Molecular Human Reproduction, Vol. 8, No. 1, 81-87,
January 2002
© 2002 European Society of Human Reproduction and Embryology
Uterine physiology |
Novel hepatocyte growth factor/scatter factor isoform transcripts in the macaque endometrium and placenta
1 Pharmaceutical Sciences Department, Texas Tech University Health Science Center School of Pharmacy, 1300 Coulter Drive, Amarillo, TX 79106 and 2 Division of Reproductive Sciences, Oregon Health Sciences University, Oregon Regional Primate Center, Beaverton, OR 97006, USA
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and methodsResultsDiscussionAcknowledgementsReferences |
|---|
Hepatocyte growth factor/scatter factor (HGF/SF) induces proliferation, motility and morphogenesis of cells that express the proto-oncogene for the tyrosine kinase receptor, c-Met. Because these cellular events occur in the endometrium during the menstrual cycle and in placenta during development, we have initiated studies of this growth factor in these tissues from macaques. Several HGF/SF alternatively spliced transcripts have been previously reported in other tissues. However, expression of HGF/SF isoforms in the endometrium has not been studied. Here we describe the relative transcript amounts of HGF/SF isoforms in the endometrium and placenta using RNase protection analyses. During these analyses, we discovered two unexpected protected bands that were found through sequence analyses to represent isoforms similar to the previously reported NK1 and NK2 except that they encode a five amino acid deletion in the first kringle domain. We designated these two isoforms as dNK1 and dNK2. Endometrium expressed all of the isoforms; however, dNK2 was consistently expressed at higher levels than NK2 transcripts. In contrast, placenta expressed NK2 and dNK2 mRNA at equal levels, and both NK1 and dNK1 were undetectable in placenta. HGF/SF function in endometrium and placenta may involve complex interactions between the isoforms of HGF/SF and those of c-Met.
endometrium/estrogen/hepatocyte growth factor (scatter factor)/placenta/progesterone
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionAcknowledgementsReferences |
|---|
Hepatocyte growth factor, also known as scatter factor (HGF/SF), binds heparin and acts in a pleiotropic manner upon cells that express the c-Met proto-oncogenic receptor (Kitamura et al., 1993
). These pleiotropic events include cell migration, mitosis, morphogenesis (Sugawara et al., 1997), angiogenesis (Rosen and Goldberg, 1997) and epithelial-to-mesenchymal transition (Fafeur et al., 1997; Fournier et al., 2000). Because these events occur during tissue remodelling and development, it has been proposed that HGF/SF plays a role in regeneration of the endometrium and in development of the placenta. HGF/SF mRNA has been localized to endometrial stromal cells and to chorioallantoic mesenchyme of the early conceptus in sheep (Chen et al., 2000). HGF/SF and c-Met transcripts have also been localized to the villous core and cytotrophoblast and decidual glands respectively in human placenta (Clark et al., 1996). Studies with HGF/SF knockout mice show that placental development requires HGF/SF expression (Uehara et al., 1995). Sugawara et al. have shown that HGF/SF promotes migration and tubule formation of isolated endometrial epithelial cells in vitro (Sugawara et al., 1997). In addition, it has been shown that HGF promotes migration of human endometrial epithelial carcinoma cell lines HEC-1A and KLE in vitro (Bae-Jump et al., 1999). However, the actual expression of HGF/SF in primate endometrium has not been shown (Sugawara et al., 1997).
