Introduction

Temporomandibular disorders (TMD) encompass a range of medical and dental disorders that affect the function and health of the temporomandibular joints (TMJ), masticatory musculature and the accompanying structures, and are manifested by limitation of jaw movements, TMJ sounds like clicking and crepitus, as well as headaches and pain.1 Various factors contribute to the development of TMD, including trauma to the joints and muscles, anatomical variations, psychosocial components, and the heightened sensitivity of nociceptive pathways.2,3 Symptoms of TMD, especially pain, was considered to be a threatening stimulus in TMD patients, and generally associated with fatigue, mood and sleep disturbances, and poor quality of life.4 Biomechanical and psychological signs and symptoms, often complicated by pain, tend to blur the fundamental cause(s) of the issue. Therefore, it is notoriously difficult to treat. Clinically, pain is often managed with symptomatic treatment, highlighting the need for more targeted treatments aimed at reducing or eliminating pain and enhancing the quality of life for these patients.5

Previously, it has been thought that the production and perception of physiologic and clinical pain primarily reflects neuronal activity alone. However, it has become increasingly apparent that the majority of pain types either originate from or are shaped by the complicated interaction between the immune system and the somatosensory system.6–8 It is considered that peripheral nociceptors can discern immune cells by recognizing the cytokines they secrete. Conversely, once activated, nociceptors release cytokines, neuropeptides, chemokines, and microRNAs from their terminals or cell bodies, thereby influencing resident immune cells such as neutrophils, macrophages, mast cells, and dendritic cells.9,10 Extensive evidence indicates that specific pro-inflammatory cytokines, including TNF-α, IL-1, and IL-6, play a role in the development of pathological pain.11–13 Additional studies have demonstrated that TNF-α is a key mediator of TMD-related pain, influencing both synaptic plasticity in central sensitization and the inflammatory pain response, which leads to increased nociceptive sensitization in periarticular tissues and trigeminal ganglia.14,15 Therefore, pain needs to be recognized as a neuroimmune condition by considering neuroimmune interactions as integrative drivers and modulators, and not just a sign of inflammatory response.

As an important element of the immune system, immune cells are the initial responders of therapeutic intervention. When faced with pathogenic factors, these cells secrete cytokines or directly interact with lymphocytes to maintain the balance of tissue microenvironment and actively participate in the process of tissue injury and regeneration.16 Despite growing interest, research exploring the relationship between immune cells and TMD-related pain remains limited, often constrained by small sample sizes and the influence of confounding variables.

Advancements in large-scale genome-wide association studies (GWAS) and Mendelian randomization (MR) techniques now allow for the investigation of causal relationships between immune traits and disease outcomes.17 Some studies have validated the use of MR in probing causal associations in TMD, offering a method to minimize confounding factors and avoid reverse causality in causal inference.18,19 Therefore, our study employs a two-sample MR approach, integrating GWAS data of immune traits and TMD-related pain, to explore the causal relationship between immune traits and TMD-related pain.

Materials and Methods Study Design

In this research, a two-sample Mendelian Randomization (MR) approach was employed to investigate the causal link between 731 immune traits and TMD-related pain. Genetic variation was leveraged as instrumental variables (IVs) in the MR analysis to test a causal link between the exposure and the outcome (Figure 1). The research began by using immune traits as the exposure to assess potential causal relationships with TMD-related pain risk. In the next step, TMD-related pain was treated as the exposure to investigate possible reverse causal effects on immune traits. The selection of these IVs was guided by three primary assumptions: (1) Relevance, meaning the IVs must be linked to the exposure; (2) Independence, which required the IVs to be unrelated to confounding factors; (3) Exclusion restriction, where the IVs were expected to be independent of the outcome, conditional on the exposure. Any instrumental variables (IVs) that violate the three core assumptions will be excluded from the analysis (Figure 2). All cases included in this study have been approved by the relevant institutional review boards, and all participants have provided their informed consent.

