Herrmann H et al. (1989), Expression of intermediate filament proteins du...

XB-ART-26933

Development 1989 Feb 01;1052:279-98.

Show Gene links Show Anatomy links

Expression of intermediate filament proteins during development of Xenopus laevis. I. cDNA clones encoding different forms of vimentin.

Herrmann H, Fouquet B, Franke WW.

???displayArticle.abstract???
To provide a basis for studies of the expression of genes encoding the diverse kinds of intermediate-filament (IF) proteins during embryogenesis of Xenopus laevis we have isolated and characterized IF protein cDNA clones. Here we report the identification of two types of Xenopus vimentin, Vim1 and Vim4, with their complete amino acid sequences as deduced from the cloned cDNAs, both of which are expressed during early embryogenesis. In addition, we have obtained two further vimentin cDNAs (Vim2 and 3) which are sequence variants of closely related Vim1. The high evolutionary conservation of the amino acid sequences (Vim1: 458 residues; Mr approximately 52,800; Vim4: 463 residues; Mr approximately 53,500) to avian and mammalian vimentin and, to a lesser degree, to desmin from the same and higher vertebrate species, is emphasized, including conserved oligopeptide motifs in their head domains. Using these cDNAs in RNA blot and ribonuclease protection assays of various embryonic stages, we observed a dramatic increase of vimentin RNA at stage 14, in agreement with immunocytochemical results obtained with antibody VIM-3B4. The significance of very weak mRNA signals detected in earlier stages is discussed in relation to negative immunocytochemical results obtained in these stages. The first appearance of vimentin has been localized to a distinct mesenchymal cell layer underlying the neural plate or tube, respectively. The results are discussed in relation to programs of de novo synthesis of other cytoskeletal proteins in amphibian and mammalian development.

???displayArticle.pubmedLink??? 2806127

Species referenced: Xenopus laevis
Genes referenced: krt12.4 mt-tr ncoa5 plec tbx2 trna tspo vim vim.2
???displayArticle.antibodies??? Krt Ab2 Vim Ab2

???attribute.lit??? ???displayArticles.show???

	Fig. 1. Immunofluorescence microscopy showing reactions of monoclonal antibody VIM-3B4 to vimentin on cultured kidney epithelial (XLKE) cells of line A6 (A-C) and a frozen section of ovary tissue of Xenopus laevis (D,E). (A) Typical fibrillar networks of vimentin IFs are seen in XLKE-A6 cells. (B,C) Epifluorescence (B) and phase-contrast (C) pictures of the same field of an XLKE-A6 cell monolayer after treatment with 10 M-colcemid for 4h, showing aggregates of collapsed IF in the perinuclear cytoplasm, as is typical for vimentin IF. (D,E) Epifluorescence (D) and phase contrast (E) pictures of a cryostat section of ovarian tissue, showing bright staining of interstitial cells as well as endothelial cells and erythrocytes of a blood vessel. Bars, 50,um.

