Reactome: a knowledgebase of biological pathways

Reactome, located at http://www.reactome.org is a curated, peer-reviewed resource of human biological processes. Given the genetic makeup of an organism, the complete set of possible reactions constitutes its reactome. The basic unit of the Reactome database is a reaction; reactions are then grouped into causal chains to form pathways. The Reactome data model allows us to represent many diverse processes in the human system, including the pathways of intermediary metabolism, regulatory pathways, and signal transduction, and high-level processes, such as the cell cycle. Reactome provides a qualitative framework, on which quantitative data can be superimposed. Tools have been developed to facilitate custom data entry and annotation by expert biologists, and to allow visualization and exploration of the finished dataset as an interactive process map. Although our primary curational domain is pathways from Homo sapiens, we regularly create electronic projections of human pathways onto other organisms via putative orthologs, thus making Reactome relevant to model organism research communities. The database is publicly available under open source terms, which allows both its content and its software infrastructure to be freely used and redistributed.

An inherited limb deformity created by insertional mutagenesis in a transgenic mouse

We have created an insertional mutation that leads to a severe defect in the pattern of limb formation in the developing mouse. The novel recessive mutation is phenotypically identical and non-complementary to two previously encountered limb deformity mutations, and is closely linked to a dominant mutation that gives rise to a related limb dysmorphism. The inserted element thus provides a molecular genetic link with the control of pattern formation in the mammalian embryo.

Characterization of proteins that interact with the cell-cycle regulatory protein Ran/TC4

The human Ras-related nuclear protein Ran/TC4 (refs 1-4) is the prototype of a well conserved family of GTPases that can regulate both cell-cycle progression and messenger RNA transport. Ran has been proposed to undergo tightly controlled cycles of GTP binding and hydrolysis, to operate as a GTPase switch whose GTP- and GDP-bound forms interact differentially with regulators and effectors. One known regulator, the protein RCC1 (refs 12, 13), interacts with Ran to catalyse guanine nucleotide exchange, and both RCC1 and Ran are components of an intrinsic checkpoint control that prevents the premature initiation of mitosis. To test and extend the GTPase-switch model, we searched for a Ran-specific GTPase-activating protein (GAP), and for putative effectors (proteins that interact specifically with Ran/TC4-GTP). We report here the identification of a Ran GAP and its use to characterize the GTP-hydrolysing properties of mutant Ran proteins, and the identification and cloning of a binding protein specific for Ran/TC4-GTP.

The SH2 domain protein GRB-7 is co-amplified, overexpressed and in a tight complex with HER2 in breast cancer

SH2 domain proteins are important components of the signal transduction pathways activated by growth factor receptor tyrosine kinases. We have been cloning SH2 domain proteins by bacterial expression cloning using the tyrosine phosphorylated C-terminus of the epidermal growth factor receptor as a probe. One of these newly cloned SH2 domain proteins, GRB-7, was mapped on mouse chromosome 11 to a region which also contains the tyrosine kinase receptor, HER2/erbB-2. The analogous chromosomal locus in man is often amplified in human breast cancer leading to overexpression of HER2. We find that GRB-7 is amplified in concert with HER2 in several breast cancer cell lines and that GRB-7 is overexpressed in both cell lines and breast tumors. GRB-7, through its SH2 domain, binds tightly to HER2 such that a large fraction of the tyrosine phosphorylated HER2 in SKBR-3 cells is bound to GRB-7. GRB-7 can also bind tyrosine phosphorylated SHC, albeit at a lower affinity than GRB2 binds SHC. We also find that GRB-7 has a strong similarity over > 300 amino acids to a newly identified gene in Caenorhabditis elegans. This region of similarity, which lies outside the SH2 domain, also contains a pleckstrin homology domain. The presence of evolutionarily conserved domains indicates that GRB-7 is likely to perform a basic signaling function. The fact that GRB-7 and HER2 are both overexpressed and bound tightly together suggests that this basic signaling pathway is greatly amplified in certain breast cancers.

Ran/TC4: a small nuclear GTP-binding protein that regulates DNA synthesis

Ran/TC4, first identified as a well-conserved gene distantly related to H-RAS, encodes a protein which has recently been shown in yeast and mammalian systems to interact with RCC1, a protein whose function is required for the normal coupling of the completion of DNA synthesis and the initiation of mitosis. Here, we present data indicating that the nuclear localization of Ran/TC4 requires the presence of RCC1. Transient expression of a Ran/TC4 protein with mutations expected to perturb GTP hydrolysis disrupts host cell DNA synthesis. These results suggest that Ran/TC4 and RCC1 are components of a GTPase switch that monitors the progress of DNA synthesis and couples the completion of DNA synthesis to the onset of mitosis.

