Expression of mRNA for a short form of heparin-binding EGF-like growth factor

In this paper we report the cloning and characterization of cDNA encoding a novel, short form of heparin-binding EGF-like growth factor (SF HB-EGF), and show expression of specific mRNA in various tissues and cell types. Our data suggest that SF HB-EGF mRNA is a product of alternative splicing. Like normal HB-EGF, SF HB-EGF contains the signal peptide, the propeptide, the heparin-binding domain and the first two conservative disulfide loops of the EGF unit. Instead of the third disulfide loop, the spacer, the transmembrane and the cytoplasmic domains, SF HB-EGF has a nine amino acid tail.     

Two forms of precursors of a heparin-binding EGF-like growth factor-diphtheria toxin receptor

A new mRNA coding for the heparin-binding EGF-like growth factor (HB-EGF) was found in Vero cells. The corresponding cDNA had C-156 in place of T, which resulted in a loss of the NheI site and a substitution of Leu-33 with Pro in the HB-EGF precursor. The known and new forms of the precursor were accordingly termed L and P. A possible conformational change in the corresponding propeptide region were assumed to affect processing of soluble secreted HB-EGF. The L and P mRNAs are differently expressed in various cell lines and have the identical 5'-untranslated sequences. Possibly, they are transcribed from one promoter and then alternatively spliced. Stimulation of resting Vero cells with tetraphorbol ester (TPA) substantially increased production of the L form, decreased production of the P form, and did not affect expression of the total HB-EGF mRNA. This was associated with an increase in binding of the diphtheria toxin, suggesting that the L HB-EGF precursor acts as its receptor.     

A short form of the heparin-binding EGF-like growth factor with a changed EGF domain

In all secreted proteins related to the epidermal growth factor (EGF), EGF domains that occur in a mature factor are each encoded by two exons, and those that do not, by one exon. During splicing, additional exon 3a can be inserted between exons 3 and 4, which code for the EGF domain of the mature heparin-binding EGF-like growth factor (HB-EGF). The resulting mRNA codes for the short form of HB-EGF (SF HB-EGF), which retains the signal peptide, the propeptide, and the heparin-binding domain. However, its EGF domain lacks the C-terminal subdomain essential for the interaction with the EGF receptor (EGFR). Structural analysis suggested that SF HB-EGF is a secreted polypeptide that has high affinity for heparin, but weakly, if at all, interacts with EGFR. Data obtained in three different systems indicated that SF HB-EGF possesses a mitogenic activity but utilizes a signal transduction pathway other than that of HB-EGF.     

A Short Form of the Heparin-binding EGF-like Growth Factor with a Changed EGF Domain

In all secreted proteins related to the epidermal growth factor (EGF), EGF domains that occur in a mature factor are each encoded by two exons, and those that do not, by one exon. During splicing, additional exon 3a can be inserted between exons 3 and 4, which code for the EGF domain of the mature heparin-binding EGF-like growth factor (HB-EGF). The resulting mRNA codes for the short form of HB-EGF (SF HB-EGF), which retains the signal peptide, the propeptide, and the heparin-binding domain. However, its EGF domain lacks the C-terminal subdomain essential for the interaction with the EGF receptor (EGFR). Structural analysis suggested that SF HB-EGF is a secreted polypeptide that has high affinity for heparin but weakly, if at all, interacts with EGFR. Data obtained in three different systems indicated that SF HB-EGF possesses a mitogenic activity but utilizes a signal transduction pathway other than that of HB-EGF.     

Two Forms of the Heparin-binding EGF-like Growth Factor Precursor,a Receptor for Diphtheria Toxin

A new mRNA coding for the heparin-binding EGF-like growth factor (HB-EGF) was found in Vero cells. The corresponding cDNA had C-156 in place of T, which resulted in a loss of the NheI site and replacement of Leu-33 with Pro in the HB-EGF precursor. The known and new forms of the precursor were accordingly termed L and P. A conformational change in the corresponding propeptide region was assumed to affect the processing of soluble secreted HB-EGF. The L and P mRNAs are differently expressed in various cell lines, have identical 5"-untranslated sequences, and are probably transcribed from one promoter and then alternatively spliced. Stimulation of resting Vero cells with tetraphorbol ester (TPA) substantially increased the production of the L form, decreased the production of the P form, and did not affect the expression of total HB-EGF mRNA. This was associated with increased binding of the diphtheria toxin, suggesting that the L HB-EGF precursor acts as its receptor.     

The short form of heparin-binding EGF-like growth factor. Part II

The structure of monkey (Chlorocebus aethiops) heparin-binding EGF-like growth factor (HB-EGF) gene has been investigated in this work in comparison with the known structure of human gene. It was shown that HB-EGF short form (SF-HB-EGF) specific exon 3a is mapped between exons 3 and 4 at distance 700 b.p. from exon 4. In a number of human and simian cell lines the main part of SF-HB-EGF mRNA does not contain HB-EGF mRNA specific exons 4 and 5. In comparison with HB-EGF mRNA in SF-HB-EGF mRNA P-form, but not L-form of is predominant, and this mRNA encodes a polypeptide with changed propeptide structure. Labeled SF-HB-EGF competes with HB-EGF and EGF for binding sites at A431 cell surface, which may be due to interaction with specific receptor. All the data suggest a specific role of SF-HB-EGF in cellular signalization.     

