Recurrence rate of intramucosal gastric cancer with positive vertical margin due to lesion damage during endoscopic submucosal dissection.

BACKGROUND AND STUDY AIM: In principle, additional surgery is performed after endoscopic submucosal dissection for early gastric cancer if the vertical margin is positive, regardless of lesion damage. The recurrence rate of vertical margin-positive lesions due to lesion damage after endoscopic submucosal dissection is unknown, and unnecessary surgeries may be performed. In this study, we investigated whether there was a difference in the recurrence rate between vertical margin-positive lesions due to lesion damage and vertical margin-negative lesions. PATIENTS AND METHODS: We included 1,294 intramucosal gastric cancer lesions that were resected by endoscopic submucosal dissection between January 2008 and December 2016, without additional surgery. The lesions were divided into the Damage and No damage groups based on vertical margin status. The Damage group had only one non-curative indication: a positive vertical margin due to lesion damage. The No damage group had no non curative indications. We compared the recurrence rate between the Damage and No damage groups. RESULTS: The recurrence rates of the Damage and No damage groups were 0% (0/23; 95% confidence interval: 0-14.8%) and 0% (0/1,271; 95% confidence interval: 0-0.003%), respectively, with no statistically significant difference. CONCLUSIONS: In intramucosal gastric cancer, the recurrence rate of vertical margin-positive lesions due to lesion damage was 0%, which did not differ from that of vertical margin-negative lesions with curative resection. Follow-up, instead of additional surgery, may be an option for patients with non-curative resection when the only non-curative indication is a positive vertical margin due to lesion damage.

Angiopoietin-like protein 2 promotes chondrogenic differentiation during bone growth as a cartilage matrix factor.

OBJECTIVE: Chondrocyte differentiation is crucial for long bone growth. Many cartilage extracellular matrix (ECM) proteins reportedly contribute to chondrocyte differentiation, indicating that mechanisms underlying chondrocyte differentiation are likely more complex than previously appreciated. Angiopoietin-like protein 2 (ANGPTL2) is a secreted factor normally abundantly produced in mesenchymal lineage cells such as adipocytes and fibroblasts, but its loss contributes to the pathogenesis of lifestyle- or aging-related diseases. However, the function of ANGPTL2 in chondrocytes, which are also differentiated from mesenchymal stem cells, remains unclear. Here, we investigate whether ANGPTL2 is expressed in or functions in chondrocytes. METHODS: First, we evaluated Angptl2 expression during chondrocyte differentiation using chondrogenic ATDC5 cells and wild-type epiphyseal cartilage of newborn mice. We next assessed ANGPTL2 function in chondrogenic differentiation and associated signaling using Angptl2 knockdown ATDC5 cells and Angptl2 knockout mice. RESULTS: ANGPTL2 is expressed in chondrocytes, particularly those located in resting and proliferative zones, and accumulates in ECM surrounding chondrocytes. Interestingly, long bone growth was retarded in Angptl2 knockout mice from neonatal to adult stages via attenuation of chondrocyte differentiation. Both in vivo and in vitro experiments show that changes in ANGPTL2 expression can also alter p38 mitogen-activated protein kinase (MAPK) activity mediated by integrin α5ß1. CONCLUSION: ANGPTL2 contributes to chondrocyte differentiation and subsequent endochondral ossification through α5ß1 integrin and p38 MAPK signaling during bone growth. Our findings provide insight into molecular mechanisms governing communication between chondrocytes and surrounding ECM components in bone growth activities.

Randomized study of orally administered fluorinated pyrimidines (capecitabine versus S-1) in women with metastatic or recurrent breast cancer: Japan Breast Cancer Research Network 05 Trial.

