Partial Peptide Mapping Research Articles

Identification, characterization, and sequence analysis of a cDNA encoding a phosphoprotein of human herpesvirus 6

Human herpesvirus 6 (HHV-6)-specific monoclonal antibody (Mab) 9A5D12 reacted with the nucleus of HHV-6 strain GS-infected cells and immunoprecipitated a phosphorylated polypeptide with an approximate size of 41 kDa, designated HHV-6 P41. A 110-kDa polypeptide was also immunoprecipitated by the MAb. These polypeptides were synthesized early in infection, and the synthesis was greatly reduced by phosphonoacetic acid. Polypeptides with identical sizes were recognized by the MAb from cells infected with an additional eight HHV-6 strains. A 2.1-kb cDNA insert was identified from an HHV-6(GS) cDNA library constructed in the lambda gt11 expression system by using MAb 9A5D12. This cDNA insert hybridized specifically with viral DNA from HHV-6 strains GS and Z-29 and with two predominant transcripts with approximate sizes of 2.5 and 1.2 kb from infected cells. The reactivity of the MAb with a fusion protein expressed in the prokaryotic vector suggested that the cDNA encodes a 62- to 66-kDa protein. Analysis of the nucleotide sequence of the cDNA insert revealed a 623-amino-acid-residue single open reading frame of 1,871 nucleotides, with an open 5' end. The predicted polypeptide is highly basic and contains a long stretch of highly hydrophobic residues localized to the carboxy terminus. The amino-terminal half of the predicted HHV-6 protein from the cDNA shows significant homology with the UL44 gene product of human cytomegalovirus, coding for the ICP36 family of early-late-class phosphoproteins. Two TATA boxes are located at nucleotide positions 668 and 722 of the cDNA. In vitro translation of RNA transcribed in vitro from the cDNA resulted in the synthesis of a 41-kDa polypeptide only. This polypeptide was readily immunoprecipitated by MAb 9A5D12, and its partial peptide map was identical to that of the 41-kDa polypeptide detected in infected cells. Together, these results indicate that the HHV-6 P41 is encoded within a gene coding for a larger protein.

A serum heterodimer from hagfish (Eptatretus stoutii) exhibits structural similarity and partial sequence identity with immunoglobulin.

We have isolated, characterized, and partially sequenced immunoglobulin from the most primitive extant nonvertebrate craniate, the hagfish, a jawless fish that may have diverged from the vertebrate lineage more than 500 million years ago. The 160-kDa protein, which is a minor serum component, is composed of two different heavy chains of 69 and 74 kDa and a light chain of 29 kDa and resembles known immunoglobulin on the basis of an equimolar ratio of heavy and light chains, N-linked glycosylation of heavy chains, presence of intra- and interchain disulfide bonds, and polydispersity of each peptide chain. High molecular mass (polymeric) as well as low molecular mass (monomeric) forms were isolated from serum. The hagfish immunoglobulin is unique in that each heterodimer is composed of two different heavy chains and two light chains. The partial peptide maps and amino acid compositions of the two heavy chains differ; the chains do not crossreact immunologically. Slight crossreactivity of the 74-kDa heavy chain with antisera against purified shark immunoglobulin and some conservation of amino acid sequences, including those surrounding a cysteine, suggest that the isolated protein is an immunoglobulin.

Heat shock protein induction in rat pancreatic islets by recombinant human interleukin 1?

