Association of HLA-E single nucleotide polymorphisms with human myeloid leukemia.

Single nucleotide polymorphisms (SNPs) of HLA-E are related to the occurrence of many diseases, but their functions remain unclear. In this study, the function of SNPs at HLA-E rs76971248 and rs1264457 on the myeloid leukemia cells was analyzed by a progressive procedure, included genotyping, mRNA transcription, regulatory element, protein expression, and anti-tumor effect. The frequencies of rs76971248 G and rs1264457 G were found higher in myeloid leukemia patients than those in healthy blood donors (p < 0.05). For myeloid leukemia, rs76971248 T was protective, while rs1264457 G was susceptible. We also found that rs76971248 affected HLA-E mRNA transcription and membrane HLA-E (mHLA-E) expression in K562 cells through differently binding to transcription factor HOXA5 (p < 0.0001), while rs1264457 affected mHLA-E expression by changing mRNA transcription and an encoding amino acid (p < 0.01). In contrast, the expression of soluble HLA-E (sHLA-E) was not influenced by both rs1264457 and rs76971248. The higher HLA-E expression was detected among myeloid leukemia patients, and the K562 cells with higher HLA-E molecules played a significant inhibitory effect on the killing activity of NK-92MI cells (p < 0.05). In conclusion, the higher HLA-E expression of myeloid leukemia cells is promoted by rs76971248 G and rs1264457 G, which helps escape from NK-92MI cells' killing.

Pseudomonas marianensis sp. nov., a marine bacterium isolated from deep-sea sediments of the Mariana Trench.

A novel marine Gram-stain-negative, aerobic, rod-shaped bacterium, designated as strain PS1T, was isolated from the deep-sea sediments of the Mariana Trench and characterized phylogenetically and phenotypically. Bacterial optimal growth occurred at 35 °C (ranging 10-45 °C), pH 6 (ranging pH 5-10) and with 11% (w/v) NaCl (ranging 0-17%). The 16S rRNA gene sequence similarity results revealed that strain PS1T was most closely related to Pseudomonas stutzeri ATCC 17588T, Pseudomonas nitrititolerans GL14T, Pseudomonas zhaodongensis NEAU-ST5-21T, Pseudomonas xanthomarina DSM 18231T and Pseudomonas kunmingensis HL22-2T with 98.3-98.7%. The digital DNA-DNA hybridization values and the average nucleotide identity between strain PS1T and the reference strains were 20.4-40.1% and 78.7-79.4%, respectively. The major respiratory quinone is ubiquinone Q-9. The major polar lipids were phosphatidylethanolamine, diphosphatidyglycerol, phosphatidylglycerol, phosphatidylcholine, aminoglycolipid, two unidentified glycolipids and one unidentified lipid. The predominant cellular fatty acids of strain PS1T were summed feature 8 (C18:1ω7c and/or C18:1ω6c), summed feature 3 (C16:1ω7c and/or C16:1ω6c), C16:0 and cyclo-C19:0 ω8c. The G + C content of the genomic DNA was 63.0%. The combined genotypic and phenotypic data indicated that strain PS1T represents a novel species of the genus Pseudomonas, for which the name Pseudomonas marianensis sp. nov. is proposed, with the type strain PS1T (= DSM 112238T = MCCC 1K05112T).

Establishment of Rapid Detection Methods for rs76971248 Related to Leukemia.

Background: The HLA-E gene is a member of the HLA-I gene family. Its genetic polymorphism is regarded as associated with numerous diseases. Establishing a rapid and accurate detection method of disease-related SNP sites in HLA-E is particularly important. Methods: Blood samples from 226 healthy blood donors and 228 leukemia patients were collected, and DNA was extracted. Three typing methods based on PCR-sequence-based typing, TaqMan genotyping, and high-resolution melting curve were established to identify rs76971248 (G>T). The Chi-square test was used for statistical analysis by SPSS. Results: Three methods based on PCR-SBT, TaqMan genotyping, and HRM were all able to identify rs76971248. The software for analyzing the results of HLA-E sequencing was easy to use, and the results were accurate. The frequency of rs76971248 in different types of leukemia patients was significantly lower than that in healthy blood donors (p < 0.05). And the frequency of the G/G genotype in leukemia patients was significantly higher than that in healthy blood donors (p < 0.05). Conclusions: For the screening of known SNP sites in large-scale populations, among the three methods, the TaqMan genotyping method had the advantage of shortest time consumption, simplest operation, and greatest specificity, which was the most appropriate method for this experiment. The analysis software for HLA-E gene sequencing needed to be further optimized. rs76971248 had a protective effect against leukemia. And the G/G genotype was a risk factor for leukemia.

