N-terminal Hydrophobic Domain Research Articles

Structural changes in a hydrophobic domain of the prion protein induced by hydration and by Ala→Val and Pro→Leu substitutions

X-ray diffraction was used to study the structure of assemblies formed by synthetic peptide fragments of the prion protein (PrP) that include the hydrophobic domain implicated in the Gerstmann-Sträussler-Scheinker (GSS) mutation (P102L). The effects of hydration on polypeptide assembly and of Ala→Val substitutions in the hydrophobic domain were characterized. Synthetic peptides included: (i) Syrian hamster (SHa) hydrophobic core, SHa106-122 (KTNMKHMAGAAAAGAVV); (ii) SHa104-122(3A-V), with A→V mutations at 113, 115 and 118 (KPKTNMKHMVGVAAVGAVV); (iii) mouse (Mo) wild-type sequence of the N-terminal hydrophobic domain, Mo89-143WT; and (iv) the same mouse sequence with leucine substitution for proline at residue number 101, Mo89-143(P101L). Samples of SHa106-122 that formed assemblies while drying under ambient conditions showed X-ray patterns indicative of 33 Å thick slab-like structures having extensive H-bonding and intersheet stacking. By contrast, lyophilized peptide that was equilibrated against 100 % relative humidity showed assemblies with only a few layers of β-sheets. The Ala→Val substitutions in SHa104-122 and Mo89-143(P101L) resulted in the formation of 40 Å wide, cross-β fibrils. Observation of similar size β-sheet fibrils formed by peptides SHa104-122(3A-V) and the longer Mo89-143(P101L) supports the notion that the hydrophobic sequence forms a template or core that promotes the β-folding of the longer peptide. The substitution of amino acids in the mutants, e.g. 3A→V and P101L, enhances the folding of the peptide into compact structural units, significantly enhancing the formation of the extensive β-sheet fibrils.

Membrane Topology and Cell Surface Targeting of Microsomal Epoxide Hydrolase: EVIDENCE FOR MULTIPLE TOPOLOGICAL ORIENTATIONS

Microsomal epoxide hydrolase (mEH) is a bifunctional membrane protein that plays a central role in the metabolism of xenobiotics and in the hepatocyte uptake of bile acids. Numerous studies have established that this protein is expressed both in the endoplasmic reticulum and at the sinusoidal plasma membrane. Preliminary evidence has suggested that mEH is expressed in the endoplasmic reticulum (ER) membrane with two distinct topological orientations. To further characterize the membrane topology and targeting of this protein, an N-glycosylation site was engineered into mEH to serve as a topological probe for the elucidation of the cellular location of mEH domains. The cDNAs for mEH and this mEH derivative (mEHg) were then expressed in vitro and in COS-7 cells. Analysis of total expressed protein in these systems indicated that mEHg was largely unglycosylated, suggesting that expression in the ER was primarily of a type I orientation (Ccyt/Nexo). However, analysis, by biotin/avidin labeling procedures, of mEHg expressed at the surface of transfected COS-7 cells, showed it to be fully glycosylated, indicating that the topological form targeted to this site originally had a type II orientation (Cexo/Ncyt) in the ER. The surface expression of mEH was also confirmed by confocal fluorescence scanning microscopy. The sensitivity of mEH topology to the charge at the N-terminal domain was demonstrated by altering the net charge over a range of 0 to +3. The introduction of one positive charge led to a significant inversion in mEH topology based on glycosylation site analysis. A truncated form of mEH lacking the N-terminal hydrophobic transmembrane domain was also detected on the extracellular surface of transfected COS-7 cells, demonstrating the existence of at least one additional transmembrane segment. These results suggest that mEH may be integrated into the membrane with multiple transmembrane domains and is inserted into the ER membrane with two topological orientations, one of which is targeted to the plasma membrane where it mediates bile acid transport.

