• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of pnasPNASInfo for AuthorsSubscriptionsAboutThis Article
Proc Natl Acad Sci U S A. Dec 24, 2002; 99(26): 17008–17013.
Published online Dec 9, 2002. doi:  10.1073/pnas.262658799
PMCID: PMC139260
Genetics, Medical Sciences

Associations between human disease genes and overlapping gene groups and multiple amino acid runs

Abstract

Overlapping gene groups (OGGs) arise when exons of one gene are contained within the introns of another. Typically, the two overlapping genes are encoded on opposite DNA strands. OGGs are often associated with specific disease phenotypes. In this report, we identify genes with OGG architecture and genes encoding multiple long amino acid runs and examine their relations to diseases. OGGs appear to be susceptible to genomic rearrangements as happens commonly with the loci of the DiGeorge syndrome on human chromosome 22. We also examine the degree of conservation of OGGs between human and mouse. Our analyses suggest that (i) a high proportion of genes in OGG regions are disease-associated, (ii) genomic rearrangements are likely to occur within OGGs, possibly as a consequence of anomalous sequence features prevalent in these regions, and (iii) multiple amino acid runs are also frequently associated with pathologies.

The study of the association between human diseases and their underlying molecular causes is of considerable medical importance. Some disease-associated genes represent essential genes whose functional impairment is deleterious. However, nonessential genes may also induce disease phenotypes by means of dominant-negative effects or gain of toxic function. Diseases caused by deletions in noncoding regions may relate to gene regulation. Several genetic disease mechanisms can be distinguished, including (i) haplo-insufficiency (1), wherein loss of one gene copy results in insufficient gene product for normal function. In general, haploinsufficiency indicates that both alleles are necessary for proper biological function. In diseases such as Down's syndrome and Charcot-Marie-Tooth disease, gene overexpression can result from trisomy. In these cases it is the number of functional gene copies which is critical. (ii) Altered chromosome structure (e.g., segmental duplications and deletions, break point clusters, inversions, and translocations) can be related to disease. (iii) Toxic gain or loss of function can arise through alteration of protein binding sites, protein misfolding, or inappropriate aggregation as occurs in the polyglutamine trinucleotide repeat diseases. Our studies emphasize diseases associated with chromosomal sequence anomalies, the occurrence of overlapping gene groups (OGGs), and genes encoding multiple long amino acid runs.

Overlapping Gene Groups

There appears to be a strong correlation between genes associated with human diseases and overlapping groups of genes and/or genes that encode multiple amino acid runs (see examples below). OGGs are distributed in the current human genome Ensembl (www.ensembl.org) annotation as shown in Table Table1.1. Here we review examples in chromosomes (Chr) 21 and 22. There are at least 10 OGGs in Chr 21, according to the Riken annotation (ref. 2; Table Table2),2), and at least 34 OGGs in Chr 22 (Sanger data release 3.1, ref. 3; Table Table3).3). Tables Tables22 and and33 also indicate associations with known diseases. More OGGs may emerge as the genome annotation is refined.

Table 1.
Numbers of overlapping gene groups in the Ensembl annotation of human chromosomes
Table 2.
Overlapping gene groups in human Chr 21 (Riken annotation)
Table 3.
Overlapping gene groups in human Chr 22 (Sanger annotation)

OGG loci may be susceptible to genomic rearrangements, as occurs with the loci of the DiGeorge syndrome (DGS) region of Chr 22. Such rearrangements may be mediated by recombination events based on region-specific low copy repeats. The DGS region of 22q11.2 is particularly rich with segmental duplications, which can induce deletions, translocations, and genomic instability (4). There are several anomalous sequence features associated with OGGs, including Alu sequences intersecting exons, pseudogenes occupying introns, and single-exon (intronless) genes that often result from a processed multiexon gene.

