Abstract
Background/Objectives
Several recent studies suggest a possible role of lymphocyte activation 3 (LAG3) protein. LAG3 can behave as an α‐synuclein ligand, and serum and cerebrospinal fluid‐soluble LAG3 levels have been proposed as a marker of Parkinson's disease (PD). In this study, we aimed to investigate whether there is an association between 3 common single‐nucleotide variations (SNVs) in the LAG3 gene and its closely related CD4 molecule gene and the risk of PD in a Caucasian Spanish population. Two of them have been previously associated with the risk of PD in Chinese females.
Methods
We analysed genotypes and allele frequencies for CD4 rs1922452, CD4 951818 and LAG3 rs870849 SNVs, by using specifically designed TaqMan assays, in a cohort composed of 629 PD patients and 865 age‐ and gender‐matched healthy controls.
Results
The frequencies of the CD4 rs1922452 A/A genotype, according to the dominant and recessive genetic models, and of the CD4 rs1922452/A allelic variant were significantly lower, and the frequencies of the CD4 rs951818 A/A genotype, according to the dominant genetic model, and of the CD4 rs951818/A allele, were significantly higher in PD patients than in controls. The differences were not significant after stratifying by sex. These two SNVs showed strong linkage. Regression models showed a lack of relation between the 3 SNVs studied and the age at onset of PD.
Conclusions
These data suggest a possible role of CD4 rs1922452 and CD4 rs951818 polymorphisms in the risk of PD.
Keywords: CD4 gene, genetics, LAG3 gene, Parkinson's disease, polymorphisms, risk factors
1. INTRODUCTION
More than 200 years after the initial description of Parkinson's disease (PD), its aetiology has not been clearly established. Data from many studies reported during the last 35 years suggest that PD should be considered a genetically complex disease, with an interplay of both environmental and genetic factors. According to the Gene Database, there have been reported at least 24 genes related to susceptibility to familial PD, named PARK1 to PARK24 (link https://www.ncbi.nlm.nih.gov/gene, last accessed November 21, 2021), but these genes only explain a low percentage of the heritability of PD. Likely, meta‐analyses from hypothesis‐driven case–control association studies on candidate genes (many of them with variable and inconsistent results) and data from hypothesis‐free Genome‐Wide Association Studies (GWAS) could provide the most reliable data regarding the possible role of genetic polymorphisms in the risk of PD, 1 although a detailed revision of these studies is out of the scope of the present article.
One of the most important pathological hallmarks of PD is the abnormal expression and aggregation of α‐synuclein (α‐syn), which is the main protein component of Lewy bodies. 2 Lymphocyte activation 3 (LAG3) protein, which belongs to the immunoglobulin superfamily and is expressed by microglia, neurons and peripheral immune cells, 3 can bind to α‐syn preformed fibrils with high affinity, and this binding initiates processes of endocytosis, transmission and toxicity induced by α‐syn preformed fibrils leading to loss of dopamine neurons. 4 Some studies reported that the absence of LAG3 in neuronal cultures 4 and depletion of LAG3 in α‐syn transgenic mice reduce aggregation of α‐syn. 5 By contrast, another study did not find expression of LAG3 in human and murine neurons and did not find that overexpression of LAG3 in cultured human neural cells caused worsening of α‐syn pathology, and did not confirm changes in survival in α‐syn transgenic mice by LAG3 depletion. 6 Serum soluble LAG3 levels are significantly increased in PD patients compared with controls in two studies. 7 , 8 Another study showed similar serum soluble LAG3 levels but increased CSF‐soluble LAG3 concentrations in PD patients compared with controls. 9 These data suggest that LAG3 could be a potential biomarker for PD, and its genetic variability can be related to the risk of presenting PD.
