Vascular endothelial growth factor and cellular chemotaxis: a possible autocrine pathway in adult T-cell leukemia cell invasion.

Our previous report (T. Hayashibara et al., Leukemia, 13: 1634-1635, 1999) revealed a possible link between high plasma vascular endothelial growth factor (VEGF) concentration and leukemic cell invasion in adult T-cell leukemia (ATL). However, the biological mechanism of this link has not been elucidated. The purpose of this study was to address that mechanism. Our present observations showed that VEGF mRNA was expressed in ATL cell lines. The corresponding protein was secreted into the extracellular environment, which suggested that the major source of plasma VEGF is ATL cells themselves. More interestingly, all of the cell lines examined were found to express the mRNA and protein for fms-like tyrosine kinase-1 (Flt-1), which is one of the receptors for VEGF. Cytofluorometric analysis demonstrated the VEGF binding potency of these cells. In clinical specimens, expression of VEGF and Flt-1 mRNAs was detected in all (100%) of 11 and 8 (73%) of 11 ATL patients, respectively. Cytofluorometric analysis revealed that VEGF effectively bound only to Flt-1-expressing cells. These findings are highly suggestive of an autocrine pathway involving VEGF operating in ATL. The proliferation of ATL cell lines was not affected by treatment with an anti-VEGF antibody or exogenous VEGF, which indicated that VEGF has no mitogenic effect on ATL cells. In contrast, we made the interesting finding that treatment with exogenous VEGF enhanced the chemotactic activities of some ATL cell lines, which may play a key role in ATL cell invasion. Collectively, these data lead us to propose a possible autocrine mechanism involving VEGF operating by way of Flt-1, in which ATL cells up-regulate their own chemotaxis to facilitate their invasion into various organs.

Localization of porcine cardiac myosin epitopes that induce experimental autoimmune myocarditis.

Porcine cardiac myosin induced severe myocarditis in Lewis rats, similar to that seen after immunization with human or rat cardiac myosin. We investigated the localization of pathogenic epitopes by using porcine cardiac myosin. Subfragment-1 (S-1) and the rod were obtained by alpha-chymotryptic digestion. The rod was further fragmented by using cyanogen bromide cleavage. Three subfragments of S-1 were prepared by tryptic digestion. All Lewis rats immunized with the rod exhibited severe myocarditis, and the immunization of the cyanogen bromide-cleaved peptide equivalent to human beta-cardiac myosin heavy chain RDCB9 (residues 1070 to 1165) induced moderate myocarditis. Although none of the rats immunized with the whole S-1 exhibited any myocardial lesions, moderate inflammatory cardiac lesions were detectable in rats immunized with the tryptic digests of S-1. Our results indicate that the myocardiogenic epitopes in Lewis rats are located in the RDCB9 (residues 1070 to 1165) of the cardiac myosin rod and that a cryptic minor epitope may reside in S-1.

Primary structure of chicken cardiac myosin S-1 heavy chain.

The sequence of the NH2-terminal 830 amino acid residues of chicken cardiac ventricular muscle myosin subfragment-1 (S-1) was determined. S-1 was obtained by limited chymotryptic digestion, and cleaved into three characteristics fragments (23, 41, and 22 kDa fragments) by limited tryptic digestion. These fragments were isolated by gel filtration on a Sephadex G-100 column, followed by cation-exchange chromatography on a CM-52 column and reverse-phase HPLC. The isolated fragments were sequenced completely. Peptides overlapping the 23 and 41 kDa fragments and also overlapping the 41 and 22 kDa fragments were obtained by cleaving S-1 with cyanogen bromide, and sequenced completely. We also obtained a minor fragment, the 20 kDa fragment, in addition to the three characteristic fragments. Amino acid compositions of the cyanogen bromide peptides of the 20 kDa fragment indicated that a portion of S-1 heavy chains had lost their COOH-terminal 21 residues during limited tryptic digestion. Methylated amino acid residues were found at four positions: epsilon-N-monomethyllysine at position 32, epsilon-N-trimethyllysine residues at 127 and 549, and 3-N-methylhistidine at 754.

Interaction between the heavy and the regulatory light chains in smooth muscle myosin subfragment 1.