Cloning and expression of a widely expressed receptor tyrosine phosphatase

We describe the identification of a widely expressed receptor-type (transmembrane) protein tyrosine phosphatase (PTPase; EC 3.1.3.48). Screening of a mouse brain cDNA library under low-stringency conditions with a probe encompassing the intracellular (phosphatase) domain of the CD45 lymphocyte antigen yielded cDNA clones coding for a 794-amino acid transmembrane protein [hereafter referred to as receptor protein tyrosine phosphatase alpha (R-PTP-alpha)] with an intracellular domain displaying clear homology to the catalytic domains of CD45 and LAR (45% and 53%, respectively). The 142-amino acid extracellular domain (including signal peptide) of R-PTP-alpha is marked by a high serine/threonine content (32%) as well as eight potential N-glycosylation sites but displays no similarity to known proteins. Genetic mapping assigns the gene for R-PTP-alpha to mouse chromosome 2, closely linked to the Il-1a and Bmp-2a loci. The corresponding mRNA (3.0 kilobases) is expressed in most murine tissues and most abundantly expressed in brain and kidney. Antibodies against a synthetic peptide of R-PTP-alpha identified a 130-kDa protein in cells transfected with the R-PTP-alpha cDNA.

Structure and evolutionary origin of the gene encoding mouse NF-M, the middle-molecular-mass neurofilament protein

We describe the complete sequence of the gene encoding mouse NF-M, the middle-molecular-mass neurofilament protein. The coding sequence is interrupted by two intervening sequences which align perfectly with the first two intervening sequences in the gene encoding NF-L (the low-molecular-mass neurofilament protein); there is no intron in the gene encoding NF-M corresponding to the third intron in NF-L. Therefore, both the number of introns and their arrangement in the genes coding NF-L and NF-M contrast sharply with the number and arrangement of introns in the genes of known sequence, encoding other members of the intermediate filament multigene family (desmin, vimentin, glial fibrillary acidic protein and the acidic and basic keratins); with the exception of a single truncated keratin gene that lacks an encoded tailpiece, these genes all contain eight introns, of which at least six are placed at homologous locations. Assuming the existence of a primordial intermediate filament gene containing most (if not all) the introns found in contemporary non-neurofilament intermediate filament genes, it seems likely that an RNA-mediated transposition event was involved in the generation of an ancestral gene encoding the NF polypeptides. A combination of insertional transposition and gene-duplication events could then explain the anomalous number and placement of introns within these genes. Consistent with this notion, we show that the genes encoding NF-M and NF-L are linked.

Chromosomal location of structural genes encoding murine immunoglobulin lambda light chains. Genetics of murine lambda light chains

To determine the chromosomal localization of murine lambda light (L) chain structural genes, DNA from a panel of 11 mouse x hamster somatic cell hybrids was scored for the presence of sequences homologous to cloned lambda DNA probe molecules. Six of the hybrids had detectable lambda I and lambda II gene sequences. In all six, the full complement of murine sequences was present, and in its germline configuration. The remaining hybrids lacked any detectable murine lambda L chain gene sequences. The only mouse chromosome present in all of the positive hybrids and absent from the negative ones was number 16, allowing the assignment of lambda L chain structural genes to this chromosome. Together with the previous assignments of the kappa L chain genes to chromosome 6 and heavy chain genes to chromosome 12, this finding completes the mapping of Ig structural genes in the mouse at the chromosomal level.

Linkage mapping of a mouse gene, iv, that controls left-right asymmetry of the heart and viscera

Inherited single gene defects have been identified in both humans and mice that lead to loss of developmental control over the left-right asymmetry of the heart and viscera. In mice the recessively inherited mutation iv leads to such apparent loss of control over situs: 50% of iv/iv mice exhibit situs inversus and 50% exhibit normal situs. The affected gene product has not been identified in these animals. To study the normal function of iv, we have taken an approach directed to the gene itself. As a first step, we have mapped iv genetically, by examining its segregation in backcrosses with respect to markers defined by restriction fragment length polymorphisms. The iv locus lies 3 centimorgans (cM) from the immunoglobulin heavy-chain constant-region gene complex (Igh-C) on chromosome 12. A multilocus map of the region suggests the gene order centromere-Aat (alpha 1-antitrypsin gene complex)-(11 cM)-iv-(3 cM)-Igh-C-(1 cM)-Igh-V (immunoglobulin heavy-chain variable-region gene complex).