LAG3 protein is encoded by the LAG3 gene (also known as CD223 gene; chromosome 12p13.31; gene ID 3902, MIM 153337), which contains 8 exons and is closely related to the CD4 molecule gene, (CD4; gene ID 920, MIM 186940; this gene encodes the CD4 membrane glycoprotein of T lymphocytes, is expressed in several brain regions, acts as a coreceptor with the T‐cell receptor to recognise antigens, participates in the early phase of T‐cell activation and mediates indirect neuronal damage in immune‐mediated and infectious diseases of the central nervous system). LAG3 gene has been described recently as a possible predictor of the development of brain regional atrophy in PD patients. 10
LAG3 (12:6881678‐6887621) and CD4 (12:6896024‐6929974) are close genes. The main single nucleotide polymorphisms (SNVs) in LAG3 and CD4 genes are rs870849 T > C (chromosomal position 12:6887020, a missense variant located in the LAG3 gene; its T allele has been related to protection against severity of primary immune thrombocytopenia), 11 rs1922452 A > G (chromosomal position 12:6896194, located in an intron within the CD4 gene; it is associated with comorbidity in multiple sclerosis), 12 and rs951818 C > A (chromosomal position 12:6896055, located in a noncoding transcript exon in the CD4 gene; it has been related to disease progression and mortality of sepsis). 13
A recent case–control study, involving 646 PD patients and 536 healthy controls from a Chinese population, described significantly higher frequencies of CD4 rs1922452/AA and CD4 rs951818/CC genotypes in PD females than in their respective controls. 9 This study aims to replicate these findings in a large cohort of Caucasian Spanish PD patients and healthy controls.
2. METHODS
2.1. Study participants
The study involved 629 patients diagnosed with idiopathic PD (all of them aged more than 18 years) according to the UK Parkinson's Disease Society Brain Bank Clinical Diagnostic Criteria, 14 all of them were recruited in the Movement Disorders Units from 4 Hospitals and examined personally by consultant neurologists specialised in this area, and 865 age‐ and sex‐matched healthy controls. Recruitment of PD patients was done by 2 of these neurologists who served as Neurology consultants in 2 different hospitals (H.A‐N and F.J.J‐J), another who served as a Neurology consultant in other 2 of these hospitals (P.P), and another 2 in a single centre (M.B. and MA) through the inclusion period, being the inclusion criteria homogeneous. Controls were recruited from the University Hospital, Badajoz, Spain (404 subjects, most of them were students of staff from the University of Extremadura), and at the Clinica Universitaria de Navarra, Pamplona, Spain (461 subjects, healthy spouses of patients visiting this Hospital). All control individuals had no family history of PD, had no systemic or neurological diseases (including PD and other movement disorders) and underwent a medical examination prior to inclusion. Participants were recruited between January 2003 and March 2018. Table 1 summarises the demographic data of both PD patients and control subjects. These individuals have participated in a previous genetic association study. 15
TABLE 1.
Demographic data
| Group | PD patients (n = 629) | Controls (n = 865) | P |
|---|---|---|---|
| Age, years, mean (SD) | 66.65 (10.53) | 65.22 (10.79) | .071 |
| Age range, years | 22–95 | 19–92 | — |
| AAO, years, mean (SD) | 57.68 (14.03) | NA | — |
| AAO range, years | 14–85 | NA | — |
| Men n (%) | 295 (46.9) | 392 (45.3) | .924 |
Abbreviations: AAO, age at onset; NA, not available; PD, Parkinson's disease; SD, standard deviation.
2.2. Ethical aspects
The principles of the Helsinki Declaration were applied to the participants' recruitment, all of them signing their written informed consent after a full explanation of the study purpose and procedure. The study was approved by the Ethics Committees of Clinical Investigation of the Clínica Universitaria de Navarra (Pamplona, Spain), the Hospital Universitari Mutua de Terrassa (Terrassa, Barcelona, Spain) and the Infanta Cristina University Hospital (Badajoz, Spain).
2.3. Genotyping of LAG3 rs870849, rs1922452 and rs951818 variants
Genotyping was performed in genomic DNA obtained from peripheral leukocytes of venous blood samples of patients diagnosed with PD and controls. The analysis was performed by using real‐time PCR (Applied Biosystems 7500 qPCR thermocycler) with specific custom‐designed TaqMan probes (Life Technologies). Since this study aims to replicate previous findings, we selected the same SNVs analysed in the previous study. These SNVs include the only missense SNVs with an allele frequency over 0.01 in the population analysed and one intronic and one noncoding transcript exonic SNVs with high allele frequencies both, in the discovery population and in the population analysed here. The SNVs analysed and the TaqMan assays are rs870849 (C___9797874_10), rs1922452 (C__11914936_10) and rs951818 (C___8921385_10).
