Associations of CYLD, JAK2 and TLR4 Genotypes with PSA Levels and Immunophenotype in Benign Prostatic Hyperplasia and Prostate Cancer

Background : Prostate cancer (PCa) is one of the most common malignant tumors of the male urinary system, and its incidence and mortality rates have been increasing worldwide. Benign prostatic hyperplasia (BPH) represents stromal and epithelial cell proliferation in the prostate in elderly males. Abnormal activation of inflammation-related signalling molecules, such as toll-like receptor 4 (TLR4) and Janus kinase/signal transducers and activators of transcription (JAK/STAT) has been linked to the initiation and progression of various human diseases including PCa and BPH. Cylindromatosis (CYLD) gene alterations are associated with PCa progression. In this study, the contribution of CYLD , JAK2 , and TLR4 gene variants to PCa and BPH risks and their associations with prostate-specific antigen (PSA) levels, immunophenotype, and clinical features in Vietnamese men were determined. Methods : A total of 102 patients with PCa, 65 with BPH, and 114 healthy controls were enrolled. The immunophenotype was analyzed by flow cytometry, cytokine secretion by enzyme-linked immunosorbent assay (ELISA), and gene variants by DNA sequencing. Results : Lower levels of transforming growth factor β (TGF-β ) and higher numbers of CD13 + CD117 − and CD56 + CD25 + cells were observed in the PCa group than in the BPH group. Genetic analysis of the CYLD gene identified five single nucleotide polymorphisms (SNPs), of which c.2351-47 C > T, c.2351-46A > T, and rs1971432171 T > G had significantly higher frequencies in PCa patients than in the control and BPH groups. Sequencing of the TLR4 gene revealed five nucleotide changes, in which the rs2149356 SNP showed an increased risk for both PCa and BPH and the c.331-206 SNP had a reduced risk for PCa. Importantly, the expansion of activated natural killer (NK) cells and higher levels of PSA were found in PCa patients carrying the CT genotype of the CYLD c.2351-47 compared to those with the wild-type genotype. Conclusion : Activation of NK cells in CYLD -sensitive PCa patients was associated with serum PSA release and the CYLD c.2351-47 variant may be a significant risk factor for prostatitis in PCa patients.


Introduction
Prostate cancer (PCa) is one of the most common malignant tumors of the male urinary system, and its incidence and mortality rates have increased considerably in recent years [1].The etiology and pathogenesis of PCa remain unclear, although several factors such as age, family history, ethnicity, and dietary and genetic mutations are known to be related to PCa risk [2].Classifications based on clinicopathological features, such as prostate cancer-specific antigen, Gleason score, and tumor, node and metastasis (TNM) system, are considered as the accepted practice standards for determining tumor stage for PCa [3].Unlike PCa, benign prostatic hyperplasia (BPH) is characterized by stromal and epithelial cell proliferation in the prostate in elderly males.Prostate inflammation is responsible for the development of BPH [4] and linked to the pathogenesis with PCa pathogenesis [5].The inflammatory response in these patients is characterized by the accumulation of immune cells, mainly T and B cells, and macrophages into the prostate tissue, and their activation results in the release of inflammatory cytokines [4][5][6].Prostate-specific antigen (PSA), a glucoproteinase produced by both normal and malignant cells of the prostate gland, is a useful serum biomarker for diagnosing and controlling PCa [2].Increased PSA levels can indicate reliable sign of prostate cancer aggressivity [7].In addition, the inflammatory cytokine interleukin 6 (IL-6) secreted by immune cells is known to promote the release of PSA in PCa patients [6].
Prostate cancer patients have elevated numbers of peripheral circulation regulatory (Treg) T [8] and myeloidderived cells, which are linked to poor outcome in PCa [9].Among myeloid-related markers, CD13 and C-kit receptor (CD117) play key roles in normal hematopoiesis and are detectable in neoplastic human tissues and at sites of prostate [10][11][12].The CD13/Aminopeptidase N membrane metallopeptidase participates in negatively regulating several signalling molecules, such as the toll-like receptor (TLR) 4 and Janus kinase/signal transducers and activators of transcription (JAK/STAT) [13].
Moreover, immunological investigations have indicated that abnormal expression of TLR4 and JAK/STAT pathways is linked to tumor development, cell migration, immune invasion, and progression of various human diseases, including PCa and BPH [14,15].TLR4 activation is triggered by bacterial lipopolysaccharide to elicit an inflammatory response in immune cells [16], and its levels are upregulated in PCa [14].TLR4 deficient mice fail to respond to viral and bacterial infection [17].Several investigations have indicated that polymorphisms within the TLR4 gene are associated with PCa [14,16].The TLR4 c.1063 A>G single nucleotide polymorphism (SNP) was shown to attenuate TLR4 activation and inflammatory responses [18].Similarly, JAK2 has recently been considered a regulator of the immune response in the pathogenesis of PCa and BPH [15].The rs10429491 SNP in JAK2 has been reported to be associated with PCa risk [19].
The association between CYLD variants and the risk of PCa and BPH is not well known, although its expression levels regulate PCa progression [20].CYLD is a deubiquitinating enzyme that functions as a negative regulator of immune reactions and tumor cell proliferation in PCa [21].CYLD is known to inhibit the survival, glucose uptake, and growth of prostate tumors [22].Mutations in CYLD lead to cyclindroma disease and myeloma [23].
In this study, we determined the contribution of CYLD, JAK2, and TLR4 gene variants to PCa and BPH risk and their associations with PSA levels, immunophenotype, and clinicopathological features in Vietnamese men.