These in-vitro experiments used the full length activated form of HGF/SF. However, there are four known alternatively spliced variants of HGF/SF transcripts. First, there is the full length transcript that encodes four plasminogen-like kringle domains, plus a serine protease-like domain. The secreted proforms of HGF/SF must be cleaved to be active, a process that is thought to occur during tissue damage (Nakamura et al., 1989; Silvagno et al., 1995). Second, there is the HGF/SF mRNA that is similar to the full length form except for a 15 bp deletion in the first kringle domain. This form is generally referred to as dHGF (Seki et al., 1990; Shiota et al., 2000). In vitro, dHGF and HGF/SF exhibit different biological properties (Shima et al., 1994). Third, a truncated HGF/SF transcript called NK2 with antagonistic properties encodes the amino terminus and first two kringle domains but lacks kringles 3 and 4 as well as the carboxy terminus (Chan et al., 1991). Finally, another truncated isoform known as NK1 encodes only the amino terminus and the first kringle domain; antagonist as well as agonist effects have been attributed to this isoform (Lokker et al., 1993; Cioce et al., 1996; Jakubczak et al., 1998). The distinct functions of these alternative-spliced HGF/SF isoforms at physiological concentrations are unknown.
To provide further insights into the role of HGF/SF expression in the endometrium and placenta, we examined the relative amounts of the various isoforms in these tissues using the highly sensitive method of RNase protection. Additionally, we analysed the various HGF/SF isoforms in the endometrium as compared to that in placenta. Of interest, we found that the endometrium expresses two novel HGF/SF isoforms, only one of which is expressed by the placenta.
|
Materials and methods |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionAcknowledgementsReferences |
|---|
Animals and treatments
Animal care during these studies was provided by the veterinary staff of the Oregon Regional Primate Research Center (ORPRC) Division of Animal Resources in accordance with the NIH Guide for Care and Use of Laboratory Animals. Eleven pigtail macaques (Macaca nemestrina) were ovariectomized and treated sequentially with estradiol (E2) and progesterone to create artificial menstrual cycles. To create these cycles, a 3 cm Silastic capsule (0.34 cm inner diameter; 0.64 cm outer diameter; Dow Corning, Midland, MI, USA) packed with crystalline E2 (Steraloids Inc., Wilton, NH, USA) was inserted s.c. at the time of ovariectomy to stimulate an artificial proliferative phase. After 14 days of E2, a 6 cm Silastic capsule containing crystalline progesterone (Steraloids Inc.) was inserted s.c. for 14 days to stimulate an artificial secretory phase. Removal of the progesterone implant (leaving the E2 implant in place) completed each cycle and induced two to three days of menstruation. The uteri were collected by mid-ventral laparotomy. Serum levels of E2 and progesterone in these animals were analysed by radioimmunoassay (see Table I
) to confirm retention and function of the implants.
|
For pigtail macaques, uteri were collected on days 1 (19 h after progesterone withdrawal), 2, 3, 4, 5, 6 and 8 after progesterone implant withdrawal (n = 1 each).
For rhesus macaques, uteri were collected from three treatment groups. Group 1 received implants of E2 alone for 14 days (n = 2); group 2 received implants of E2 for 14 days, then E2 plus implants of progesterone for 14 days (n = 2); and group 3 received E2 plus progesterone for 14 days then the progesterone implant was removed for 8 days (n = 1).
All placentae were obtained from rhesus macaques undergoing either Caesarean sections or natural births on the days of gestation indicated in Figure 3A and B.
|
Riboprobe preparation
[32P]UTP-labelled RNA probes were prepared for RNase protection analysis. Using T7 and SP6 RNA polymerases, the 606 bp HGF monkey-specific riboprobe corresponding to nucleotides 395–1000 of the human HGF cDNA sequence (accession no. E03331) was synthesized. The riboprobe used for normalization was that for the monkey-specific S10 ribosomal protein cDNA which corresponds to the human version (nucleotides 171–290, accession no. U14972). The riboprobe corresponding to the 3' untranslated region of the human NK1 cDNA, nucleotides 1108–1294 (accession no. U46010) (Rubin et al., 1991
), was synthesized from a monkey-specific template. All templates were cloned in our laboratory or at the Molecular Core Facility at the Oregon Regional Primate Research Center. All monkey-specific constructs were confirmed by sequencing. An RNA ladder template (Century; Ambion, Inc.) was used to generate RNA-specific size markers. We used the in-vitro transcription protocol as described in Current Protocols in Molecular Biology (Ausubel et al., 1998). For RNase protection, full length probes were purified by electrophoresing on a 6% polyacrylamide denaturing gel. Gel slices containing each probe were mashed in 100 µl of diethylpyrocarbonate (DEPC)-treated water and the probe was eluted by two incubations in 600 µl of 1x proteinase K (PK) buffer (0.3 mol/l NaCl, 0.5% SDS, 10 mmol/l Tris pH 7.5, 200 µg/ml PK, and 20 µg/ml tRNA) for 5–15 min at 37°C. The suspended probe from both incubations was filtered (0.45 µm; Acrodisc), then chloroform/phenol extracted and ethanol precipitated.
