Immunohistochemical Markers of Temporomandibular Disorders: A Review of the Literature

Temporomandibular disorders (TMD) are a group of internal derangements encompassing dysfunction, displacement, degeneration of the temporomandibular joints and surroundings muscles of mastication, often accompanied by pain. Relationships between TMD and various chemical biomarkers have been examined throughout the years. This paper aims to gather evidence from the literature regarding other biomarkers and presenting them as one systematic review to investigate the potential links between TMD and different biochemical activity. To identify relevant papers, a comprehensive literature search was carried out in MEDLINE/PubMED, EMBASE, Web of Science and a manual search was performed in the International Journal of Oral and Maxillofacial Surgery, Journal of Oral and Maxillofacial surgery, and Journal of Cranio-Maxillo-Facial Surgery. The literature review produced extensive results relating to the biochemical and immunohistochemical markers of TMD. Many enzymes, inflammatory markers, proteoglycans, and hormones were identified and organized in tables, along with a brief description, study design, and conclusion of each study. Through this review, recurring evidence provides confidence in suggesting involvement of certain biomarkers that may be involved in this complex pathogenesis, in addition to pointing to differences in gender prevalence of TMD. However, more organized research on large human samples needs to be conducted to delve deeper into the understanding of how this disease develops and progresses.


Introduction
The temporomandibular joint (TMJ) is a ginglymoarthrodial joint acting as a union of the mandibular fossa of the temporal bone and the head of the mandibular condyle. It is used in the mastication processes via translational (gliding) and rotational (hinging) movements. Its primary components include the articular disc, the articular capsule, innervation and vascularization, ligaments, and the muscles of mastication [1].
The inner region of the TMJ contains an articular disc composed of fibrous connective tissue devoid of any blood vessels and nerves. It is attached to the capsular ligament anteriorly, posteriorly, medially, and laterally to separate the TMJ into two cavities. Posteriorly, loose connective tissue is attached to the connective tissue and is highly vascularized and innervated. Superiorly, the fibroelastic fascia attaches with elastic fibers to the tympanic plate. Inferiorly, at the posterior edge of the disc is the inferior retrodiscal lamina composed of fibrous, collagenous fibers. The disc is attached to the condylar formation at the medial and lateral aspects. The remaining body of the retrodiscal tissue is attached posteriorly to a large venous plexus that fills with blood as the condyle moves forward.
The TMJ is surrounded by the articular capsule, a capsular ligament with specialized endothelial cells that form a synovial lining which provides boundary and weeping lubrication. These cells provide synovial fluid, which act as a medium to prevent friction in the moving joint (boundary lubrication), as well as the ability of the articular surface to absorb

Search Protocol
The MEDLINE/PubMed, EMBASE, and Web of Science databases were searched between January 2020 and March 2020. The manual search was performed in the three main journals of the field (International Journal of Oral and Maxillofacial Surgery, Journal of Oral and Maxillofacial Surgery, and Journal of Cranio-Maxillo-Facial Surgery), as well as from the reference list of studies included in this systematic review.
For the construction of the search strategy, MeSH (Medical Subject Heading) terms were used, which are considered descriptors of controlled subjects for searching MEDLINE, Web of Science, and Cochrane. To make the search more sensitive, vocabularies not controlled by the use of keywords were included. Entree terms were used for the search strategy in EMBASE.
The search was carried out by combining the terms ("MeSH" and keywords) for population and intervention, with the help of the Boolean operators "OR" and "AND". The strategy adopted for the MEDLINE/PubMed base was as follows: (TMJ OR temporomandibular joint OR TMD OR temporomandibular dysfunction OR temporomandibular joint dysfunction OR temporomandibular joint derangement) "AND" (immunohistochemistry OR immune antibody OR proteomic OR protein expression). In addition to these terms, the following keywords were used: disc displacement OR TMJ internal derangement OR TMD. The PRISMA 2020 statement was used as reference in reporting this systematic review.  10,89.) The inclusion criteria was: human and animal research with a definitive clinical diagnosis or specific signs and symptoms for TMD, and protein expression results obtained through either immunohistochemistry OR immune antibody OR proteomic OR protein expression. The exclusion criteria were as follows: studies not in English, full text not available, studies not related to TMDs or where there was no mention of specific diagnoses or signs and symptoms, studies without the use of control to compare sample with and without protein expression.
A total of 107 studies were analyzed but only 86 were in accordance with PRISMA's criteria. The excluded studies either did not have a clear diagnosis of TMD or an adequate methodology. Three studies did not have the full text available, and one was not written in English. (Figure 1).