At least 28 genes in Chr 21 are related to diseases, as characterized in the GeneCards database (5), as are 64 genes in Chr 22. Specific disorders that have been mapped to genes on Chr 21 and that involve OGG structures include: amyotrophic lateral sclerosis (ALS, Lou Gehrig's disease), linked to the GRIK1 ionotrophic kainate 1 glutamate receptor gene at 21q22 (6, 7); homocystinuria, a metabolic disorder linked to the cystathionine beta-synthase (CBS) gene (8); genes of the Down's Syndrome Critical Region (DSCR) (9–11); and the gene for amyloid beta (A4) precursor protein (APP) at location 21q21, associated with Alzheimer's disease (12).

We illustrate examples of OGGs in Fig. Fig.1.1. The overlapping structure of the GRIK1 gene is shown in the first example: ORF41 (two exons) and ORF9 (two exons) overlap GRIK1 (17 exons) in the intron between exons 8 and 9, and in the intron between exons 1 and 2, respectively. There is some overlap between the first exon of ORF41 and exon 8 of GRIK1. The GRIK1 (GLUR5) locus at 21q 22.1 (13) coincides with the localization of the mutant gene causing ALS. The structure and function of the glutamate receptor subunits GLUR2, GLUR5, and GLUR6 are altered by RNA editing, converting the codon CAG (coding for glutamine) to the codon CGG (arginine), which may be important in controlling the rate of calcium flux in different states of the brain (13). Another prominent example in Chr 21 is the overlap between the genes CBS and PKNOX1. All 14 exons of the 30-kb-long CBS gene are located in the last intron of the gene for the homeobox protein PKNOX1 (11 exons). The gene U2AF1 (eight exons) is situated 5′ to the CBS gene in the same long intron of PKNOX1. The metabolic disorder homocystinuria is due to cystathionine beta-synthase deficiency and manifests as disorders of the eyes, central nervous system, skeletal systems, and vascular systems. The exons of overlapping genes tend to lie within large introns, usually the boundary (first or last) introns of another gene structure.

Fig 1.
Examples of OGGs in Chrs 21 and 22.

In Chr 22, two OGGs are associated with genes of the DGS region: CLTCL1/DVL1L1 (clathrin heavy polypeptide-like 1/human homolog to the 3′ end of Drosophila dishevelled segment-polarity gene) and TR/COMT (thioredoxin reductase beta/catechol-O-methyltransferase) (14, 15). DGS is related to one or more large deletions from Chr 22 apparently generated by recombination at meiosis. The 22q11 region of Chr 22 is susceptible to rearrangements associated with several genetic disorders and malignant tumors. These include the cat eye syndrome (CES), part of the velocardiofacial syndrome (VCFS), DGS, and the der(22) chromosomal translocation (16). Many VCFS/DGS patients have a similar 3-Mb deletion and some have sporadically dispersed short deletions or translocations. DGS apparently results from haplo-insufficiency effects, and, in particular, the transcription factor Tbox-1 gene (TBX1) has been documented as one major contributing factor in congenital heart defects (4). How the observed OGG contributes to any DGS-related phenotype is unknown. In general, as in most gene deletion syndromes, a large majority of patients with DGS and Smith-Magenis syndrome have a common deletion interval, which may reflect meiotic unequal crossing-over mediated by flanking low copy number repeats. However, although patients with these conditions have almost identical deletions, there is substantial clinical variability. Galili et al. (14) verified synteny between a 150-kb region on mouse Chr 16 and the portion of 22q11 most commonly deleted in DGS.

Another major OGG of Chr 22 connects TIMP3 (tissue inhibitor of metalloproteinase; refs. 17 and 18) with SYN3 (synapsin-III, a membrane protein possibly involved in regulating neurotransmitter release) and is shown in Fig. Fig.11 Lower. TIMP3 is associated with Sorsby fundus dystrophy and is a zinc-binding endopeptidase localized to the extracellular matrix that is expressed in many tissues, but is especially abundant in the placenta. Further OGG examples include: TR and COMT (putatively connected with schizophrenia); SERPIND1 (heparin cofactor II associated with thrombophilia) and PIK4CA (phosphatidylinositol 4-kinase α-subunit); and RTDR1 (rhabdoid tumor deletion region protein 1) and GNAZ (guanine nucleotide binding protein α-z).