RNase protection analyses
Total RNA from tissues was prepared as previously described either by guanidinium isothiocyanate lysis and centrifugation over a caesium chloride cushion or by a single-step acid guanidinium thiocyanate–phenol–chloroform extraction and quantified by measuring optical density. RNase protection analyses were performed as previously described (Ausubel et al., 1998) with minor modifications. Briefly, RNA samples, and tRNA as a negative control, were precipitated with excess amounts of the appropriate gel-purified [32P]UTP-labelled probes. The pellet was resuspended in 30 µl of annealing buffer [40 mmol/l PIPES (pH 6.4), 0.4 mol/l NaCl, 1 mmol/l EDTA, 80% formamide] and allowed to hybridize overnight at 42°C. Unhybridized RNA was digested with RNase A (50 µg/ml) and RNase T1 (4 µg/ml) for 30 min at 37°C. RNases were then removed by treatment with proteinase K and extraction with phenol/chloroform/isoamyl alcohol. After ethanol precipitation, the RNA pellet was resuspended in 90% formamide loading buffer, denatured at 85°C for 10 min and electrophoresed on a 6% polyacrylamide denaturing gel. The dried gel was exposed to film at –70°C for the time periods indicated in the figure legend.
Densitometry
Relative levels of HGF isoform expression were derived as compared to the expression level of ribosomal protein RNA, S10. Densitometry was performed using the Bio-Rad Molecular Analyst Software program and Densitometer 700 Scanner. The optical density of each band was obtained and background for that lane was subtracted. Graphs were derived using the Microsoft Excel program.
RT–PCR
Reverse transcription was performed with total RNA from macaque endometrium in the proliferative phase. Briefly, 2 µg of total RNA in 11 µl DEPC-treated H2O was heated to 70°C for 10 min and placed on ice to maintain the denatured state. The following was added: 1x 1st strand buffer (Clontech, Palo Alto, CA, USA), 10 mmol/l dithiothreitol, 2 mmol/l dNTP, 20 IU RNasin, 0.5 µmol/l penultimate 3' primer and incubated for 1 h at 56°C for NK1-specific 3' untranslated region (UTR) primer and at 53°C for NK2-specific 3' UTR primer. NK1 and NK2 transcripts each possess unique 3' UT regions (Chan et al., 1991; Cioce et al., 1996). After incubation, 100 IU of reverse transcriptase (MMLV; Promega, Madison, WI, USA) was added and the mixture was incubated for 45 min at 42°C with inactivation at 65°C for 10 min.
PCR for 35 cycles was performed with the Advantage® cDNA PCR kit by Clontech. Briefly, 1x cDNA PCR reaction buffer, ~1.0 ng cDNA template from RT mixture, 0.2 mmol/l dNTP mix, 0.2 µmol/l 5' primer, 0.2 µmol/l 3' nested primer, 1x Advantage® polymerase mix were combined, denatured at 94°C for 30 s, and cycled under the following conditions: 94°C for 40 s, annealed at 56°C for 1 min, and 68°C for 2 min. Products were electrophoresed on a 2% gel containing 0.5 µg/ml ethidium bromide. The very bottom of each expected size band, ~365 bp for NK1 and ~534 bp for NK2, was excised and reamplified using the same PCR protocol and primers. The resulting products were then subcloned into the pCRII vector (Invitrogen) and sequenced.