	Fig. 3. Identification of the polypeptides encoded by the cDNA clones pXenViml and pXenVim4. Coelectrophoresis of the high-salt-buffer-resistant cytoskeletal proteins of XLKE-A6 cells (same system and symbols as in Fig. 2C,D) with products of in vitro translation from hybrid-selected embryonal stage-18 mRNA (A,B) and from mRNA obtained by in vitro transcription/translation of pXenViml (C,D). Coelectrophoresis of the mixed products obtained from in vitro transcription/translation of pXenViml and pXenVim4, respectively (E,F). (A,B) Coomassie-brilliantblue- stained gel (A) and corresponding autoradiograph (B), showing that the [ SJmethionine-labelled in vitro translation products from hybrid-selected mRNA comigrate with unlabelled vimentin (bracket) of XLKE-A6 cells. (C,D) Coomassie-brilliant-blue-stained gel (C) and corresponding autoradiograph (D) of the product obtained by in vitro transcription and translation, showing that the [ SJmethionine-labelled translation product comigrates with unlabelled vimentin (bracket) of XLKE-A6 cells. Note that some modification, probably phosphorylation, also takes place in the rabbit reticulocyte assay used for in vitro translation, resulting in the appearance of a minor, more acidic variant. (E,F) Coomassie-brilliant-blue-stained gel (E) and corresponding autoradiograph (F) of mixed in vitro transcription/translation products of Xenopus vimentin clones pXenViml and pXenVim4. As revealed from parallel gels of individual [35S]methionine-labelled translation products, pXenVim4 is translated into a polypeptide slightly more acidic and less mobile (triangle) than pXenViml (bracket). Fluorography was for 4h. After prolonged exposure acidic variants were also visible.
	Fig. 4. Nucleotide sequence and deduced amino acid sequence of cloned Xenopus laevis vimentins. (A) Combined sequence of clones pXenViml, containing nucleotides 1-1688 (indicated by an arrow), and pXenVim3, containing nucleotide 1367 (indicated by an arrowhead) to the 3'-end of the clone. The clones are identical in the overlapping region. Differences found in the pXenVim2 sequence are indicated by an upward triangle (insertion of an AGC triplet coding for serine) and a downward triangle (substitution of T for C without change of the coded amino acid). (B) Sequence of clone pXenVim4. Major differences of the nucleotide sequence to pXenViml are overlined and the amino acid changes are encircled.
	Fig. 5. Amino acid sequence comparison of Xenopus vimentin Vim 1 (Xen), chicken vimentin (Chk; taken from Zehner et al. 1987) and hamster vimentin (Ham; taken from Quax et al. 1983). Bold-faced letters denote amino acids identical in Xenopus and at least one of the other two species. Amino acid sequences have been aligned for maximal homology, insertions introduced for this purpose are denoted by horizontal bars. The downward arrow demarcates the start and the upward arrow the end of the a'-helical rod domain. The dots represent positions a and d of the heptade convention to maximize coiled-coil configuration. The rod domain contains two non-o--helical interruptions of 11 and 43 amino acids, respectively, giving rise to coiled-coil subdomains 1A (CIA), IB (C1B) and 2 (C2). The arrowhead indicates an alteration in the heptade pattern that probably results in a 'stutter' in coil 2.
	Fig. 6. Comparison of the 3'-untranslated sequences of vimentin mRNAs of Xenopus (Xen), chicken (Chk; from Zehner et al. 1987), hamster (Ham; from Quax et al. 1983) and human (Hum; this study) as determined from cDNAs and genes, respectively. Sequences have been aligned for maximal homology and insertions introduced for this purpose are denoted by horizontal bars. The three stars mark the ends of the coding region. Regions of relatively high homology are boxed and numbered by roman numerals. Bold-faced letters denote nucleotides present in all four sequences. The apparent consensus sequences are indicated underneath each box; Lower case letters indicate the presence of at least three identical nucleotides. P, pyrimidine nucleotide; R, purine nucleotide. The arrowheads indicate the probable polyadenylation sites in the genomic sequences for chicken (Zehner & Paterson, 1983) and hamster vimentin (Quax et al. 1983). Note that the presumptive polyadenylation signal (overlined) in box VII is flanked by a highly conserved sequence.