2.4. Statistical analysis
We analysed the Hardy–Weinberg equilibrium by using the online application http://ihg.gsf.de/cgi‐bin/hw/hwa1.pl, and the statistical package PLINK 16 was run to analyse the allele and the genotype frequencies. Finally, the spss 20.0 package (SPSS Inc.) was used to carry out the rest of the statistical analyses. χ 2 or, when appropriate, Fisher tests were used to perform intergroup comparisons and the false discovery rate procedure 17 to carry out the correction for multiple testing. Crude and corrected values (P and Pc, respectively) were obtained for each intergroup comparison, including all genotypes or all alleles. Haplotype analysis was carried out by using the snpstats software (https://www.snpstats.net/).
To calculate the sample size, that was carried out based on the minor allele frequencies in control individuals, we analysed the frequency for carriers of the risk gene, using a relative risk (RR) value equal to 1.5 (P < .05). 18 , 19 The calculation of the statistical power for variant alleles in this study, based on the allele frequencies observed in healthy controls and in patients was, respectively, for one‐tailed and two‐tailed associations: rs870849 99.9% and 99.9%, rs1922452 99.9% and 99.9%, and rs951818 99.9% and 99.9%. The negative predictive values were calculated as described elsewhere. 20 Finally, a T‐test for independent samples was used for the comparison of the age at PD onset across genotype categories for the 3 SNVs studied.
Reporting of the study conforms to broad EQUATOR guidelines, 21 specifically STROBE and STREGA checklists are summarised in Table S1.
3. RESULTS
The genotypes of the 3 allelic variants studied were in Hardy–Weinberg's equilibrium, both in PD patients and controls. The P‐values for the Hardy–Weinberg's equilibrium in the SNVs analysed were equal to 0.870, 0.940 and 0.600 for the SNVs rs1922452 A/G, rs951818 A/C and rs870849 C/T, respectively. Table 2 shows the genotypes of PD patients and control subjects according to different genetic models. According to these results, the model providing the strongest predictive capacity is the dominant model. According to this model, the SNVs rs1922452 A/G and rs951818 A/C present statistically significant differences when comparing patients and control subjects. Such differences remain significant after false discovery rate (FDR) correction for multiple comparisons. The comparison of the allele frequencies, which is summarised in Table 3 indicated a statistically significant effect of the alleles rs1922452/G and rs951818/A, both being more frequent among patients with PD than in control individuals. The statistical significance is higher with the rs1922452/G variant but, for both SNVs, the differences remain significant after FDR correction for multiple comparisons.
TABLE 2.
Genotypes of patients with PD and healthy volunteers (control subjects) according to different genetic models
| Variant | Genotype, n (%) | Codominant OR (95% CI) | P, Pc | Dominant OR (95% CI) | P, Pc | Recessive OR (95% CI) | P, Pc | Overdominant OR (95% CI) | P, Pc | ||
|---|---|---|---|---|---|---|---|---|---|---|---|
| rs1922452 | AA | AG | GG | ||||||||
| Control | 140 (16.2) | 414 (47.9) | 311 (36.0) | ||||||||
| PD | 74 (11.8) | 294 (46.7) | 261 (41.5) |
0.85 (0.68–1.06) 0.63 (0.45–0.87) |
.018, .054 | 0.79 (0.64–0.98) | .030, .045 | 0.69 (0.51–0.93) | .015, .045 | 0.96 (0.78–1.17) | .670, .670 |
| rs951818 | AA | AC | CC | ||||||||
| Control | 299 (34.6) | 426 (49.2) | 140 (16.2) | ||||||||
| PD | 252 (40.1) | 291 (46.3) | 86 (13.7) |
0.81 (0.65–1.01) 0.73 (0.53–1.00) |
.075, .113 | 0.79 (0.64–0.98) | .030, .045 | 0.82 (0.61–1.10) | .180, .270 | 0.89 (0.72–1.09) | .250, .670 |
| rs870849 | CC | CT | TT | ||||||||
| Control | 340 (39.3) | 399 (46.1) | 126 (14.6) | ||||||||
| PD | 225 (35.8) | 301 (47.9) | 103 (16.4) |
1.14 (0.91–1.43) 1.24 (0.91–1.68) |
.330, .330 | 1.16 (0.94–1.44) | .160, .160 | 1.15 (0.87–1.52) | .340, .340 | 1.07 (0.87–1.32) | .510, .670 |
Abbreviation: PD, Parkinson's disease. Statistically significant values are marked in bold.
TABLE 3.
