Single Nucleotide Polymorphisms And Expression Of Duffy Antigen Receptor For Chemokines In Colorectal Cancer

Background and objectiveColorectal cancer (CRC) is the second most common malignant tumor in females andthe third in males worldwide. Some chemokines and their receptors play an importantrole in regulating CRC progression, angiogenesis, and metastasis. Duffy antigenreceptor for chemokines (DARC), also called Duffy antigen is a glycosylated membraneprotein located at red blood cell and endothelial cell which belongs to the subfamily ofsilent chemokine receptors. DARC gene consists of two major alleles: FY*A and FY*Bresulted from a single nucleotide polymorphism （SNP）at rs12075. The FY*A andFY*B encode Duffy blood group antigen Fyaand Fybrespectively. DARC is unable toinduce conventional signaling responses after binding angiogenic chemokines andappears to be a scavenger of chemokines. The angiogenic chemokines play an importantrole in colorectal cancer growth, angiogenesis, metastasis and prognosis, which suggestthat the expression of DARC may be related to tumorigenesis and progression of CRC.CRC genetic susceptibility factors include chromosomal instability and microsatelliteinstability as well as molecular mutations of some oncogenes and tumor suppressorgene and SNP. It has been determined that a number of SNPs are associated with CRCgenetic susceptibility. A recent study indicated that a SNP，rs12075in DARC gene isassociated with metastasis of breast cancer. It suggests that rs12075in DARC might beassociated with CRC. The discovery of the molecular basis of DARC has promoted the development ofseveral molecular methods. So far, several methods have been described to genotypeDARC, such as polymerase chain reaction with restriction fragment lengthpolymorphism (PCR-RFLP), polymerase chain reaction with allele-specific primers(PCR-ASP). However, there are several disadvantages in these methods. For example,post-PCR manipulations such as agarose gel electrophoresis and so on are required. Soit is very necessary to develop a simple, rapid, and high-throughput method which willprovide a useful tool to study association of DARC with diseases and genotype Duffyblood group in transfusion medicine.Our goals were to investigate the role of DARC in CRC by detecting expression ofDARC, matrix metalloproteinase (MMP)-9and CD34in CRC and to develop areal-time PCR with TaqMan-MGB probes for genotyping of rs12075in DARC. We alsoinvestigated association of rs12075in DARC with CRC using developed method.Furthermore，in order to illustrate the mechanism DARC gene polymorphisms affectcolorectal cancer we analyzed association of DARC genotype with red blood cell (RBC)chemokine scavenging and expression of DARC on RBC.Methods1. CRC and unaffected tissues were collected. The mRNA expression of DARCwas detected using quantitative real-time PCR in CRC and unaffected tissues.The protein expression of DARC, MMP-9and CD34were detected usingimmunohistochemistry. Microvessel density （MVD） was assessed byimmunohistochemistry of CD34.2. Primers and TaqMan-MGB probes were designed and synthesized to genotypers12075in DARC locus. A total of120samples were genotyped by using thenew real-time PCR with TaqMan-MGB probes and conventional PCR-ASP. Theresults obtained by the two methods were compared. Moreover, the repeatabilityand limit of this real-time PCR with TaqMan-MGB probes were determined.3. Blood samples from patient with CRC and control group were collected and thengenotyped using real-time PCR with TaqMan-MGB probes. Furthermore，the expression of DARC on RBC and chemokine scavenging of RBC were detectedusing flow cytometry and ELISA respectively.Results1. Expression of DARC in CRC and unaffected tissuesPositive staining for DARC was detected in66out of90(73.3%) patients with CRC.All paired unaffected tissues were DARC positive (64out of64,100%).Theexpression rate of DARC is significantly lower in CRC than in the unaffected tissues(P<0.001). DARC mRNA expression is significantly decreased in CRC comparedwith the unaffected tissues (P=0.012) and median decrease was4-fold. The DARCprotein expression is positively correlated with DARC mRNA, no matter in CRC orunaffected tissues (P<0.001).2. Correlation of DARC expression with clinical pathological featuresThe expression of DARC was negatively correlated with lymph node metastasis andTNM stage (P＜0.05), and positively correlated with tumor differentiation (P＜0.001). However, no significant correlation was observed between DARC expressionand other clinical pathological features, such as age, gender, tumor size, localization,depth of penetration and distant metastasis.3. Expression of MMP-9in CRC and unaffected tissuesPositive staining for MMP-9was detected in72out of90(80.0%) patients with CRC.The unaffected tissues showed positive MMP-9in33out of64(51.6%). Theexpression rate of MMP-9is significantly higher in CRC than in the unaffectedtissues (P<0.001).4. Correlation of MMP-9expression with clinical pathological featuresThe expression of MMP-9was positively correlated with lymph node metastasis andTNM stage (P＜0.05), and negatively correlated