Interleukin 1 beta, potentiated by tumour necrosis factor alpha, is cytotoxic to pancreatic Beta cells in vitro. We have hypothesized that interleukin 1 beta induces oxygen free radicals in Beta cells. Since cytotoxicity induced by free radicals and by heat may activate the same cellular repair mechanism (the heat shock response), the aim of this study was to investigate the pattern of protein synthesis in isolated islets after exposure to interleukin 1 beta (150 pg/ml, 24 h), tumour necrosis factor alpha (50 ng/ml, 24 h) heat shock (43 degrees C, 30 min) and H2O2 (0.1 mmol/l, 20 min). By polyacrylamide gel electrophoresis, autoradiography, Western-blot analysis and partial peptide mapping of 35S-methionine labelled islets, interleukin 1 beta was found to induce a 73 kilodalton protein belonging to the heat shock protein family heat shock protein 70, a heat shock protein 90, and haem oxygenase. A minor induction of heat shock protein 73 and haem oxygenase was seen after H2O2. Interleukin 1 beta did not induce heat shock proteins in rat thyroid cells, rat mesangial cells or in human monocytes. Tumour necrosis factor alpha did not induce selective protein synthesis. Pre-exposure of islets to heat, tumour necrosis factor alpha, or H2O2 did not prevent the impairment of glucose-stimulated insulin release seen after 24 h of interleukin 1 beta exposure. The data are compatible with free radical induction by interleukin 1 beta. However, the heat shock response is not specific for oxidative injury, and previous studies have shown discrepant effects as to a protective effect of free radical scavengers against interleukin 1 beta-mediated beta-cytotoxicity.(ABSTRACT TRUNCATED AT 250 WORDS)

Agonists and Antagonists Recognize Different but Overlapping Populations of A1 Adenosine Receptors: Modulation of Receptor Number by MgCl2, Solubilization, and Guanine Nucleotides

A1 selective agonist and antagonist radioligands bind to the same A1 adenosine receptor binding subunit, as documented by photoaffinity labelling and partial peptide maps. In this study we document that although these radioligands recognize the same A1 adenosine receptor (A1AR), they recognize different numbers of A1ARs in bovine brain membranes, with agonist number being greater than antagonist number. Neither addition of guanine nucleotides nor removal of Mg2+ ions enhanced antagonist binding in membranes. On solubilization, agonists still recognized a greater number of A1ARs but addition of guanine nucleotides or removal of Mg2+ substantially increased the number of receptors detected with antagonist radioligands. The effects of Mg2+ and guanine nucleotides were not additive, suggesting that formation of a "low agonist-receptor-G protein state" by either modulating agent was sufficient to alter the receptor conformation such that it could be recognized by antagonist. These studies suggest that a proportion of the "precoupled A1AR-G protein complex" in membranes are in a conformation that cannot be recognized by antagonists and that membrane constraints are such that ions or guanine nucleotides cannot sufficiently modulate the conformation to allow it to recognize antagonists. On removal of membrane structure by solubilization, these constraints are removed.

DNA polymerases alpha, delta, and epsilon: three distinct enzymes from HeLa cells.

DNA polymerases alpha, delta, and epsilon have been purified and characterized from the same HeLa cell extract in order to determine their relationship by comparing them from the same cell type. The catalytic properties and the primary structures of the large subunits of the DNA polymerases as compared by partial peptide mapping with N-chlorosuccinimide are different. Likewise, the small subunit of DNA polymerase epsilon appears to be distinct from the large subunit of the same polymerase and from the smaller subunits of DNA polymerase alpha. HeLa DNA polymerase delta is processive only when HeLa proliferating cell nuclear antigen is present, whereas DNA polymerase epsilon is quite processive in its absence. Inhibitor and activator spectra of DNA polymerases alpha, delta, and epsilon also distinguish the three enzymes. These results and immunologic comparisons published elsewhere support the premise that HeLa DNA polymerases alpha, delta, and epsilon are distinct enzymes that have common properties with yeast DNA polymerases I, III, and II, respectively.

