Association of Toll-like receptors polymorphisms with the risk of acute lymphoblastic leukemia in the Brazilian Amazon

Acute lymphoblastic leukemia (ALL) is the most common hematologic malignancy in children in childhood. Single-nucleotide polymorphism (SNPs) in key molecules of the immune system, such as Toll-like receptors (TLRs) and CD14 molecules, are associated with the development of several diseases. However, their role in ALL is unknown. A case–control study was performed with 152 ALL patients and 187 healthy individuals to investigate the role of SNPs in TLRs and the CD14 gene in ALL. In this study, TLR6 C > T rs5743810 [OR: 3.20, 95% CI: 1.11–9.17, p = 0.003) and TLR9 C > T rs187084 (OR: 2.29, 95% CI: 1.23–4.26, p = 0.000) seems to be a risk for development of ALL. In addition, the TLR1 T > G rs5743618 and TLR6 C > T rs5743810 polymorphisms with protection against death (OR: 0.17, 95% IC: 0.04–0.79, p = 0.008; OR: 0.48, 95% IC: 0.24–0.94, p = 0.031, respectively). Our results show that SNPs in TLRs genes may be involved in the pathogenesis of ALL and may influence clinical prognosis; however, further studies are necessary to elucidate the role of TLR1, TLR4, TLR5, TLR6, TLR9 and CD14 polymorphisms in this disease.

Leukemia (acute and chronic) represents the 10th most frequent cause of cancer worldwide 1 . In Brazil, for the triennium 2020-2022, approximately 10,810 new cases in women and men are expected and approximately 300 new cases in North region of Brazil, where leukemia (acute and chronic) is the fifth most frequent cancer 2,3 . The most common type of leukemia in childhood is acute lymphoblastic leukemia (ALL) with a prevalence up to 25% of cancers in children who are under the age of 15 years 4 .
The neoplastic process results from genetic errors that contribute to blocking cell maturation and accumulation of leukemic clones (blasts) in the bone marrow microenvironment. Its etiology is still unknown; however, some risk factors are associated, including environmental, genetic and infectious factors 5,6 . Evidence from previous studies suggests that ALL is related to a deficit in immune system regulation in early childhood [7][8][9][10][11][12] . Furthermore, it is suggested that polymorphisms or genetic variations in the genes of molecules that are important in the development and progression of diseases may be important factors in the increase of intrinsic biological differences, influencing clinically distinct results and conferring genetic susceptibility to cancer 13,14 .
Toll-like receptors make up the main family of pattern recognition receptors (PRRs) of the innate immune system, and are involved in fighting pathogens and inflammation, and recognizing pathogen-associated molecular patterns (PAMPs) and damage-associated molecular patterns (DAMPs), which thus modulates the immune response via the activation of cells that mediate the immune response. In addition, TLRs are vital molecules www.nature.com/scientificreports/ in regulating the activation of adaptive immunity and are essential in preventing and curing cancer [15][16][17][18] . CD14 normally acts as a receptor for lipopolysaccharide (LPS) and LPS-binding proteins, which are responsible for IL-1, IL-6, and TNF-α inflammatory cytokine production. Studies have reported that CD14 may be involved in development of cancers such as acute lymphoblastic leukemia 19 . Single-nucleotide polymorphism (SNPs) in TLRs and CD14 genes have been identified in different types of tumors, both in solid tumors and hematologic malignancies 13,[20][21][22][23] . However, the role of immunogenic genes that play a crucial role in ALL is still poorly understood. In this study, we demonstrated the role of TLR1, TLR4, TLR5, TLR6, TLR9, and CD14 co-receptor polymorphisms in patients diagnosed with ALL.

Results
General characteristics of the study population. As shown in Table 1, the male gender was predominant in both groups with the median age 30 years for the control group and 12 years for case study group, 133 (71%) and 95 (62%), respectively. Most of the ALL patients had B-ALL (84%) and lived in Manaus (87%). A total of 64 (42%) had infectious comorbidities on diagnosis, 121 (80%) relapsed after induction therapy and 70 (46%) died. These patients also showed a median hemoglobin concentration of 8.6 g/dL, hematocrit of 25.1 g/dL, WBC 9.435/mm 3 and platelets 58.000/mm 3 . TLR6 C > T rs5743810 and TLR9 C > T rs5743836 polymorphisms predict risk of acute lymphoblastic leukemia. In our analyses, only TLR5 R > S rs5744105 (OR: 1.39, 95% CI: 1.03-1.89, p = 0.000) and TLR6 C > T rs5743810 (OR: 0.57, 95% CI: 0.34-0.95, p = 0.033) in case group deviated from Hardy-Weinberg equilibrium. In Table 2 Table 2. We performed a regression model by ancestrally but no difference was observed.
Association of TLR1 T > G rs5743618 and TLR6 C > T rs5743810 polymorphisms with death in acute lymphoblastic leukemia. In Table 3