The small nuclear GTPase Ran: how much does it run?

Ran is one of the most abundant and best conserved of the small GTP binding and hydrolyzing proteins of eukaryotes. It is located predominantly in cell nuclei. Ran is a member of the Ras family of GTPases, which includes the Ras and Ras-like proteins that regulate cell growth and division, the Rho and Rac proteins that regulate cytoskeletal organization and the Rab proteins that regulate vesicular sorting. Ran differs most obviously from other members of the Ras family in both its nuclear localization, and its lack of sites required for post-translational lipid modification. Ran is, however, similar to other Ras family members in requiring a specific guanine nucleotide exchange factor (GEF) and a specific GTPase activating protein (GAP) as stimulators of overall GTPase activity. In this review, the multiple cellular functions of Ran are evaluated with respect to its known biochemistry and molecular interactions.

A mouse homeo box gene is expressed in spermatocytes and embryos

The MH-3 gene, which contains a homeo box that is expressed specifically in the adult testis, was identified and mapped to mouse chromosome 6. By means of in situ hybridization with adult testis sections and Northern blot hybridization with testis RNA from prepuberal mice and from Sl/Sld mutant mice, it was demonstrated that this gene is expressed in male germ cells during late meiosis. In the embryo, MH-3 transcripts were present at day 11.5 post coitum, a stage in mouse development when gonadal differentiation has not yet occurred. The MH-3 gene may have functions in spermatogenesis and embryogenesis.

Cellular functions of TC10, a Rho family GTPase: regulation of morphology, signal transduction and cell growth

The small Ras-related GTPase, TC10, has been classified on the basis of sequence homology to be a member of the Rho family. This family, which includes the Rho, Rac and CDC42 subfamilies, has been shown to regulate a variety of apparently diverse cellular processes such as actin cytoskeletal organization, mitogen-activated protein kinase (MAPK) cascades, cell cycle progression and transformation. In order to begin a study of TC10 biological function, we expressed wild type and various mutant forms of this protein in mammalian cells and investigated both the intracellular localization of the expressed proteins and their abilities to stimulate known Rho family-associated processes. Wild type TC10 was located predominantly in the cell membrane (apparently in the same regions as actin filaments), GTPase defective (75L) and GTP-binding defective (31N) mutants were located predominantly in cytoplasmic perinuclear regions, and a deletion mutant lacking the carboxyl terminal residues required for post-translational prenylation was located predominantly in the nucleus. The GTPase defective (constitutively active) TC10 mutant: (1) stimulated the formation of long filopodia; (2) activated c-Jun amino terminal kinase (JNK); (3) activated serum response factor (SRF)-dependent transcription; (4) activated NF-kappaB-dependent transcription; and (5) synergized with an activated Raf-kinase (Raf-CAAX) to transform NIH3T3 cells. In addition, wild type TC10 function is required for full H-Ras transforming potential. We demonstrate that an intact effector domain and carboxyl terminal prenylation signal are required for proper TC10 function and that TC10 signals to at least two separable downstream target pathways. In addition, TC10 interacted with the actin-binding and filament-forming protein, profilin, in both a two-hybrid cDNA library screen, and an in vitro binding assay. Taken together, these data support a classification of TC10 as a member of the Rho family, and in particular, suggest that TC10 functions to regulate cellular signaling to the actin cytoskeleton and processes associated with cell growth.

Characterization of four novel ras-like genes expressed in a human teratocarcinoma cell line

A mixed-oligonucleotide probe was used to identify four ras-like coding sequences in a human teratocarcinoma cDNA library. Two of these sequences resembled the rho genes, one was closely related to H-, K-, and N-ras, and one shared only the four sequence domains that define the ras gene superfamily. Homologs of the four genes were found in genomic DNA from a variety of mammals and from chicken. The genes were transcriptionally active in a range of human cell types.

Chromosomal assignment of the mouse kappa light chain genes

Mouse-hamster somatic cell hybrids containing a variable number of mouse chromosomes have been used in experiments to determine which mouse chromosome carries the immunoglobulin kappa light chain genes. It has been shown by nucleic acid hybridization that the kappa constant gene and the genes for at least one variable region subgroup are on mouse chromosome 6. This somatic cell genetic mapping procedure appears to be general and can be applied to any expressed or silent gene for which an appropriate nucleic acid probe exists.