Peflin, a Novel Member of the Five-EF-Hand-Protein Family, Is Similar to the Apoptosis-Linked Gene 2 (ALG-2) Protein but Possesses Nonapeptide Repeats in the N-Terminal Hydrophobic Region

The calpain small subunits of sorcin, grancalcin, and ALG-2 constitute a family of the Ca2+-binding proteins with five EF-hand-like motifs (penta-EF-hand domain or PEF domain) in their C-terminal regions and hydrophobic domains with variable lengths in their N-terminal regions. Searching the human DNA data base of expressed sequence tags (EST) revealed novel partial sequences similar to, but distinct from, the sequences of the previously known PEF proteins. We isolated a cDNA clone of near full length by 5′- and 3′-RACE (rapid amplification of cDNA end) methods and compared the predicted amino acid sequence (284 residues) of the novel EF-hand protein, named peflin, with those of known PEF proteins. The PEF domain of peflin is most similar to ALG-2 (40.9% identity) among the family, particularly in EF-1 (46.2%) and EF-3 (57.1%) regions. Peflin has a longer N-terminal hydrophobic domain than any other member of the family, and it contains nine nonapeptide (A/PPGGPYGGP) repeats. Western blot analysis demonstrated that peflin (30 kDa) was expressed in various nonadherent and adherent cultured human cell lines, including Jurkat, HL60, HeLa, and HT1080. Peflin may play basic roles in Ca2+ signaling irrespective of cell types.

The Transmembrane Domains of the Human Multidrug Resistance P-glycoprotein Are Sufficient to Mediate Drug Binding and Trafficking to the Cell Surface

The human multidrug resistance P-glycoprotein (P-gp) is organized in two tandem repeats with each repeat consisting of an N-terminal hydrophobic domain containing six potential transmembrane segments followed by a hydrophilic domain containing a nucleotide-binding fold. A series of deletion mutants together with an in vivo drug-binding assay were used to test whether the deletion mutants interacted with substrates or were transported to the cell surface. We found that a deletion mutant consisting of only the transmembrane domains (residues 1-379 plus 681-1025) retained the ability to interact with drug substrates. In the absence of drug substrates, the deletion mutant was sensitive to trypsin and endoglycosidase H. Expression in the presence of verapamil, vinblastine, capsaicin, or cyclosporin A, however, resulted in a mutant protein that was resistant to trypsin and endoglycosidase H. The mutant was then detected at the cell surface and was sensitive to digestion by endoglycosidase F. By contrast, the N-terminal transmembrane domain (residues 1-379) alone did not interact with drug substrates, since it was sensitive to only endoglycosidase H and was not detected at the cell surface. These results show that the nucleotide-binding domains are not required for interaction of P-gp with substrate or for trafficking of P-gp to the cell surface.

Expression of Human Kidney 11β-Hydroxysteroid Dehydrogenase (11-HSD2) in Bacteria

The kidney isozyme of 11β-hydroxysteroid dehydrogenase (11-HSD2) protects the mineralocorticoid receptor from spurious activation by glucocorticoids. To explore structure-function relationships, human 11-HSD2 cDNA was subcloned into the bacterial expression vector, pET25b.E. colitransformed with wild-type cDNA produced active enzyme that retained biochemical characteristics of the native protein. The addition of 6 histidine residues to the C-terminus of the wild-type enzyme (11-HSD2/His) increased activity 2-fold. Whereas wild-type activity was almost completely sedimented following 100,000gcentrifugation, 10-30% of total activity of 11-HSD2/His remained in the supernatant. The 11-HSD2 isozyme normally contains three N-terminal hydrophobic domains. Mutant 11-HSD2/His possessing a single hydrophobic domain retained partial activity, but elimination of all domains inactivated the enzyme. Thus, the N-terminal hydrophobic domains are essential for complete activity of 11-HSD2 but association with an intact cell membrane is not.

Interactions of ribosome nascent chain complexes of the chloroplast-encoded D1 thylakoid membrane protein with cpSRP54.