Patients and Control Subjects
Fresh peripheral blood samples were collected from untreated 102 prostate cancer patients and 65 prostatic hyperplasia patients at the National Institute of Hematology and Blood Transfusion, 103 and K Hospitals, Ha Noi, Vietnam.None of the patients had received prior hormonal therapy or radiotherapy.Prostate tumors were staged using the American Joint Committee on Cancer (AJCC) tumornode-metastasis (TNM) staging system and graded using the Gleason score [3].The control group consisted of 114 healthy individuals.No individuals in the control population took any medication or suffered from any known acute or chronic diseases.All patients and volunteers gave writ-ten consent to participate in the study.Person care and experimental procedures were performed according to the Vietnamese law for the welfare of humans and were approved by the Research Ethics Committee, Military Hospital 103, no.86/CNChT-HDDD.

Isolation of Peripheral Blood Mononuclear Cells
Whole blood samples from PCa and BPH patients and healthy donors were collected by venipuncture and transferred to sterile tubes containing Ethylenediaminetetraacetic acid (EDTA) as an anticoagulant.Peripheral blood mononuclear cells (PBMCs) were isolated by Ficoll density gradient centrifugation (Ficoll-Paque Plus, GE Healthcare, Chicago, IL, USA).The cells were then counted in a Neubauer chamber, washed with PBS, and analyzed for further experiments.

Immunostaining and Flow Cytometry
All flow cytometry data acquisition was conducted using the instrument software FACSAria Fusion (BD Biosciences, San Jose, CA, USA) as previously described [25].Immunostaining was performed using antibodies against CD45, CD3, CD4, CD13, CD25, CD40, CD56, CD117, and FoxP3 (all from eBioscience, Waltham, MA, USA) at a concentration of 10 µg/mL.After incubation with the antibodies for 60 min at 4 °C, the cells were determined by flow cytometry (BD FACSLyric, BD Biosciences, San Jose, CA, USA).

Statistics
The genotype frequencies among PCa and BPH patients and the control group were calculated using Fisher's exact test.Statistical analyses were performed using SPSS (version 24, IBM Corp, Armonk, NY, USA) and GraphPad Prism (Version 8.4.2,GraphPad Software, LLC, Boston, MA, USA).Statistical significance of the differences was determined using the Mann-Whitney U test.For all statistical analyses, the level of significance was set at p < 0.05.