[Study on the Relationship between the Level of Soluble HLA-E Molecules in Plasma and Gene Polymorphism and Leukemia].

OBJECTIVE: To explore the relationship between the level of soluble HLA-E (sHLA-E) molecules in plasma and gene polymorphism and leukemia in Shenzhen of China. METHODS: Enzyme-linked immunosorbent assay was used to detect sHLA-E level in plasma of 103 leukemia patients and 113 healthy blood donors. PCR-SBT was used to identify the HLA-E genotype of 73 leukemia patients and 76 healthy blood donors. RESULTS: The level of plasma sHLA-E of 103 leukemia patients was significantly higher than that of 113 healthy blood donors (P<0.001); And the level of plasma sHLA-E in 77 myeloid leukemia patients was also significantly higher (P<0.001). The percentage of patients with plasma sHLA-E concentration of 0-199 ng/ml in leukemia and myeloid leukemia patients was 37.86% and 32.47%, respectively, which was significantly lower than 53.98% of healthy donors, the difference was statistically significant (P<0.05, P<0.01); While, when the plasma sHLA-E concentration was more than 400 ng/ml, the percentage was 33.01% and 36.36%, respectively, which was significantly higher than 13.28% of healthy donors, the difference was also statistically significant (P=0.001, P<0.001). There was no significant difference in the level of plasma sHLA-E among different HLA-E genotypes (P>0.05), whether healthy blood donors or leukemia patients. CONCLUSION: The level of plasma sHLA-E in patients with leukemia (especially myeloid leukemia) is significantly higher than that of healthy blood donors, but different HLA-E genotypes do not affect the level of plasma sHLA-E. A cut-off value for the concentration of plasma sHLA-E (recommended risk value >400 ng/ml) can be set to assess the risk of certain pre-leukemia patients.

Correlation of Human Leukocyte Antigen-E Genomic Polymorphism with Leukemia and Functional Study of Human Leukocyte Antigen-E Different Type Promoters.

Human leukocyte antigen (HLA)-E is one of the least polymorphic nonclassical major histocompatibility complex (MHC) I genes; its nucleotide variability can affect immune response. In this study, we assess the correlation between HLA-E polymorphism and leukemia and further study the transcriptional activity of promoter variation at nucleotide position-26. A total of 142 healthy blood donors and 111 leukemia patients were collected. The genomic sequence of HLA-E was amplified by high-fidelity reaction system and identified by Sanger and cloning sequencing. The dual luciferase reporter gene assay was used to detect the transcription activity of promoter variation at nucleotide position-26. In the HLA-E genomic sequence results, a total of 16 alleles and 32 genotypes were detected; the HLA-E*01:01:01:06 allele had a significantly lower frequency in leukemia patients than in healthy participants (p = 0.026 < 0.05). And the HLA-E*01:03:02:01, *01:03:02:01 genotype showed the greatest difference in frequency between the two groups of participants (p = 0.028 < 0.05). Eight HLA-E alleles were first reported worldwide in Chinese individuals. The results of the dual luciferase reporter gene experiment showed that the transcription activity of the mutant-type promoter (HLA-E*01:01:01:06 with "T" allele at nucleotide position-26) was significantly lower compared with the wild-type promoter (HLA-E*01:01:01:01 with "G" allele at nucleotide position-26) (p = 0.0242 < 0.05). HLA-E*01:01:01:06 allele has a protective effect against leukemia through decreasing transcription activity by "T" variation at nucleotide position-26.

A Chinese leukemia patient, typed as HLA-E*01:01:01:11, a novel HLA-E allele.

There is a single nucleotide different in intron 1 between E*01:01:01:01 and E*01:01:01:11.

[Distribution of KIR/HLA alleles among ethnic Han Chinese patients with hepatocellular carcinoma from southern China].