Human nucleolus organizers on nonhomologous chromosomes can share the same ribosomal gene variants

The distributions of three human ribosomal gene polymorphisms among individual chromosomes containing nucleolus organizers were analyzed by using mouse--human hybrid cells. Different nucleolus organizers can contain the same variant, suggesting the occurrence of genetic exchanges among ribosomal gene clusters on nonhomologous chromosomes. Such exchanges appear to occur less frequently in mice. This difference is discussed in terms of the nucleolar organization and chromosomal location of ribosomal gene clusters in humans and mice.

Chromosomal location of the structural gene cluster encoding murine immunoglobulin heavy chains

To determine the chromosomal location of mouse immunoglobulin heavy chain structural genes unambiguously, a panel of somatic cell hybrids was scored for the presence of DNA sequences homologous to gamma 2b-, mu-, and alpha-heavy chain-constant region DNA probe molecules. The hybrids, formed between mouse and hamster cells, contained various combinations of mouse chromosomes plus a full set of hamster chromosomes. Hybrids that retained mouse chromosome 12 reacted with the probes, whereas hybrids that had lost the chromosome, or its distal half, failed to react. These results indicate that structural genes for the gamma 2b-, mu-, and alpha-heavy chain-constant regions map to the distal half of this chromosome.

Localization of the casein gene family to a single mouse chromosome

A series of mouse-hamster somatic cell hybrids containing a variable number of mouse chromosomes and a constant set of hamster chromosomes have been used to determine the chromosomal location of a family of hormone-inducible genes, the murine caseins. Recombinant mouse cDNA clones encoding the alpha-, beta-, and gamma-caseins were constructed and used in DNA restriction mapping experiments. All three casein cDNAs hybridized to the same set of somatic cell hybrid DNAs isolated from cells containing mouse chromosome 5, while negative hybridization was observed to ten other hybrid DNAs isolated from cells lacking chromosome 5. A fourth cDNA clone, designated pCM delta 40, which hybridized to an abundant 790 nucleotide poly(A)RNA isolated from 6-d lactating mouse mammary tissue, was also mapped to chromosome 5. The chromosomal assignment of the casein gene family was confirmed using a mouse albumin clone. The albumin gene had been previously localized to mouse chromosome 5 by both breeding studies and analogous molecular hybridization experiments. An additional control experiment demonstrated that another hormone-inducible gene, specifying a 620 nucleotide abundant mammary gland mRNA, hybridized to DNA isolated from a different somatic cell hybrid line. These studies represent the first localization of a peptide and steroid hormone-responsive gene family to a single mouse chromosome.

A linkage map of mouse chromosome 1 using an interspecific cross segregating for the gld autoimmunity mutation

An interspecific backcross was used to define a high resolution linkage map of mouse Chromosome (Chr) 1 and to analyze the segregation of the generalized lymphoproliferative disease (gld) mutation. Mice homozygous for gld have multiple features of autoimmune disease. Analysis of up to 428 progeny from the backcross [(C3H/HeJ-gld x Mus spretus)F1 x C3H/HeJ-gld] established a map that spans 77.6 cM and includes 56 markers distributed over 34 ordered genetic loci. The gld mutation was mapped to a less than 1 cM segment on distal mouse Chr 1 using 357 gld phenotype-positive backcross mice. A second backcross, between the laboratory strains C57BL/6J and SWR/J, was examined to compare recombination frequency between selected markers on mouse Chr 1. Significant differences in crossover frequency were demonstrated between the interspecific backcross and the inbred laboratory cross for the entire interval studied. Sex difference in meiotic crossover frequency was also significant in the laboratory mouse cross. Two linkage groups known to be conserved between segments of mouse Chr 1 and the long arm of human Chrs 1 and 2 where further defined and a new conserved linkage group was identified that includes markers of distal mouse Chr 1 and human Chr 1, bands q32 to q42.

Somatic cell genetics and gene families

The utility of somatic cell genetic analysis for the chromosomal localization of genes in mammals is well established. With the development of recombinant DNA probes and efficient blotting techniques that allow visualization of single-copy cellular genes, somatic cell genetics has been extended from the level of phenotypes expressed by whole cells to the level of the cellular genome itself. This extension has proved invaluable for the analysis of genes not readily expressed in somatic cell hybrids and for the study of multigene families, especially pseudogenes dispersed in different chromosomes throughout the genome.