The interaction between the heavy and the regulatory light chains within chicken gizzard myosin heads was investigated by using a zero-length chemical cross-linker, 1-ethyl-3-[3-(dimethylamino)-propyl]carbodiimide (EDC). The chicken gizzard subfragment 1 (S-1) used was treated with papain so that the heavy chain was partly cleaved into the NH2-terminal 72K and the COOH-terminal 24K fragments and the regulatory light chain into the 16K fragment. S-1 was reacted with EDC either alone or in the presence of ATP or F-actin. In all cases, the 16K fragment of the regulatory light chain formed a covalent cross-link with the 24K heavy chain fragment but not with the 72K fragment. The 38K cross-linked peptide, which was the product of cross-linking between the 16K light chain and the 24K heavy chain fragments, was isolated and further cleaved with cyanogen bromide and arginylendopeptidase. Smaller cross-linked peptides were purified by reverse-phase HPLC and then characterized by amino acid analysis and sequencing. The results indicated that cross-linking occurred between Lys-845 in the heavy chain and Asp-168, Asp-170, or Asp-171 in the regulatory light chain. The position of the cross-linked lysine was only three amino acid residues away from the invariant proline residue mapped as the S-1-rod hinge by McLachlan and Karn [McLachlan, A. D., & Karn, J. (1982) Nature (London) 299, 226-231]. We propose that the COOH-terminal region of the regulatory light chain is located in the neck region of myosin and that this region and the phosphorylation site of the regulatory light chain together may play a role in the phosphorylation-induced conformational change of gizzard myosin.

The primary structure of skeletal muscle myosin heavy chain: I. Sequence of the amino-terminal 23 kDa fragment.

Subfragment-1 was prepared from adult chicken pectoralis myosin by limited digestion with alpha-chymotrypsin, and an amino-terminal 23 kDa fragment of the heavy chain was obtained by digesting the subfragment-1 with trypsin. The 205-residue sequence of the fragment was determined by sequencing its cyanogen bromide, tryptic, and chymotryptic peptides. The amino-terminal alpha-amino group of the fragment was acetylated, and two methylated lysines; epsilon-N-monomethyllysine and epsilon-N-trimethyllysine were recognized at the 35th and 130th positions, respectively, as in rabbit skeletal myosin. Comparing the 205-residue sequence of the skeletal myosin with those of cardiac, and gizzard myosins from chicken, considerable differences are recognized, especially in the amino-terminal region, but strong homologies are observed around the reactive lysine residue, around the epsilon-N-trimethyllysine residue, and around the consensus sequence of GXXGXGKT for nucleotide-binding proteins. On the other hand, only 12 amino acid substitutions are recognized between adult and embryonic skeletal myosins, allowing for the post-translational methylation.

The primary structure of skeletal muscle myosin heavy chain: II. Sequence of the 50 kDa fragment of subfragment-1.

The complete amino acid sequence of the 50 kDa fragment of subfragment-1 from adult chicken pectoralis muscle myosin was determined. It contained 431 residues including an epsilon-N-trimethyllysine at position 346. The 431-residue sequence corresponds to the sequence of residues 206 to 639 of chicken embryonic breast muscle myosin heavy chain which was predicted from the nucleotide sequence of the cDNA by Molina et al. [Molina, M. I., Kropp, K.E., Gulick, J., & Robbins, J. (1987) J. Biol. Chem. 262, 6478-6488]. Comparing the two sequences, 23 amino acid substitutions and three deletions/insertions are recognized.

The primary structure of skeletal muscle myosin heavy chain: III. Sequence of the 22 kDa fragment and the alignment of the 23 kDa, 50 kDa, and 22 kDa fragments.