Figure 1 The flowchart of the study. GWAS, genome-wide association study.

Abbreviations: AC, absolute cell counts; RC, relative cell counts; MFI, median fluorescence intensity; MP, morphological parameters; TMD, Temporomandibular disorder; IVs, instrumental variables; LD, linkage disequilibrium; IVW, inverse-variance weighted; WME, Weighted Median; MR, Mendelian randomization; MR-PRESSO, Mendelian randomized polymorphism residual sum and outlier.

Figure 2 Schematic diagram of the assumption of two-sample Mendelian randomization.

Abbreviation: IVs, instrumental variables.

Genome-Wide Association Study (GWAS) Data Sources for TMD-Related Pain

The data from Genome-wide association study (GWAS) for TMD-related pain (https://r9.risteys.finngen.fi/endpoints/DENTAL_TMD) was obtained from the FinnGen Research Consortium. The diagnosis criteria followed International Classification of Diseases (ICD-10) codes M79.1, K07.60 and K07.63, which encompass various forms of TMD associated with pain, including myalgia. The dataset captured a broad range of TMD-related pain conditions without further classification into specific subtypes. This database is R9 version and conducted a genetic analysis on 377,277 Europeans, among which there were 10,303 cases (Ncase) and 366,974 controls (Ncontrol).

Genome-Wide Association Study (GWAS) Data Sources for Immune Cell Traits

The statistical summary of immune traits from GWAS was sourced from the GWAS Catalog (GWAS IDs from GCST0001391 to GCST0002121).20 The catalog includes 731 immune traits, comprising 389 median fluorescence intensities (MFI) reflecting surface antigen levels, 192 relative cell counts (RC), 118 absolute cell counts (AC), and 32 morphological parameters (MP). Specific immune traits in the database were categorized into Treg panels, mature T cell panels, TBNK panels (including T cells, B cells, and NK cells), dendritic cell (DC) panels, mature B cell panels, myeloid panels and monocytes. The genomic analysis of immune traits was investigated using data from 3757 Europeans by collecting peripheral blood from blood donors, which underwent flow cytometry analysis after antibody staining, with estimates derived from the gene sequences of the Sardinian population.21

Selection of Instrumental Variables

For the selection of IVs for immune traits, according to recent studies,20,22 the significance of immune traits was set to P<1×10−5. To ensure the independence of the SNPs, the linkage disequilibrium (LD) analysis was conducted using the PLINK software, and the LD was defined with r2 threshold of less than 0.1 within a 500 kb distance. The r2 for LD was calculated using the available data from the 1000 Genomes Project.23

In the selection of IVs associated with TMD-related pain, the significance level was set to P<5×10−6, after excluding IVs with low F-statistics (F<10), and retained 18 IVs for the MR analysis. This rigorous selection process ensured that our instrumental variables are both statistically significant and robust for the subsequent MR analysis.

Statistical Analysis

In this research, five models were employed for MR analysis to precisely analyze causal link between immune traits and TMD-related pain, include Inverse Variance Weighting (IVW), MR-Egger, Weighted Median Method (WME), Weighted mode and Simple mode. The IVW method was employed as major MR analysis that leveraged the impact of Single Nucleotide.

Polymorphisms (SNPs) on the exposure for weighting, providing an estimate of the causal relationship,24 and the False Discovery Rate (FDR) was also applied in IVW method for correction to prevent an increase in Type I errors due to multiple testing.25 Other methods were used as complementary methods.

Moreover, several sensitivity analyses were performed for the result to analyze the horizontal pleiotropy and heterogeneity. Heterogeneity among SNPs was evaluated with the Cochran’s Q test. Horizontal pleiotropy of SNPs was assessed using the MR-Pleiotropy Residual Sum and Outlier (MR-PRESSO) method26 and MR-Egger regression (Egger-intercept) method.27 Leave-one-out analysis was also conducted to assess the influence of each individual genetic variant.