with tumor differentiation(P=0.037).However, no significant correlation was observed between MMP-9expression andother clinical pathological features, such as age, gender, tumor size, localization,depth of penetration and distant metastasis.5. Correlation of MVD with clinical pathological features The MVD was positively correlated with lymph node metastasis and TNM stage (P＜0.01), and negatively correlated with tumor differentiation(P=0.009). However, nosignificant correlation was observed between MMP-9expression and other clinicalpathological features, such as age, gender, tumor size, localization, depth ofpenetration and distant metastasis.6. Correlation of DARC expression with MMP-9and MVDThe expression of DARC was negatively correlated with MMP-9and MVD(P＜0.001). The expression of MMP-9was positively correlated with MVD(P＜0.001).7. The comparison of real-time PCR with TaqMan-MGB probes and PCR-ASPThere was a complete concordance of results for all samples genotyped by real-timePCR with TaqMan-MGB probes and PCR-ASP.8. The repeatability and detection limit of real-time PCR with TaqMan-MGBprobesThe five FY*A/FY*A，five FY*A/FY*B and one FY*B/FY*B samples werereexamined. The retesting results were consistent with those of the initial testing. Thedetection limit was at least50pg of genomic DNA for homozygous FY*A/FY*A andFY*B/FY*B, and100pg of genomic DNA for heterozygous FY*A/FY*B.9. The mismatch probability of Fya and Fyb antigen in random transfusion and theFY*A and FY*B allelic frequencies in the Chinese population in DalianThe mismatch probability of Fya and Fyb is0.1123and0.0044respectively inrandom transfusion. The combination of Fya and Fyb is0.1167. FY*A allelicfrequencies in the Dalian Chinese population is93.3%and higher than FY*B（6.7%）.The FY*A and FY*B allelic frequencies in the Dalian Chinese Han population aresimilar to those reported in the other areas of China, but differ significantly from thatin Chinese She population in China.10. The comparison of DARC genotype distribution between CRC and control groupNumber of FY*A/FY*A、FY*A/FY*B and FY*B/FY*B genotype were266,39and1respectively in control group and286,38and1respectively in CRC. The associationgenotype of DARC with CRC was not observed. 11. Correlation of genotype DARC with clinical pathological featuresCombination of FY*A/FY*B and FY*B/FY*B genotype frequencies in lymph nodenegative patients was higher than that in lymph node positive patients. However，FY*A/FY*A genotype frequency in lymph node negative patients was lower than thatin lymph node positive patients. CRC patients with FY*A/FY*A genotype havehigher lymph node metastasis risk（2.137-fold）than patients without FY*A/FY*Agenotype.12. Correlation of genotype of DARC with RBC chemokine scavengingRBC CXCL8scavenging in FY*A/FY*A samples was markedly lower than that inFY*A/FY*B samples（P=0.016）. Similar result was also observed in RBC CCL2scavenging（P=0.038）.13. Correlation of genotype of DARC with expression of DARC on RBC surfaceThe expression of DARC on RBC surface in FY*A/FY*A samples was markedlyhigher than that in FY*A/FY*B samples (P=0.021).Conclusions1. Expression of DARC is down-regulated in CRC. This down-regulation might beinvolved in tumorigenesis of CRC.2. The expression of DARC was negatively correlated with lymph node metastasis,TNM stage, MVD and MMP-9, and positively correlated with tumordifferentiation3. DARC inhibited angiogenesis of CRC, perhaps via a decreasing angiogenicchemokines and MMP-9in microenvironment and subsequently suppressed CRCprogression and metastasis.4. DARC might be a negative regulator in CRC tumorigenesis, progression,angiogenesis, and metastasis. It could be a potential prognosis marker in CRCand a molecular therapeutic target in treating CRC.5. Real-time PCR with TaqMan-MGB probes for genotyping of DARC is simple,rapid, reliable, reproducible, sensitive, and high-throughput. This method issuperior to conventional PCR-ASP 6. FY*A allelic frequencies in the Dalian Chinese population is higher than FY*B.The FY*A and FY*B allelic frequencies in the Dalian Chinese Han populationare similar to those reported in the other areas of China, but differ significantlyfrom that in Chinese She population in China. It suggests that DARC allelicfrequency was associated with race.7. The combination of mismatch probability of Fya and Fyb antigen is0.1167inrandom transfusion. It suggests that risk of immune hemolytic reactions resultedfrom mismatch of Duffy antigens should be noted.8. Polymorphism of rs12075in DARC was not associated with CRC tumorigenesis.9. Polymorphism of rs12075in DARC was correlated with lymph node metastasisand TNM stage. CRC patients with FY*A/FY*A genotype have higher risk oflymph node metastasis than that with non-FY*A/FY*A genotype10. Polymorphism of rs12075in DARC affected the expression of DARC on RBC.11. Polymorphism of rs12075in DARC affected RBC chemokine scavenging.12. Polymorphism of rs12075in DARC affected lymph node metastasis of CRC,probably via modifying structures of DARC on RBC，subsequently affected RBCangiogenic chemokine scavenging and further affected lymph node metastasis ofCRC.