The amino acid sequence of the 197-residue 22 kDa fragment from chicken pectoralis muscle was determined to be as follows: K-K-G-S-S-F-Q-T-V-S-A-L-F-R-E-N-L-N-K-L- M-A-N-L-R-S-T-H-P-H-F-V-R-C-I-I-P-N-E-T-K-T-P-G-A-M-E-H-E-L-V-L-H-Q-L-R- C-N-G-V- L-E-G-I-R-I-C-R-K-G-F-P-S-R-V-L-Y-A-D-F-K-Q-R-Y-R-V-L-N-A-S-A-I-P-E-G-Q- F-M-D-S- K-K-A-S-E-K-L-L-G-S-I-D-V-D-h-T-Q-Y-R-F-G-H-T-K-V-F-F-K-A-G-L-L-G-L-L-E- E-M-R-D- D-K-L-A-E-I-I-T-R-T-Q-A-R-C-R-G-F-L-M-R-V-E-Y-R-R-M-V-E-R-R-E-S-I-F-C-I- Q-Y-N-V-R-S-F-M-N-V-K-H-W-P-W-M-K-L-F-F-K, where h stands for 3-N-methylhistidine. The amino acid sequences of the 22 kDa fragment and its equivalent fragment from chicken ventricle and gizzard muscle myosins were also determined by our group. Predicted secondary structures of these 22 kDa fragment regions and of the reported chicken embryo myosin revealed some possible structural differences.(ABSTRACT TRUNCATED AT 250 WORDS)

The primary structure of skeletal muscle myosin heavy chain: IV. Sequence of the rod, and the complete 1,938-residue sequence of the heavy chain.

In the preceding paper [Maita, T., Miyanishi, T., Matsuzono, K., Tanioka, Y., & Matsuda, G. (1991) J. Biochem. 110, 68-74], we reported the amino-terminal 837-residue sequence of the heavy chain of adult chicken pectoralis muscle myosin. This paper describes the carboxyl terminal 1,097-residue sequence and the linkage of the two sequences. Rod obtained by digesting myosin filaments with alpha-chymotrypsin was redigested with the protease at high KCl concentration, and two fragments, subfragment-2 and light meromyosin, were isolated and sequenced by conventional methods. The linkage of the two fragments was deduced from the sequence of an overlapping peptide obtained by cleaving the rod with cyanogen bromide. The rod contained 1,039 amino acid residues, but lacked the carboxyl-terminal 58 residues of the heavy chain. A carboxyl-terminal 63-residue peptide obtained by cleaving the whole heavy chain with cyanogen bromide was sequenced. Thus, the carboxyl terminal 1,097-residue sequence of the heavy chain was completed. The linkage of subfragment-1 and the rod was deduced from the sequence of an overlapping peptide between the two which was obtained by cleaving heavy meromyosin with cyanogen bromide. Comparing the sequence of the adult myosin thus determined with that of chicken embryonic myosin reported by Molina et al. [Molina, M.I., Kropp, K.E., Gulick, J., & Robbins, J. (1987) J. Biol. Chem. 262, 6478-6488], we found that the sequence homology is 94%.

Lys-65 and Glu-168 are the residues for carbodiimide-catalyzed cross-linking between the two heads of rigor smooth muscle heavy meromyosin.

We have previously demonstrated that the two heads of chicken gizzard heavy meromyosin (HMM) in a rigor complex with rabbit skeletal F-actin could be cross-linked by the water-soluble carbodiimide 1-ethyl-3-[3-(dimethylamino)propyl]carbodiimide. Here, we report the location of the cross-linked sites in the amino acid sequence of the HMM heavy chain. One of the cross-linked residues was identified as Glu-168 by sequencing the CN1.CN6 cross-linked peptide containing residues 1-77 (CN1) and 164-203 (CN6). This site is located close to the ATP-binding site of HMM. Since the other site was further into the amino acid sequence of CN1, another cross-linked peptide corresponding to residues 53-66 and 145-182 was isolated from the 1-ethyl-3-[3-(dimethylamino)propyl]carbodiimide-treated acto-tryptic gizzard HMM digested further by other proteolytic enzymes. The amino acid sequence of this peptide and its cyanogen bromide fragment indicated that the cross-linking occurred between Glu-168 and Lys-65. Our results suggests that these two amino acid side chains are in contact with each other in the acto-gizzard HMM rigor complex and participate in the electrostatic interaction between the two HMM heads bound to F-actin. Based on the head-to-head contact, we propose a three-dimensional model for the attachment of gizzard HMM heads to F-actin.

Evidence for the association between two myosin heads in rigor acto-smooth muscle heavy meromyosin.