Allele frequencies patients with PD and healthy volunteers (control subjects)
| Variant | Allele, n (%) | OR (95% CI) | P | Pc | NPV (95% CI) | |
|---|---|---|---|---|---|---|
| rs1922452 | A | G | ||||
| Control | 694 (40.1) | 1036 (59.9) | ||||
| PD | 442 (35.1) | 816 (64.9) | 0.81 (0.70–0.94) | .006 | .018 | 0.56 (0.55–0.57) |
| rs951818 | A | C | ||||
| Control | 1024 (59.2) | 706 (40.8) | ||||
| PD | 795 (63.2) | 463 (36.8) | 1.18 (1.02–1.38) | .027 | .041 | 0.60 (0.58–0.63) |
| rs870849 | C | T | ||||
| Control | 1079 (62.4) | 651 (37.6) | ||||
| PD | 751 (59.7) | 507 (40.3) | 0.90 (0.77–1.04) | .139 | .139 | 0.56 (0.54–0.59) |
Note: Test for trend for rs1922452: OR = 1.25; chi‐square = 7.71; P = .005. Test for trend for rs951818: OR = 0.85; chi‐square = 4.94; P = .026.
Abbreviations: NPV, negative predictive value; P, crude probability; Pc, probability after multiple comparisons; PD, Parkinson's disease.
We calculated the linkage disequilibrium between the three SNVs analysed. There is a strong linkage between the SNVs rs1922452 A/G and rs951818 A/C, with a D′ value equal to 0.950, an r‐square value equal to 0.929 and a P‐value equal to .0001. This is in agreement with the already described linkage for these two SNVs in the Iberian population in Spain, with D′ value equal to 1.000, r‐square equal to 1.000 and P‐value equal to .0001 according to the online tool LDpair from the National Cancer Institute (https://ldlink.nci.nih.gov/?tab=ldpair). In contrast, the SNV rs870849 C/T is not linked to the previously mentioned SNVs, with a D′ value equal to 0.313 and 0.34, an r‐square value of .195 and −.194, and a P‐value equal to .068 and .072 as compared to the SNVs rs1922452 A/G and rs951818 A/C, respectively. This is in agreement with the data obtained in LDpair for Iberian subjects: D′ equal to 0.231 and 0.231, r‐square equal to .019 and .019, and P‐value equal to .044 and .044, for the SNVs rs1922452 A/G and rs951818 A/C, respectively. Therefore our findings suggest that the two SNVs that are at linkage disequilibrium are associated with the risk of developing PD.
To refine the potential of the SNV genotyping analyses to identify the risk alleles, we analysed the putative role of the haplotypes with the risk and the results are summarised in Table 4. We identified 8 haplotypes in the population studied, but none of these had a strong association with the risk, thus suggesting that the putative effect of the SNVs is not accumulative.
TABLE 4.
Haplotype analysis
| Haplotype (rs1922452 AG, rs951818 AC, rs870849 CT) | PD patients (%) | Controls (%) | OR (95% CI) | P |
|---|---|---|---|---|
| G A C | 0.327 | 0.319 | Reference | |
| G A T | 0.296 | 0.265 | 1.08 (0.87–1.34) | .480 |
| A C C | 0.245 | 0.290 | 0.81 (0.65–1.01) | .066 |
| A C T | 0.097 | 0.097 | 0.95 (0.71–1.27) | .730 |
| G C C | 0.018 | 0.012 | 1.47 (0.76–2.83) | .250 |
| Rare haplotypes | 0.017 | 0.022 | 0.72 (0.41–1.26) | .250 |
Note: Rare haplotypes (combining any haplotypes with frequencies under 0.01) include the combinations GCT, AAC and AAT.
Abbreviation: PD, Parkinson's disease.
Table 5 shows the genotypes of PD patients and control subjects, stratified by sex, according to different genetic models. The differences that were observed in the whole study group (Table 2) were not statistically significant when participants were stratified by sex. Neither were significant the allele frequencies as shown in Table 6. The trends towards the risk of genotypes and allele frequencies were similar to those reported for the whole series, but the results were not statistically significant because of the sample size when patients were subdivided.
TABLE 5.