The mechanisms of targeting, insertion and assembly of the chloroplast-encoded thylakoid membrane proteins are unknown. In this study, we investigated these mechanisms for the chloroplast-encoded polytopic D1 thylakoid membrane protein, using a homologous translation system isolated from tobacco chloroplasts. Truncated forms of the psbA gene were translated and stable ribosome nascent chain complexes were purified. To probe the interactions with the soluble components of the targeting machinery, we used UV-activatable cross-linkers incorporated at specific positions in the nascent chains, as well as conventional sulfhydryl cross-linkers. With both cross-linking approaches, the D1 ribosome nascent chain was photocross-linked to cpSRP54. cpSRP54 was shown to interact only when the D1 nascent chain was still attached to the ribosome. The interaction was strongly dependent on the length of the nascent chain that emerged from the ribosome, as well as the cross-link position. No interactions with soluble SecA or cpSRP43 were found. These results imply a role for cpSRP54 in D1 biogenesis.

Hmg-coA reductase gene family in cotton (Gossypium hirsutum L.): unique structural features and differential expression of hmg2 potentially associated with synthesis of specific isoprenoids in developing embryos.

As a first step towards understanding the biosynthesis of isoprenoids that accumulate in specialized pigment glands of cotton at the molecular level, two full-length genes (hmg1 and hmg2) were characterized encoding hmg-coA reductase (HMGR; EC 1.1.1.34), the enzyme that catalyzes the formation of a key isoprenoid precursor. Cotton hmgr genes exhibited features typical of other plant genes, however, hmg2 encodes the largest of all plant HMGR enzymes described to date. HMG2 contains several novel features that may represent functional specialization of this particular HMGR isoform. Such features include a unique 42 amino acid sequence located in the region separating the N-terminal domain and C-terminal catalytic domain, as well as an N-terminal hydrophobic domain that is not found in HMG1 or other HMGR enzymes. DNA blot analysis revealed that hmg1 and hmg2 belong to small subfamilies that probably include homeologous loci in allotetraploid cotton (Gossypium hirsutum L.). Ribonuclease protection assays revealed that hmg1 and hmg2 are differentially expressed in a developmentally- and spatially-modulated manner during morphogenesis of specialized terpenoid-containing pigment glands in embryos. Induced expression of hmg2 coincided with a possible commitment to sesquiterpenoid biosynthesis in developing embryos, although other developmental processes also requiring HMGR cannot be excluded.

A chicken homolog of mammalian interleukin-1 beta: cDNA cloning and purification of active recombinant protein.

Upon induction with lipopolysaccharide (LPS) the chicken macrophage cell line HD-11 secretes an activity that stimulates the synthesis of a CXC chemokine in the chicken fibroblast cell line CEC-32. We used a cDNA expression cloning strategy in COS cells to characterize this activity. The isolated cDNA clone codes for a polypeptide of 267 amino acids which lacks a hydrophobic N-terminal domain that could serve as secretory signal. Sequence homology and structural features indicate that this protein is the chicken homolog of mammalian interleukin-1 beta (ChIL-1 beta). Northern blot analysis showed that ChIL-1 beta RNA is quickly induced in blood monocyte-derived macrophages reaching maximal levels within one hour after onset of LPS treatment. To test for biological activity of putative mature ChIL-1 beta, a cDNA fragment comprising amino acids 106 to 267 of the open reading frame was expressed in Escherichia coli so that the resulting polypeptide carried a histidine tag at its N-terminus for easy purification by nickel chelate affinity chromatography. Purified His-ChIL-1 beta potently induced CXC chemokine RNA synthesis in CEC-32 cells. When injected intravenously into adult chickens, it quickly induced a transient increase in serum corticosterone levels.

AlphaKAP is an anchoring protein for a novel CaM kinase II isoform in skeletal muscle.