We used the following primers for NK1 from accession no. u46010 (Rubin et al., 1991). 5' primer: 5'-ACTGCATCATTGGTAAAGGAC-3' (nucleotides 434–454); 3' penultimate primer: 5'-CTTGTCAGCCATTCAGTTTTCC-3' (nucleotides 1372–1393); 3' nested primer: 5'-TGCATTTGCACGAACAAC-3' (nucleotides 782–799).
We used the following primers for NK2 from accession no. m77227 (Chan et al., 1991): 5' primer: 5'-GCATCATTGGTAAAGGAC-3' (nucleotides 465–482); 3' penultimate primer: 5'-TCAGAAAAGCTGGGTAAG-3' (nucleotides 1161–1178); 3' nested primer: 5'-TGTCACTCACCAGAAGAAG-3' (nucleotides 981–999).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionAcknowledgementsReferences |
|---|
Endometrial HGF/SF isoform expression
We used RNase protection analyses to determine the relative expression of the different HGF/SF isoforms. Endometrial samples from eight different animals were analysed three times by this assay. As can be seen in the representative analysis in Figure 1A
, the lowest level of endometrial HGF/SF isoform expression was evident at the end of the artificial luteal phase, when progesterone levels were high, and the highest levels were found in estrogen-treated animals after progesterone was withdrawn. Serum hormone levels of the macaques used in this RNase protection analysis are shown in Table I
. We detected the expected sized protected bands for full length HGF/SF (~606 bp), dHGF (~502 bp), NK2 (~456 bp) and NK1 (255 bp). Both HGF/SF and dHGF were expressed in equal amounts and were generally expressed at 3- and 7-fold higher levels than the NK2 and NK1 isoforms respectively (Figure 1B). In addition, two unexpected protected bands (367 and 151 bp) were observed. These same protected bands were also seen in estrogen + progesterone, E-8dP and E-14dP rhesus macaque endometrial RNA samples (data not shown). We named these two unexpected bands dNK2 (367 bp) and dNK1 (151 bp) because we hypothesized they had deletions in the first kringle domain of NK2 and NK1 respectively. The resulting band representing dNK1 or dNK2 in our assays was always the larger of the two protected fragments on either side of the 15 bp deletion. In other words, the region 5' of the 15 bp deletion would run at the bottom or off the gel because of its small size (86 bp). These results show that dNK2 was consistently expressed at higher levels than the previously reported NK2 (Figure 1B). When detectable, levels of the various isoforms relative to full length HGF transcripts did not appear to change with hormonal treatment.
|
Placental HGF/SF isoform expression
In all six placentae from different animals, levels of full length HGF/SF and dHGF mRNA were essentially equal as seen in the endometrium. However, in the placenta, the truncated isoforms tended to be even lower than in endometrium or undetectable. NK2 and dNK2 were expressed at similar levels but NK1 and dNK1 transcripts were not detected (Figure 2A
). In the RNase protection analyses comparing endometrium with placenta (Figure 2B), NK1 and dNK1 transcripts were not detectably expressed in placenta, whereas NK2 was expressed at comparable levels and dNK2 at lower levels to those detected in endometrium (Figure 2C and D). Consistently, an unknown isoform band at ~520 bp in endometrium was absent in placental RNA samples. This protected 520 bp band is not described in this report. However, we did evaluate the unexpected dNK1 and dNK2 bands (see below).
|
Isolation and sequencing of two novel isoforms
We performed RT–PCR as described in Materials and methods using RNA isolated from estrogenized endometrium. Figure 3A
shows the products from NK1- or NK2-specific primers. Fortunately, these two isoforms have distinct 3' untranslated regions to which we could design isoform-specific primers. Because it is difficult to isolate such closely related bands (of 15 bp difference) by electrophoresis, we excised the lower half of the major bands that were the predicted size (arrows) for NK1 and NK2 bands and subcloned each into the pCRII vector. The spurious bands were probably due to the primers binding non-specifically. However, this was the annealing temperature at which we received adequate specific product to isolate from the gel, purify, and subclone.