There are several OGG-like structures involving sequences related to BCR (breakpoint cluster region fusion gene), which connects the distal part of Chr 22 to the q-arm of Chr 9 in the Philadelphia translocation, causing chronic myeloid leukemia. BCR itself overlaps with the F-box protein pseudogene FBXW3. In addition, there are seven BCR-like pseudogenes on Chr 22, of which two appear in OGGs, as shown in Table Table4.4. There are five non-OGG BCR-like pseudogenes: AP000550.6, AP000552.3, BCRL4, AP000354.4, and BCRL6. It is striking that the eight BCR-related sequences of Chr 22 cluster within a 6-Mb stretch.

Table 4.
Three OGGs associated with BCR-like genes/ pseudogenes in Chr 22

OGGs and Anomalous Sequence Features

There is a conspicuous association of disease genes with OGGs involving intrusions of Alu, and/or pseudogene, and/or single-exon (intronless) sequences (see Tables Tables22 and and3).3). Also, two single-exon genes, CLDN17 (claudin) and CLDN8, that contribute to tight junction formations are immediately 5′ to GRIK1. SOD1 (superoxide dismutase) follows the OGG of TIAM1 and a pseudogene (BTRC2P) in Chr 21, and is within 2 Mb of GRIK1. Reduction in SOD1 activity might be expected to lead to an accumulation of toxic superoxide radicals, which can cause familial ALS (19). Allelic variants of GRIK1 further contribute to the pathogenesis of juvenile absence epilepsy (13). In the CBS/PKNOX1 OGG, an Alu sequence overlaps with an exon of the PKNOX1 gene. These may predispose the gene to detrimental rearrangements. It is further documented that the CBS gene can undergo alternative splicing in its 5′ UTR (8). Another gene, CRYAA (crystalline), which can produce a cataract phenotype, is directly 3′ to PKNOX1.

In Chr 21, only 34 (of 12,168) Alu elements overlap exons. Twenty Alu elements are either totally within or envelop a complete exon, four of which are internal exons. Four Alus overlap internal exons, whereas the other 10 overlap boundary exons mostly in UTRs. In Chr 22 (23,675 Alus), there are only 165 instances, involving 87 genes, of an Alu overlapping an exon. Of these, 5 involve an internal coding exon, 98 are with a noncoding exon, and 62 are with boundary exons that contain a translation initiation or termination codon. In 3 cases, the Alu completely envelops an exon, in 141 it is contained within an exon, and in 21 cases the Alu and exon overlap. However, there are 12,367 instances of an Alu being contained in an intron of a gene on Chr 22, involving 408 of the 546 coding genes. These results are broadly consistent with other studies of the occurrence of transposable elements in coding genes (20). Although Alu insertions and other transposition events have been shown to generate null alleles through insertional transposition, these appear to be uncommon mechanisms for human diseases (21), except possibly in the context of OGG structures. In Chr 21, there are no pseudogenes overlapping exons. In Chr 22, there are 11 pseudogenes that show some overlap with exons of coding genes. In eight of these cases, the overlap is between an exon and an intron of a multiexon pseudogene. There is one intronless pseudogene that partially overlaps with a gene exon sequence, and two intronless pseudogenes contained within exons; however, all of the exons of the genes involved are untranslated. No pseudogenes overlap with coding exons. Pseudogene sequences are biased toward highly expressed genes, emphasizing ribosomal protein genes (22, 23).