Short form of α9 promotes α9β1 integrin-dependent cell adhesion by modulating the function of the full-length α9 subunit

The α9β1 integrin is a multifunctional receptor that interacts with a variety of ligands including vascular cell adhesion molecule 1, tenascin-C, and osteopontin. A 2.3-kb truncated form of α9 integrin subunit cDNA was identified by searching the Medline database. This splice variant, which we called the short form of α9 integrin (SFα9), encodes a 632-aa isoform lacking transmembrane and cytoplasmic domains, and its authentic expression was verified by PCR and Western blotting. SFα9 is expressed on the cell surface but cannot bind ligand in the absence of the full-length α9 subunit. Over-expression of SFα9 in cells expressing full-length α9 promotes α9-dependent cell adhesion. This promoting effect of SFα9 requires the authentic cytoplasmic domain of the co-expressed full-length α9 subunit. Thus, SFα9 is a novel functional modulator of α9β1 integrin by inside-out signaling.     

Nucleotide and deduced amino acid sequence of a human cDNA (NQO2) corresponding to a second member of the NAD(P)H:quinone oxidoreductase gene family. Extensive polymorphism at the NQO2 gene locus on chromosome 6.

NAD(P)H:quinone oxidoreductases (NQOs) are flavoproteins that catalyze the oxidation of NADH or NADPH by various quinones and oxidation-reduction dyes. We have previously described a complementary DNA that encodes a dioxin-inducible cytosolic form of human NAD(P)H:quinone oxidoreductase (NQO1). In the present report we describe the nucleotide sequence and deduced amino acid sequence for a cDNA clone that is likely to encode a second form of NAD(P)H:quinone oxidoreductase (NQO2) which was isolated by screening a human liver cDNA library by hybridization with a NQO1 cDNA probe. The NQO2 cDNA is 976 nucleotides long and encodes a protein of 231 amino acids (Mr = 25,956). The human NQO2 cDNA and protein are 54% and 49% similar to human liver cytosolic NQO1 cDNA and protein, respectively. COS1 cells transfected with NQO2 cDNA showed a 5-7-fold increase in NAD(P)H:quinone oxidoreductase activity as compared to nontransfected cells when either 2,6-dichlorophenolindophenol or menadione was used as substrate. Western blot analysis of the expressed NQO1 and NQO2 cDNA proteins showed cross-reactivity with rat NQO1 antiserum, indicating that NQO1 and NQO2 proteins are immunologically related. Northern blot analysis shows the presence of one NQO2 mRNA of 1.2 kb in control and 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) treated human hepatoblastoma Hep-G2 cells and that TCDD treatment does not lead to enhanced levels of NQO2 mRNA as it does for NQO1 mRNA. Southern blot analysis of human genomic DNA suggests the presence of a single gene approximately 14-17 kb in length. The NQO2 gene locus is highly polymorphic as indicated by several restriction fragment length polymorphisms detected with five different restriction enzymes.(ABSTRACT TRUNCATED AT 250 WORDS)     

Molecular cloning and characterization of Scotch pine defensin 2

A 252 bp cDNA fragment that corresponds to defensin 2 (PsDef2) was amplified from a cDNA library from seven-day plantlets of Pinus sylvestris L. This fragment encodes a protein that consists of 83 amino acid residues. The protein contains an N-terminal signal peptide, which includes 33 amino acid residues. A mature form of defensin 2 of Scotch pine contains a gamma-thionine domain and it is also characterized by specific conservative residues that are common to all plant defensins.     

Cloning and expression of sterol Delta 14-reductase from bovine liver.

Biosynthesis of cholesterol represents one of the fundamental cellular metabolic processes. Sterol Delta 14-reductase (Delta 14-SR) is a microsomal enzyme involved in the conversion of lanosterol to cholesterol in mammals. Amino-acid sequence analysis of a 38-kDa protein purified from bovine liver in our laboratory revealed > 90% similarity with a human sterol reductase, SR-1, encoded by the TM7SF2 gene, and with the C-terminal domain of human lamin B receptor. A cDNA encoding the 38-kDa protein, similar to human TM7SF2, was identified by analysis of a bovine expressed sequence tag (EST) database. The cDNA was synthesized by RT-PCR, cloned, and sequenced. The cDNA encodes a 418 amino-acid polypeptide with nine predicted transmembrane domains. The deduced amino-acid sequence exhibits high similarity with Delta 14-SR from yeasts, fungi, and plants (55-59%), suggesting that the bovine cDNA encodes Delta 14-SR. Northern blot analysis of bovine tissues showed high expression of mRNA in liver and brain. The polypeptide encoded by the cloned cDNA was expressed in COS-7 cells. Immunofluorescence analysis of transfected cells revealed a distribution of the protein throughout the ER. COS-7 cells expressing the protein exhibited Delta 14-SR activity about sevenfold higher than control cells. These results demonstrate that the cloned bovine cDNA encodes Delta 14-SR and provide evidence that the human TM7SF2 gene encodes Delta 14-SR.     