In this study, statistical software R 4.3.2, Two Sample MR package 0.5.7, and MR-PRESSO package 1.0 were used, P-value of less that 0.05 was regarded as statistically significant.

Result Exploration of Causal Impacts of Immune Traits on TMD-Related Pain

In this study, we conducted a two-sample MR analysis, mainly employing the IVW method. At a nominal significance level, we revealed the causal role between TMD-related pain and 75 types of immune traits (IVW P<0.05). Specifically, 33 types were found to decrease the risk of TMD-related pain, 41 types increased the risk, while 1 type appeared to have no effect (Figure 3). After the FDR adjusting in the IVW method, we identified a causal relationship in 2 types of immune traits: Hematopoietic Stem Cell absolute count (Myeloid panel), HLA DR+ CD4+ T cell (TBNK panel). The IVW method showed that Hematopoietic Stem Cell absolute count could decrease the level of TMD-related pain (OR=0.95, 95% CI= 0.93~0.98, PFDR= 0.019, P = 4.99E-05) and HLA DR+ CD4+ T cell could increase the level of TMD-related pain (OR=1.04, 95% CI=1.02~1.06, PFDR =0.024, P = 1.27E-04). We also observed similar results using four other analytical approaches. For the immune trait Hematopoietic Stem Cell, the results were as follows: the MR Egger result was OR=0.98, 95% CI= 0.95~1.01, P =0.189; the WME result was OR=0.96, 95% CI= 0.93~0.99, P=0.017; the Weighted mode result was OR=0.96, 95% CI= 0.93~1.00, P =0.04; and the Simple mode result was OR= 0.95, 95% CI= 0.90~1.00, P =0.072. For the immune trait HLA DR+ CD4+ T cell, the results were as follows: the MR Egger result was OR=1.03, 95% CI= 1.00~1.06, P =0.034; the WME result was OR=1.03, 95% CI= 1.00~1.06, P =0.059; the Weighted mode result was OR=1.03, 95% CI= 1.01~1.06, P =0.009; and the Simple mode result was OR=1.07, 95% CI= 1.02~1.14, P =0.018 (Figure 4). Scatter plots in five methods were showed in (Figure 5). Furthermore, the sensitivity analyses of the Cochran Q, MR-PRESSO, Egger-intercept, LOO analysis were showed in (Table 1 and Figure 6).

Table 1 Sensitivity Analyses of Causal Impacts of Immune Traits on TMD-Related Pain

Figure 3 Forest plots showed the causal impacts of immune traits on TMD-related pain at a nominal significance level, indicating the direction of effect (OR) and statistical significance (95% CI). OR >1 suggests increased risk, while OR <1 indicates a potential protective effect.

Abbreviations: SNPs, single nucleotide polymorphisms; OR, odds ratio; CI, confidence interval; IVW, inverse-variance weighted.

Figure 4 Forest plots showed the causal impacts of immune traits on TMD-related pain after FDR adjustment, indicating the direction of effect (OR) and statistical significance (95% CI). OR >1 suggests increased risk, while OR <1 indicates a potential protective effect.

Abbreviations: SNPs, single nucleotide polymorphisms; OR, odds ratio; CI, confidence interval; FDR, false discovery rate; HLA, human leukocyte antigen; CD, cluster of differentiation; IVW, inverse-variance weighted.

Figure 5 Scatterplot of SNP effects on the absolute counts of hematopoietic stem cells and HLADR+CD4+ T cell expression levels versus TMD-related pain. The slopes of the lines indicate the estimated causal effects, with a negative trend suggesting an inverse relationship between immune traits and TMD-related pain. The spread of the data points provides insight into the variability and consistency of these estimates.

Abbreviations: TMD, Temporomandibular disorder; MR, Mendelian randomization; SNP, single nucleotide polymorphism; HLA, human leukocyte antigen; CD, cluster of differentiation.