Concurrence of OGGs, Pseudogenes, and Disease Genes

It appears that genes containing pseudogenes in introns or Alu elements overlapping with exons have a strong disposition for disease on Chr 22. There are 546 genes annotated (Sanger data), with 64 disease genes [GeneCards (5)], and 34 OGGs involving 71 genes, of which 13 are disease-associated. Thus, the fraction of genes with an associated disease from among the OGG collection is 13/71 = 0.18. There are 64 − 13 = 51 remaining known disease genes of a total of 546 − 71 = 475 genes not associated with OGGs. The fraction of disease genes in non-OGG surroundings is thus (64 − 13)/(546 − 71) = 0.10. An analogous calculation indicates that disease genes seem to have a higher chance of overlapping with pseudogenes: 49 genes overlap with pseudogenes in Chr 22, including 12 of the 64 known disease genes. Thus 12/64 = 0.19 of disease genes in Chr 22 are associated with pseudogenes, compared with (49 − 12)/(546 − 64) = 0.08 of genes with no known disease association.

The OGG of the gene combination TIMP3 and SYN3 is conserved in mouse and in Drosophila (17). A number of OGG structures based on the Ensembl data collection connect additional TIMP subunits and synapsin subunits in the human and mouse genomes. These are displayed in Table Table5.5. Another OGG present in both human Chr 22 and mouse is the gene pair TR and COMT (see Table Table3).3).

Table 5.
OGGs of TIMP and synapsin in human and mouse

Multiple Amino Acid Runs

There are 192 human protein sequences [of 10,651 ≥ 200 aa long, extracted from RefSeq (www.ncbi.nlm.nih.gov/LocusLink/refseq.html)] that have multiple amino acid runs (24). More than 40% of these proteins are associated with diseases, as identified in OMIM (25). All established human CAG triplet repeat (polyglutamine) diseases (26), together with some potential new ones, qualify as having multiple runs, not just of glutamine. In addition, many proteins related to leukemia and other cancers have multiple runs: 14 cancer-related proteins (e.g., adenomatous polyposis coli, breast carcinoma-associated antigen, and matrix metalloproteinase 24); 10 leukemia-related proteins often resulting from chromosomal translocations (listed in Table Table6);6); 14 channel proteins, mainly voltage-gated Ca2+ and K+ channel proteins; 6 proteases, including acrosin, calpain 4, and some metalloproteinases; and a variety of disease syndrome-related proteins (e.g., Wiskott-Aldrich syndrome and cat eye syndrome). A key aspect of 82 of the 192 human protein sequences is their role in transcription, translation, and developmental regulation. Strikingly, many of these proteins are homeotic homologs of Drosophila developmental sequences and transcription factors, including timeless, trithorax, frizzled, dead ringer (retained), and diaphanous 3.

Table 6.
Leukaemia-related proteins containing multiple amino acid runs

In marked contrast, no metabolic enzymes (e.g., glycolysis, tricarboxylic acid cycle, and pentose phosphate pathway), structural proteins (e.g., actin, myosin, and troponin 1), or housekeeping proteins contain multiple runs. However, several structural-regulatory proteins do have multiple runs, including ankyrin 3, nucleolin, SMARCA2 (actin-dependent regulator of chromatin), and synapsin II, which may function in the regulation of neurotransmitter release.

Prokaryote protein analogs/homologs in the human genome do not have multiple amino acid runs. On this basis, multiple runs in human proteins may be a recent evolutionary outcome, concomitant with complex brain or heart development. Multiple runs are, however, substantially conserved between human and mouse proteins. Of 56 SWISS-PROT mouse proteins that have multiple runs, 52 have a human homolog. In 43 cases (83%), the human homolog also has multiple runs; in 10%, the human homolog has more than one run but does not meet the criterion for multiple runs; and in the remaining 7% [DDX9 (ATP-dependent RNA helicase A), DUS8 (neuronal tyrosine threonine phosphatase 1), HOXD9 (homeobox protein), and UBF1 (nucleolar transcription factor 1)], the human protein has one or no runs. Examples of human/mouse proteins that share multiple runs are CREB-binding protein, diaphanous and even-skipped homologs, anaplastic lymphoma kinase, myc-associated zinc finger (MAZ), and two zinc finger proteins of the cerebellum (ZIC2 and ZIC3). The disease genes meningioma 1 (MN1), Ran GTPase activating protein 1 (RANGAP1), and the cat eye syndrome region (CECR) of Chr 22 encode proteins with an abundance of multiple long homopeptides, multiple charge clusters, and a large count of multiplets (amino acid doublets, triplets, etc.). These sequence properties could induce neurological phenotypes (24).

