academy of Sciences of the United States of America, vol. 91, no. 24, pp. 11373–11377, 1994.
- M. Gearing, G. W. Rebeck, B. T. Hyman, J. Tigges, and S. S. Mirra, “Neuropathology and apolipoprotein E profile of aged chimpanzees: implications for Alzheimer disease,” Proceedings of the National Academy of Sciences of the United States of America, vol. 91, no. 20, pp. 9382–9386, 1994.
- A. G. Clark, S. Glanowski, R. Nielsen et al., “Inferring nonneutral evolution from human-chimp-mouse orthologous gene trios,” Science, vol. 302, no. 5652, pp. 1960–1963, 2003.
- G. Koopman, A. G. Haaksma, J. ten Velden, C. E. Hack, and J. L. Heeney, “The relative resistance of HIV type 1-infected chimpanzees to AIDS correlates with the maintenance of follicular architecture and the absence of infiltration by CD8+ cytotoxic T lymphocytes,” AIDS Research and Human Retroviruses, vol. 15, no. 4, pp. 365–373, 1999.
- C. Kornfeld, M. J. Ploquin, I. Pandrea et al., “Antiinflammatory profiles during primary SIV infection in african green monkeys are associated with protection against AIDS,” Journal of Clinical Investigation, vol. 115, no. 4, pp. 1082–1091, 2005.
- A. Varki, “A chimpanzee genome project is a biomedical imperative,” Genome Research, vol. 10, no. 8, pp. 1065–1070, 2000.
- D. M. Macdonald, E. C. Holmes, J. C. Lewis, and P. Simmonds, “Detection of hepatitis B virus infection in wild-born chimpanzees (pan troglodytes verus): phylogenetic relationships with human and other primate genotypes,” Journal of Virology, vol. 74, no. 9, pp. 4253–4257, 2000.
- J. C. Venter, M. D. Adams, E. W. Myers et al., “The sequence of the human genome,” Science, vol. 291, no. 5507, pp. 1304–1351, 2001.
- E. S. Lander, L. M. Linton, B. Birren et al., “Initial sequencing and analysis of the human genome,” Nature, vol. 409, no. 6822, pp. 860–921, 2001.
- Z. Cheng, M. Ventura, X. She et al., “A genome-wide comparison of recent chimpanzee and human segmental duplications,” Nature, vol. 437, no. 7055, pp. 88–93, 2005.
- A. Harris, J. Rogers, and A. Milosavljevic, “Human-specific changes of genome structure detected by genomic triangulation,” Science, vol. 316, no. 5822, pp. 235–237, 2007.
- H. Kehrer-Sawatzki and D. N. Cooper, “Molecular mechanisms of chromosomal rearrangement during primate evolution,” Chromosome Research, vol. 16, no. 1, pp. 41–56, 2008.
- T. Marques-Bonet, M. Caceres, J. Bertranpetit, T. M. Preuss, J. W. Thomas, and A. Navarro, “Chromosomal rearrangements and the genomic distribution of gene-expression divergence in humans and chimpanzees,” Trends in Genetics, vol. 20, no. 11, pp. 524–529, 2004.
- T. Marques-Bonet, J. Sanchez-Ruiz, and L. Armengol, “On the association between chromosomal rearrangements and genic evolution in humans and chimpanzees,” Genome Research, vol. 8, p. R230, 2007.
- T. Marques-Bonet and A. Navarro, “Chromosomal rearrangements are associated with higher rates of molecular evolution in mammals,” Gene, vol. 353, no. 2, pp. 147–154, 2005.
- Y. Hahn and B. Lee, “Identification of nine human-specific frameshift mutations by comparative analysis of the human and the chimpanzee genome sequences,” Bioinformatics, vol. 21, supplement 1, pp. i186–i194, 2005.
- Y. Hahn and B. Lee, “Human-specific nonsense mutations identified by genome sequence comparisons,” Human Genetics, vol. 119, no. 1-2, pp. 169–178, 2006.
- X. S. Puente, G. Velasco, A. Gutierrez-Fernandez, J. Bertranpetit, M. C. King, and C. Lopez-Otin, “Comparative analysis of cancer genes in the human and chimpanzee genomes,” BMC Genomics, vol. 7, p. 15, 2006.
- A. Wetterbom, U. Gyllensten, L. Cavelier, and T. F. Bergstrom, “Genome-wide analysis of chimpanzee genes with premature termination codons,” BMC Genomics, vol. 10, p. 56, 2009.