Identification and Characterization of the Small Nuclear Ribonucleoprotein Particle D' Core Protein

The addition of urea to sodium dodecyl sulfate (SDS)-polyacrylamide gels has allowed the identification and characterization of the small nuclear ribonucleoprotein particle (snRNP) D' protein and has also improved resolution of the E, F, and G snRNP core proteins. In standard SDS-polyacrylamide gels, the D' and D snRNP core proteins comigrate at approximately 16 kilodaltons. The addition of urea to the separating gel caused the D' protein to shift to a slower electrophoretic mobility that is distinct from that of the D protein. The shift to a slower electrophoretic mobility in the presence of urea suggests that the D' protein has extensive secondary structure that is not totally disrupted by SDS alone. Both N-terminal sequencing and partial peptide maps indicate that the D and D' proteins are distinct gene products, and the sequence data have identified the faster moving of the two proteins as the previously cloned D protein (L. A. Rokeach, J. A. Haselby, and S. O. Hoch, Proc. Natl. Acad. Sci. USA 85:4832-4836, 1988). In the cytoplasm, the D protein is found primarily in the small-nuclear-RNA-free 6S protein complexes, while the D' protein is found primarily in the 20S protein complexes. Like the D protein, the D' protein is an autoantigen in patients with systemic lupus erythematosus and is recognized by some of the Sm class of autoimmune antisera.

Identification and characterization of the small nuclear ribonucleoprotein particle D' core protein.

The addition of urea to sodium dodecyl sulfate (SDS)-polyacrylamide gels has allowed the identification and characterization of the small nuclear ribonucleoprotein particle (snRNP) D' protein and has also improved resolution of the E, F, and G snRNP core proteins. In standard SDS-polyacrylamide gels, the D' and D snRNP core proteins comigrate at approximately 16 kilodaltons. The addition of urea to the separating gel caused the D' protein to shift to a slower electrophoretic mobility that is distinct from that of the D protein. The shift to a slower electrophoretic mobility in the presence of urea suggests that the D' protein has extensive secondary structure that is not totally disrupted by SDS alone. Both N-terminal sequencing and partial peptide maps indicate that the D and D' proteins are distinct gene products, and the sequence data have identified the faster moving of the two proteins as the previously cloned D protein (L. A. Rokeach, J. A. Haselby, and S. O. Hoch, Proc. Natl. Acad. Sci. USA 85:4832-4836, 1988). In the cytoplasm, the D protein is found primarily in the small-nuclear-RNA-free 6S protein complexes, while the D' protein is found primarily in the 20S protein complexes. Like the D protein, the D' protein is an autoantigen in patients with systemic lupus erythematosus and is recognized by some of the Sm class of autoimmune antisera.

Characterization and nucleosomal core localization of Achlya histones involved in stress-induced chromatin condensation

To better understand the basis for heat shock-induced chromatin condensation in Achlya, a further characterization of the histones of this organism was carried out. The nucleosomal location (i.e., core vs linker), partial peptide map, and electrophoretic behavior of each Achlya histone was determined and compared to the well-characterized histones of rabbit kidney. The results of this and previous studies suggest that in Achlya, no nucleosome linker-associated histone analogous to histone H1 of higher eucaryotes is observed and that the Achlya histone designated α is a novel nucleosomal core histone. These observations may reflect the existence of a mechanism of stress-induced chromatin condensation which does not involve histone H1.

Erythrophagocytosis induces heat shock protein synthesis by human monocytes-macrophages.

Exposure of cells to elevated temperatures and other environmental stresses results in the expression of specific genes encoding the so-called heat shock proteins (HSPs). Since exogenous H2O2 induces in human monocytes the synthesis of HSPs, and previous induction of HSPs protects these cells from oxidative injury, we investigated whether HSP synthesis was also induced during generation of reactive oxygen species by the phagocyte itself during phagocytosis. As a model system, we analyzed the effects of erythrophagocytosis on protein synthesis by the human premonocytic line U937, in which phagocytosis is induced during differentiation with 1,25-dihydroxyvitamin D3. Exposure to whole erythrocytes, but not to erythrocyte ghosts, induced in the phagocytic cells only the synthesis of the 70- and 83- to 90-kDa HSPs and a 32-kDa oxidation-related stress protein identical by partial peptide mapping to heme oxygenase. The radioprotective aminothiol N-(2'-mercaptoethyl)-1,3-propanediamine (WR-1065), which can substitute for glutathione as hydrogen donor, prevented this induction. These results suggest that oxygen free radicals generated in the presence of hemoglobin-derived iron and consecutive glutathione depletion are involved in induction of stress protein synthesis during erythrophagocytosis. HSPs synthesized during phagocytosis may play a role in the phagocyte's defense mechanisms and in protective immunity.