The rigor complexes that formed between rabbit skeletal muscle F-actin and chicken gizzard heavy meromyosin (HMM), in which the heavy chains had been cleaved with trypsin into 24K, 50K, and 68K fragments, were examined by using the zero-length chemical cross-linker 1-ethyl-3-[3-(dimethylamino)propyl]carbodiimide (EDC). Two cross-linked products of approximate Mr 115K and 60K were generated. These products were not obtained by EDC treatment of HMM in the absence of F-actin. The HMM fragments that participated in cross-linking were identified by fluorescent labeling and amino acid composition studies. The 115K peptide was determined to be a covalently cross-linked complex that formed between actin and the COOH-terminal 68K fragment of the HMM heavy chain. Our results are in agreement with a previous study which proposed that the site of cross-linking between HMM and F-actin resides within the COOH-terminal 22K fragment of the myosin subfragment 1 heavy chain [Marianne-Pépin, T., Mornet, D., Bertrand, R., Labbé, J.-P., & Kassab, R. (1985) Biochemistry 24, 3024-3029]. The 60K peptide, however, was not a product of cross-linking between HMM and F-actin. On the basis of its amino acid composition, we concluded that this 60K peptide was a cross-linked dimer of the NH2-terminal 24K fragments of the HMM heavy chain. The cross-linking of acto-gizzard HMM significantly increased the Mg-ATPase activity of gizzard HMM without any observable phosphorylation of the regulatory (20K) light chains.(ABSTRACT TRUNCATED AT 250 WORDS)

Carbodiimide-catalyzed cross-linking sites in the heads of gizzard heavy meromyosin attached to F-actin.

In the rigor complex between rabbit skeletal muscle F-actin and chicken gizzard heavy meromyosin (HMM), the direct contact between two HMM heads was demonstrated by using a zero-length cross-linker 1-ethyl-3-[3-(dimethylamino)propyl]maleimide (EDC) [Onishi, H., Maita, T., Matsuda, G., & Fujiwara, K. (1989) Biochemistry (preceding paper in this issue)]. Here, the 60K peptide which was a product of the EDC cross-linking between two 24K heavy chain (tryptic) fragments of HMM was further fragmented with cyanogen bromide, and the location of the cross-linking sites on the amino acid sequence of the HMM heavy chain was investigated. The result showed that one site resided within the 77-residue peptide region (residues 1-77) on one head of HMM, whereas the other site belonged to the 40-residue peptide region (residues 164-203) on the other head. This finding suggests that the two HMM heads are in contact with each other at different sites. Ultracentrifugal fractionation revealed that the head-to-head cross-linked gizzard HMM could be reversibly released from F-actin in the presence of Mg-ATP. The yield of the head-to-head cross-linking was not significantly changed with the acto-HMM complex between actin/HMM head molar ratios of 1 and 4, and it was very slightly decreased even at a molar ratio of 8, where HMM molecules were attached sparsely to actin filaments.(ABSTRACT TRUNCATED AT 250 WORDS)

Amino acid sequence of the regulatory light chain of clam foot muscle myosin.

The amino acid sequence of the regulatory light chain of foot muscle myosin from surf-clam (Spisula sachalinensis) was determined by conventional methods. It was: xS-D-D-K-K-A-K-A-A-T-S-S-V-L-T-K-F-T-Q-N-Q-I-Q-E-M-K-E-A-F-T-M-I-D-Q-N-R -D-G-L- I-D-V-S-D-L-K-E-M-Y-S-N-L-G-T-A-P-Q-D-S-V-L-Q-A-M-V-K-E-A-P-Q-M-N-F-T-G- F-L-S-L- F-S-E-K-M-S-G-T-D-P-E-E-T-L-R-N-A-F-Q-M-F-D-S-D-N-T-G-Y-I-P-E-E-Y-M-K-D- L- L-E-N-M-G-D-N-F-S-K-D-E-V-R-Q-T-W-K-E-A-P-I-A-G-G-K-V-D-Y-N-A-F-V-S-K-I- K- G-K-E-Q-D-D-A. The alpha-amino group of the light chain was blocked, and a typical calcium binding structure was recognized at the 33rd to 44th residues, as in other myosin regulatory light chains. The sequences of regulatory light chains from various muscle myosins were arranged according to the well known four-domain structure, and structural homologies were obtained for each of the domains. Based on the homologies, the relationships between the structure, function, and molecular evolution were discussed.

Amino acid sequence of the essential light chain of adductor muscle myosin from Ezo giant scallop, Patinopecten yessoensis.