Genotypes of patients with PD and healthy volunteers (control subjects), stratified by sex, according to different genetic models
| Variant | Genotype, n (%) | Codominant OR (95% CI) | P, Pc | Dominant OR (95% CI) | P, Pc | Recessive OR (95% CI) | P, Pc | Overdominant OR (95% CI) | P, Pc | ||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Women | |||||||||||
| rs1922452 | AA | AG | GG | ||||||||
| Control | 79 (16.7) | 225 (47.6) | 169 (35.7) | ||||||||
| PD | 42 (12.6) | 153 (45.8) | 139 (41.6) |
0.87 (0.63–1.20) 0.61 (0.37–1.00) |
.140, .300 | 0.81 (0.59–1.10) | .170, .210 | 0.66 (0.42–1.04) | .071, .213 | 0.98 (0.73–1.33) | .910, .910 |
| rs951818 | AA | AC | CC | ||||||||
| Control | 168 (35.5) | 229 (48.4) | 76 (16.1) | ||||||||
| PD | 134 (40.1) | 155 (46.4) | 45 (13.5) |
0.77 (0.55–1.07) 0.71 (0.45–1.13) |
.200, .300 | 0.75 (0.55–1.03) | .076, .210 | 0.83 (0.54–1.27) | .380, .430 | 0.85 (0.63–1.15) | .280, .795 |
| rs870849 | CC | CT | TT | ||||||||
| Control | 185 (39.1) | 221 (46.7) | 67 (14.2) | ||||||||
| PD | 122 (36.5) | 160 (47.9) | 52 (15.6) |
1.19 (0.85–1.66) 1.30 (0.83–2.04) |
.430, .430 | 1.22 (0.89–1.67) | .210, .210 | 1.18 (0.79–1.78) | .430, .430 | 1.10 (0.81–1.49) | .530, .795 |
| Men | |||||||||||
| rs1922452 | AA | AG | GG | ||||||||
| Control | 61 (15.6) | 189 (48.2) | 142 (36.2) | ||||||||
| PD | 32 (10.8) | 141 (47.8) | 122 (41.4) |
0.83 (0.61–1.12) 0.65 (0.42–1.00) |
.120, .360 | 0.78 (0.58–1.04) | .090, .270 | 0.72 (0.48–1.07) | .100, .300 | 0.93 (0.70–1.23) | .620, .740 |
| rs951818 | AA | AC | CC | ||||||||
| Control | 131 (33.4) | 197 (50.3) | 64 (16.3) | ||||||||
| PD | 118 (40.0) | 136 (46.1) | 41 (13.9) |
0.85 (0.63–1.15) 0.74 (0.48–1.14) |
.340, .510 | 0.82 (0.62–1.10) | .180, .270 | 0.81 (0.55–1.21) | .310, .465 | 0.92 (0.70–1.22) | .570, .740 |
| rs870849 | CC | CT | TT | ||||||||
| Control | 155 (39.5) | 178 (45.4) | 59 (15.1) | ||||||||
| PD | 103 (34.9) | 141 (47.8) | 51 (17.3) |
1.10 (0.81–1.49) 1.18 (0.77–1.81) |
.720, .720 | 1.12 (0.84–1.49) | .460, .460 | 1.12 (0.75–1.66) | .580, .580 | 1.05 (0.79–1.39) | .740, .740 |
Abbreviation: PD, Parkinson's disease.
TABLE 6.
Alleles of patients with PD and healthy volunteers (control subjects) stratified by sex
| Variant | Allele, n (%) | OR (95% CI) | P | Pc | NPV (95% CI) | |
|---|---|---|---|---|---|---|
| Women | ||||||
| rs1922452 | A | G | ||||
| Control | 383 (40.5) | 563 (59.5) | ||||
| PD | 237 (35.5) | 431 (64.5) | 0.81 (0.66–0.99) | .042 | .126 | 0.57 (0.55–0.59) |
| rs951818 | A | C | ||||
| Control | 565 (59.7) | 381 (40.3) | ||||
| PD | 423 (63.3) | 245 (36.7) | 1.16 (0.95–1.43) | .144 | .216 | 0.61 (0.58–0.64) |
| rs870849 | C | T | ||||
| Control | 591 (62.5) | 355 (37.5) | ||||
| PD | 404 (60.5) | 264 (39.5) | 0.92 (0.75–1.13) | .417 | .417 | 0.57 (0.54–0.61) |
| Men | ||||||
| rs1922452 | A | G | ||||
| Control | 311 (39.7) | 473 (60.3) | ||||
| PD | 205 (34.7) | 385 (65.3) | 0.81 (0.65–1.01) | .062 | .137 | 0.55 (0.53–0.57) |
| rs951818 | A | C | ||||
| Control | 459 (58.5) | 325 (41.5) | ||||
| PD | 372 (63.1) | 218 (36.9) | 1.21 (0.97–1.51) | .091 | .137 | 0.60 (0.57–0.63) |
| rs870849 | C | T | ||||
| Control | 488 (62.2) | 296 (37.8) | ||||
| PD | 347 (58.8) | 243 (41.2) | 0.87 (0.70–1.08) | .197 | .197 | 0.55 (0.52–0.58) |
Abbreviations: NPV, negative predictive value; P, crude probability; Pc, probability after multiple comparisons; PD, Parkinson's disease.
