| PART1 EXPRESSION PROFILE-BASED IN SILICO SCREENING OF CANDIDATE GENES FOR CHILDHOOD HEMOPHAGOCYTIC LYMPHOHISTIOCYTOSIS AND RELATED EXPERIMENTSObjective The aim of this study was to screen candidate genes of hemophago-cytic lymphohistiocytosis (HLH) in children using bioinformatics strategy and to elucidate the genetic mechanisms of development and prognosis in childhood HLH.Methods (1) The GSE26050 datasets was downloaded from GEO database. Sig-nificance analysis of microarrays was performed using packages in R language, and differentially expressed genes (DEGs) were regarded as "test gene set". Genes associated with childhood HLH, obtained by data mining tools Genecards and Fable were defined as "training gene set". Finally, candidate genes of child-hood HLH were screened by the tool "Toppgene".(2) One-hundred unrelated pediatric subjects and 146 healthy controls were re-cruited. Genomic DNA was extracted from peripheral blood samples. SNP ana-lysis was performed by using PCR and SNaPshot sequencing.(3) The chi-square test or Fisher’s exact test were used to compare frequencies of SNPs alleles, genotypes and haplotypes between pediatric HLH patients and controls. All the differences were regarded as significant when aP<0.05 was ob-tained. Odds ratios (ORs) and 95% confidence intervals (CIs) were calculated for the estimated risk of these potential factors for developing HLH. The data were analyzed using the SPSS for Windows version 17.0. Deviations of the fre- quencies of SNPs alleles from the HWE and LD were tested in Haploview v.4.2 software. Haplotypes and their frequencies were estimated by the PHASE ver-sion 2.1 software.(4) Patients’clinical information was obtained from medical records. Overall su-rvival (OS) was measured from the date of diagnosis to the date of last follow-up or death. Whether and when a patient died were obtained from inpatient and outpatient records, or patients’families through follow-up telephone calls. OS was analyzed using the Kaplan-Meier curve method. Survival comparisons were done by the Log-rank test. All the differences were regarded as significant when a P<0.05 was obtained.Results (1) A total of 86 genes were regarded as "test gene set", and 73 genes associated with childhood HLH were defined as "training gene set". Forty-five candidate genes were screened out by Toppgene, including 42 up-regulated genes and 3 down-regulated genes (TNFRSF17, CX3CR1 and CCR2).(2) The frequency of rs3743591GG genotype in TNFRSF17 gene was significa-ntly higher in patients than that in controls (10.0% vs.1.4%, P=0.002), suggest-ing a risk association (OR=8.000,95% CI=1.714-37.349); the frequency of rs3743591 G allete was significantly higher in patients than that in controls (20.5% vs.10.6%, P=0.002), suggesting a risk association (OR=2.171,95% CI= 1.308-3.603). The frequency of rs2017662AA genotype in TNFRSF17 gene was significantly higher in patients than that in controls (11.0% vs.1.4%, P= 0.001), suggesting a risk association (OR=8.899,95% CI=1.928-41.082); the frequency of rs2017662A allete was significantly higher in patients than that in controls (21.0% vs.10.3%, P=0.001), suggesting a risk association (OR=2.322,95% CI= 1.396-3.860).(3) The frequency of rs2853712TT genotype in CX3CR1 gene was significantly higher in patients than that in controls (72.0% vs.50.0%, P=0.001), suggesting a risk association (OR=2.571,95% CI=1.493-4.430).(4) The frequency of rs1799865TT genotype in CCR2 gene was significantly hi-gher in patients than that in controls (39.0% vs.20.5%, P=0.002), suggesting a risk association (OR=2.472,95% CI=1.401-4.363); the frequency of rsl799865 T allete was significantly higher in patients than that in controls (53.0% vs. 37.0%, P=0.000), suggesting a risk association (OR=1.921,95% CI=1.333-2.769).