We conclude that the majority of OGGs and genes encoding significantly many amino acid long runs are potentially associated with disease. We also hypothesize that OGGs increase the potential for genomic rearrangements and/or disruption of transcription regulation, and may predispose these gene groups to contain disease-related genes at substantially higher frequency than non-OGG genes. The presence of an OGG may cause difficulties in transcription, in fostering complex gene rearrangements, and redundancies in propensity to mutations and mutational hot spots (partly dependent on the presence of Alus), and in generating gene dosage imbalances. Alu (and other transposable elements) are innately mobile and, like pseudogenes, are heavily prone to mutation (21). Also, many single-exon genes, like pseudogenes, derive often from the processing of multiexon genes (see ref. 22). Thus, human disease genes tend to be associated with disrupting Alu sequences, and/or pseudogenes, and/or proximal single-exon genes. Extant OGGs and consequent rearrangements appear as a novel configuration of many disease genes. Experimental studies are required to confirm these observations and elucidate the underlying mechanisms.

Acknowledgments

This work was supported in part by National Institutes of Health Grants 5R01GM10452-38 and 5R01HG00335-15 (to S.K.).

Abbreviations

  • Chr, chromosome
  • ALS, amyotrophic lateral sclerosis
  • CBS, cystathionine beta-synthase
  • DGS, DiGeorge syndrome
  • OGG, overlapping gene group
  • Ψg, pseudogene