Age at onset of PD was similar for the 3 possible genotypes of CD4 rs1922452, CD4 rs951818 and LAG3 rs870849 polymorphisms (Table 7). Regression models including sex and the age at onset were carried out. When the dependent variable was the risk of developing PD, three independent variables were significant: sex (P < .001), rs1922452 (P = .005) and rs951818 (P = .036). When the dependent variable was the age at onset, only sex (P < .001) was a significant factor. Although previous studies did not identify sex‐related differences in the age at onset of PD, 22 , 23 , 24 in our cohort the age at onset is slightly higher in men (58.66 years, SD 11.89) than in women (56.28 years, SD 12.68); two‐tailed T‐test P = .036. This may be due to chance but would explain the finding suggesting that sex is a confounder in this particular study.
TABLE 7.
Age at onset of PD according to the genotypes
| Age at onset (SD) years | |||
|---|---|---|---|
| Two‐Tailed T‐Test compared with A/A | Two‐Tailed T‐Test compared with A/G | ||
| rs1922452 AA | 56.53 (12.68) | ||
| rs1922452 AG | 57.69 (11.97) | 0.547 | |
| rs1922452 GG | 57.35 (12.25) | 0.677 | 0.776 |
| Two‐Tailed T‐Test compared with A/A | Two‐Tailed T‐Test compared with A/C | ||
| rs951818 AA | 57.31 (12.15) | ||
| rs951818 AC | 57.57 (12.29) | 0.836 | |
| rs951818 CC | 56.45 (11.89) | 0.633 | 0.537 |
| Two‐Tailed T‐Test compared with C/C | Two‐Tailed T‐Test compared with C/T | ||
| rs870849 CC | 58.22 (11.35) | ||
| rs870849 CT | 56.74 (11.91) | 0.220 | |
| rs870849 TT | 57.32 (14.21) | 0.589 | 0.717 |
Abbreviation: PD, Parkinson's disease.
4. DISCUSSION
Since the previously commented data suggest that LAG3 could be a reliable marker for PD, 7 , 8 , 9 it seems reasonable to address the possible association between the most relevant polymorphisms in the LAG3 gene and in its closely related CD4 gene with the risk of developing PD.
In this replication study, we found a modest risk increase for PD in carriers of the CD4 rs951818/A allelic variant, while carriers of the CD4 rs1922452/A had a modestly decreased risk of this disease. This result was not shown when analysing male and female individuals separately, a fact that should likely be related to an effect size. By contrast, we did not find an association between LAG3 rs870849 SNV and the risk of PD.
It should be stated that the variant allele frequencies are similar in control individuals regarding the SNV rs870849/C in our study (0.624) and that described in East Asian individuals according to the gnomAD database (0.644). Also are similar the frequencies for rs951818/A (0.401 in our study and 0.352 in East Asians). In contrast, the frequencies for rs1922452/G are higher (0.599) in our study as compared with that of East Asians (0.352), thus increasing the chance of findings individuals with variant genotypes and therefore increasing the statistical power for this SNV in our study.
To our knowledge, only a previous study on Chinese subjects addressed the possible relationship between these polymorphisms and the risk of PD. 9 In such study, data on the whole series did not show an association between CD4 rs1922452, CD4 rs951818 and LAG3 rs870849 and PD. However, in accordance with our findings, an association between CD4 rs951818 and increased risk of PD (although in this study the association was restricted to females) was reported. 9 By contrast, our data on Caucasians suggested that CD rs1922452/A might be associated with decreased risk of developing PD. No information on such SNVs in the PD GWAS Locus Browser (https://pdgenetics.shinyapps.io/GWASBrowser/) was identified. This is expected since, although we identified statistically significant association in this study (Tables 2, 3 and 6) the strength of the association (P‐values) are well below the threshold necessary to identify associations in GWAS.