Ca2+/calmodulin-dependent protein kinase II (CaM kinase II) is present in a membrane-bound form that phosphorylates synapsin I on neuronal synaptic vesicles and the ryanodine receptor at skeletal muscle sarcoplasmic reticulum (SR), but it is unclear how this soluble enzyme is targeted to membranes. We demonstrate that alphaKAP, a non-kinase protein encoded by a gene within the gene of alpha-CaM kinase II, can target the CaM kinase II holoenzyme to the SR membrane. Our results indicate that alphaKAP (i) is anchored to the membrane via its N-terminal hydrophobic domain, (ii) can co-assemble with catalytically competent CaM kinase II isoforms and target them to the membrane regardless of their state of activation, and (iii) is co-localized and associated with rat skeletal muscle CaM kinase II in vivo. alphaKAP is therefore the first demonstrated anchoring protein for CaM kinase II. CaM kinase II assembled with alphaKAP retains normal enzymatic activity and the ability to become Ca2+-independent following autophosphorylation. A new variant of beta-CaM kinase II, termed betaM-CaM kinase II, is one of the predominant CaM kinase II isoforms associated with alphaKAP in skeletal muscle SR.

Domain Structure, Intracellular Trafficking, and β2-Microglobulin Binding of a Major Histocompatibility Complex Class I Homolog Encoded by Molluscum Contagiosum Virus

The MC80R gene of molluscum contagiosum virus (MCV) type 1 encodes a major histocompatibility complex (MHC) class I homolog that lacks several amino-acid residues critical for peptide binding by MHC molecules, contains an unusually long N-terminal hydrophobic domain possibly derived by triplication of a signal peptide, and has a C-terminal transmembrane domain with two glutamate residues. All of these features were present in the orthologous gene of MCV type 2. The MC80R gene was expressed as two glycosylated polypeptides of Mr 47,000 and 42,000. Pulse-chase experiments indicated that the larger polypeptide was a precursor of the shorter one and that the entire N-terminal domain was slowly removed, consistent with its function as a long signal peptide. The protein was largely sequestered in the endoplasmic reticulum and Golgi membranes, remained endoglycosidase-H sensitive, and was not detected on the cell surface. In addition, a genetically modified form of the MC80R protein lacking the transmembrane and cytoplasmic domains was not secreted. The roles of the MC80R protein domains were investigated by constructing chimera between the viral protein and the MHC class I protein HLA-A2. Expression studies confirmed that the N- and C-terminal hydrophobic regions of the MC80R protein served as signal and transmembrane domains, respectively. The central portion of the MC80R protein, corresponding to the α1–α3 extracellular domains of HLA-A2, was largely responsible for sequestering the protein in the endoplasmic reticulum or Golgi compartments. The MC80R protein, as well as HLA-A2 chimera with the central region of MC80R, formed stable intracellular complexes with β2-microglobulin. Complex formation, however, was detected only by overexpression of the MC80R protein or β2-microglobulin.

The ethylene-receptor family from Arabidopsis: structure and function.

The gaseous hormone ethylene regulates many aspects of plant growth and development. Ethylene is perceived by a family of high-affinity receptors typified by the ETR1 protein from Arabidopsis. The ETR1 gene codes for a protein which contains a hydrophobic N-terminal domain that binds ethylene and a C-terminal domain that is related in sequence to histidine kinase-response regulator two-component signal transducers found in bacteria. A structural model for the ethylene-binding domain is presented in which a Cu(I) ion is coordinated within membrane-spanning alpha-helices of the hydrophobic domain. It is proposed that binding of ethylene to the transition metal would induce a conformational change in the sensor domain that would be propagated to the cytoplasmic transmitter domain of the protein. A total of four additional genes that are related in sequence to ETR1 have been identified in Arabidopsis. Specific missense mutations in any one of the five genes leads to ethylene insensitivity in planta. Models for signal transduction that can account for the genetic dominance of these mutations are discussed.