(5) The frequency of haplotype A-T-G-G-G-C-T-T-T-T-T-A-G-A-C consisted by SNPs rs3743591, rs11570151, rs2017662, rs2071336, rs2669850, rs17793056, rs13088991,rs13062158, rs2853712, rs2669841, rs2853711, rs3762823, rs3092963, rs3092962, and rs1799865 was significantly higher in patients than that in controls (5.7% vs.1.3%, P=0.003), indicating a risk association (OR= 5.025,95% CI=1.530-16.505). The frequency of haplotype G-T-A-G-G-T-C-C-T-C-G-A-G-A-C consisted by SNPs rs3743591, rsl 1570151, rs2017662, rs2071336, rs2669850, rs17793056, rs13088991, rs13062158, rs2853712, rs2669841, rs2853711, rs3762823, rs3092963, rs3092962, and rs1799865 was significantly higher in patients than that in controls (5.2% vs.0%, P=4.43 × 10"5), also suggesting a risk association.(6) The median survival of patients carrying the rs3743591GG genotype was significantly shorter than those with rs3743591 AA or AG genotype (86.0 months vs.118.0 months, P=0.009). The median survival of patients carrying the rs2017662AA genotype was significantly shorter than those with rs2017662 GG or GA genotype (88.0 months vs.118.0 months, P=0.004). The average su-rvival between patients carrying the rs2853712TT genotype and those with rs2853712 CC or CT genotype showed no significant difference (96.5 months vs. 108.6 months, P=0.124). The median survival of patients carrying the rs1799865 TT genotype was significantly shorter than those with rs1799865CC or CT ge-notype (92.0 months vs.107.0 months, P=0.021).Conclusions (1) Toppgene method used in this study identified 45 candidate genes of childhood HLH, including 42 up-regulated genes and 3 down-regulated genes.(2) Three independent genes, including TNFRSF17 gene, CX3CR1 gene, and CCR2 gene, were down-regulated in pediatric HLH, indicating an association with childhood HLH.(3) The TNFRSF17 gene rs3743591, rs2017662, CX3CR1 gene rs2853712, and CCR2 gene rs 1799865 polymorphisms are associated with the development of childhood HLH.(4) The A-T-G-G-G-C-T-T-T-T-T-A-G-A-C haplotype and G-T-A-G-G-T-C-C-T-C-G-A-G-A-C haplotype consisted by SNPs rs3743591, rs11570151, rs2017662, rs2071336, rs2669850, rs17793056, rs13088991, rs13062158, rs2853712, rs2669841, rs2853711, rs3762823, rs3092963, rs3092962, and rs1799865 are associated with the development of childhood HLH.(5) The TNFRSF17 gene rs3743591, rs2017662, and CCR2 gene rs1799865 polymorphisms could serve as molecular markers in evaluating the prognosis of childhood HLH.PART2 SCREENING THE SH2D1A AND XIAP GENE MUTATIONS IN CHINESE CHILDREN WITH HEMOPHAGOCYTIC LYMPHOHISTIOCYTOSISObjective This study aimed to investigate the prevalence of mutations and se-quence variations of SH2D1A/XIAP gene in Chinese pediatric patients with HLH and to explore the possible relationship between gene mutations and cli-nical manifestations.Methods One-hundred pediatric patients diagnosed with HLH and 100 controls were recruited. Genomic DNA was extracted from peripheral blood samples. Nucleotide sequences of all exons and their flanking intronic sequences of SH2D1A/XIAP were amplified by PCR followed by direct sequencing. Bio-informatics methods were used to predict whether SH2D1A/XIAP gene muta-tions affected their protein function. The patients’ clinical features were assessed by retrieval of data from medical records.Results (1) One novel missense variant was discovered in one HLH child, Phe27Ile cited in exon 1 of XIAP gene, which was neither found in the control population nor reported previously. The mutation incidence was 1%(1/100).(2) The Phe27Ile mutation in XIAP gene was expected to affect its protein function by PolyPhen2 and SIFT programs.(3) Five SNPs were found both in case group and control group. There was no significant difference in SNPs genotypes and alletes between HLH patients and normal controls.(4) The patient showing the Phe27Ile mutation was a 13 year-old girl; she had no familiar history of HLH; she exhibited recurrent episodes of HLH, and died from multiple system organ failure (MSOF) at 16 years of age.Conclusions One novel missense variant, Phe27Ile in XIAP gene, was identified; this gene mutation may be associated with the development and poor outcome of pediatric HLH. |
PDF Full Text Request