After subcloning and plating, we isolated five clones from each plate, prepared DNA, and had sequencing performed. We took the resulting sequences and compared them to the respective NK1 or NK2 sequence already in the database. We used the program Bestfit in GCG; representative comparisons are seen in Figure 3B and C. At least three out of the five clones for each isoform contained the same 15 bp deletion.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionAcknowledgementsReferences |
|---|
This paper reports the discovery of two new truncated HGF/SF isoforms that encode a deletion in the first kringle domain and are truncated after either the first kringle (dNK1) or the second kringle (dNK2). In previous reports on truncated forms, RNase protection analyses were not used and therefore the different bands were not evident from Northern analyses. The 15 bp deletion found in dNK1 and dNK2 is the same deletion found in dHGF. Estrogenized endometrium consistently expressed dNK2 at higher levels than NK2 mRNA, whereas NK1 and dNK1 levels were approximately equal in their degree of expression. In contrast, the placenta, which did not express either NK1 or dNK1 at detectable levels, expressed NK2 and dNK2 in approximately equal amounts.
In previous reports on deletions in the first kringle, the functions of deleted forms are unclear. One function attributed to this deletion sequence is heparin binding. Another family of heparin binding proteins is the fibroblast growth factor (FGF) family. The amino acid sequence deleted in the first kringle of dNK1 and dNK2, FLPSS, is one of the conserved sequences in all seven members of the FGF family, and its deletion in basic FGF modulates heparin-binding activity (Seno et al., 1990). Also, both the hairpin loop and the first kringle domain have been shown to play a role in heparin binding by HGF/SF. HGF/SF and NK1 require heparin sulphate glycosaminoglycan (HSGAG) to bind c-Met efficiently as seen in studies of HSGAG-deficient CHO cells. Shima et al. have reported that dHGF elutes slightly earlier than HGF/SF on heparin-affinity high performance chromatography (Shima et al., 1991), suggesting that the 15 bp deletion does change the heparin binding property. Heparin has been localized to basement membrane surrounding glands and blood vessels in the endometrium (Aplin et al., 1988). We have localized c-Met mRNA and protein to glandular epithelium and to blood vessels in primate endometrium (data not shown), while in ovine endometrium c-Met mRNA has been localized exclusively to luminal and glandular epithelium (Chen et al., 2000). As a result, the deletion in the first kringle domain may affect the localization of these isoforms to heparinized regions and their proximity to the c-Met receptor in the endometrium. In addition, HGF/SF, both in its active and precursor forms, as well as NK1 and NK2, bind to the extracellular matrix components, thrombospondin-1 and fibronectin (Lamszus et al., 1996), though this study did not examine the effects of the deletion in the first kringle and its effects on binding to these proteins. Kringle structures have also been shown to facilitate protein–protein interactions (Patthy et al., 1984). Therefore, the conformational changes caused by the lack of five amino acids in the first kringle probably affects the binding of HGF/SF isoforms to several molecules in the extracellular matrix, including heparin as well as unknown proteins.
The functions of NK1 and NK2 have been elusive. NK2 was the first described truncated HGF/SF isoform. Unlike HGF/SF, NK2 does not induce DNA synthesis in B5/589 human mammary epithelial cells. However, a 10–20-fold molar excess of NK2 over HGF/SF can inhibit DNA synthesis by 50% (Chen et al., 1991). Therefore, NK2 could act as an antagonist for HGF/SF-induced DNA synthesis. For this DNA synthesis assay (Chan et al., 1991), NK2 was isolated from SK-LMS-1 mammalian cells that normally produce this protein and then purified over a heparin–Sepharose column. It is conceivable that dNK2 does not elute from this column at the same time as NK2 due to a possible change in affinity for heparin. Therefore, it is not clear whether this assay used NK2 or dNK2 or both. Conditioned media from a naturally NK2-producing cell line, SBC-5, causes cell scattering on SBC-1 and SBC-2 cell lines. However, a sensitive method of HGF/SF detection was not used in this protocol (Itakura et al., 1994). Therefore, this cell scattering could have been due to contamination with HGF/SF. Transfected cDNA for NK2 does not cause angiogenesis in vivo when compared to HGF/SF (Silvagno et al., 1995). However, NK2 causes c-Met phosphorylation and can induce cell dissociation of, but not mitosis in, MDCK cells, although it has a 30-fold lower specific activity than HGF/SF (Hartmann et al., 1992). Therefore, it appears that NK2 activates the c-Met receptor and stimulates breakdown of cell–cell contacts in epithelial cells in culture. Whether or not dNK2 also binds to heparin, c-Met, or acts as a stimulator of cell scattering remains to be determined.