- S. J. Hsieh, C. Y. Lin, N. H. Liu, W. Y. Chow, and C. Y. Tang, “GeneAlign: a coding exon prediction tool based on phylogenetical comparisons,” Nucleic Acids Research, vol. 34, pp. W280–W284, 2006.
- F. O. Desmet, D. Hamroun, M. Lalande, G. Collod-Beroud, M. Claustres, and C. Beroud, “Human splicing finder: an online bioinformatics tool to predict splicing signals,” Nucleic Acids Research, vol. 37, p. e67, 2009.
- A. Churbanov, I. B. Rogozin, J. S. Deogun, and H. Ali, “Method of predicting splice sites based on signal interactions,” Biology Direct, vol. 1, p. 10, 2006.
- G. X. Yu, E. E. Snyder, S. M. Boyle et al., “A versatile computational pipeline for bacterial genome annotation improvement and comparative analysis, with Brucella as a use case,” Nucleic Acids Research, vol. 35, no. 12, pp. 3953–3962, 2007.
- M. V. Olson and A. Varki, “Sequencing the chimpanzee genome: insights into human evolution and disease,” Nature Reviews Genetics, vol. 4, no. 1, pp. 20–28, 2003.
- M. V. Olson, “When less is more: gene loss as an engine of evolutionary change,” American Journal of Human Genetics, vol. 64, no. 1, pp. 18–23, 1999.
- H. H. Stedman, B. W. Kozyak, A. Nelson et al., “Myosin gene mutation correlates with anatomical changes in the human lineage,” Nature, vol. 428, no. 6981, pp. 415–418, 2004.
- H. Winter, L. Langbein, M. Krawczak et al., “Human type I hair keratin pseudogene phihHaA has functional orthologs in the chimpanzee and gorilla: evidence for recent inactivation of the human gene after the Pan-Homo divergence,” Human Genetics, vol. 108, no. 1, pp. 37–42, 2001.
- G. E. Folk Jr. and A. Semken Jr., “The evolution of sweat glands,” International Journal of Biometeorology, vol. 35, no. 3, pp. 180–186, 1991.
- M. Pagel and W. Bodmer, “A naked ape would have fewer parasites,” Proceedings of the Royal Society B, vol. 270, supplement 1, pp. S117–S119, 2003.
- J. F. Caceres and A. R. Kornblihtt, “Alternative splicing: multiple control mechanisms and involvement in human disease,” Trends in Genetics, vol. 18, no. 4, pp. 186–193, 2002.
- B. Modrek and C. Lee, “A genomic view of alternative splicing,” Nature Genetics, vol. 30, no. 1, pp. 13–19, 2002.
- P. W. Holland, J. Garcia-Fernandez, N. A. Williams, and A. Sidow, “Gene duplications and the origins of vertebrate development,” Development Supplement, pp. 125–133, 1994.
- D. Steinke, S. Hoegg, H. Brinkmann, and A. Meyer, “Three rounds (1R/2R/3R) of genome duplications and the evolution of the glycolytic pathway in vertebrates,” BMC Biology, vol. 4, p. 16, 2006.
- G. Cannarozzi, A. Schneider, and G. Gonnet, “A phylogenomic study of human, dog, and mouse,” PLoS Computational Biology, vol. 33, no. 1, pp. 9–14, 2007.
- E. Birney, M. Clamp, and R. Durbin, “Gene wise and genomewise,” Genome Research, vol. 14, no. 5, pp. 988–995, 2004.
- E. Birney and R. Durbin, “Dynamite: a flexible code generating language for dynamic programming methods used in sequence comparison,” in Proceedings of the International Conference Intelligent Systems for Molecular Biology, vol. 5, pp. 56–64, 1997.
- V. Curwen, E. Eyras, T. D. Andrews et al., “The Ensembl automatic gene annotation system,” Genome Research, vol. 14, no. 5, pp. 942–950, 2004.
- Y. Sakoyama, K. J. Hong, S. I. Byun et al., “Nucleotide sequences of immunoglobulin epsilon genes of chimpanzee and orangutan: DNA molecular clock and hominoid evolution,” Proceedings of the National Academy of Sciences of the United States of America, vol. 84, no. 4, pp. 1080–1084, 1987.
- O. Emanuelsson, S. Brunak, G. von Heijne, and H. Nielsen, “Locating proteins in the cell using TargetP, SignalP and related tools,” Nature Protocols, vol. 2, no. 4, pp. 953–971, 2007.