	Fig. 7. Detection of vimentin-specific mRNA in various tissues and cells of adult animals and in embryonic stages of Xenopus laevis by RNA blot analysis (A-D) and RNase protection assays (E-H). (A) RNA was prepared from various tissues (Davis et al. 1986), and 20 //g total RNA (lanes 1, 4 and 5) or 1 fig poly(A)+RNA (lanes 2, 3, 6 and 7) were applied for formaldehyde/agarose gel electrophoresis. Hybridization was carried out with a random-primed, 32P-labelled (O-SxlO^tsmin"1 ml"1), 5'- end-specific probe of pXenViml. Lane 1, oesophagus; lane 2, XLKE-A6 cells; lane 3, skin; lane 4, skeletal muscle; lane 5, cardiac muscle; lane 6, liver; lane 7, ovary. Horizontal bars indicate the positions of Xenopus 28 S and 18 S rRNA. Exposure: 24 h. (B) RNA was prepared from ovary of adult frogs and embryos of the indicated developmental stages, 20 ^g total glyoxylated RNA were loaded per lane. RNA was blotted to Genescreen plus and hybridized with a 32P-labelled (3xl06ctsmin~1 ml"1) antisense RNA probe from pXenViml. The final washes were with 0-lxSSC, 0-5% SDS at 72Â°C. Lane 0, ovary; lane 1, unfertilized eggs; lane 2, stage 6-5 (morula); lane 3, stage 9 (fine cell blastula); lane 4, stage 11 (gastrula); lane 5, stage 14 (neural plate stage); lane 6, stage 18 (neural groove stage); lane 7, stage 28; lane 8, stage 39; lane 9, stage 42 (swimming tadpole). Horizontal bars indicate the positions of bovine 28 S, E. coli 23 S and bovine 18 S rRNAs. Exposure: 4h. Because of the higher activity of the riboprobe, the signal obtained for ovarian tissue appears enhanced, compared to that shown in A, lane 7). (C) RNA as loaded in B was run on a formaldehyde/agarose gel, blotted to nitrocellulose and hybridized with P-labelled probe as in B. The final washes were with 0-lxSSC, 0-1 % SDS at 65Â°C. Lane 1, unfertilized eggs; lane 2, stage 6-5; lane 3, stage 9; lane 4, stage 11; lane 5, stage 14; lane 6, stage 18; lane 7, stage 28; lane 8, stage 39; lane 9, stage 42. Horizontal bars indicate the positions of Xenopus 28S and 18S rRNA. Exposure: 5h. After 64 h exposure, the signals were considerably enhanced as shown in lanes 1' and 2' which correspond to lanes 1 and 2. Note the reaction with 18S and 28S rRNAs. (D) Blot shown in C after ribonuclease A treatment (20/igmn1 for 20min at 25Â°C followed by 65 Â°C washes with OlxSSC, 01 % SDS). Exposure: 10 days. Note that the signal at the position of the rRNA disappeared whereas that at the position of vimentin mRNA persisted (lanes 2-9). (E) Autoradiogram of a ribonuclease protection assay with 1 /ig sense RNA generated by in vitro transcription from pXenViml (lanes 10 and 12) and from the cytokeratin clone pXenCK 1/8 (lanes 9 and 11), showing that the sense RNA of pXenViml, but not that of pXenCKl/8, specifically protects the 32P-labelled anti-sense vimentin probes (arrows in lanes 10 and 12). Hybridizations shown here were carried out at 45Â°C (lanes 1, 2, 7 and 8) or 60Â°C (all other lanes) for 15h with lxlO5 ctsmin"1 (~lfmole) of the uniformly labelled 3'-specific probe (lanes 2, 4, 6, 8, 11 and 12) or the 'TYRKLEGE-probe' (see Materials and methods; lanes 1, 3, 5, 7, 9 and 10). To control for stringency, the probes were hybridized with 10 fig tRNA (lanes 1-8) and either subjected to RNAse treatment (lanes 5-8) or processed further without RNAse treatment (lanes 1-4). Lane M shows f/pall-restricted pBR 322 size markers (from top to bottom: 622, 527, 403, 309, 242, 238, 217, 201, 190, 180, 160, 147, 123, 110, 90 nucleotides). Exposure: lh. (F) Autoradiogram of a ribonuclease protection assay with 5 /Ig total Xenopus RNA from unfertilized eggs (odd numbers) and stage 18 (even numbers) using the 3'-specific probe (lanes 1-4) or the 'TYRKLEGE probe' (lanes 