Structural requirements of iron-responsive elements for binding of the protein involved in both transferrin receptor and ferritin mRNA post-transcriptional regulation

The synthesis of both transferrin receptor (TfR) and ferritin is regulated post-transcriptionally by iron. This is mediated by iron responsive elements (IREs) in the 5'- and 3'-untranslated regions, respectively, of TfR and ferritin mRNAs. Although these IREs have different sequences, they both form a characteristic stem-loop. We used competition assays and partial peptide mapping of UV-crosslinked ferritin and TfR IRE-protein complexes to show that the cytosolic protein binding to the ferritin 5'-IRE, the iron-responsive element binding protein (IRE-BP), also binds to TfR 3'-IREs. To identify the structural requirements necessary for RNA-protein binding, ferritin IRE RNAs were synthesized which contained altered secondary structures and base substitutions. Affinities of these RNAs for IRE-BP were assayed in RNA-protein binding gels. Substitutions disrupting base-pairing of the stem prevented IRE-BP binding. Substitutions which restored base-pairing also restored IRE-BP binding. We conclude that the IRE-BP binds to both ferritin and TfR IREs and recognizes a particular IRE conformation.

Identification and distribution of two forms of the interleukin 1 receptor

Using affinity crosslinking techniques, we have biochemically characterized the interleukin-1 (IL1) receptor and investigated its distribution on a range of murine and human cell lines. We show that two forms of IL1 receptor can be identified on the basis of specific crosslinking with 125I-IL1 alpha and 125I-IL1 beta. The two receptor forms have an approximate molecular mass of approximately 80 and approximately 60 kDa, and were found on both murine and human cells. Their relative distribution shows no clear cell lineage restriction and does not correlate with preferential binding of IL1 alpha or IL1 beta. Some cells, such as the T helper cell line D10.G4.1, express both forms of the receptor. Iodine 125-IL1 was crosslinked to the two receptor forms and a partial peptide map analysis of the two receptor/ligand complexes was performed. Comigration of the major partial peptide fragments suggests that the approximately 80 and approximately 60 kDa forms of the receptor may be differentially processed forms of the same protein. Treatment of the approximately 60 kDa IL1 receptor on Raji cells with N-glycanase reduced its molecular mass by 12 kDa, showing that this lower molecular mass form is a glycoprotein; glycosylation differences alone probably do not account for the difference in mass between the two forms.

The promastigote surface protease (gp63) of Leishmania is expressed but differentially processed and localized in the amastigote stage

The expression, processing and localization of the promastigote surface glycoprotein, gp63, in the amastigote form of Leishmania mexicana was examined. Metabolically labeled protein was immunoprecipitated from promastigotes and amastigotes. The isolated proteins were subjected to deglycosylation and partial peptide mapping. The cleavage products generated migrated similarly in sodium dodecyl sulfate-polyacrylamide gel electrophoresis, indicating that the proteins were closely related. The majority of gp63 in amastigotes was inaccessible to surface-labeling procedures, and lacked the phosphatidylinositol membrane anchor. Immunolocalization of this subpopulation of gp63 revealed it to be present within the parasite's flagellar pocket. Despite the relative paucity of ‘membrane-form’ gp63, isolation and analysis of surface proteins from lesion amastigotes indicated that gp63 was the most abundant protein on the amastigote surface.