OBJECTIVE: To assess the association of KIR/HLA alleles with hepatocellular carcinoma (HCC) and hepatitis B virus (HBV) infection among ethnic Han Chinese patients from southern China. METHODS: For 95 patients with HCC and 171 healthy controls, the genotype of HLA-C alleles was determined with a PCR sequence-specific oligonucleotides typing method on an Illumina GenDx NGSgo platform. Genotypes comprised of HLA-C and KIR gene alleles were also subjected to statistical analysis. RESULTS: In total 16 KIR genes (2DL2, 2DS2, 2DS3, 2DS5, 3DS1, 2DS1, 2DL5, 2DS4, 3DL1, 3DP1, 2DL3, 2DP1, 3DL3, 2DL1, 3DL2 and 2DL4) were discovered in the two groups. The frequencies of KIR2DL3 alleles and combinational genotypes of KIR2DL3/HLA-C1C2 were significantly lower in the patient group compared with the controls (0.9368 vs. 0.9883, χ²>3.84; P<0.05, OR = 0.1; 0.0112 vs. 0.2663, χ²>3.84; P<0.05, RR = 0.03). The frequency of HLA-C2C2 genotype of the patient group was significantly lower than that of the controls (0.0316 vs. 0.2690, P<0.05, RR = 0.09), while the frequency of HLA-C1C2 genotype was significantly higher than that of the controls (0.2316 vs. 0.0058, P<0.05, RR = 51.23). CONCLUSION: Above results suggested that the KIR2DL3 allele is associated with lower risk for HCC. There may be individual difference in patients with HCC and HBV infection but various combinations of KIR/HLA alleles.

Polyphasic taxonomic analysis of Parasedimentitalea marina gen. nov., sp. nov., a psychrotolerant bacterium isolated from deep sea water of the New Britain Trench.

A novel Rhodobacteraceae bacterium, strain W43T, was isolated from a deep-sea water sample from the New Britain Trench. Strain W43T exhibited the highest 16S rRNA gene sequence similarity of 96.5% to Sedimentitalea nanhaiensis DSM 24252T, Phaeobacter gallaeciensis DSM 26640T, Phaeobacter inhibens DSM 16374T, and Phaeobacter porticola P97T. Phylogenetic analysis of the 16S rRNA gene and phylogenomic analysis of the genome showed that strain W43T formed an independent monophyletic branch within the family Rhodobacteraceae. Strain W43T was Gram-stain-negative, rod-shaped, and grew optimally at 16-20°C, pH 6.5-7.0 and 2% (w/v) NaCl. The principal fatty acids were C18:1ω7c/C18:1ω6c, major respiratory quinone was ubiquinone-10 and predominant polar lipids were phosphatidylethanolamine, phosphatidylglycerol and diphosphatidylglycerol. The 5 080 916 bp long genome, comprising a circular chromosome and four plasmids, exhibits a G + C content of 55.9 mol%. The combined phenotypic, chemotaxonomic and molecular data show that strain W43T represents a novel species of a novel genus, for which the name Parasedimentitalea marina gen. nov. sp. nov. is proposed (type strain W43T = MCCC 1K03532T = KCTC 62635T).

Clinical significance of HLA-E genotype and surface/soluble expression levels between healthy individuals and patients with acute leukemia.

Human leukocyte antigen (HLA)-E is a nonclassical HLA molecule with limited polymorphisms. Genotype frequency and expression of HLA-E were examined here for the first time in acute leukemia patients and healthy controls. The frequency of HLA-E*01:03/*01:03 individuals was significantly higher (p = .008, OR = 1.845), while the frequency of HLA-E*01:01/*01:01 individuals was much lower in the patient group (p = .002, OR = .363) than in control group. The surface expression on HLA-E*01:03/*01:03 individuals was found to be significantly higher than on HLA-E*01:01/*01:01 individuals in both of acute leukemia and control groups, but no significant difference was observed between the corresponding genotypes in two groups. However, the level of expression of soluble HLA-E is significantly higher in patients than in the control group, but there was no genotype-specific expression in either group. These findings indicate that soluble HLA-E secretion and HLA-E*01:03/*01:03 genotype that brings higher surface expression might play important roles in the mechanisms underlying tumor escape in acute leukemia.

Allelic diversity of KIR3DL1/3DS1 in a southern Chinese population.