The current study has several limitations that include the possibility of a selection bias (perhaps related to the fact that patient recruitment was done in a hospital setting), the relatively low sample size, despite our previous calculation of statistical power, which was adequate for odds‐ratio (OR) detection of 1.5 and the lack of similar previous studies in Caucasians. These limitations warrant replication studies. In addition, it could not be excluded the possibility that some healthy controls subjects who participated in the study might develop PD in the future, but taking into account the PD incidence rates in subjects older than 65 in Spain 25 and the proportion of healthy controls carrying the risk genotype that might eventually develop PD, it is unlikely that this fact had a significant influence on the results of this study.
In summary, this study suggests a weak, although statistically significant, association between CD4 rs1922452 and CD4 rs951818 polymorphisms and the risk of PD in the Caucasian Spanish population.
AUTHORS' CONTRIBUTIONS
All authors fulfil the criteria of authorship, and no one else who fulfils the criteria has been excluded. All of them have approved the final submitted version. EGM involved in drafting/revising the manuscript for content, including medical writing for content, study concept or design, acquisition of data, interpretation of data, study supervision and coordination, and obtaining funding. PP involved in drafting/revising the manuscript for content, including medical writing for content, study concept or design, acquisition of data and interpretation of data. JGT involved in drafting/revising the manuscript for content, including medical writing for content and acquisition of data. HAN involved in drafting/revising the manuscript for content, including medical writing for content, study concept or design, acquisition of data, interpretation of data, study supervision and coordination. IA involved in drafting/revising the manuscript for content, including medical writing for content and acquisition of data. MB involved in drafting/revising the manuscript for content, including medical writing for content and acquisition of data. MOCA involved in drafting/revising the manuscript for content, including medical writing for content and acquisition of data. MA involved in drafting/revising the manuscript for content, including medical writing for content and acquisition of data. JAGA involved in drafting/revising the manuscript for content, including medical writing for content, study concept or design, acquisition of data, statistical analysis and interpretation of data, study supervision and coordination, and obtaining funding. FJJJ involved in drafting/revising the manuscript for content, including medical writing for content, study concept or design, acquisition of data, analysis or interpretation of data, study supervision and coordination.
CONFLICT OF INTEREST
All authors declare that there is no financial or nonfinancial conflict of interest.
Supporting information
Tables S1–S2
ACKNOWLEDGMENTS
This work was supported in part by Grants RETICS RD16/0006/0004 (ARADyAL), PI18/00540 and PI21/01683 from Fondo de Investigación Sanitaria, Instituto de Salud Carlos III, Madrid, Spain, and IB20134 and GR21073 from Junta de Extremadura, Mérida, Spain. Partially funded with FEDER funds.
García‐Martín E, Pastor P , Gómez‐Tabales J, et al. Association between LAG3/CD4 gene variants and risk of Parkinson's disease. Eur J Clin Invest. 2022;52:e13847. doi: 10.1111/eci.13847
Contributor Information
José A. G. Agúndez, Email: jagundez@unex.es.
Félix Javier Jiménez‐Jiménez, Email: fjavier.jimenez@salud.madrid.org, Email: felix.jimenez@sen.es.
REFERENCES
- 1. Jiménez‐Jiménez FJ, Alonso‐Navarro H, García‐Martín E, Agúndez JA. Advances in understanding genomic markers and pharmacogenetics of Parkinson's disease. Expert Opin Drug Metab Toxicol. 2016;12(4):433‐448. [DOI] [PubMed] [Google Scholar]
- 2. He S, Zhong S, Liu G, Yang J. Alpha‐synuclein: the interplay of pathology, neuroinflammation, and environmental factors in Parkinson's disease. Neurodegener Dis. 2020;20(2–3):55‐64. [DOI] [PubMed] [Google Scholar]