The cDNA cloning of the hamster homologue of the human L6 gene

The cDNA cloning and sequencing of a Syrian hamster gene that produces a surface antigen on cells transformed by the SV40 or BK virus are described. Sequence analysis showed that this gene is a member of the transmembrane 4 superfamily (TM4SF) and encodes the Syrian hamster counterpart of the human and the murine L6 antigens. The amino-acid sequences are highly conserved, being 71% identical among three species.     

Molecular cloning of three distinct forms of the Na+,K+-ATPase alpha-subunit from rat brain

Rat brain and kidney cDNA libraries were constructed and screened with a cDNA insert corresponding to the mRNA for the sheep kidney Na+,K+-ATPase catalytic subunit. The alpha-subunit cDNAs isolated from the kidney library were derived from a single class of messenger RNA, and the brain cDNAs were derived from three classes of messenger RNA. The most abundant brain cDNA, which spans 5.1 kilobases, encodes the alpha(+) form of the enzyme. The second most abundant brain cDNA, which spans 3.65 kilobases, is identical with that of the kidney form and therefore encodes the alpha isoform. The third class of cDNA, which spans 3.55 kilobases, was present at low abundance and encodes an isoform of the alpha-subunit, designated alpha III, which has not been identified previously. The complete nucleotide sequence and deduced amino acid sequence for each of the brain and kidney cDNAs have been determined. In addition, we have identified a lysine-rich sequence that may function as a movable, ion-selective gate during cation binding and occlusion and have also identified several amino acid sequence variations that appear to explain some of the well-known species and tissue differences in cardiac glycoside sensitivity.     

Induction of alpha-caveolin-1 (alphaCAV1) expression in bovine granulosa cells in response to an ovulatory dose of human chorionic gonadotropin

Caveolins are implicated in endocytosis, cholesterol trafficking and signal transduction. A cDNA fragment corresponding to caveolin-1 (CAV1) was identified in a mRNA profiling expression study in bovine granulosa cells (GC) following human chorionic gonadotropin (hCG)-induced ovulation. Thus, we have characterized CAV1 cDNA and studied its spatio-temporal expression pattern in bovine ovarian follicles. The full-length bovine alphaCAV1 cDNA was cloned and encodes a putative 22 kDa protein. Expression of alphaCAV1 was studied in bovine GC obtained from follicles at different developmental stages: small follicles (SF: 2-4 mm), dominant follicles (DF), ovulatory follicles (OF: 24 hr post-hCG), and corpus luteum (CL). Semiquantitative RT-PCR analysis showed a 6.5-fold increase in alphaCAV1 mRNA in GC of OF versus DF (P < 0.0001), whereas CAV2 mRNA was increased by only twofold (P < 0.0007). Temporal expression of alphaCAV1 mRNA from OF recovered at 0, 6, 12, 18, and 24 hr after hCG injection showed an 8.5-fold increase of alphaCAV1 mRNA after 24 hr compared to 0 hr (P < 0.0018) whereas no significant variation was detected for CAV2. Immunoblot demonstrated an initial increase in alphaCAV1 protein level 12 hr post-hCG, reaching a maximum at 24 hr. Immunohistochemical localization of CAV1 was observed in GC of OF isolated 18 and 24 hr after hCG injection, whereas no signal was detected in GC of DF and SF. The induction of alphaCAV1 in GC of OF suggests that alphaCAV1 likely contributes to control the increase in membrane signaling that occurs at the time of ovulation and luteinization.     

Soluble form of heparin-binding EGF-like growth factor contributes to retinoic acid-induced epidermal hyperplasia

Heparin-binding EGF-like growth factor (HB-EGF), a member of the EGF-family, is thought to be important for keratinocyte functions. HB-EGF is first synthesized as a membrane-anchored form, and its soluble form is released by ectodomain shedding. Here we investigate the role of HB-EGF in epidermal hyperplasia induced by all-trans retinoic acid (tRA) treatment. HB-EGF is normally expressed in epidermis of normal adult mice at very low levels, but topical tRA treatment results in epidermal hyperplasia, concomitant with the strong induction of HB-EGF expression in the suprabasal layer. tRA-induced epidermal hyperplasia was reduced both in the keratinocyte-specific HB-EGF null mice (K5-HB(del/del)) and knock-in mice expressing the uncleavable mutant form of HB-EGF (HB(uc/uc)), as compared with wild-type HB-EGF knock-in mice (HB(lox/lox)). Among ErbB tyrosine kinase receptors, EGF receptor (EGFR) and ErbB2 were selectively activated by tRA treatment in skin from wild-type mice, while the activation of these ErbB receptors was significantly reduced in the skin of HB-EGF null mice. These results indicate that expression of HB-EGF and generation of its soluble form, followed by activation of EGFR and ErbB2, are pivotal processes in tRA-induced epidermal hyperplasia.