5-8). Hybridization was at 45Â°C (lanes 3, 4, 7 and 8) or 60Â°C (lanes 1, 2, 5 and 6). Arrows mark the fully protected probes without polylinker sequences. The arrowhead in lane 4 marks a prominent protected fragment of 160 nucleotides. Exposure: 3 days. (G) Autoradiogram of a ribonuclease protection assay with 5 /tg total RNA from oocytes (lane 1), unfertilized eggs (lane 2), stage-6-5 (lane 3), stage-18 (lane 4) and stage-28 embryos (lane 5) hybridized to the 3'-specific probe at 65Â°C. Exposure: 3 days. (H) Autoradiogram of a ribonuclease protection assay with 5/<g total RNA from oocytes (lane 1), unfertilized eggs (lane 2), stage-6-5 (lane 3), stage-9 (lane 4) and stage-28 embryos (lane 5) hybridized to the 3'-specific probe at 65Â°C in buffer containing only 50 % formamide as compared to 80% as used in panels E-G. Exposure: 3 days.
	Fig. 8. Immunofiuorescence microscopy performed on cryostat sections of frozen ovarian tissue from Xenopus laevis using monoclonal antibody VIM-3B4. (A,B) The same field is shown in epifluorescence (A) and phase-contrast (B) optics. Note that only the interstitial cells show a bright fluorescence, whereas the oocytes are negative. (C,D) In a region with previtellogenic oocytes (C, epifluorescence; D, phase contrast) the oocytes are negative whereas the interstitial cells, the endothelial cells and the erythrocytes of the blood vessel show intense fluorescence, v, blood vessel. Bars. 50um.
	Fig. 9. Survey picture montage of immunofluorescence micrographs of sections through snap-frozen stage- 14 embryos (neural plate stage) of Xenopus laevis (whole embryos were sectioned and documented in overlapping micrographs). (A) Reaction of monoclonal antibody VIM-3B4, showing positive reaction only in a distinct and thin mesenchymal cell layer (demarcated by brackets) whereas the neural plate (np), notochord (no) and somites (s) are negative. ec, ectoderm; en, endoderm. (B) Reaction of monoclonal cytokeratin antibody lu-5, showing an area corresponding to that shown in A in a step section. Note that all cell layers are positive for cytokeratins. In particular, the notochord (no) and the ectoderm (ec) are intensely stained. The mesenchymal cell layer (brackets) shown to be positive for vimentin in A, is also positive for cytokeratin. Symbols are as in A. Bars, 50,um.
	Fig. 10. Immunofluorescence microscopy performed on cryostat sections of snap-frozen stage-18 (neural groove stage) embryos of Xenopus laevis using monoclonal antibodies to vimentin (A) and cytokeratin (B). (A) Immunofluorescence micrograph of the neural groove region of a stage-18 embryo, showing the reaction of monoclonal antibody VIM-3B4. The reaction is restricted to a single cell layer (brackets) beneath the neural groove (n). All other cell layers are negative. ec, ectoderm; s, somite; n, neural groove; no, notochord. (B) Reaction of monoclonal antibody lu-5, on a step section of the same region shown in A. Note that all cell layers show a positive reaction, the notochord (no) and the ectoderm (ec) being brightly stained and that the thin layer of cells shown to be positive for vimentin in A is also positive for cytokeratin (brackets). Symbols are as in A. Bars, 50 um.
	Fig. H- Analysis of various fractions from oocytes and unfertilized eggs of Xenopus laevis for the presence of vimentin by SDS-PAGE and immunoblotting with different antibodies. (A) Proteins transferred to nitrocellulose were stained by Ponceau S after the alkaline phosphatase immunostaining reaction shown in B. Lane 1, proteins recovered in the 100 000 g supernatant from total egg extracts (HSS proteins), concentrated by precipitation with acetone, 30-egg equivalents; lane 2, HSS proteins directly loaded, 12-egg equivalents; lane 3, high-salt-buffer-resistant Â« a fraction of the low-speed pellet from egg homogenates; lane " 4, insoluble fraction from high-salt-buffer-extracted material 1 2 3 4 5 6 7 1 2 3 4 5 6 7 pelleted at lOOOOg after an initial 800g centrifugation; lane 5, high-salt-buffer-resistant material pelleted at 