Chromosomal distribution of ribosomal protein genes in the mouse

The chromosomal distributions of five families of mouse r-protein genes (S16, L18, L19, L30 and L32/33) were studied by Southern blot analysis of DNa from a panel of mouse-hamster hybrid cells containing various complements of mouse chromosomes. Our results indicated that members of a particular family are often located on more than one chromosome, that extensive clustering of many r-protein gene families on a few chromosomes is unlikely, and that there is no obligatory linkage of r-protein and rRNA genes.

A linkage map of mouse chromosome 12: localization of Igh and effects of sex and interference on recombination

Inheritance of restriction fragment length polymorphisms associated with four anonymous DNA markers (D12Nyu1, 2, 3 and 4), the Fos proto-oncogene, the Mtv-9 viral integration site, and the alpha 1-antitrypsin (Aat-1) and immunoglobulin heavy chain (Igh) gene families in the mouse has been followed in a backcross experiment. A Bayesian multilocus map-building strategy yielded the map: centromere-D12Nyu2-10 cM-D12Nyu1-2 cM-D12Nyu3-15 cM-Fos-1 cM-D12Nyu4-2 cM-Mtv-9-8 cM-Aat-1-17 cM-Igh-C. A map constructed from male meiotic data was substantially shorter than one constructed from female meiotic data. Significant interference was observed for the linkage group. Two groups of markers studied in recombinant inbred strains of mice could be interpolated into the map: Es-25, D12Nyu10, D12Nyu7 and Apob form a cluster proximal to D12Nyu2, and Ly-18, Ah, and D12Nyu5 form a cluster between D12Nyu2 and D12Nyu1. These data establish an unambiguously ordered linkage group including Igh and Aat-1 that spans most of chromosome 12.

Interleukin-1 alpha and beta genes: linkage on chromosome 2 in the mouse

Two interleukin-1 polypeptides, alpha and beta, are known, and cDNAs corresponding to each have been described. Genomic cloning and Southern blotting experiments suggest that in the mouse each is encoded by a gene present in one copy per haploid genome. Analysis of a panel of somatic cell hybrids carrying various mouse chromosomes on a constant Chinese hamster background indicates that both genes map to mouse chromosome 2. Further, analysis of the inheritance of DNA restriction fragment length polymorphisms associated with each gene in recombinant inbred strains of mice shows the two loci to be tightly linked to one another, and to lie approximately 4.7 centimorgans distal to B2m (beta-2 microglobulin). We have named the locus encoding IL-1 alpha Il-1 alpha and the locus encoding IL-1 beta Il-1b.

Role of surface modulating assemblies in growth control of normal and transformed fibroblasts

Cellular microtubules, microfilaments, and surface receptors have been postulated to form a surface modulating assembly that regulates surface receptor mobility and cell growth. To test this hypothesis, we examined three agents known to affect cell growth [colchicine, concanavalin A (Con A), and the src gene product of Rous sarcoma virus] for their effects on chick embryo fibroblasts. Individual cells from serum-starved normal fibroblast populations became committed to enter S phase at various times over a 12 hr period after exposure to serum. Colchicine and other microtubule-disrupting agents blocked entry into S phase at a point close to the commitment point for each cell. The lectin Con A also blocked entry into the S phase when present in doses sufficient to modulate surface receptor mobility. In contrast, succinyl-Con A, which does not induce surface modulation, had no effect. Both Con A and colchicine blocked the appearance of cytoplasmic factors capable of stimulating DNA replication in a cell-free system. To study endogenous effects on the surface modulating assembly, we infected fibroblasts with a Rous sarcoma virus (tsNY68) having a temperature-sensitive mutation in the transforming (src) gene. We have previously shown that microtubular and microfilamentous structures of the surface modulating assembly are direct or indirect targets of the src gene product with consequent reduction in the capacity of Con A to induce surface modulation. TsNY68-infected fibroblasts shifted to the non-permissive temperature acquired normal microtubular morphology more rapidly (2 hr) than cells grown at the permissive temperature in the presence of protein synthesis inhibitors (7.5 hr). This suggests that the src gene product acts directly on the surface modulating assembly rather than via the nucleus or at the level of protein synthesis. Furthermore, "transformation" of the surface modulating assembly was partly blocked by treatment of the infected cells with Con A but not succinyl-Con A. Both Con A and colchicine inhibited entry into the S phase following a shift from nonpermissive to permissive growth conditions. All of these observations are in accord with the hypothesis that the surface modulating assembly acts as a signal regulator in growth control.