PURPOSE: Capecitabine and S-1 are orally administered fluorinated pyrimidines with high-level activity against metastatic breast cancer (MBC). This randomized, multicenter, phase II study compared the activities and safeties of the oral fluoropyrimidines, capecitabine and S-1, in breast cancer patients. METHODS: Patients with MBC were randomly assigned to receive capecitabine 825 g/m(2) twice daily on days 1-21 every 4 weeks or S-1 40-60 mg twice daily, according to body surface area, on days 1-28 every 6 weeks. The primary endpoint was progression-free survival (PFS). RESULTS: A total of 142 patients were enrolled and randomized to either capecitabine (N = 73) or S-1 (N = 69). Median PFS (progression-free survival) was 1.2 years for capecitabine and 1.3 years for S-1, with a hazard ratio (S-1/capecitabine) of 0.85 (95 % confidence interval [CI] 0.52-1.38) (P = 0.48 by log-rank). The confirmed objective response rates were 24.0 % for capecitabine and 23.1 % for S-1 (P = 0.938). The most common treatment-related adverse events were grade 1-2 in intensity. Thrombocytopenia (S-1: 9.2 %, capecitabine: 1.4 %; P = 0.040) and nausea (S-1: 26.2 %, capecitabine: 14.1 %; P = 0.079) were more frequent in the S-1 group, while hand-foot syndrome occurred more often in the capecitabine group (S-1: 10.8 %, capecitabine: 25.4 %; P = 0.029). CONCLUSIONS: The results of the current study demonstrate that both S-1 and capecitabine are effective and well-tolerated treatments in patients with MBC, while their adverse events were different. They are both convenient, orally administered drugs, making them attractive agents for use in outpatient treatment.

Phase II trial of preoperative chemotherapy for breast cancer: Japan Breast Cancer Research Network (JBCRN)-02 trial.

BACKGROUND: Neoadjuvant chemotherapy (NAC) is one of the main strategies for patients with locally advanced breast cancer. In our previous study, biological markers such as estrogen receptor (ER), progesterone receptor (PgR), and HER2 were essential predictors of the effectiveness of NAC to help individualize treatment. This study examined the effect of NAC on the disease-free survival (DFS) of breast cancer patients. Furthermore, the study was expanded by adding Ki-67 as a biological marker, and examined the correlation between Ki-67 and the prognosis. PATIENTS AND METHODS: Between September 2005 and September 2007, 43 patients with breast cancer received NAC and surgery. Four cycles of DC (doxorubicin: 60 mg/m(2) and cyclophosphamide: 500 mg/m(2)) were administered intravenously (i.v.) on day 1 every 21 days, followed by 12 cycles of paclitaxel i.v. (80 mg/m(2)) every 7 days, prior to surgery. The primary endpoint was the pathological complete response (pCR) rate and the secondary endpoint was DFS; the pCR rate was estimated for each groups stratified by the presence or absence of different factors (PcR, ER/PgR, and Ki-67). RESULTS: The clinical response (cCR+cPR) rate was 81.0%, and the pCR rate was 25.6%. The pCR rate was 75, 50, 9 and 0% in HER2(+)/ER(-), HER2(+)/ER(+), HER2(-)/ER(-), and HER2(-)/ER(+) patients, respectively. The 4-year DFS rate was estimated at 78% for all patients. The HER2 status was an independent predictor of pathological complete response (pCR). The DFS rate of patients with lower Ki-67 values (<15%) was higher than that of patients with higher Ki-67 values (≥15%). The treatment-related adverse events were manageable: the majority were mild, but five patients experienced grade 3 (neutropenia and sensory neuropathy) adverse events. CONCLUSION: DC followed by weekly paclitaxel is an active and manageable preoperative regimen for breast cancer patients. HER2 overexpression may be a good predictive marker of pCR, and the Ki-67 value after NAC may be a prognostic factor for DFS.

Efficacy of S-1 in patients with capecitabine-resistant breast cancer-Japan Breast Cancer Research Network (JBCRN) 04-1 trial.

BACKGROUND: S-1 is an orally administered fluorinated pyrimidine with high activity in metastatic breast carcinoma (MBC) and in chemotherapy-pretreated metastatic breast carcinoma. PATIENTS AND METHODS: Forty patients with MBC who did not respond to capecitabine-based chemo-therapy and then received S-1 were identified from our data base of records between 2006 and 2008. The clinico-pathological data and outcomes of these patients were then reviewed. RESULTS: The overall response rate was 27.8%. The median survival was 19.2 months, and the median time to disease progression was 6.2 months. The most common treatment-related adverse events (all grades) were hand-foot syndrome (15%), nausea (15%), vomiting (7.5%), disorder of taste (7.5%), and diarrhea (5%). However, the majority were mild to moderate in intensity, and only one patient experienced grade 3 (according to the National Cancer Institute of Canada Common Toxicity criteria) adverse events. Myelosuppression and alopecia were rare, and there were no reported treatment-related deaths. CONCLUSION: The results of the current study demonstrate that S-1 is an effective and well-tolerated treatment in patients with capecitabine-resistant MBC. In addition, it is a convenient, orally administered drug, which makes it an attractive agent for use in outpatient treatment.