Figure 6 The leave-one-out analysis of causal impacts of immune traits on TMD-related pain. The Y-axis shows the SNP ID excluded from the analysis, and the X-axis presents the MR odds ratios. The red line reflects the MR estimate calculated with all SNPs included.

Abbreviations: TMD, Temporomandibular disorder; MR, Mendelian randomization; HLA, human leukocyte antigen; CD, cluster of differentiation.

Exploration of Causal Impacts of TMD-Related Pain on Immune Traits

To investigate the reverse causality of TMD-related pain on immune traits, a two-sample MR analysis was employed, mainly using the IVW method. At a nominal significance level (IVW P<0.05), evidence of a causal relationship was identified between TMD-related pain and 24 types of immune cells. Specifically, TMD-related pain was observed to elevate the level of 10 type of immune traits, while decreasing the levels of 14 other types (Figure 7). The results showed no immune traits after FDR adjustment.

Figure 7 Forest plots showed the causal impacts of TMD-related pain on immune traits at a nominal significance level, indicating the direction of effect (OR) and statistical significance (95% CI). OR >1 suggests increased risk, while OR <1 indicates a potential protective effect.

Abbreviations: SNPs, single nucleotide polymorphisms; OR, odds ratio; CI, confidence interval; IVW, inverse-variance weighted.

Discussion

By using publicly accessible genetic data, we conducted a systematic examination of the potential causal relationship between 731 immune cell traits and TMD-related pain, employing MR techniques. For all we know, this study represents the first application of MR to investigate a broad spectrum of immune traits in relation to TMD-related pain. We found a causal relationship in 2 types of immune traits: Hematopoietic Stem Cell absolute count and HLA DR+ CD4+ T cell. Our result offers further evidence for a causal relationship between immune traits and TMD-related pain. However, reverse MR analysis failed to identify significant causal associations.

Interestingly, our results indicate that higher absolute cell counts of HSC may be linked with reduced TMD-related pain. While there is limited-existing literature on this, the established roles of HSC such as self-renewal and the capacity to give rise to a variety of progenitor cells, including lymphoid and myeloid stem cells, as well as the ability to further differentiate into progenitor T cells, progenitor B cells and NKs, may shed light on our observations. Mast cells, also derived from HSCs, are an essential component of the innate immune system and are highly responsive to activation of the hypothalamic-pituitary-adrenal (HPA axis), the major stress response system, and their activation leads to downstream secretion of cortisol and suppression of immune responses. These findings implied that stress and psychological factors might be crucial in the pathogenesis of TMD pain and aligned with previous research on the relationship between trait anxiety and TMD.18

Further, our data showed elevated expression of HLA DR+ CD4+ T cell may increase the risk of TMD pain. This may be consistent with the findings of previous research the reduced expression of HLA-DR on monocytes during chronic inflammation highlights the anti-inflammatory role of HLA-DR molecules.28 There are few studies to explore its association with TMD pain risk and little is known about the role of HLA DR+ CD4+ T cell in its progression. When HLA-DR is low expressed, CD4 + T lymphocytes are not activated, resulting in the body can neither stimulate B cells to produce specific antibodies, nor increase the production of CD8 + T lymphocytes, and this immune pathway is impaired and can lead to immune suppression. The mechanism of immune cells related in the TMD pain may be very complex, more experimental research should be carried out to further explore the mechanism behind HLA DR+ CD4+ T cell and TMD pain risk.

Despite careful selection of IVs that met various modeling assumptions and extensive sensitivity analyses to minimize the effects of possible confounders, there are limitations to our study. Firstly, this study relied on a European population database, limiting the ability to generalize the results to other ethnic groups and reducing the broader applicability of our findings. Second, the absence of detailed individual data hindered our ability to stratify the population further. Third, due to the limited sample size, the threshold for SNP selection was loose, which may result in some degree of false positives. Therefore, we interpret the outcomes produced in this study cautiously.