The inhibitory KIR3DL1 and the activating KIR3DS1 segregate as alleles of the same locus. KIR3DL1 is highly diversified at the allele level and KIR3DL1 alleles exhibit varied levels of expression and ligand binding affinity resulting in varied degrees of NK cell inhibition. Previous studies have shown that the KIR3DL1/3DS1 polymorphism associated with viral infection, cancer and transplantation. However, little is known about the population distribution of KIR3DL1/3DS1 alleles in Chinese. The present study examined allelic diversity of KIR3DL1/3DS1 in a southern Chinese population (N=306) using PCR-SSP and sequencing based typing. The presence of KIR3DL1 and KIR3DS1 were detected in 97.1% and 34.0% of the tested individuals respectively. A total of 10 KIR3DL1 alleles (including 2 novel ones) and 6 KIR3DS1 alleles (including 5 novel ones) were identified. Common KIR3DL1 alleles (>10%) were KIR3DL1*01502 (74.8%), KIR3DL1*00501 (23.9%) and KIR3DL1*00701 (15.7%). KIR3DS1*01301 was the predominant KIR3DS1 allele with other KIR3DS1 alleles only sporadically observed. The knowledge of the allelic polymorphism of KIR3DL1/3DS1 may help to better understand the role played by KIR3DL1/3DS1 in associated diseases and clinical transplantation in southern Chinese.

Optimal selection of natural killer cells to kill myeloma: the role of HLA-E and NKG2A.

Immunotherapy with allogeneic natural killer (NK) cells offers therapeutic perspectives for multiple myeloma patients. Here, we aimed to refine NK cell therapy by evaluation of the relevance of HLA-class I and HLA-E for NK anti-myeloma reactivity. We show that HLA-class I was strongly expressed on the surface of patient-derived myeloma cells and on myeloma cell lines. HLA-E was highly expressed by primary myeloma cells but only marginally by cell lines. HLA-E(low) expression on U266 cells observed in vitro was strongly upregulated after in vivo (bone marrow) growth in RAG-2(-/-) γc(-/-) mice, suggesting that in vitro HLA-E levels poorly predict the in vivo situation. Concurrent analysis of inhibitory receptors (KIR2DL1, KIR2DL2/3, KIR3DL1 and NKG2A) and NK cell degranulation upon co-culture with myeloma cells revealed that KIR-ligand-mismatched NK cells degranulate more than matched subsets and that HLA-E abrogates degranulation of NKG2A+ subsets. Inhibition by HLA-class I and HLA-E was also observed with IL-2-activated NK cells and at low oxygen levels (0.6 %) mimicking hypoxic bone marrow niches where myeloma cells preferentially reside. Our study demonstrates that NKG2A-negative, KIR-ligand-mismatched NK cells are the most potent subset for clinical application. We envision that infusion of high numbers of this subclass will enhance clinical efficacy.

[Sequence analysis of a novel allele HLA-C*01:78].

OBJECTIVE: To investigate the genetic basis for a novel allele HLA-C*01:78. METHODS: Genomic DNA was extracted from peripheral blood using a QIAGEN quick DNA extraction kit. The regions encompassing HLA-C from exon 1 to intron 3 and intron 3 to 3'UTR were amplified and cloned using a cloning sequencing kit in order to split the two alleles apart. Selected clones were sequenced to include exons 2 to 4. RESULTS: Sequencing results have indicated the HLA-C alleles of the proband to be a novel C*03:04 allele. The sequence has been submitted to GenBank (KF049216). BLAST analysis has confirmed the novel allele to have one nucleotide difference as C*01:03 at genomic nt316C>A (codon 82CGC>AGC) in exon 2, which has resulted in replacement of one amino acid (82R>S). CONCLUSION: The novel allele has been officially named as C*01:78 by the WHO Nomenclature Committee. The HLA allele type of the proband was therefore A*02:07, 24:02; B*40:01, 46:01; C*01:78, 03:04; DQB1*05:02, 05:02; DRB1*16:02, 16:02.

Genetic profile of KIR and HLA in southern Chinese Han population.