- 3. Angelopoulou E, Paudel YN, Villa C, Shaikh MF, Piperi C. Lymphocyte‐activation gene 3 (LAG3) protein as a possible therapeutic target for Parkinson's disease: molecular mechanisms connecting neuroinflammation to α‐synuclein spreading pathology. Biology. 2020;9(4):86. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Mao X, Ou MT, Karuppagounder SS, et al. Pathological α‐synuclein transmission initiated by binding lymphocyte‐activation gene 3. Science. 2016;353(6307):aah3374. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Gu H, Yang X, Mao X, et al. Lymphocyte activation gene 3 (Lag3) contributes to α‐synucleinopathy in α‐synuclein transgenic mice. Front Cell Neurosci. 2021;15:656426. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Emmenegger M, De Cecco E, Hruska‐Plochan M, et al. LAG3 is not expressed in human and murine neurons and does not modulate α‐synucleinopathies. EMBO Mol Med. 2021;13(9):e14745. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Cui SS, Du JJ, Liu SH, Meng J, et al. Serum soluble lymphocyte activation gene‐3 as a diagnostic biomarker in Parkinson's disease: a pilot multicenter study. Mov Disord. 2019;34:138‐141. [DOI] [PubMed] [Google Scholar]
- 8. Roy A, Choudhury S, Banerjee R, Basu P, Kumar H. Soluble LAG‐3 and toll‐interacting protein: novel upstream neuro‐inflammatory markers in Parkinson's disease. Parkinsonism Relat Disord. 2021;91:121‐123. [DOI] [PubMed] [Google Scholar]
- 9. Guo W, Zhou M, Qiu J, et al. Association of LAG3 genetic variation with an increased risk of PD in Chinese female population. J Neuroinflammation. 2019;16(1):270. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Freeze B, Acosta D, Pandya S, Zhao Y, Raj A. Regional expression of genes mediating trans‐synaptic alpha‐synuclein transfer predicts regional atrophy in Parkinson disease. Neuroimage Clin. 2018;18:456‐466. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Wang S, Zhang X, Leng S, et al. Immune checkpoint‐related gene polymorphisms are associated with primary immune thrombocytopenia. Front Immunol. 2021;11:615941. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Al‐Eitan L, Qudah MA, Qawasmeh MA. Association of multiple sclerosis phenotypes with single nucleotide polymorphisms of IL7R, LAG3, and CD40 genes in a Jordanian population: a genotype‐phenotype study. Biomolecules. 2020;10(3):356. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Mewes C, Alexander T, Büttner B, et al. Effect of the lymphocyte activation gene 3 polymorphism rs951818 on mortality and disease progression in patients with sepsis‐a prospective genetic association study. J Clin Med. 2021;10(22):5302. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Hughes AJ, Daniel SE, Kilford L, Lees AJ. Accuracy of clinical diagnosis of idiopathic Parkinson's disease: a clinico‐pathological study of 100 cases. J Neurol Neurosurg Psychiatry. 1992;55(3):181‐184. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. García‐Martín E, Diez‐Fairen M, Pastor P, et al. Association between the missense alcohol dehydrogenase rs1229984T variant with the risk for Parkinson's disease in women. J Neurol. 2019;66(2):346‐352. [DOI] [PubMed] [Google Scholar]
- 16. Purcell S, Neale B, Todd‐Brown K, et al. PLINK: a tool set for whole‐genome association and population‐based linkage analyses. Am J Hum Genet. 2007;81(3):559‐575. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Benjamini Y, Hochberg Y. Controlling the false discovery rate: a practical and powerful approach to multiple testing. J Roy Statist Soc Ser B. 1995;57:289‐300. [Google Scholar]
- 18. Daly AK, Day CP. Candidate gene case‐control association studies: advantages and potential pitfalls. Br J Clin Pharmacol. 2001;52(5):489‐499. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Pértegas Díaz S, Pita FS. Cálculo del poder estadístico de un estudio. Cad Aten Primaria. 2003;10:59‐63. [Google Scholar]
- 20. Altman DG, Bland JM. Diagnostic tests 2: predictive values. BMJ. 1994;309(6947):102. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Simera I, Moher D, Hoey J, Schulz KF, Altman DG. A catalogue of reporting guidelines for health research. Eur J Clin Investig. 2010;40(1):35‐53. [DOI] [PubMed] [Google Scholar]
- 22. Baba Y, Putzke JD, Whaley NR, Wszolek ZK, Uitti RJ. Gender and the Parkinson's disease phenotype. J Neurol. 2005;252(10):1201‐1205. [DOI] [PubMed] [Google Scholar]
- 23. Hu T, Ou R, Liu H, et al. Gender and onset age related‐differences of non‐motor symptoms and quality of life in drug‐naïve Parkinson's disease. Clin Neurol Neurosurg. 2018;175:124‐129. [DOI] [PubMed] [Google Scholar]
- 24. Crispino P, Gino M, Barbagelata E, et al. Gender differences and quality of life in Parkinson's disease. Int J Environ Res Public Health. 2020;18(1):198. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Benito‐León J, Bermejo‐Pareja F, Morales‐González JM, et al. Incidence of Parkinson disease and parkinsonism in three elderly populations of Central Spain. Neurology. 2004;62(5):734‐741. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Tables S1–S2