Clinical Associations
A total of 102 patients with PCa and 65 patients with BPH were enrolled, and the clinical characteristics of the study subjects are summarized in Table 1.The mean ages of patients with PCa and BPH were 69.11 and 69.87 years, respectively.A comprehensive correlation analysis showed that both patient groups had higher glucose levels than normal values.Moreover, urea levels were higher, whereas hemoglobin levels were lower in PCa patients than in BPH patients.Importantly, TGF-β levels were significantly reduced in both patient groups; however, there was no difference in its expression levels between PCa and BPH patients (Fig. 1).The frequencies of all important clinical parameters, including PSA level at the time of diagnosis, Gleason score, and TNM staging, are detailed in Table 1.All patients with PCa (100%) presented with PSA values >12 ng/mL and had a diagnosis of intermediate or highly aggressive disease with Gleason score >7 (17.65%) or ≥8 (82.35%), respectively.Of the 102 PCa patients analyzed, 72 had M1 stage disease, 71 had N1 stage disease, and 30 had T4 stage disease.

Analysis of Immunophenotypic Profiles in Prostate Cancer and Hyperplasia Patients
Next, changes in the expression of CD4 T, regulatory T (Treg), NK, and CD13 + cells in PCa and BPH cells were investigated.

Genotype Frequency of CYLD Gene in Prostate Cancer and Hyperplasia Patients
Sequencing of the CYLD gene identified a p.Q732R (c.2436 A>G) SNP in exon 15 and three intronic nucleotide changes, including c.2351-47 C>T, c.2351-46 A>T, and rs1971432171 T>G in intron 14 and c.2483+188 G>C in intron 15 (Table 2 and Fig. 3).The distribution of genotype frequencies of the five SNPs, except for c.2483+188 G>C, was consistent with Hardy-Weinberg genetic balance (p > 0.05, Table 3).The three SNPs in intron 14 had significantly higher frequencies in PCa than in the control and BPH groups, whereas only patients carrying the SNP c.2351-47 C>T were linked to an enhanced risk of BPH (Table 2).In addition, SNP c.2483+188 G>C had a protective effect against both prostatic cancer and hyperplasia (Table 2).

Genotype Frequency of JAK2 Gene in Prostate Cancer and Hyperplasia Patients
Next, sequencing of JAK2 identified three nucleotide changes, including rs994555780 T>C in intron 12 and rs4495487 T>C and rs10974947 G>A in intron 13 (Table 2 and Fig. 4).The distribution of genotype frequencies of the three SNPs was consistent with the HWE (p > 0.05, Table 3); however, the carrier frequencies of the three SNPs were not different between the controls and the two patient groups (Table 2).

Genotype Frequency of TLR4 Gene in Prostatic Cancer and Hyperplasia
Sequencing of TLR4 identified five nucleotide changes, including rs2149356 T>G, c.331-337 A>G, rs911685299 A>G, c.331-206 A>G, and c.331-180 T>A in intron 3 (Table 2 and Fig. 5).The genotype distribution of the five SNPs, except for rs2149356, was in agreement with HWE (p > 0.05, Table 3).The rs2149356 showed  a significantly increased risk for both PCa and BPH (Table 2).In addition, we observed a significantly reduced PCa risk in carriers of c.331-206 (Table 2).