Conclusion

In summary, through stringent bidirectional MR analyses, we identify a potential causal relationship between immune cells and TMD pain, eliminating reverse causality. These discoveries significantly enhance our knowledge of the interaction between immune traits and TMD-related pain, opening new possibilities for designing treatment from an immunological perspective. Further research is required to clarify the specific role of immune cells and the resulting pain in the pathogenesis of TMD.

Ethics Statement

According to local legislation (Article 32 of the document No.4 of 2023 “Notice on of the Ethical Review of Life Sciences and Medical Research Involving Humans” issued by National Health Science and Education Development) and institutional requirements, the study utilized only publicly accessible data and did not involve direct human participation, thus exempting it from the need for ethical approval or consent.

Acknowledgments

The authors acknowledge the participants and investigators of the FinnGen study.

Disclosure

The authors report no conflicts of interest in this work.

References

1. Schiffman E, Ohrbach R, Truelove E, et al. Diagnostic criteria for temporomandibular disorders (DC/TMD) for clinical and research applications: recommendations of the international RDC/TMD consortium network* and orofacial pain special interest group†. J Oral Facial Pain Headache. 2014;28(1):6–27. doi:10.11607/jop.1151

2. DTS L, Leung YY. Temporomandibular disorders: current concepts and controversies in diagnosis and management. Diagnostics. 2021;11(3). doi:10.3390/diagnostics11030459

3. Sun R, Zhang S, Si J, et al. Association between oral behaviors and painful temporomandibular disorders: a cross-sectional study in the general population. J Pain Res. 2024;17:431–439. doi:10.2147/JPR.S449377

4. Busse JW, Casassus R, Carrasco-Labra A, et al. Management of chronic pain associated with temporomandibular disorders: a clinical practice guideline. BMJ. 2023;383:e076227. doi:10.1136/bmj-2023-076227

5. Yao L, Sadeghirad B, Li M, et al. Management of chronic pain secondary to temporomandibular disorders: a systematic review and network meta-analysis of randomised trials. BMJ. 2023;383:e076226. doi:10.1136/bmj-2023-076226

6. Jain A, Hakim S, Woolf CJ. Immune drivers of physiological and pathological pain. J Exp Med. 2024;221(5). doi:10.1084/jem.20221687

7. Tanaka T, Okuda H, Isonishi A, et al. Dermal macrophages set pain sensitivity by modulating the amount of tissue NGF through an SNX25-Nrf2 pathway. Nat Immunol. 2023;24(3):439–451. doi:10.1038/s41590-022-01418-5

8. Ghasemlou N, Chiu IM, Julien JP, Woolf CJ. CD11b+Ly6G- myeloid cells mediate mechanical inflammatory pain hypersensitivity. Proc Natl Acad Sci U S A. 2015;112(49):E6808–6817. doi:10.1073/pnas.1501372112

9. Chen O, Donnelly CR, Ji RR. Regulation of pain by neuro-immune interactions between macrophages and nociceptor sensory neurons. Curr Opin Neurobiol. 2020;62:17–25. doi:10.1016/j.conb.2019.11.006

10. Baral P, Udit S, Chiu IM. Pain and immunity: implications for host defence. Nat Rev Immunol. 2019;19(7):433–447. doi:10.1038/s41577-019-0147-2

11. Villa G, Ceruti S, Zanardelli M, et al. Temporomandibular joint inflammation activates glial and immune cells in both the trigeminal ganglia and in the spinal trigeminal nucleus. Mol Pain. 2010;6:89. doi:10.1186/1744-8069-6-89

12. Monasterio G, Castillo F, Rojas L, et al. Th1/Th17/Th22 immune response and their association with joint pain, imagenological bone loss, RANKL expression and osteoclast activity in temporomandibular joint osteoarthritis: a preliminary report. J Oral Rehabil. 2018;45(8):589–597. doi:10.1111/joor.12649

13. Pinto Fiamengui LMS, Furquim BD, De la Torre Canales G, et al. Role of inflammatory and pain genes polymorphisms in temporomandibular disorder and pressure pain sensitivity. Arch Oral Biol. 2020;118:104854. doi:10.1016/j.archoralbio.2020.104854