A conserved BURP domain defines a novel group of plant proteins with unusual primary structures.

We have identified a new class of plant proteins containing a common C-terminal region, which we have termed the BURP domain. These proteins are defined not only by the BURP domain, but also by the overall similarity in their modular construction. The BURP domain proteins consist of either three or four modules: (i) an N-terminal hydrophobic domain -- a presumptive transit peptide, joined to (ii) a short conserved segment or other short segment, (iii) an optional segment consisting of repeated units which is unique to each member, and (iv) the C-terminal BURP domain. These individual modules appear to be combined to form two main classes of BURP domain proteins. The BURP domain proteins, despite the similarities in their primary structural features, show no obvious similarities in the tissues or conditions under which they are expressed. The presence of the conserved BURP domain in diverse plant proteins suggests an important and fundamental functional role for this domain.

Human chondroitin 6-sulfotransferase: cloning, gene structure, and chromosomal localization

Chondroitin 6-sulfotransferase (C6ST) is the key enzyme in the biosynthesis of chondroitin 6-sulfate, a glycosaminoglycan implicated in chondrogenesis, neoplasia, atherosclerosis, and other processes. C6ST catalyzes the transfer of sulfate from 3′-phosphoadenosine 5′-phosphosulfate to carbon 6 of the N-acetylgalactosamine residues of chondroitin. Based on the previously published avian sequence, we searched the database of expressed sequence tags (dbEST) and obtained partial-length cDNAs that we completed by 5′-RACE using human chondrosarcoma and endothelial-cell RNA as template. Stable transfection of our full-length expression construct into CHO-K1 cells resulted in marked increases in C6ST and keratan sulfate sulfotransferase (KSST) enzymatic activities in cell homogenates. The predicted 411 amino acid sequence of human C6ST contains an N-terminal hydrophobic domain consistent with membrane insertion, four potential sites for N-linked glycosylation, several consensus sequences for protein phosphorylation, and one RGD sequence. The human and chick C6ST cDNA share 51% nucleotide identity, 40% amino acyl identity, and 75% amino acyl conservation. The human C6ST gene structure has been elucidated and exhibits an intron-less coding region, and the gene has been mapped to human chromosome 11 by radiation hybrid panel mapping.

Refinement of 3D structure of bovine lens αA-crystallin

In absence of 3D structures for α-crystallin subunits, αA and αB, we utilized a number of experimental and molecular modeling techniques to generate working 3D models of these polypeptides (Farnsworth et al., 1994. In Molecular Modeling: From Virtual Tools to Real Problems (Eds. Kumosinski, T.F. and Liebman, M.N.) ACS Symposium Series 576, Ch. 9:123–134, 1994, ACS Books, Washington DC). The refinement of the initial bovine αA model was achieved using a more accurate estimation of secondary structure by new/updated methods for analyzing the far UV-CD spectra and by neural network secondary structure predictions in combination with database searches. The spectroscopic study reveals that α-crystallin is not an all β-sheet protein but contains ∼17% α-helices, ∼33% β-structures and ∼50% turns and coils. The refinement of the αA structure results in an elongate, asymmetric amphipathic molecule. The hydrophobic N-terminal domain imparts the driving force for subunit aggregation while the more flexible, polar C-terminal domain imparts aggregate solubility. In our quaternary structure of the aggregate, the monomer is the minimal cooperative subunit. In bovine αA, the highly negatively charged C-terminal domain has three small positive areas which may participate in dimer or tetramer formation of independently expressed C-terminal domains. The electrostatic potential of positive areas is modulated and become more negative with phosphorylation and ATP binding. The refined bovine αA model was used to construct αA models for the human, chick and dogfish shark. A high degree of conservation of the three dimensional structure and the electrostatic potential was observed. Our proposed open micellar quaternary structure correlates well with experimental data accumulated over the past several decades. The structure is also predictive of the more recent data.