Observation of the distribution of Zn protoporphyrin IX (ZPP) in Parma ham by using purple LED and image analysis.

We investigated the distribution of Zn protoporphyrin IX (ZPP) in Parma ham by using purple LED light and image analysis in order to elucidate the mechanism of ZPP formation. Autofluorescence spectra of Parma ham revealed that ZPP was present in both lean meat and fat, while red emission other than that of ZPP was hardly detected. Although ZPP was found to be distributed widely in Parma ham, it was more abundant in intermuscular fat and subcutaneous fat than in lean meat. The intensity of red emission was weak in muscles that were exposed during the processing. ZPP in both lean meat and subcutaneous fat tended to be more abundant in the inner region than in the outer region. It was thought that ZPP is transferred from lean meat to fat tissue during the processing, resulting in the small amount of ZPP in the lean meat adjacent to subcutaneous fat. Our results led to a completely new hypothesis that ZPP is formed in lean meat and transferred to fat tissue.

A new method to immortalize primary cultured rat hepatocytes.

BACKGROUND: In conventional methods of establishing hepatocyte cell lines, the immortalizing gene alone is introduced into hepatocytes. We designed a new method in which not only the immortalizing gene, the simian virus-40 large T-antigen (SV-40 Tag) gene, but also a drug-resistant gene, under the control of an albumin enhancer/promoter, were introduced into hepatocytes to efficiently obtain immortalized hepatocyte cell lines. METHODS: The plasmid pAPUR contains the puromycin-resistant gene under the control of an albumin enhancer/promoter, and the pSVTag contains the early region of SV-40 enhancer/promoter and the SV-40 Tag gene. Both pAPUR and pSVTag were transferred into isolated rat hepatocytes by electroporation. After these cells were cultured on a collagen-coated dish for 24 hours, puromycin selection was started. Expression levels of albumin, alpha-fetoprotein (AFP), SV-40 Tag, and cytokeratin 19 (CK 19) in the transformed cells were evaluated by western analysis, immunocytochemical staining, and RT PCR. RESULTS: Approximately 3 weeks after transfection, five or six colonies appeared on the dish. Twenty strains were obtained by cloning these cells. All strains that were similar to immature hepatocytes expressed albumin and SV-40 Tag, although CK 19 was not detected. AFP expression was detected in 33% of these strains. CONCLUSIONS: All clones cotransfected by pAPUR and pSVTag expressed albumin. Our new method may be useful to establish hepatocyte cell lines.

Function of the human insulin promoter in primary cultured islet cells.

Pancreatic islet beta cells regulate the rate of insulin gene transcription in response to a number of nutrients, the most potent of which is glucose. To test for its regulation by glucose, the promoter sequence was isolated from the human insulin gene. When linked to chloramphenicol acetyltransferase and transfected into primary islet cultures, the human insulin promoter is activated by glucose. In parallel islet transfections, glucose also activates the L-pyruvate kinase and islet amyloid chain ketoacid dehydrogenase E1a promoter, but it does not affect the beta cell glucose kinase promoter. Using deletion and substitution mutations of the proximal human insulin promoter, we mapped a metabolic response element to the E box, E1, at -100 base pairs relative to the transcription start site. Although the isolated E1 element responds to glucose, inclusion of either of two AT-rich sequences, A1 or A2/C1 on either side of E1, results in dramatic synergistic activation. Inclusion of A2/C1 also increases the response to glucose. The A2-E1-A1 region alone, however, does not explain all of the activity of the human insulin promoter in cultured islets, and other transcriptionally important elements likely to contribute to the glucose response as well.

Pancreatic beta cells express a diverse set of homeobox genes.

Homeobox genes, which are found in all eukaryotic organisms, encode transcriptional regulators involved in cell-type differentiation and development. Several homeobox genes encoding homeodomain proteins that bind and activate the insulin gene promoter have been described. In an attempt to identify additional beta-cell homeodomain proteins, we designed primers based on the sequences of beta-cell homeobox genes cdx3 and lmx1 and the Drosophila homeodomain protein Antennapedia and used these primers to amplify inserts by PCR from an insulinoma cDNA library. The resulting amplification products include sequences encoding 10 distinct homeodomain proteins; 3 of these proteins have not been described previously. In addition, an insert was obtained encoding a splice variant of engrailed-2, a homeodomain protein previously identified in the central nervous system. Northern analysis revealed a distinct pattern of expression for each homeobox gene. Interestingly, the PCR-derived clones do not represent a complete sampling of the beta-cell library because no inserts encoding cdx3 or lmx1 protein were obtained. Beta cells probably express additional homeobox genes. The abundance and diversity of homeodomain proteins found in beta cells illustrate the remarkable complexity and redundancy of the machinery controlling beta-cell development and differentiation.