References

1. Seidman J. G. & Seidman, C. (2002) J. Clin. Invest. 109, 451-455. [PMC free article] [PubMed]
2. Hattori M., Fujiyama, A., Taylor, T. D., Watanabe, H., Yada, T., Park, H. S., Toyoda, A., Ishii, K., Totoki, Y., Choi, D. K., et al. (2000) Nature 405, 311-319. [PubMed]
3. Dunham I., Shimizu, N., Roe, B. A., Chissoe, S., Hunt, A. R., Collins, J. E., Bruskiewich, R., Beare, D. M., Clamp, M., Smink, L. J., et al. (1999) Nature 402, 489-495. [PubMed]
4. Edelmann L., Pandita, R. K., Spiteri, E., Funke, B., Goldberg, R., Palanisamy, N., Chaganti, R. S., Magenis, E., Shprintzen, R. J. & Morrow, B. E. (1999) Hum. Mol. Genet. 8, 1157-1167. [PubMed]
5. Rebhan M., Chalifa-Caspi, V., Prilusky, J. & Lancet, D. (1998) Bioinformatics 14, 656-664. [PubMed]
6. Eubanks J. H., Puranam, R. S., Kleckner, N. W., Bettler, B., Heinemann, S. F. & McNamara, J. O. (1993) Proc. Natl. Acad. Sci. USA 90, 178-182. [PMC free article] [PubMed]
7. Paschen W. & Djuricic, B. (1994) Cell. Mol. Neurobiol. 14, 259-270. [PubMed]
8. Avramopoulos D., Cox, T., Kraus, J. P., Chakravarti, A. & Antonarakis, S. E. (1993) Hum. Genet. 90, 566-568. [PubMed]
9. Fuentes J.-J., Pritchard, M. A., Planas, A. M., Bosch, A., Ferrer, I. & Estivill, X. (1995) Hum. Mol. Genet. 4, 1935-1944. [PubMed]
10. Nakamura A., Hattori, M. & Sakaki, Y. (1997) J. Biochem. 122, 872-877. [PubMed]
11. Korenberg J. R., Chen, X. N., Schipper, R., Sun, Z., Gonsky, R., Gerwehr, S., Carpenter, N., Daumer, C., Dignan, P., Disteche, C., et al. (1994) Proc. Natl. Acad. Sci. USA 91, 4997-5001. [PMC free article] [PubMed]
12. Yanker B. A. (1996) Neuron 16, 921-932. [PubMed]
13. Sander T., Hildmann, T., Kretz, R., Furst, R., Sailer, U., Bauer, G., Schmitz, B., Beck-Mannagetta, G., Wienker, T. F. & Janz, D. (1997) Am. J. Med. Genet. 74, 416-421. [PubMed]
14. Galili N., Baldwin, H. S., Lund, J., Reeves, R., Gong, W., Wang, Z., Roe, B. A., Emanuel, B. S., Nayak, S., Mickanin, C., et al. (1997) Genome Res. 7, 17-26. [PubMed]
15. Schinke M. & Izumo, S. (2001) Nat. Genet. 27, 238-240. [PubMed]
16. Tapia-Páez I., Kost-Alimova, M., Hu, P., Roe, B. A., Blennow, E., Fedorova, L., Imreh, S. & Dumanski, J. P. (2001) Hum. Genet. 109, 167-177. [PubMed]
17. Weber B. H., Vogt, G., Pruett, R. C., Stohr, H. & Felbor, U. (1994) Nat. Genet. 8, 352-356. [PubMed]
18. Stohr H., Roomp, K., Felbor, U. & Weber, B. H. (1995) Genome Res. 5, 483-487. [PubMed]
19. Estevez A. G., Crow, J. P., Sampson, J. B., Reiter, C., Zhuang, Y., Richardson, G. J., Tarpey, M. M., Barbeito, L. & Beckman, J. S. (1999) Science 286, 2498-2500. [PubMed]
20. Nekrutenko A. & Li, W.-H. (2001) Trends Genet. 17, 619-621. [PubMed]
21. Deininger P. L. & Batzer, M. A. (1999) Mol. Genet. Metab. 67, 183-193. [PubMed]
22. Chen C., Gentles, A. J., Jurka, J. & Karlin, S. (2002) Proc. Natl. Acad. Sci. USA 99, 2930-2935. [PMC free article] [PubMed]
23. Harrison P. M., Hegyi, H., Balasubramanian, S., Luscombe, N. M., Bertone, P., Echols, N., Johnson, T. & Gerstein, M. (2002) Genome Res. 12, 272-280. [PMC free article] [PubMed]
24. Karlin S., Brocchieri, L., Bergman, A., Mrazek, J. & Gentles, A. J. (2002) Proc. Natl. Acad. Sci. USA 99, 333-338. [PMC free article] [PubMed]
25. Hamosh A., Scott, A. F., Amberger, J., Bocchini, C., Valle, D. & McKusick, V. A. (2002) Nucleic Acids Res. 30, 52-55. [PMC free article] [PubMed]
26. Zoghbi H. Y. & Orr, H. T. (2000) Annu. Rev. Neurosci. 23, 217-247. [PubMed]

Articles from Proceedings of the National Academy of Sciences of the United States of America are provided here courtesy of National Academy of Sciences
PubReader format: click here to try

Formats:

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...

Links

  • Gene (nucleotide)
    Gene (nucleotide)
    Records in Gene identified from shared sequence links
  • MedGen
    MedGen
    Related information in MedGen
  • Nucleotide
    Nucleotide
    Published Nucleotide sequences
  • OMIM
    OMIM
    OMIM record citing PubMed
  • Protein
    Protein
    Published protein sequences
  • PubMed
    PubMed
    PubMed citations for these articles
  • Substance
    Substance
    PubChem Substance links
  • Taxonomy
    Taxonomy
    Related taxonomy entry
  • Taxonomy Tree
    Taxonomy Tree

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...