NK1 has been isolated and analysed by several laboratories. At a 50-fold molar increase over HGF/SF, NK1 phosphorylates the c-Met receptor and acts as a competitive HGF/SF antagonist but lacks intrinsic mitogenic activity in primary rat hepatocyte cultures (Lokker and Godowski, 1993). At an 80-fold molar increase over HGF/SF, NK1 stimulates DNA synthesis (Cioce et al., 1996). Overexpression of NK1 in transgenic mice has been shown to have similar yet reduced effects to overexpression of HGF/SF in transgenic mice. These effects include liver hyperplasia, kidney epithelial hyperplasia, aberrant striated muscle formation and increased tumorigenesis in older mice. Increased NK1 expression may facilitate oligomerization with HGF/SF and c-Met dimerization and activation. However, NK1 was overexpressed in many of these transgenic mouse tissues at levels equal to HGF/SF (Jakubczak et al., 1998) which are not physiological levels. As a result, it is unclear how NK1 functions in vivo.
We have localized HGF/SF in macaque placenta (not shown), consistent with that previously reported for human placenta (Clark et al., 1996). It has been proposed that HGF/SF expressed in the villous core acts on the trophoblastic cells in a paracrine manner to cause trophoblast growth as well as migration into the decidua (Lail-Trecker et al., 1998). In the placenta, the absence of NK1 and dNK1, which can antagonize HGF/SF action, may favour unopposed HGF/SF action. However, in the endometrium, the presence of NK1 and dNK1 may provide a moderating effect during estrogen-stimulated HGF/SF expression. Northern blot analyses of placental expression of HGF/SF isoforms has been reported previously (Kitamura et al., 1993). However, our work is the first report that NK1 and dNK1 mRNA are not detectable in placenta. Also, this is the first report showing that endometrium expresses all previously reported isoforms including the two novel isoforms we detected by RNase protection analysis and confirmed by RT–PCR, cloning and sequencing. This difference in HGF/SF isoform expression may be due in part to a difference in expression of alternative splicing proteins between endometrium and placenta, such as that described for the splicing factor SC35 (Nie et al., 2000).
In conclusion, we have isolated two novel, alternatively spliced, truncated variants of HGF/SF, designated as dNK1 and dNK2. The placenta lacked NK1 and dNK1. Endometrium expressed higher levels of dNK2 than the previously described NK2. Like dHGF, dNK1 and dNK2 may have different solubilities and tertiary structures when compared with NK1 and NK2. Also, since dHGF and HGF/SF require cleavage to be active and the truncated isoforms do not possess this cleavage region, the truncated isoforms may be continually active in both endometrium and placenta. There may be oligomerization of different isoforms and complex binding patterns to the c-Met isoforms which play important roles in estrogen-dependent HGF signalling. Whether the endometrial signalling pathways induced by these truncated isoforms differ from that of the full length forms, and whether they play important roles in endometrial and placental physiology, are matters for future research.
|
Acknowledgements |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionAcknowledgementsReferences |
|---|
We thank Kuni Mah for her technical assistance and Tosh Lyons for his assistance in preparation of the manuscript. This work was supported by the NIH grant HD 19182.