KIR and their HLA ligands are encoded by two of the most diverse gene families in the human genome. The function of KIR on the NK cell is highly dependent on the normal expression of class I HLA on the target cell. Previous population studies in southern Chinese have been focused on the KIR framework genes and genotypes but little is known about the compound profiles of KIR/HLA. The present study examined 503 unrelated individuals from southern Chinese Han population for the polymorphism of KIR and class I HLA genes. All 16 KIR genes were detected in the study population and the four framework genes KIR3DL2, 3DL3, 3DP1, and 2DL4 were present in all individuals. Thirty unique KIR gene profiles were found reflecting a rather limited number of KIR haplotypes in this population. KIRAA1 was the most common profile observed in 54.7% of the samples. Among the AA1 individuals, 15.6% were homozygous for the deleted KIR2DS4. Haplotype A (74.8%) was more common than haplotype B (25.2%). HLA-C1 was a much more common ligand for 2D KIRs than C2. Bw4-80I, Bw4-80T, and the Bw4-bearing HLA-A alleles were detected at similar frequencies. The matched KIR+HLA pairs 2DL2/3+C1 (98.1%), 3DL1+Bw4 (73.3%), 3DL2+A3/11 (60.0%) were the most common ones whereas 3DS1+Bw4-80I was the least common (9.4%). A total of 193 unique compound profiles of KIR-HLA were identified in 480 informative individuals, 130 of the profiles being detected only once. The study provided a comprehensive analysis of the KIR/HLA profiles in southern Chinese in regards of the presence/absence of KIR genes, HLA ligands, matched KIR+HLA pairs, and KIR/HLA compound profiles. The results could help to better understand the role played by KIR/HLA interaction in associated diseases and clinical transplantation in southern Chinese.

HLA-C polymorphisms and PCR dropout in exons 2 and 3 of the Cw*0706 allele in sequence-based typing for unrelated Chinese marrow donors.

To evaluate the accuracy of SBT protocols for HLA-C and to better understand the HLA-C polymorphism in Chinese, 1795 unrelated CMDP donors were typed at exons 2, 3, and 4 of the HLA-C gene using the Atria commercial kit. Of the study subjects, 1768 showed conclusive typing results, whereas the other 27 showed inconclusive results. Subsequent full-length cloning and haplotype sequencing showed that 11 of the 27 inconclusive results could be explained by the presence of nine novel alleles identified: Cw*0130, 0624, 070206, 075602, 0766, 0767, 0820, 0821, and 0827. These novel alleles were generated by a total of 10 coding-region substitutions, eight of them being located in the antigen-binding groove. Cw*0766 and Cw*075602 were detected three and two times, respectively, in the 1795 donors. The other 16 inconclusive samples were retested by SBT using our in-house PCR primers; all of them were found to carry Cw*0706, which dropped out in exons 2 and 3 in the initial PCR using the commercial primers amplifying from 5' UTR to intron 3. Our results showed the importance of the full-length genomic sequence and intronic SNPs for the development of more accurate SBT. The allele distribution and novel alleles detected in this study also provide further insights into the HLA-C polymorphism in the Chinese Han population.

Molecular interactions of pesticides at the soil-water interface.

High-resolution magic angle spinning (HR-MAS) NMR spectroscopy combined with saturation-transfer double difference (STDD) NMR can be used to analyze the molecular-level interactions of pesticides and whole soils occurring at the soil-water interface. Here 1H HR-MAS STDD NMR has been applied to some common pesticides (trifluralin, acifluorfen, and (4-nitro-3-(trifluoromethyl) phenol) and a pesticide degradation product (1-naphthol). Results indicate that dipolar interactions, H-bonding, hydrophobic associations, and potentially pi-pi interactions are the predominant sorption mechanisms for these molecules at the soil-aqueous interface. It is evident that the physical and chemical characteristics of soil are highly influential in determining the mechanisms of pesticide sorption, as they significantly affect soil conformation. In particular, different binding mechanisms were observed for 1-naphthol in soil swollen using a buffer versus D2O, indicating that the K(oc) alone may not be enough to accurately predict the behavior of a molecule in a real soil environment. Preliminary kinetic-based studies suggest that both the swelling solvent and soil moisture content significantly influence the sequestration of trifluralin. These studies demonstrate that HR-MAS and STDD NMR are powerful and versatile tools which can be applied to expand our knowledge of the mechanistic interactions of agrochemicals at the molecular level.

Application of saturation transfer double difference NMR to elucidate the mechanistic interactions of pesticides with humic acid.