800g; lane 6, high-salt-buffer-resistant material pelleted at lOOOOg. Lanes 3-6, 20-egg equivalents. The prominent high molecular weight band in lanes 3-6 contains the abundant major yolk protein. Lane 7, total cytoskeletal proteins of XLKE-A6 cells, 2 fig, representing approximately 0-2 ng vimentin (see Fig. 2A). (B) Immunoreaction corresponding to A, developed until background staining appeared, as shown by the unspecific reaction of the Mr â€”100000 yolk protein band (lanes 3-6) which was also seen when blots were only incubated with second antibody. Specific vimentin reaction is only seen in XLKE-A6 cells (lane 7). (C,D) Calibration of the sensitivity of antibody VIM-3B4 for the detection of Xenopus vimentin. (C) For quantification, Xenopus vimentin synthesized in E. coli transfected with clone pXenViml (cf. Magin et al. 1987) was dissolved in electrophoresis sample buffer, and a 10 uL sample was applied to the gel (lane 2). For comparison, bovine serum albumin was loaded as follows: lane 1, 1-4 fig; lane 3, l-2^g; lane 4, l^g; lane 5, 0-8jUg; lane 6, 0-6fÂ«g; lane 7, 0-4//g. As estimated after staining with Coomassie brilliant blue, the applied amount of vimentin corresponded to approximately l^g of bovine serum albumin. (D) Immunoblot analysis of recombinant Xenopus vimentin diluted to various degrees: lane 1, l^g; lane 2, 500 ng; lane 3, 100 ng; lane 4, 50 ng; lane 5, 10 ng; lane 6, 5ng; lane 7, 1 ng. Note faint reaction in lane 6, showing that 5 ng are detectable. The band of somewhat higher SDS-PAGE mobility is proteolytically trimmed vimentin of Mr ~38000. Blots shown in B and D have been processed in parallel. (E,F) Examination of cytoskeletal fraction from unfertilized eggs with guinea pig antibodies to vimentin (E), in comparison with reference vimentin (F). (E) Autoradiogram of an immunoblot analysis of proteins from 15 unfertilized eggs separated by two-dimensional gel j - a l a electrophoresis, followed by incubation of nitrocellulose blots with guinea pig antibodies to vimentin and 125Ilabelled protein A. (F) Parallel experiment to that shown in E with cytoskeletal proteins from Chinese hamster ovary fibroblasts containing approximately l,ug vimentin. Both blots were exposed for 3 days at 70Â°C using intensifying screens. (G-L) Examination of the presence of vimentin in oocytes of Xenopus by two-dimensional gel electrophoresis, followed by immunoblotting. (G) Cytoskeletal proteins from 150 mg oocytes were separated by two-dimensional gel electrophoresis, transferred to nitrocellulose and stained with Ponceau S. The bracket denotes the major oocytes type II cytokeratin l/8, the three arrowheads denote a triplet of unidentified cytoskeletal proteins of Mr 56000-60000. Positions of cytokeratin 2 and 3, corresponding to human cytokeratin 18 and 19 and a group of basic cytokeratins of Mr â€”56000 are also indicated. (H) The same amount of protein as shown in G was mixed with cytoskeletal proteins from XLKE-A6 cells for the identification of the position of Xenopus vimentin (arrow) relative to the oocyte proteins. Symbols as in G. (I) Immunoblot analysis of cytoskeletal oocyte proteins with monoclonal antibody PK VI. Note absence of staining at the position of vimentin (arrows in H and J). Weak staining of five polypeptides of Mr 56000-60000 is indicated by arrowheads. (J) Immunoblot analysis of a mixture of cytoskeletal proteins from oocytes and XLKE-A6 cells with monoclonal antibody PK VI. Note staining of cytokeratin 1/8 and vimentin but not component X. (K) Immunoblot analysis of cytoskeletal oocyte proteins with antibody anti-IFA. Note staining of cytokeratin 1/8 but absence of staining at the position of vimentin. Note weak staining of triplet polypeptides of Mr 56000-60000. (L) Immunoblot analysis of a mixture of cytoskeletal proteins from oocytes and XLKE-A6 cells with anti-IFA. Vimentin as well as several cytokeratins are heavily stained, thus demonstrating that anti-IFA reacts with Xenopus vimentin. Comparison of L and K shows absence of detectable vimentin in K.