|
Notes |
|---|
3 To whom correspondence should be addressed. E-mail: suzanne{at}cortex.ama.ttuhsc.edu
|
References |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionAcknowledgementsReferences |
|---|
Aplin, J.D., Charlton, A.K. and Ayad, S. (1988) An immunohistochemical study of human endometrial extracellular matrix during the menstrual cycle and first trimester of pregnancy. Cell Tiss. Res., 253, 231–240.[ISI][Medline]
Ausubel, F.M., Brent, R., Kingston, R.E., Moore, D.D., Seidman, J.G., Smith, J.A. and Struhl, K. (1998) In Current Protocols in Molecular Biology. Greene Publishing, Wiley–Interscience, New York.
Bae-Jump, V., Segreti, E.M., Vandermolen, D. et al. (1999) Hepatocyte growth factor (HGF) induces invasion of endometrial carcinoma cell lines in vitro. Gynecol. Oncol., 73, 265–272.[ISI][Medline]
Chan, A.M., Rubin, J.S., Bottaro, D.P. et al. (1991) Identification of a competitive HGF antagonist encoded by an alternative transcript. Science, 251, 802–804.
Chen, C., Spencer, T.E. and Bazer, F.W. (2000) Expression of hepatocyte growth factor and its receptor c-met in the ovine uterus. Biol. Reprod., 62, 1844–1850.
Chomczynski, P. and Sacchi, N. (1987) Single-step method of RNA isolation by acid guanidinium thiocynate–phenol–chloroform extraction. Analyt. Biochem., 162, 156–159.
Cioce, V., Csaky, K.G., Chan, A.M.L. et al. (1996) Hepatocyte growth factor (HGF)/NK1 is a naturally occurring HGF/Scatter factor variant with partial agonist/antagonist activity. J. Biol. Chem., 271, 13110–13115.
Clark, D.E., Smith, S.K., Sharkey, A.M. et al. (1996) Hepatocyte growth factor/scatter factor and its receptor c-met: localisation and expression in the human placenta throughout pregnancy. J. Endocrinol., 151, 459–467.[Abstract]
Fafeur, V., Tulasne, D., Queva, C. et al. (1997) The ETS1 transcription factor is expressed during epithelial–mesenchymal transitions in the chick embryo and is activated in scatter factor-stimulated MDCK epithelial cells. Cell Growth Diff., 8, 655–665.[Abstract]
Fournier, T.M., Lamorte, L., Maroun, C.R. et al. (2000) Cbl-transforming variants trigger a cascade of molecular alterations that lead to epithelial mesenchymal conversion. Mol. Biol. Cell, 11, 3397–3410.
Hartmann, G., Naldini, L., Weidner, K.M. et al. (1992) A functional domain in the heavy chain of scatter factor/hepatocyte growth factor binds the c-Met receptor and induces cell dissociation but not mitogenesis. Proc. Natl. Acad. Sci. USA, 89, 11574–11578.
Itakura, Y., Yamamoto, T., Matsumoto, K. et al. (1994) Autocrine stimulation of motility in SBC-5 human lung carcinoma cells by a two-kringle variant of HGF. Cancer Lett., 83, 235–243.[ISI][Medline]
Jakubczak, J.L., LaRochelle, W.J. and Merlino, G. (1998) NK1, a natural splice variant of hepatocyte growth factor/scatter factor, is a partial agonist in vivo. Mol. Cell. Biol., 18, 1275–1283.
Kitamura, N., Miyazawa, K., Uehara, Y. et al. (1993) Gene expression and regulation of HGF-SF. Experientia, 65 (Suppl.), 49–65.
Lail-Trecker, M., Fulati, R. and Peluso, J.J. (1998) A role for hepatocyte growth factor/scatter factor in regulating normal and neoplastic cells of reproductive tissues. J. Soc. Gynecol. Invest., 5, 114–121.[ISI][Medline]
Lamszus, K., Joseph, A., Jin, L. et al. (1996) Scatter factor binds to thrombospondin and other extracellular matrix components. Am. J. Pathol., 149, 805–818.[Abstract]
Lokker, N.A. and Godowski, P.J. (1993) Generation and characterization of a competitive antagonist of human hepatocyte growth factor, HGF/NK1. J. Biol. Chem., 268, 17145–17150.