Associations of the SNPs in CYLD, JAK2 and TLR4 Genes with Clinical Characteristics and Immunophenotype in Prostatic Cancer and Hyperplasia
Association analysis of the CYLD, JAK2, and TLR4 genes with immunophenotype indicated that the numbers of activated NK (CD56 + CD25 + ) cells were expanded in PCa patients carrying the CT genotype of CYLD c.2351-47 compared to those with wild-type genotypes (Fig. 6A,B).These results indicate that the c.2351-47 in CYLD gene might be partially linked to the inflammatory response in PCa patients.
Next, PSA is used as a prostate disease marker to sufficiently predict prostate enlargement to serve as a thera-peutic and reliable indicator of cancer aggressivity.As expected, the number of PCa patients with higher PSA levels of 12 ng/mL was 102 (100%), while the number of BPH patients with PSA levels above the clinical cutoff of 4 ng/mL.In addition, percentage of PCa patients had higher PSA levels (100 ng/mL), was 62/102 (60.8%) (Table 4).Importantly, PCa carriers of the CT genotype of CYLD c.2351-47 and the CC genotype of JAK2 rs4495487 were associated with elevated levels of PSA ≥100 ng/mL (Table 4).Although CYLD c.2483+188 was indicated as a protective variant in both patient groups, carriers of the GC genotype of CYLD c.2483+188 had higher PSA levels than normal values in BPH cases.In addition, carriers of JAK2 rs10974947 and TLR4 rs2149356 were at a reduced risk of elevated PSA levels (Table 4).

Discussion
Our study provides evidence for the association of CYLD sequence variants (c.2351-47, c.2351-46, and rs1971432171) with PCa risk in the Vietnamese study population.In this study, PCa-sensitive SNPs were identified for the first time.Mutations in the CYLD gene are known to be linked to abnormal cellular function in mice [26] and humans [23,25] and the development of cyclindroma disease and myeloma [23].Downregulated expression of CYLD is also associated with PCa progression [20].In contrast, a rs12324931 SNP in CYLD is associated with the risk of inflammatory bowel disease [27].In our recent study, p.W736G in the CYLD gene was indicated as the pathogenic variant in patients with polycythemia vera [28]; however, it was not found in all cases in this study as well as in patients with leukemia, including 352 patients with myeloid leukemia, and 145 patients with lymphocytic leukemia (unpublished data).
Moreover, PCa cases carrying the CT genotype of CYLD c.2351-47 had an expanded number of activated NK cells and an increased risk of PSA levels of more than 100 ng/mL (Fig. 6 and Table 4).High pretreatment serum PSA levels are associated with worse PCa outcome [29].CYLD functions as a negative regulator of the immune response, including the activation of NK cells [26], and its expression is inhibited by high glucose [30].Moreover, CYLD inhibits cancer cell proliferation, glucose uptake, and growth of prostate tumor cells [22].A recent study indicated that the release of cytokines and chemokines by activated NK cells induces the recruitment of accessory immune cells such as T cells [31].In addition, high PSA levels are linked to inflammatory responses [6].Evidence suggests that activation of NK cells in CYLD-sensitive PCa patients is associated with the release of serum PSA.
Functional studies of the CYLD gene revealed that it is a negative regulator of proinflammatory gene expression through TLR4 signalling [32].Unlike CYLD, carriers of TLR4 rs2149356 were associated with significantly elevated risks of both PCa and BPH compared with healthy controls.In contrast, the c.331-206 SNP in this gene was associated with a reduced risk of prostatic cancer.Changes in TLR4 expression are linked to prostate cancer risk and tumorigenesis by inducing inflammation [5].Consistently, the rs2149356 SNP was found in Caucasian and South Asian patients with prostate cancer [33].In contrast, the TLR4 rs11536889 SNP in the 3' UTR is associated with the risk of prostate cancer in Korea [16] and Sweden [34], but not in other populations [35].
The effects of JAK2 expression on prostate cancer cell function have been revealed in a recent study [36].Unlike this study, a known rs10429491 SNP in JAK2 is associated with PCa risk [19].Importantly, carriers of the CC genotype of JAK2 rs4495487 were associated with higher levels of PSA in PCa patients, but not in BPH patients (Table 4).
Growing evidence indicates that chronic inflammation, which results from the infiltration of immune cells into prostate tissue and blood circulation, increases the risk of high-grade PCa development [8,37].Similarly, PCa patients have an elevated number of circulating and tumorinfiltrating Tregs [8].Recently, NK cells isolated from the peripheral blood of patients with PCa displayed increased expression of the surface antigens CD56, CD9, and  CD49a [38].In contrast, we observed that the number of CD56 + CD25 + expressing cells was higher in patients with PCa than in patients with BPH and healthy controls.Loss of CYLD expression leads to NK cell activation [26].Moreover, activated NK cells were expanded in PCa patients carrying the CT genotype of the CYLD c.2351-47 variant compared to those with the wild-type genotype (Fig. 6).Consistently, there was no significant difference in the percentage of NK cells and their activation between BPH patients and healthy controls [39].CD25 + NK cells display functional and metabolic activities under regulatory T cell-mediated suppression in cancer patients [40].
Lastly, CD13 + CD117 − cells were recruited into the circulation in patients with PCa but not in patients with BPH and healthy controls.CD13 and CD117 are expressed in human prostate and neoplastic tissues [10][11][12].CD13 is also highly expressed in myeloid cells [13], and its activation is mediated through the TLR4 signalling pathway to elicit an inflammatory response [13].CD13 expression is elevated in several JAK2-positive blood cancers, such as polycythemia vera [28,41].Unlike activated NK cells, the number of CD13 + CD117 − cells was unaltered in patients with CYLD variants.