Separation of X- and Y-chromosome-bearing murine sperm by free-flow electrophoresis: evaluation of separation using PCR.

The effectiveness of separation of murine X- and Y-bearing sperm by free-flow electrophoresis was evaluated by the polymerase chain reaction (PCR). The ratio of X- and Y-bearing sperm from cauda epididymis was analyzed before and after free-flow electrophoresis. A Y-chromosome-specific sequence (pY353/B) and an autosomal sequence (myogenin) were used to estimate the ratio between X- and Y-sperm in the separated fractions. Cauda epididymal mice sperm were separated into two peak fractions under the electric field. Each peak fraction contained sperm of normal shape, however, the motility of the sperm was extremely diminished after separation by electrophoresis. DNA was extracted from 10(4) sperm from each fraction and from the unseparated sperm, and Y-chromosome specific PCR was performed. The PCR experiment revealed that fraction No. 16 (the peak near the cathode) was a Y-sperm rich fraction, whereas fraction No. 22 (the peak near the anode) was a Y-sperm poor one. These results suggested that murine X- and Y-sperm could be successfully separated by free-flow electrophoresis. Analysis of the chromosome-specific sequence by PCR was demonstrated to be a direct and adequate method to evaluate the separation of X- and Y-sperm.

K-sam-related gene, N-sam, encodes fibroblast growth factor receptor and is expressed in T-lymphocytic tumors.

We recently reported the isolation of the K-sam complementary DNA (cDNA), which was amplified preferentially in poorly differentiated types of stomach cancer and codes for one of the heparin-binding growth factor or fibroblast growth factor (FGF) receptor families. The K-sam-related gene, N-sam (NCC-IT-cell-derived sam), was isolated by screening of the cDNA libraries of human immature teratoma cells, NCC-IT. Sequence analysis of the N-sam cDNAs showed that N-sam encodes a human FGF receptor, the FLG protein. N-sam was expressed in lymphocytic leukemia/lymphoma cells, predominantly in the thymic T-cell phenotype. In a T-cell leukemia line, MOLT3, N-sam mRNA expression was markedly enhanced by 12-O-tetradecanoylphorbol-13-acetate treatment and was also up-regulated by basic FGF exposure. These results indicate that N-sam expression is regulated during T-cell ontogeny and modulated by its putative ligand exposure. The results also suggested that interaction between immature T-cell and marrow or thymic interstitial cells might be mediated by N-sam and basic FGF stored in the extracellular matrix of stromal cells.

[Clinical evaluation of four tumor markers (CEA, TPA, CA50 and CA72-4) in colorectal cancer].

In a total of 194 cases, consisting of 86 cases of colorectal cancer undergone operation later, 7 cases of non resectable cancer, 34 cases of recurrence, 43 cases of NED (no evidence of a recurrence after radical surgery for colorectal cancer) and 24 cases of benign colorectal disease, serum CEA, TPA, CA50 and CA72-4 levels were determined. The positivity rate was high for all four markers in stage V cases among 86 cases of colorectal cancer, and in cases of non resectable cancer and cases of recurrence. The highest positive rate was obtained with CEA. On the contrary, in cases of stage I to IV the positivity rates of these four tumor markers were as low as 0 to 34.8%. Out of 127 cases of colorectal cancer excluded of 43 NED cases, 52 cases were negative for all four tumor markers and 14 cases were positive only for CEA. In 49 cases, CEA and at least one of the other three tumor markers gave positive results. In 12 cases, CEA gave negative results and at least one of the other three tumor markers positive results. In conclusion, measurement of blood levels of these tumor markers is limited of its usefulness in early diagnosis of colorectal cancer. However, in the diagnosis of advanced stage and of recurrence during the postoperative follow-up period the measurement of tumor markers provides useful information. CEA is most sensitive among the four tumor markers tested and any combination of these four markers is not advantageous because of an increase in false positivity rate.

[Investigation of 5-fluorouracil (5-FU) levels and pyrimidine nucleoside phosphorylase activities in the tissues from patients with uterine cervical and ovarian cancers after oral administration of 5'-DFUR].