The amino acid sequence of the essential light chain (abbreviated as SHLC) of adductor muscle myosin from Ezo giant scallop (Patinopecten yessoensis) was determined by conventional methods. The light chain was composed of 156 amino acid residues with proline and lysine as its amino and carboxyl termini, respectively. Comparing this sequence with that of the SHLC from bay scallop (Aquipecten irradians), only 5 amino acid substitutions were recognized. The sequence homology between scallop and squid SHLCs was 53.7%. On the other hand, a partially fragmented SHLC "modified SHLC" reported by Konno and Watanabe (J. Biochem. 98, 141-148 (1985) was prepared by chymotryptic digestion of the scallop myosin in the presence of EDTA, and was assigned as the carboxyl-terminal 106-residue peptide of the SHLC. This may suggest that the regulatory light chain covers the amino-terminal region of the SHLC in the myosin molecule.

Amino acid sequence of the regulatory light chain of squid mantle muscle myosin.

The amino acid sequence of the regulatory light chain of mantle muscle myosin from squid (Todarodes pacificus) was determined by conventional methods. It was: xA-E-E-A-P-R-R-V-K-L-S-Q-R-Q-M-Q-E-L-K-E-A-F-T-M-I-D-Q-D-R-D-G-F-I-G-M- E-D-L-K-D-M-F-S-S-L-G-R-V-P-P-D-D-E-L-N-A-M-L-K-E-C-P-G-Q-L-N-F-T- A-F-L-T-L-F-G-E-K-V-S-G-T-D-P-E-D-A-L-R-N-A-F-S-M-F-D-E-D-G-Q-G-F-I-P- E-D-Y-L-K-D-L-L-E-N-M-G-D-N-F-S-K-E-E-I-K-N-V-W-K-D-A-P-L-K-N-K-Q-F- N-Y-N-K-M-V-D-I-K-G-K-A-E-D-E-D. The alpha-amino group of this light chain was blocked, and a typical calcium-binding structure was recognized at the sequence of residue 26 to residue 37, like those in other myosin regulatory light chains.

Amino acid sequence of the amino-terminal 24 kDa fragment of the heavy chain of chicken gizzard myosin.

Chicken gizzard myosin was modified with N-iodoacetyl-N'-(5-sulfo-1-naphthyl)-ethylenediamine (IAEDANS) in the presence of ATP and in 0.15 M KCl, where the myosin assumed 10S conformation. From the tryptic digest of the modified myosin, a fluorescent fragment (24 kilodaltons) was isolated by gel filtration on a Sephadex G-100 column followed by chromatography on a CM 52 column. The amino acid sequence of the fragment was analyzed by conventional methods, and was: (S,Z)K-P-L-S-D-D-E-K-F-L-F-V-D-K-N-F-V-N-N-P-L-A-Q-A-D-W-S-A-K-K- L-V-W-V-P-S-E-K-H-G-F-E-A-A-S-I-K-E-E-K-G-D-E-V-T-V-E-L-Q-E-N-G-K-K- V-T-L-S-K-D-D-I-Q-K-M-N-P-P-K-F-S-K-V-E-D-M-A-E-L-T-C-L-N-E-A-S-V-L- H-N-L-R-E-R-Y-F-S-G-L-I-Y-T-Y-S-G-L-F-C-V-V-I-N-P-Y-K-Q-L-P-I-Y-S-E-K-I- I-D-M-Y-K-G-K-K-R-H-E-M-P-P-H-I-Y-A-I-A-D-T-A-Y-R-S-M-L-Q-D-R-E-D-Q- S-I-L-C-T-G-E-S-G-A-G-K-T-E-N-T-K-K-V-I-Q-Y-L-A-V-V-A-S-S-H-K-G-K. The amino-terminus was blocked, and the fragment was assigned as an amino-terminal part of the heavy chain of gizzard myosin. Position 127 was occupied by epsilon-N-trimethyllysine. Trp-130 of rabbit skeletal myosin heavy chain, which was reported to cross-link to an azide derivative of ATP by Okamoto and Yount (Proc. Natl. Acad. Sci. U.S. 82, 1575-1579 (1985], was replaced by glutamine in gizzard myosin. Cys-93 of the fragment is the amino acid residue whose reaction with IAEDANS alters the ATPase activity of gizzard myosin (Onishi, H. (1985) J. Biochem. 98, 81-86).

The primary structure of the myosin head.