Structure of the puf operon of the obligately aerobic, bacteriochlorophyll alpha-containing bacterium Roseobacter denitrificans OCh114 and its expression in a Rhodobacter capsulatus puf puc deletion mutant.

Roseobacter denitrificans (Erythrobacter species strain OCh114) synthesizes bacteriochlorophyll a (BChl) and the photosynthetic apparatus only in the presence of oxygen and is unable to carry out primary photosynthetic reactions and to grow photosynthetically under anoxic conditions. The puf operon of R. denitrificans has the same five genes in the same order as in many photosynthetic bacteria, i.e., pufBALMC. PufC, the tetraheme subunit of the reaction center (RC), consists of 352 amino acids (Mr, 39,043); 20 and 34% of the total amino acids are identical to those of PufC of Chloroflexus aurantiacus and Rubrivivax gelatinosus, respectively. The N-terminal hydrophobic domain is probably responsible for anchoring the subunit in the membrane. Four heme-binding domains are homologous to those of PufC in several purple bacteria. Sequences similar to pufQ and pufX of Rhodobacter capsulatus were not detected on the chromosome of R. denitrificans. The puf operon of R. denitrificans was expressed in trans in Escherichia coli, and all gene products were synthesized. The Roseobacter puf operon was also expressed in R. capsulatus CK11, a puf puc double-deletion mutant. For the first time, an RC/light-harvesting complex I core complex was heterologously synthesized. The strongest expression of the R. denitrificans puf operon was observed under the control of the R. capsulatus puf promoter, in the presence of pufQ and pufX and in the absence of pufC. Charge recombination between the primary donor P+ and the primary ubiquinone Q(A)- was observed in the transconjugant, showing that the M and L subunits of the RC were correctly assembled. The transconjugants did not grow photosynthetically under anoxic conditions.

The FLG motif in the N-terminal region of glucoprotein 41 of human immunodeficiency virus type 1 adopts a type-I beta turn in aqueous solution and serves as the initiation site for helix formation.

NMR and CD studies were carried out on a peptide representing the hydrophobic N-terminal domain of envelope glycoprotein of human immunodeficiency virus type-1 in solutions of varying polarity. It was found that in aquaeous solution the amide proton of glycine in the FLG motif resonated at a considerably high field and its chemical shift, within the limit of experimental precision, had a temperature coefficient of zero in the range studied. The upfield shift of NH of the glycine could be largely attributed to the ring-current effect of phenylalanine in the FLG motif that participated in a type-1 beta turn with a short Cbeta(i)-NH(i+2) distance. The slower proton-deuterion exchange for the glycine amide proton relative to that of other glycines was consistent with a folded structure for the motif in aquaeous solution. Results of the molecular simulation showed that this proton was shielded from the solvent by non-polar side chains of the amino acid residues surrounding the turn stabilized by hydrophobic interactions, thus explaining the zero temperature coefficient of the proton chemical shift. The structural stabilizing effect of the hydrophobic interaction was supported by the behavior of the proton in less polar Me2SO solution, in which the anomaly in the chemical shift and its temperature coefficient was less prominent. Detailed secondary-structure analysis suggested that the beta turn of the FLG motif may act as an initiation core for helix formation, probably because the turn readily transforms into helical form.

Septal localization of the SpoIIIE chromosome partitioning protein in Bacillus subtilis

The 787 amino acid SpoIIIE protein of Bacillus subtilis is required for chromosome partitioning during sporulation. This process differs from vegetative chromosome partitioning in that it occurs after formation of the septum, apparently by transfer of the chromosome through the nascent septum in a manner reminiscent of plasmid conjugation. Here we show that SpoIIIE is associated with the cell membrane, with its soluble C-terminal domain located inside the cell. Immunofluorescence microscopy using affinity-purified anti-SpoIIIE antibodies shows that SpoIIIE is targeted near the centre of the asymmetric septum, in support of a direct role for SpoIIIE in transport of DNA through the septum. We also report on the isolation of a mutation affecting the N-terminal hydrophobic domain of SpoIIIE that interferes with targeting to the septum and blocks DNA transfer. This mutation also causes de-localization of the activity of the normally prespore-specific sigma factor, sigmaF, consistent with the notion that SpoIIIE can form a seal between the chromosomal DNA and the leading edge of the division septum.