The usefulness of 5'-DFUR in both patients with uterine cervical and ovarian cancers was investigated by determining pyrimidine nucleoside phosphorylase (PyNPase) activities and 5-FU levels in cancerous and normal tissues resected from them after oral administration of 5'-DFUR. In uterine cervical cancer, each group of 9 cases administered single dose of 400 mg of 5'-DFUR and 7 cases administered 400 mg of 5'-DFUR 3 times a day continuously for 7 days was investigated. In ovarian cancer, all of 9 cases were administered 400 mg of 5'-DFUR 3 times a day continuously for 7 days. In conclusion, PyNPase activities in the tissues of uterine cervical and ovarian cancers were higher than those in the normal tissues. 5-FU tissue levels in the cancerous tissues were significantly higher than in the normal tissues and blood as well. This tendency was observed in each of the single and continuous administration groups. These results suggest that the tumor selectivity which is one of characteristics of 5'-DFUR could be expected also for cancer in the field of gynecology.

Effects of orally administered 5-fluorouracil and its derivative 5'-deoxy-5-fluorouridine on 7,12-dimethylbenz[a]anthracene-induced breast carcinomas in rats.

Antitumour activity of 20 mg/kg 5-fluorouracil (5-FU) and 100 or 200 mg/kg 5'-deoxy-5-fluorouridine (5'-DFUR) was assessed by tumour regression, light and electron microscopic changes, and immunohistochemical variations in broxuridine labelling in 7,12-dimethyl-benz[a]anthracene- (DMBA-) induced breast carcinomas in rats. The relative regression of tumours was significantly (P less than 0.05) greater in animals receiving 5-FU or 5'-DFUR than in control animals. Light microscopic changes became apparent after 3 or 4 days' treatment and were more evident after 4-21 days. Electron microscopic degeneration of tumours was observed after 1 day. The broxuridine labelling index decreased significantly (P less than 0.05) in rats receiving 5-FU and especially in those receiving 5'-DFUR. Concentrations of 5-FU in DMBA-induced breast carcinomas were significantly (P less than 0.05) higher in rats treated with 5'-DFUR than in those treated with 5-FU, and the histological degeneration was significantly (P less than 0.05) more advanced in 5'-DFUR- than in 5-FU-treated animals, possibly due to the higher pyrimidine nucleoside phosphorylase activity in the former treatment group.

K-sam, an amplified gene in stomach cancer, is a member of the heparin-binding growth factor receptor genes.

DNA fragments amplified in a stomach cancer-derived cell line, KATO-III, were previously identified by the in-gel DNA renaturation method, and a 0.2-kilobase-pair fragment of the amplified sequence was subsequently cloned. By genomic walking, a portion of the exon of the gene flanking this 0.2-kilobase-pair fragment was cloned, and the gene was designated as K-sam (KATO-III cell-derived stomach cancer amplified gene). The K-sam cDNAs, corresponding to the 3.5-kilobase K-sam mRNA, were cloned from the KATO-III cells. Sequence analysis revealed that this gene coded for 682 amino acid residues that satisfied the characteristics of the receptor tyrosine kinase. The K-sam gene had significant homologies with bek, FLG, and chicken basic fibroblast growth factor receptor gene. The K-sam gene was amplified in KATO-III cells with the major transcript of 3.5-kilobases in size. This gene was also expressed in some other stomach cancer cells, a small cell lung cancer, and germ cell tumors.

Enhancing effect of bromovinyldeoxyuridine on antitumor activity of 5'-deoxy-5-fluorouridine against adenocarcinoma 755 in mice. Correlation with pharmacokinetics of plasma 5-fluorouracil levels.