The sequence of the NH2-terminal 808 amino acid residues of chicken pectoralis muscle myosin head was determined. Three characteristic 20-, 23-, and 50-kDa fragments were isolated from a digest of myosin subfragment 1 (S1) by gel filtration on a Sephadex G-100 column in the presence of 5 M guanidine hydrochloride, followed by anion-exchange chromatography on a QAE-Sephadex A-50 column in the presence of 8 M urea. The fragments were sequenced completely by conventional methods. Peptides overlapping the 23- and 50-kDa fragments and also overlapping the 50- and 20-kDa fragments were obtained by cleaving S1 with cyanogen bromide. Comparison of the 23-kDa and 50-kDa sequences with that of the overlapping peptide indicated that no additional amino acid exists between the 23- and 50-kDa fragments and that 5 amino acids exist between the 50- and 20-kDa fragments of S1. Methylated amino acid residues were found at four positions: epsilon-N-monomethyllysine at position 35, epsilon-N-trimethyllysine residues at 130 and 550, and 3-N-methylhistidine at 754.

Amino acid sequence of the 203-residue fragment of the heavy chain of chicken gizzard myosin containing the SH1-type cysteine residue.

A fluorescent fragment of Mr = 23,800 was obtained by the papain digestion of N-iodoacetyl-N'-(5-sulfo-1-naphthyl)ethylene diamine (abbreviated as IAEDANS)-modified chicken gizzard myosin. The fragment was isolated by gel filtration on a Sephadex G-100 column in the presence of 5 M guanidine-HCl followed by anion exchange chromatography on a QAE Sephadex A-50 column. This fragment contained 203 amino acid residues which could be assigned as a COOH-terminal part of the S-1 heavy chain based on the homology with the known sequence of rabbit skeletal myosin fragment. The amino acid sequence was K-G-M-F-R-T-V- G-Q-L-Y-K-E-Q-L-T-K-L-M-T-T-L-R-N-T-N-P-N-F-V-R-C-I-I-P-N-H-E-K-R-A- G-K-L-D-A-H-L-V-L-E-Q-L-R-C-N-G-V-L-E-G-I-R-I-C-R-Q-G-F-P-N-R-I-V-F-Q- E-F-R-Q-R-Y-E-I-L-A-A-N-A-I-P-K-G-F-M-D-G-K-Q-A-C-I-L-M -I-K-A-L-E-L- D-P-N-L-Y-R-I-G-Q-S-K-I-F-F-R-T-G-V-L-A-H-L-E-E-E-R-D-L-K- I-T-D-V-I-I-A- F-Q-A-Q-C-R-G-Y-L-A-R-K-A-F-A-K-R-Q-Q-Q-L-T-A-M-K-V-I-Q-R-N-C-A -A-Y-L-K-L-R-N-W-Q-W-W-R-L-F-T-K-V-K-P-L-L-Q-V-T-R. The cysteine residue which was modified with IAEDANS was of the SH1 type (Cys-65). Pro-197 was suggested to be the NH2-terminal boundary of the alpha-helical coiled-coil rod sequence of gizzard myosin, based on the homology with the nematode sequence reported by MacLachlan and Karn (Proc. Natl. Acad. Sci. U.S. 80, 4253-4257 (1983)). Three different COOH-terminal peptides (Val-Lys-Pro-Leu-Leu-Gln-Val-Thr-Arg, Val-Lys-Pro-Leu-Leu-Gln, and Val-Lys-Pro-Leu-Leu) were isolated from the tryptic digest of this fragment.(ABSTRACT TRUNCATED AT 400 WORDS)

Amino-acid sequence of the alpha and beta chains of adult hemoglobin of the harbor seal, Phoca vitulina.

The complete amino-acid sequences of the alpha and beta chains of adult hemoglobin of harbor seal, Phoca vitulina that belong to carnivora were determined as follows. The alpha and beta chains isolated by chromatography on a CM-cellulose column were digested with trypsin after S-carboxymethylation. Amino-acid sequences of the tryptic peptides derived from both chains were analysed. Comparing the primary structures of the alpha and beta chains of the seal hemoglobin with those of human, dog, bear, badger and cat, 19, 12, 12, 11, and 16 substitutions, respectively, were recognized in the alpha chain, and 12, 10, 4, 6, and 19 (22) in the beta chain.