Elucidation of mechanistic interactions of anthropogenic chemicals is critical to understanding and eventually predicting their behavior in the environment Here, a recently developed technique, saturation transfer double difference (STDD) NMR spectroscopy is employed to determine the interactions of pesticides with humic acid (HA) at the molecular level. The degree of interaction at each NMR observable nucleus in the pesticide can be quantified in the form of an epitope map, which depicts the mechanism of the pesticide-HA interaction. Our results indicate that, at pH 7, halogen atoms (F and Cl) in water-soluble pesticides (diflufenzopyr, acifluorfen, and chlorsulfuron) play a dominant role in influencing binding to HA, whereas carboxyl groups likely play a secondary role when halogen atoms are also present in the molecule, as observed with diflufenzopyr and acifluorfen. However, when present on its own, the carboxyl group dominates in binding affinityto HA (e.g., imazapyr). Electronegativity and electron density appear to play a key role in the mechanism of binding and results suggest that polar bonds are the primary points of HA contact in the water soluble pesticides investigated. Likely interactions may include hydrogen bonding and dipole-dipole interactions.

Human leukocyte antigen class II alleles and risk of cervical cancer in China.

Human leukocyte antigen (HLA) class II alleles have been associated with an increased or decreased risk of developing cervical cancer through infection with oncogenic forms of human papillomavirus (HPV). To verify whether HLA class II allelic polymorphism is related to cervical cancer in China, 133 cervical cancers and 98 healthy controls were analyzed for HLA typing. Our results showed that DPB1*1301 allele frequency was significantly higher in the cervical cancers compared with healthy controls (OR, 3.793; p = 0.002; Pc = 0.04). A significant relationship was found between DRB1*150101-DQB1*0602 haplotype (OR, 0.180; p < 0.0001; Pc < 0.003), DRB1*070101-DQB1*0201 haplotype (OR, 0.110; p = 0.001; Pc = 0.03), and decreased risk for cervical cancer. Similar tendencies were observed for DRB1*150101-DQB1*0602 haplotype with HPV16 positive cervical cancers (OR, 0.182; p = 0.001; Pc = 0.021), and for DRB1*070101-DQB1*0201 haplotype (OR, 0.144; p =0.003; Pc = 0.063). These results indicate that HLA-DPB1*1301 may confer susceptibility to cervical cancer, and the haplotypes DRB1*150101-DQB1*0602 and DRB1*070101-DQB1*0201 may contribute to the resistance to the development of cervical cancer among Chinese women. The study suggests that specific HLA class II alleles and haplotypes may influence the immune response to specific HPV-encoded epitopes and affect the risk of cervical cancer in a Chinese population from an area with a high incidence of this neoplasia.

HPV16 E6 variants and HLA class II polymorphism among Chinese women with cervical cancer.

Infection with human papillomavirus type 16 (HPV16) confers a high risk for the development of cervical cancer. Variants of this virus may interact differentially with host genetic factors, possibly affecting the disease pathogenesis. This study was designed to investigate the association between HPV16 E6 variants and human leukocyte antigen (HLA) polymorphism within a Chinese population. Peripheral blood from HPV16 positive Chinese women with cervical carcinoma, who had previously been tested for HPV16 E6 variants, was used for HLA class II typing. It was found that there was a significant positive association between DQB1*060101 allele and HPV16 As variant-positive cervical cancers (OR, 4.47; Pc=0.0018). A negative relationship was found between DRB1*150101-DQB1*0602 haplotype and decreased risk for HPV16 As variant-positive cervical cancers (OR=0.31; P=0.037). Similar tendency was observed for the haplotype DRB1*070101-DQB1*0201 with HPV16 As variant-positive cervical cancers (OR=0.16, P=0.024). Additionally, as for the HPV16 E6 prototype-positive cervical cancers, a significant positive association was found in DQB1*060101 allele (OR=5.95; P=0.002; Pc=0.036), and similar trends were observed for DQB1*030201 (OR=10.87, P<0.0001; Pc=0.0018), and DPB1*1301(OR=7.40, P=0.002; Pc=0.04). It was found that there was no significant association between DRB1-DQB1 haplotype and HPV16 prototype-positive cervical cancers. These data indicate that host genetic factors, such as HLA polymorphism, may determine the potential oncogenicity of the HPV16 E6 variant. The results suggest that a specific match between E6 variant proteins and HLA class II alleles may contribute to HPV16-related cervical carcinogenesis in a certain Chinese population.