Demonstration of Distinct Agonist and Antagonist Conformations of the A1 Adenosine Receptor

A1 adenosine receptor-binding subunits can be visualized using high affinity antagonist and agonist photoaffinity radioligands. In the present study, we examined whether agonists and antagonists bind to the same receptor-binding subunit and if agonists and antagonists induce different conformational states of the receptor in intact membranes. It was demonstrated that several agonist and antagonist photoaffinity receptor-binding subunit. When the agonist and antagonist photoaffinity labeled peptides were denatured and subjected to partial peptide map analysis using a two-dimensional gel electrophoresis system similar peptide fragments were generated from each specifically labeled protein. This suggests that both classes of ligand label and incorporate into the same binding subunit. Proteolytic digestions of agonist- and antagonist-occupied receptors in native intact membranes revealed distinct and different peptide fragments depending on whether the ligand was an agonist or an antagonist. Manipulation of incubation conditions to perturb ligand-receptor interactions alter the pattern of peptide fragments generated with each specific protease. These data suggest that agonist and antagonist photoaffinity probes interact with an incorporate into the same binding subunit but that agonist binding is associated with a unique and detectable receptor conformation.

Partial purification and peptide mapping of ubiquitin-phytochrome conjugates from oat

Phytochrome is rapidly degraded in vivo following photoconversion from the relatively stable red light absorbing form Pr to the far red light absorbing form Pfr. In etiolated seedlings from several species, this photoconversion also induces the accumulation of ubiquitin-phytochrome conjugates (Ub-P), suggesting that Pfr is degraded via a ubiquitin-dependent proteolytic pathway. To understand why Pfr is preferentially conjugated with ubiquitin, Ub-P were partially purified and characterized with the ultimate goal of mapping ubiquitin attachment sites. Ub-P were partially purified by poly(ethylene imine) and ammonium sulfate precipitations followed by hydroxyapatite chromatography and separated from unmodified phytochrome by size-exclusion chromatography. Up-P had an apparent native molecular size of approximately 600 kDa, substantially larger than that of the unmodified Pfr dimer (365 kDa). Ub-P retained the property of spectral photoreversibility. The initial digestion patterns of Ub-P were similar to unmodified phytochrome. In an attempt to identify ubiquitin attachment site(s), Ub-P were probed with a library of anti-oat phytochrome monoclonal antibodies. Of the 16 different anti-phytochrome monoclonal antibodies tested, 3 (O76C, O19F, and O311B) poorly recognized all size classes of Ub-P, indicating that the corresponding epitopes were masked either directly or indirectly as a result of ubiquitin ligation. These epitopes are located between residuesmore » 90 and 180 (O76C), 558 and 668 (O19F), and 747 and 830 (O311B) of oat phytochrome. Because the regions recognized by O19F and O311B are near the C-terminal domain containing at least one ubiquitin attachment site, and near amino acid residues that become more accessible when the chromoprotein is in the Pfr form, these regions may be important in the Pfr-dependent ubiquitination of phytochrome.« less

Biochemical and immunological characterization of the flagellar-associated regulatory subunit of a type II cyclic adenosine 5′-monophosphate-dependent protein kinase

We have shown previously that the regulatory subunit (RII) of a type II cyclic AMP (cAMP)-dependent protein kinase is tightly associated with mammalian sperm flagella (J. A. Horowitz et al. (1984) J. Biol. Chem. 259, 832–838; J. A. Horowitz et al. (1988) J. Biol. Chem. 263, 2098–2104). In the present study the flagellar RII was compared to other well-characterized RIIs using biochemical and immunological methods. Flagellar polypeptides were screened by immunoblot analysis with monoclonal antibodies directed against the RII α and RII β isoforms. An RII β monoclonal antibody failed to cross-react with any flagellar polypeptide. In contrast, mAB 622, an RII α RII β monoclonal antibody, cross-reacted with a 57,000 Da polypeptide. However, another RII α RII β monoclonal antibody interacted weakly with the flagellar RII, suggesting that the epitope for this antibody is modified in flagellar RII. Partial peptide mapping of 8-azido-[ 32P]cAMP-labeled RIIs revealed that although heart and testis generated similar fragmentation patterns, there were differences in the maps from flagellar RII. Two-dimensional sodium dodecyl sulfate-gel electrophoresis of 8-azido-[ 32P]cAMP-labeled RII from rat flagella and bovine heart showed that the former possessed a considerably more acidic isoelectric point. Partial proteolysis of the flagellar RII by either endogenous or exogenous proteases resulted in the cleavage of RII to a 40,000 M r fragment. Complete release of this fragment from the flagellum was achieved if proteolysis was performed in the presence of thiol reducing agents. In their absence, approximately 50% of the fragment remained bound to the flagellum. The soluble proteolytic fragment was shown to be monomeric by native high-resolution gel-permeation chromatography and contained a functional cAMP-binding site(s).