Discussion
Genetic variations in the genes of molecules that are important in the immune response that is involved with the progression of diseases may prove to be important factors in the amplification of intrinsic biological differences, thus influencing clinically distinct results and conferring genetic susceptibility to cancer 13,14 . Single-variant polymorphisms in Toll-like receptors generally altered the TLRs ability to recognize pathogens in two forms: first, by increasing the response against pathogens that contribute to persistent inflammation and cancer and, second, by decreasing their response that contributes to infection susceptibility, as we see in chronic lymphoblastic leukemia (CLL) 13,18,[24][25][26] . The state of Amazonas, located in the Amazonian region of Brazil, is known for its tropical climate and a highly heterogeneous population that is exposed to several pathogens. In addition, it has different endemic areas for infectious diseases, which could contribute to the modulation of the immune response and consequently Table 2. Analysis of the association TLR6 C > T rs5743810 and TLR9 C > T rs187084 with the risk of acute lymphoblastic leukemia. a OR: Odds Ratio; b 95% confidence interval; c p value: < 0.05; d AIC: Akaike information criterion value; e Adj: Adjusted for sex and age. www.nature.com/scientificreports/ the triggering of physiological, genetic, and hematological changes 27 . The biological profile of ALL is still little known, and has different patterns according to the geographic and ethnic regions of the world [27][28][29][30] .
It is important to note that the population in study is exclusively composed by mixed population. Variations in Brazilian population have been clearly described in literature with a high degree of admixture from Americans, African and/ or European ancestors 31 . Besides, the Amazon Region population has a high degree of inter-ethnic admixture due to the intense miscegenation process and the strong indigenous influence. In this study, it was found that the allelic frequency of TLR1 T > G rs5743618, CD14 C > T rs2569191, TLR4 A > G rs4986790, TLR4 C > T rs4986791 was similar to the American population, TLR6 C > T rs5743810 and TLR9 C > T rs187084 to European population, TLR9 C > T rs5743836 to African population and TLR5 R > S rs5744105 to East Asian population 32 . Previous studies demonstrated that children with admixed ancestry have a high risk of developing ALL because of the Native American ancestry 33 , which is highly concentrated in the north of the country, within the Amazon region 34 and might be one of the reasons to explain the number of cases in the region. However, more genetic studies in this population are required to confirm this hypothesis.
To our knowledge, this is the first report to investigate the possible association between polymorphisms in TLR and CD14 genes and the susceptibility to acute lymphoblastic leukemia in the Brazilian Amazon region. In our study, TLR6 C > T rs5743810 and TLR9 C > T r5743836 polymorphisms seem to be a risk for developing acute lymphoblastic leukemia. TLR-6 is a partner of TLR-2, which is responsible for PAMP recognition of invading pathogens and induces inflammation through myeloid differentiation primary response protein 88 (MyD88) and TRAF6 mediated activation of NF-κappa-B (NF-kB). In addition, in most studies, TLR6 C > T rs5743810 is associated with infectious diseases like tuberculosis in the African population and the induction of resistance to Table 3. Analysis of the association of single nucleotide polymorphisms (SNPs) in study with the risk of death in acute lymphoblastic leukemia patients. a OR: Odds Ratio; b 95% confidence interval; c p value: < 0.05; d AIC: Akaike information criterion value; e Adj: Adjusted for sex and age. Significant values are in italics. www.nature.com/scientificreports/ asthma in children 35 . In CLL, high TLR6 expression was observed in cells of the patients 36 , but it is still not clear what the role of this polymorphism is in leukemia. TLR-9 is found in the endoplasmic reticulum membrane and functions via the MyD88-dependent pathway, which leads to NF-kB activation, cytokine secretion, and inflammatory response. A meta-analysis showed that TLR9 polymorphisms are associated with increased risk for cancer and some hematological neoplasms, such as non-Hodgkin's lymphoma and acute myeloid leukemia (AML) [37][38][39] . In AML, the [C] allele of TLR9 C > T rs187084 and [T] TLR9 C > T rs5743836 are associated with disease development, as is the TT genotype of the TLR9 C > T rs5743836 polymorphism with relapse episodes 40 . Additionally, the TLR9 gene is significantly expressed in chronic lymphocytic leukemia cells 36 .
Since the exposure of hematopoietic stem cells (HSCs) to TLR ligands influences the cycling, differentiation, and their functions, chronic TLR stimulation occurs, which leads to the impairment of normal HSC repopulating activity. Furthermore, high TLR expression and signaling are associated with myelodysplastic syndromes (MDS), which are a group of hematological neoplasms with a high risk of transformation to acute leukemias 13,41 . Therefore, these studies suggested that TLRs seem to be involved in the chronic neoplastic process, but further studies are necessary to confirm this.
A study involving the Turkish population showed no association between TLR4 A > G rs4986790 and TLR4 C > T rs4986791 and the risk for ALL 42 . Interestingly, In a cohort of AML patients, both these SNPs were independent risk factors for the development of sepsis and pneumonia 43 . In our study, we did not observed association between SNPs in TLRs with comorbidities infectious and, unfortunately, it was not possible to determine the association between the SNPs with cause of death because of lack information. Polymorphisms in TLR4 are responsible for decreased recognition of ligands (e.g., LPS). Since homozygotes and heterozygotes are hyporesponsive to LPS stimulation, this polymorphism may be associated with episodes of sepsis 44 . In addition, in B-chronic lymphoblastic leukemia (B-CLL), the reduced TLR4 expression was associated with increased risk of disease progression and a poor outcome, infectious episodes, and evolution to autoimmune diseases, which suggests an impaired innate immunity 45 . In the literature, this SNP is associated with the risk of cervical 46 , rectal and neck cancer 47 .
It is important to note that our study has some limitations: (i) the sample size is a factor that influences the non-significance of the variants under study, however, studies with a larger sample size are required to increase the statistical power; (ii) the cytokines quantification and gene expression would allow us to better understand the influence of these variants in acute lymphoblastic leukemia patients but we understand that their absence not necessary weak our conclusions; (iii) our control group was composed by blood donors (> 18 years). Although we have included this topic as a limitation of the study, we understand that when using children as a control group, results can be obtained that do not reflect reality, since they could develop the disease after the study occurs; (iv) we did not have access to cause of the death in patients during chemotherapy (disease progression, infection, toxicity and other reason); (v) our clinical data collection was retrospective, so it is possible, for example, that not all patients with infections were identified, since we considered as infected only those who had requests for the described pathogens. Another point is that this was not always requested or the diagnosis was not available. Moreover, the patient could have an infection with some untested pathogen. This leads us to think that the number of patients with infectious comorbidities may be higher.