14. Wang Y, Bao M, Hou C, Wang Y, Zheng L, Peng Y. The role of TNF-α in the pathogenesis of temporomandibular disorders. Biol Pharm Bull. 2021;44(12):1801–1809. doi:10.1248/bpb.b21-00154

15. Campello CP, Lima ELS, Fernandes RSM, Porto M, Muniz MTC. TNF-α levels and presence of SNP-308G/A of TNF-α gene in temporomandibular disorder patients. Dental Press J Orthod. 2022;27(1):e2220159. doi:10.1590/2177-6709.27.1.e2220159.oar

16. Tomar N, De RK. A brief outline of the immune system. Methods Mol Biol. 2014;1184:3–12.

17. Xu W, Shen Y, Sun J, Wei D, Xie B, Song X. Causal role of immune cells in alopecia areata: a two-sample Mendelian randomization study. Skin Res Technol. 2024;30(1):e13579. doi:10.1111/srt.13579

18. Xiang Y, Song J, Liang Y, Sun J, Zheng Z. Causal relationship between psychiatric traits and temporomandibular disorders: a bidirectional two-sample Mendelian randomization study. Clin Oral Investig. 2023;27(12):7513–7521. doi:10.1007/s00784-023-05339-x

19. Chen X, Xu J, Cheng Z, Wang Q, Zhao Z, Jiang Q. Causal relationship between hypothyroidism and temporomandibular disorders: evidence from complementary genetic methods. BMC Oral Health. 2024;24(1):247. doi:10.1186/s12903-024-03999-z

20. Orru V, Steri M, Sidore C, et al. Complex genetic signatures in immune cells underlie autoimmunity and inform therapy. Nat Genet. 2020;52(10):1036–1045. doi:10.1038/s41588-020-0684-4

21. Sidore C, Busonero F, Maschio A, et al. Genome sequencing elucidates Sardinian genetic architecture and augments association analyses for lipid and blood inflammatory markers. Nat Genet. 2015;47(11):1272–1281. doi:10.1038/ng.3368

22. Yu XH, Yang YQ, Cao RR, Bo L, Lei SF. The causal role of gut microbiota in development of osteoarthritis. Osteoarthritis Cartilage. 2021;29(12):1741–1750. doi:10.1016/j.joca.2021.08.003

23. Auton A, Brooks LD, Durbin RM, et al. A global reference for human genetic variation. Nature. 2015;526(7571):68–74. doi:10.1038/nature15393

24. Burgess S, Small DS, Thompson SG. A review of instrumental variable estimators for Mendelian randomization. Stat Methods Med Res. 2017;26(5):2333–2355. doi:10.1177/0962280215597579

25. Burgess S, Bowden J, Fall T, Ingelsson E, Thompson SG. Sensitivity analyses for robust causal inference from Mendelian randomization analyses with multiple genetic variants. Epidemiology. 2017;28(1):30–42. doi:10.1097/EDE.0000000000000559

26. Verbanck M, Chen CY, Neale B, Do R. Detection of widespread horizontal pleiotropy in causal relationships inferred from Mendelian randomization between complex traits and diseases. Nat Genet. 2018;50(5):693–698. doi:10.1038/s41588-018-0099-7

27. Bowden J, Del Greco MF, Minelli C, Davey Smith G, Sheehan N, Thompson J. A framework for the investigation of pleiotropy in two-sample summary data Mendelian randomization. Stat Med. 2017;36(11):1783–1802. doi:10.1002/sim.7221

28. Mahmoodpoor A, Paknezhad S, Shadvar K, et al. Flow cytometry of CD64, HLA-DR, CD25, and TLRs for diagnosis and prognosis of sepsis in critically ill patients admitted to the intensive care unit: a review article. Anesth Pain Med. 2018;8(6):e83128. doi:10.5812/aapm.83128