Growth hormone-releasing hormone stimulates and somatostatin inhibits the release of a novel protein by cultured rat pituitary cells.

We have reported that the secretion of at least 17 distinct peptides [including rat (rGH)] GH by cultured rat pituitary cells was stimulated by GH-releasing hormone and inhibited by somatostatin, when analyzed by two-dimensional polyacrylamide gel electrophoresis. Three of these peptides (no. 23, 24, and 25) were not rGH immunoreactive. In order to determine whether these three peptides are fragments, degradation products or posttranscriptionally modified forms of rGH, rGH and peptide no. 23 were characterized structurally. From partial peptide maps of rGH and peptide no. 23 by V8 protease or chymotrypsin, it appeared that these peptides were not related to each other. By N-terminal microsequencing of two-dimensional polyacrylamide gel electrophoresis purified peptide, we have obtained the sequence of 24 N-terminal amino acid residues of peptide no. 23. This sequence has no significant homology with rGH or any other reported protein sequence. Antiserum was generated against a synthetic oligopeptide corresponding to amino acid residues 3-24 of peptide no. 23. The antiserum cross-reacted with peptides no. 23, 24, and 25 upon Western blot analysis. These results indicate that peptide no. 23 has a novel structure unrelated to other pituitary hormones. Since its secretion is influenced by GH-releasing hormone and somatostatin, peptide no. 23 may represent a previously unrecognized structurally unique growth factor.

Disulfide linkage of the zeta and eta chains of the T cell receptor. Possible identification of two structural classes of receptors.

The T cell antigen receptor (TCR) is a multisubunit membrane complex. It consists of two disulfide-linked polymorphic chains (either alpha-beta or gamma-delta heterodimers) which are noncovalently linked to five invariant chains. The CD3-gamma and CD3-delta chains bear N-linked carbohydrates and the CD3-epsilon and zeta chains are nongly-cosylated. Further analysis of the zeta chain in murine T cells demonstrates that it can exist as either a homodimer or disulfide linked to an additional protein with an apparent Mr of 22,000. The partial peptide map of this 22-kDa protein is different than zeta and all of the CD3 components. Like zeta, it has no apparent N-linked carbohydrate chains. We have chosen to refer to this subunit as the eta chain of the TCR. Ninety percent of zeta in cloned and nonclonal populations of T cells exist as a homodimer, and the remainder is found linked to the eta chain. The tight regulation of the zeta-zeta to zeta-eta ratio suggests an important functional role for these structural components of the TCR.

High-yield purification of a pp60 c- src related protein-tyrosine kinase from human platelets