Material and methods
Patients and controls. A total of 152 pediatric ALL patients diagnosed according to the criteria of the World Health Organization (WHO) 48 , who received treatment at Fundação Hospitalar de Hematologia e Hemoterapia do Amazonas (HEMOAM), 0 to 18 years old, either gender and unrelated were included in the case study group. In the control group, 187 blood donor candidates of either gender and age 18 to 65 years old and who were considered healthy according to the Brazilian Ministry of Health technical standards were included 49 Table 1. PCR-RFLP generated fragments were separated using 3-4% agarose gel electrophoresis stained with GelRed™ nucleic acid gel stain (Biotium, Hayward, CA, USA) and visualized on a UV light gel + DocXR system transilluminator (Bio-Rad Corporation, Hercules, CA, USA) with a photographic documentation system.  55 were included and, for death, patients who died within 5 years were included.
Descriptive and statistical analysis. Comparison between groups was performed using Fisher's exact test using GraphPad Prism v.5.0, with a significance level of 5%. Allele analysis was performed using the website https:// ihg. helmh oltz-muenc hen. de/ ihg/ snps. html. A logistic regression analysis that was adjusted for sex and age was performed to find associations between the genotype frequencies and ALL, infectious comorbidities, and death using the package "SNPassoc" version 1.9-2 (https:// cran. rproj ect. org/ web/ packa ges/ SNPas soc/ index. html) for R software version 3.4.3 (www.r-proje ct. org). The best genetic model was performed using the Akaike information criterion (AIC). Hardy-Weinberg equilibrium was evaluated for all SNPs. Bonferroni correction for multiple tests was also performed and the p-value adjusted was reported in bold italic characters. The results were shown as the odds ratio (OR) and 95% confidence intervals (95% CI) from multivariate logistic regression analyses.
Ethics approval and consent to participate. All protocols and consent forms were approved by the Research Ethics Committee at the HEMOAM Foundation (CEP/HEMOAM process 3.335.123/2019). The informed consent form was obtained from all patients and parents/legal guardians for minors in study. This study was carried out following the guidelines of the Declaration of Helsinki and Resolution 466/12 of the Brazilian National Health Council for research involving human beings.

Conclusion
To our knowledge, this is the first study to describe the frequency of SNPs in the TLR1, TLR4, TLR5, TLR6, TLR9 and CD14 genes in patients with ALL in the Brazilian Amazon region. Our study demonstrated that TLR6 C > T rs5743810 and TLR9 C > T rs5743836 polymorphisms are associated with the risk of acute lymphoblastic leukemia and TLR1 T > G rs5743618 and TLR9 C > T rs5743810 is involved with death. Further studies are necessary to elucidate the role of the TLR1, TLR4, TLR5, TLR6, TLR9, and CD14 polymorphisms in the pathogenesis of leukemia.

Data availability
All data generated or analysed during this study are included in this published article.