- S. Lucas, E. De Plaen, and T. Boon, “MAGE-B5, MAGE-B6, MAGE-C2, and MAGE-C3: four new member of the MAGE family with tumor-specific expression,” International Journal of Cancer, vol. 87, no. 1, pp. 55–60, 2000.
- D. Jäger, E. Stockert, M. J. Scanlan et al., “Cancer-testis antigens and ING1 tumor suppressor gene product are breast cancer antigens: characterization of tissue-specific ING1 transcripts and a homologue gene,” Cancer Research, vol. 59, no. 24, pp. 6197–6204, 1999.
- I. Garkavtsev, A. Kazarov, A. Gudkov, and K. Riabowol, “Suppression of the novel growth inhibitor p33ING1 promotes neoplastic transformation,” Nature Genetics, vol. 14, no. 4, pp. 415–420, 1996.
- S. Lee, S. H. Lee, T. G. Chung et al., “Molecular and cytogenetic characterization of two azoospermic patients with X-autosome translocation,” Journal of Assisted Reproduction and Genetics, vol. 20, no. 9, pp. 385–389, 2003.
- K. Vandepoele, N. Van Roy, K. Staes, F. Speleman, and F. Van Roy, “A novel gene family NBPF: intricate structure generated by gene duplications during primate evolution,” Molecular Biology and Evolution, vol. 22, no. 11, pp. 2265–2274, 2005.
- A. L. Gurney, S. A. Marsters, A. Huang et al., “Identification of a new member of the tumor necrosis factor family and its receptor, a human ortholog of mouse GITR,” Current Biology, vol. 9, no. 4, pp. 215–218, 1999.
- W. J. Kim, E. M. Bae, Y. J. Kang et al., “Glucocorticoid-induced tumour necrosis factor receptor family related protein (GITR) mediates inflammatory activation of macrophages that can destabilize atherosclerotic plaques,” Immunology, vol. 119, no. 3, pp. 421–429, 2006.
- E. M. Bae, W. J. Kim, K. Suk et al., “Reverse signaling initiated from GITRL induces NF-kappaB activation through ERK in the inflammatory activation of macrophages,” Molecular Immunology, vol. 45, no. 2, pp. 523–533, 2008.
- H. J. Melichar, K. Narayan, S. O. Der et al., “Regulation of gammadelta versus alphabeta T lymphocyte differentiation by the transcription factor SOX13,” Science, vol. 315, no. 5809, pp. 230–233, 2007.
- S. Krief, J. F. Faivre, P. Robert et al., “Identification and characterization of cvHsp. A novel human small stress protein selectively expressed in cardiovascular and insulin-sensitive tissues,” Journal of Biological Chemistry, vol. 274, no. 51, pp. 36592–36600, 1999.
- P. X. Xu, I. Woo, H. Her, D. R. Beier, and R. L. Maas, “Mouse Eya homologues of the Drosophila eyes absent gene require Pax6 for expression in lens and nasal placode,” Development, vol. 124, no. 1, pp. 219–231, 1997.
- P. Blanco, C. A. Sargent, C. A. Boucher, M. Mitchell, and N. A. Affara, “Conservation of PCDHX in mammals; expression of human X/Y genes predominantly in brain,” Mammalian Genome, vol. 11, no. 10, pp. 906–914, 2000.
- X. Yang, M. W. Chen, S. Terry et al., “A human- and male-specific protocadherin that acts through the Wnt signaling pathway to induce neuroendocrine transdifferentiation of prostate cancer cells,” Cancer Research, vol. 65, no. 12, pp. 5263–5271, 2005.
- K. Iida and V. Nussenzweig, “Complement receptor is an inhibitor of the complement cascade,” Journal of Experimental Medicine, vol. 153, no. 5, pp. 1138–1150, 1981.
- K. Iida and V. Nussenzweig, “Functional properties of membrane-associated complement receptor CR1,” Journal of Immunology, vol. 130, no. 4, pp. 1876–1880, 1983.
- M. R. Dorwart, N. Shcheynikov, Y. Wang, S. Stippec, and S. Muallem, “SLC26A9 is a channel regulated by the WNK kinases,” Journal of Physiology, vol. 584, no. 1, pp. 333–345, 2007.
- A. D. Gruber and B. U. Pauli, “Molecular cloning and biochemical characterization of a truncated, secreted member of the human family of -activated channels,” Biochimica et Biophysica Acta, vol. 1444, no. 3, pp. 418–423, 1999.
Copyright © 2009 GongXin Yu. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.