A large open reading frame (orf1995) in the chloroplast DNA of Chlamydomonas reinhardtii encodes an essential protein.

An open reading frame potentially encoding a protein of 1995 amino acids (orf1995) has been found in the chloroplast genome of the green alga Chlamydomonas reinhardtii. Besides having a short hydrophobic N-terminal domain with five putative transmembrane helices, the predicted orf1995 product is highly basic. orf1995 might be a homologue of the ycf1 gene in land plants, whose function has not yet been determined. Mutants of C. reinhardtii transformed with a disruption of orf1995 remain heteroplasmic for the wild-type and disrupted alleles of this gene, indicating that the orf1995 product is essential for cell survival.

The Amino Terminal Lectin-Like Domain of Thrombomodulin Is Required for Constitutive Endocytosis

AbstractThrombomodulin (TM) is a multidomain protein that serves as a cofactor in a major natural anticoagulant system. To further characterize the structure-function of TM, we have transfected COS cells with different truncated forms of TM. In the first form, COS cells expressing TM that lacks the putative signal peptide (17 residues); the lectin-like, hydrophobic N-terminal domain (226 residues); and 12 residues of the first epidermal growth factor (EGF )-like repeat (COSdel.238 cells) were found to function normally with respect to TM transport to the cell surface and thrombin-dependent protein C activation. However, in contrast to wild-type TM, as visually studied by immunofluorescence and immunogold electron microscopy, the COSdel.238 cells did not constitutively internalize anti-TM–TM or thrombin-TM complexes. To identify the region responsible for mediating the endocytic process, deletant forms of TM lacking either the lectin-like region (residues 2-155) or the hydrophobic region of the N-terminal domain (residues 161-202) were expressed in COS cells (COSdel.2-155 and COSdel.161-202, respectively). Protein C cofactor activity was maintained in both cells. Although the COSdel.161-202 cells behaved similarly to wild-type TM-transfected cells, visual studies showed a lack of constitutive internalization of thrombin-TM or anti-TM–TM complexes in the COSdel.2-155 cells. We conclude that the lectin-like domain of human TM serves to regulate cell surface expression of TM via the endocytic route and therefore may also play a major physiologic role in controlling intracellular and extracellular accumulation of thrombin in a variety of biologic systems.

Use of an alternate splice donor site in the human GAP gene is responsible for synthesis of the p100 isoform

GAP (GTPase-activating protein), a negative regulator of the receptor tyrosine kinase signal transduction pathway, exists as two isoforms: a ubiquitous, p120 form and a primate placenta-specific p100 form lacking the N-terminal hydrophobic domain. The cDNA species encoding p120 and p100 GAP are identical except that p100 GAP cDNA contains a 65-bp insert not present in p120 cDNA. The purpose of this study was to locate the 65-bp insert in the genomic GAP sequence, thereby determining the mechanism by which alternate splicing produces the two mRNA species. It was found that the 65-bp insert is located just 3′ to the sequence encoding the hydrophobic domain, indicating that the p100 form of GAP results from utilization of an alternate splice donor site. In addition, the sequence encoding the hydrophobic domain was found to be contained within a single large exon. The intron separating this exon from the exon encoding the 5′-portion of the SH2-N domain was determined to be at least 40 kb in length. Finally, it was found that the sequence encoding the SH2-N domain contains an intron 1006 bp long, and the sequence of this intron has been deduced. It is anticipated that the data presented in this paper will provide the basis for elucidating RNA processing mechanisms responsible for preferential expression of p100 GAP in the human placenta.