A protein-tyrosine kinase (PTK, EC 2.7.1.112) from human platelets was purified with high yield. Purification of the enzyme involved sequential chromatography on casein-agarose, tyrosine-agarose, heparin-Sepharose and hydroxylapatite. The procedure resulted in substantially enriched 54 52 kDa polypeptides on SDS-polyacrylamide gel electrophoresis and a yield of about 25% in PTK activity. About 250 μg of purified protein could be obtained from 1 g of cell protein. The purification factor varied between 1000 and 1500. Determination of the molecular mass of the purified PTK under nondenaturating conditions by molecular sieve chromatography revealed that the enzyme is a monomer of about 50 kDa. Among various protein substrates tested, casein was most prominently phosphorylated. All substrates were exclusively phosphorylated at tyrosine residues. Autophosphorylation at tyrosine residues of the 54 52 kDa proteins was observed in the presence of Mg 2+ or Mn 2+. At each purification step, the 54 52 kDa proteins were precipitated by sera from tumor-bearing rabbits immunoprecipitating pp60 src , but not by control sera. The amount of the immunoprecipitated purified 54 52 kDa phosphoproteins was directly proportional to the amount of antiserum used. Partial peptide mapping by V8 proteinase showed a 26 kDa tyrosine-phosphorylated fragment for the 54 and the 52 kDa proteins as well as for the pp60 c- src molecules of intact platelets. All these data indicated that purified PTK is closely related to pp60 c- src of human platelets. Using casein as a substrate for the purified enzyme, the K m for ATP was 4 μM and the V max for the reaction was 2.0 nmol/min per mg.

Cloning of the Human Keratin 18 Gene and Its Expression in Nonepithelial Mouse Cells

Human keratin 18 (K18) and the homologous mouse protein, Endo B, are intermediate filament subunits of the type I keratin class. Both are expressed in many simple epithelial cell types including trophoblasts, the first differentiated cell type to appear during mouse embryogenesis. The K18 gene was identified and cloned from among the 15 to 20 similar sequences identified within the human genome. The identity of the cloned gene was confirmed by comparing the sequence of the first two exons to the K18 cDNA sequence and transfecting the gene into various murine cell lines and verifying the encoded protein as K18 by immunoprecipitation and partial peptide mapping. The transfected K18 gene was expressed in mouse HR9 parietal endodermal cells and mouse fibroblasts even though the fibroblasts fail to express endogenous Endo B. S1 nuclease protection analysis indicated that mRNA synthesized from the transfected K18 gene is initiated at the same position as authentic K18 mRNA found in both BeWo trophoblastoma cells and HeLa cells. Pulse-chase experiments indicated that the human K18 protein is stable in murine parietal endodermal cells (HR9) which express EndoA, a complementary mouse type II keratin. Surprisingly, however, K18 was degraded when synthesized in cells which lack a type II keratin. This turnover of K18 may be an important mechanism by which epithelial cells maintain equal molar amounts of both type I and II keratins. In addition, the levels of the endogenous type I Endo B in parietal endodermal cells were compensatingly down regulated in the presence of the K18 protein, while the levels of the endogenous type II Endo A were not affected in any of the transfected cell lines.

Cloning of the human keratin 18 gene and its expression in nonepithelial mouse cells.

Human keratin 18 (K18) and the homologous mouse protein, Endo B, are intermediate filament subunits of the type I keratin class. Both are expressed in many simple epithelial cell types including trophoblasts, the first differentiated cell type to appear during mouse embryogenesis. The K18 gene was identified and cloned from among the 15 to 20 similar sequences identified within the human genome. The identity of the cloned gene was confirmed by comparing the sequence of the first two exons to the K18 cDNA sequence and transfecting the gene into various murine cell lines and verifying the encoded protein as K18 by immunoprecipitation and partial peptide mapping. The transfected K18 gene was expressed in mouse HR9 parietal endodermal cells and mouse fibroblasts even though the fibroblasts fail to express endogenous Endo B. S1 nuclease protection analysis indicated that mRNA synthesized from the transfected K18 gene is initiated at the same position as authentic K18 mRNA found in both BeWo trophoblastoma cells and HeLa cells. Pulse-chase experiments indicated that the human K18 protein is stable in murine parietal endodermal cells (HR9) which express EndoA, a complementary mouse type II keratin. Surprisingly, however, K18 was degraded when synthesized in cells which lack a type II keratin. This turnover of K18 may be an important mechanism by which epithelial cells maintain equal molar amounts of both type I and II keratins. In addition, the levels of the endogenous type I Endo B in parietal endodermal cells were compensatingly down regulated in the presence of the K18 protein, while the levels of the endogenous type II Endo A were not affected in any of the transfected cell lines.