5'-Deoxy-5-fluorouridine (DFUR), whether or not combined with (E)-5-(2-bromovinyl)-2'-deoxyuridine (BVDU) was pursued in BDF1 mice from both a pharmacokinetic viewpoint, following a single oral dose administration, and an anticancer viewpoint, following 5 daily oral doses in mice inoculated subcutaneously with adenocarcinoma 755 tumor cells. Half-life (t1/2) values for the elimination of DFUR and 5-fluorouracil (5-FU) from plasma following DFUR (100 mg/kg) administration were about 0.80 and 0.39 hr, respectively. Plasma 5-FU AUC (area under the curve) values following oral DFUR (100 mg/kg) was 0.224 micrograms.hr/ml. If DFUR (100 mg/kg) was combined with BVDU (10 mg/kg) the t1/2 and AUC values for 5-FU increased from 0.39 to 1.24 hr, and from 0.224 to 1.699 micrograms.hr/ml, respectively. Thus, BVDU significantly increased the plasma levels of 5-FU. It had no effect on the plasma levels of DFUR. At 100 mg/kg, DFUR did not show a significant antitumor activity. At 500 mg/kg it effected a 90% inhibition in tumor growth. When combined with BVDU (10 mg/kg), DFUR at 100, 200 and 300 mg/kg reduced tumor growth by 96, 100 and 100%, respectively. The antitumor activity achieved by DFUR, in the presence or absence of BVDU, correlated highly significantly with the AUC values for plasma 5-FU.

Determination of 5-fluorouracil in plasma and liver after oral administration of 5'-deoxy-5-fluorouridine using gas chromatography-mass spectrometry.

A method for determination of 5-fluorouracil (5-FU) has been established for pharmacokinetic study of 5'-deoxy-5-fluorouridine (5'-DFUR), a newly developed prodrug of 5-FU. 5-FU was extracted with diethyl-ether containing 1-propanol and methylated with CH3I in the presence of (CH3)4NOH. The methylated 5-FU was analyzed by gas chromatography-mass spectrometry using 15N2-5-FU as an internal standard. Quantitation was carried out by selected-ion monitoring of molecular ions of N,N-dimethyl-5-FU (m/z 158) and the internal standard (m/z 160). The sensitivity (greater than 2 ng/g sample), specificity and precision of the method were satisfactory for application to clinical studies of this drug. After an oral administration of 5'-DFUR (800 mg/body) to patients with cancer, 5-FU in plasma peaked within 1 h and eliminated with an apparent half life of about 1 h.

[Studies on hst, a transforming gene which belongs to a new superfamily of growth factors].

The hst gene was originally identified in surgically obtained human gastric mucosae as a transforming gene which could transform NIH3T3 cells morphologically. The hst cDNA clone was synthesized from mRNA of one of the NIH3T3 transformants. A human leukocyte genomic library was screened with this cDNA clone, and an hst genomic fragment was obtained. This genomic fragment itself had transforming activity, and the protein coding sequences were proved to be completely identical to those of the cDNA clone prepared from mRNA of the NIH3T3 transformant. This fact suggests that rearrangement or other structural alterations in the coding sequence are not required for the activation of the hst gene. The predicted hst protein consists of 206 amino acids and has a significant homology (40-50%) to fibroblast growth factors and int-2 protein. They together make up a new superfamily of growth factors and transforming genes.

Cloned hst gene from normal human leukocyte DNA transforms NIH3T3 cells.

The hst gene was originally identified as a transforming gene in DNAs from stomach cancers and a noncancerous portion of stomach mucosa by transfection assays using NIH3T3 cells (1,2). Subsequently, the hst gene obtained directly from leukocyte DNA of a leukemia patient was sequenced (3,4). Here, cosmid clones containing the hst gene were isolated directly from normal human leukocyte DNA and from T361-2nd-1 cells, a secondary transformant of NIH3T3 cells induced by transfection of DNA from a stomach cancer. All clones containing the hst gene from these different sources transformed NIH3T3 cells with similar efficiency. Restriction map of the hst gene from normal leukocyte DNA was identical with that from leukocyte DNA of a leukemia patient, while the hst gene from T361-2nd-1 cells was rearranged at the 168th nucleotide upstream of the TATA box.

Genomic sequence of hst, a transforming gene encoding a protein homologous to fibroblast growth factors and the int-2-encoded protein.

hst is a transforming gene first identified from transformed NIH 3T3 cells that were transfected with DNA of a human stomach cancer. A genomic fragment of hst obtained directly from a human genomic library also has transforming activity. This fragment has a coding sequence identical to that of the hst cDNA prepared from an NIH 3T3 transformant induced by DNA from a stomach cancer. The deduced amino acid sequence of the hst protein is 43%, 38%, and 40% homologous, respectively, to human basic fibroblast growth factor, human acidic fibroblast growth factor, and mouse int-2 protein in selected regions. This suggests that hst encodes a protein related to fibroblast growth factors, which are wide-spectrum mitogens, and to the int-2 protein, a potential oncogene product implicated in murine mammary carcinogenesis.