Nakamura, T., Nishizawa, T., Hagiya, M. et al. (1989) Molecular cloning and expression of human hepatocyte growth factor. Nature, 342, 440–443.[Medline]
Nie, G.-Y., Li, Y., Batten, L. et al. (2000) Uterine expression of alternatively spliced mRNAs of mouse splicing factor SC35 during early pregnancy. Mol. Hum. Reprod., 6, 1131–1139.
Patthy, L., Trexler, M., Vali, Z. et al. (1984) Kringles: modules specialized for protein binding. Homology of the gelatin-binding region of fibronectin with the kringle structures of proteases. FEBS Lett., 171, 131–136.[ISI][Medline]
Rosen, E.M. and Goldberg, I.D. (1997) Regulation of angiogenesis by scatter factor. Experientia, 79 (Suppl.), 193–208.
Rubin, J.S., Chan, A.M.L., Bottaro, D.P. et al. (1991) A broad-spectrum human lung fibroblast-derived mitogen is a variant of hepatocyte growth factor. Proc. Natl. Acad. Sci. USA, 88, 415–419.
Seki, T., Sugimura, A., Shimonishi, M. et al. (1990) Isolation and expression of cDNA for different forms of Hepatocyte Growth Factor from human leukocyte. Biochem. Biophys. Res. Commun., 172, 321–327.[ISI][Medline]
Seno, M., Sasada, R., Kurokawa, T. et al. (1990) Carboxyl-terminal structure of basic fibroblast growth factor significantly contributes to its affinity for heparin. Eur. J. Biochem., 188, 239–245.[ISI][Medline]
Shima, N., Nagao, M., Ogaki, F. et al. (1991) Tumor cytotoxic factor/hepatocyte growth factor from human fibroblasts: cloning of its cDNA, purification and characterization of recombinant protein. Biochem. Biophys. Res. Commun., 180, 1151–1158.[ISI][Medline]
Shima, N., Tsuda, E., Goto, M. et al. (1994) Hepatocyte growth factor and its variant with a deletion of five amino acids are distinguishable in their biological activity and tertiary structure. Biochem. Biophys. Res. Commun., 200, 808–815.[ISI][Medline]
Shiota, A., Yamashita, Y., Fujise, N. et al. (2000) A deleted form of human hepatocyte growth factor stimulates hepatic lipogenesis and lipoprotein synthesis in rats. Pharmacol. Res., 42, 443–452.[ISI][Medline]
Silvagno, F., Follenzi, A., Arese, M. et al. (1995) In vivo activation of met tyrosine kinase by heterodimeric hepatocyte growth factor molecule promotes angiogenesis. Arteriosc. Thromb. Vasc. Biol., 15, 1857–1865.
Sugawara, J., Fukaya, T., Murakami, T. et al. (1997) Hepatocyte growth factor stimulated proliferation, migration and lumen formation of human endometrial epithelial cells in vitro. Biol. Reprod., 57, 936–942.[Abstract]
Uehara, Y., Minowa, O., Mori, C. et al. (1995) Placental defect and embryonic lethality in mice lacking hepatocyte growth factor/scatter factor. Nature, 373, 702–705.[Medline]
van der Voort, R., Keehnen, R.M.J., Smit, L. et al. (1997) Paracrine regulation of germinal center B cell adhesion through the c-Met-Hepatocyte growth factor/scatter factor pathway. J. Exp. Med., 185, 2121–2131.
Wilcox, J.N. (1993) Fundamental principles of in situ hybridization. J. Histochem. Cytochem., 41, 1725–1733.[Abstract]
Submitted on February 16, 2001; accepted on September 28, 2001.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||