Conclusion
Prostate cancer carriers of the CYLD c.2351-47 variant were associated with the recruitment of activated NK cells into the blood circulation and had serum PSA more than 100 ng/mL.Therefore, the CYLD c.2351-47 variant may be a significant risk factor for prostatitis in PCa patients.

Fig. 1 .
Fig. 1.TGF-β concentrations in prostate cancer and hyperplasia patients.A graph indicates TGF-β concentrations in prostate cancer (PCa) and BPH patients and control individuals, each dot represents a single sample.*** (p < 0.001) shows significant difference from healthy individuals (Mann-Whitney U test).

Fig. 4 .
Fig. 4. Polymorphisms of JAK2 gene in prostate cancer and hyperplasia patients.DNA sequencing chromatograms of JAK2 gene from wildtype (1st panels) and variant (2nd and 3rd panels) genotypes of rs994555780, rs4495487 and rs10974947 SNPs are shown.Arrows indicate the location of the base changes.

Fig. 6 .
Fig. 6.Association of the CYLD c.2351-47 variant with immunophenotype in prostate cancer and hyperplasia patients.(A) Representative dot plots of CD56 + CD25 + expressing cells in PCa patients carrying the CC and CT genotypes of the CYLD c.2351-47 variant.(B) A graph indicates the percentage of CD56 + CD25 + expressing cells in PCa patients carrying the CC and CT genotypes of the CYLD c.2351-47 variant.** (p < 0.01) shows a significant difference from the CC genotype (Mann-Whitney U test).

Table 1 . Clinical characteristics of prostate cancer and hyperplasia patients.
ALT, alanine aminotransferase; AST, aspartate transaminase; PSA, prostate specific antigen; TGF, transforming growth factor; WBC, white blood cell; BPH, benign prostatic hyperplasia.p < 0.05 (in bold) indicates statistical significance between the prostate cancer (PCa) and BPH groups.

Table 2 . Genotype distribution of SNPs in CYLD, JAK2 and TLR4 genes in prostate cancer and hyperplasia patients.
Statistically significant results were represented in bold style.NC, not calculated for sparse data.SNPs, single nucleotide polymorphisms; TLR, Toll-like receptor.

Table 3 . General information of CYLD, JAK2 and TLR4 variants in prostate cancer and hyperplasia patients.
Position refers to the GRCh38.p10assembly; MAF, Minor allele frequency; HWE, Hardy-Weinberg equilibrium was checked by Chi-squared test; N/A, Not available.

Table 4 . Associations of the SNPs in CYLD, JAK2 and TLR4 genes with the serum PSA levels in prostate cancer and hyperplasia patients.
Statistically significant results were represented in bold style.NC, not calculated for sparse data.