Data Analysis
The search protocol that was used was designed to be as inclusive as possible to encompass as many potential biomarkers that may be involved in the pathogenesis of TMD. As such, the methodology of the included studies varies widely, from utilization of different detection techniques in analyzing biomarker involvement, such as polymerase chain reaction (PCR), DNA extraction, immunohistochemical staining (IHC), among many others, to subject species-such as detection of biomarkers in human, rabbit or mouse tissues. With such variation in study design, a direct analysis of the data obtained through the included studies was unable to be conducted. Furthermore, the quality of each study, such as looking at statistical power, was not considered beyond the impact of the study based on factors such as sample size and the species from which the tissue was harvested in the study.

Data Analysis
The search protocol that was used was designed to be as inclusive as possible to encompass as many potential biomarkers that may be involved in the pathogenesis of TMD. As such, the methodology of the included studies varies widely, from utilization of different detection techniques in analyzing biomarker involvement, such as polymerase chain reaction (PCR), DNA extraction, immunohistochemical staining (IHC), among many others, to subject species-such as detection of biomarkers in human, rabbit or mouse tissues. With such variation in study design, a direct analysis of the data obtained through the included studies was unable to be conducted. Furthermore, the quality of each study, such as looking at statistical power, was not considered beyond the impact of the study based on factors such as sample size and the species from which the tissue was harvested in the study.

Results
The literature review produced extensive results relating to the biochemical and immunohistochemical markers of TMD. The most studied are: enzymes, such as matrix

Results
The literature review produced extensive results relating to the biochemical and immunohistochemical markers of TMD. The most studied are: enzymes, such as matrix metalloproteinases (MMP), aggrecanase, and cyclooxygenase (COX); cytokines and inflammatory markers, such as interleukins (IL), tumor necrosis factor-alpha (TNF-alpha), and vascular endothelial growth factor (VEGF); proteoglycans, such as tenascin, vimentin and fibronectin; hormones, such as estrogen, progesterone and relaxin; and miscellaneous markers that did not fit into the other categories, such as aquaporin, heat shock protein, and b-cell lymphoma-2 gene (BCL-2). Each study, along with a brief description, study design, and the conclusions, are organized in the tables below.

Enzymes
Matrix metalloproteinases (MMPs) have been a large target of research attempts to understand the underlying pathogenesis of TMD. As previously mentioned, MMPs are a family of 26 proteins that are involved in extracellular matrix and basement membrane remodeling through the breakdown of collagen, gelatin, proteoglycans, and other components of the basement membrane. One of the first studies looking at MMPs in the discs of the TMJ was conducted by Kapila, using various methods, such as Western blotting and immunohistochemical staining, to identify the expression of MMPs (gelatinases, collagenases and stromelysin) in the discs of rabbit TMJ [4]. Many studies followed and showed the strongest evidence for the involvement of MMP-2 and MMP-9, with some evidence showing the involvement of MMP-7 (Kapila, [5], Tanaka, [6], Yoshida, [7], Loreto, [8,9], Almeida et al. [10]). Perroto [11], however, did not identify a significant difference in levels of MMP-13 between samples of human discs with TMD when compared to control samples. This points to the complex nature of basement-membrane-remodeling regulation involved in TMD. Evidence suggests that MMP-2 and MMP-9 are involved in the pathogenesis of TMD; however, more research needs to be conducted to support the involvement of the other suggested MMPs (MMP-1, MMP-7) and to definitively exclude those that have not been shown to be involved thus far (MMP-13). (Table 1). The presence of a MMP-9 (92-kD gelatinase) in TMJ disc and articular cells likely function in the degradative process. Additionally, this enzyme is under the control of pro-inflammatory cytokines whereby TGFbeta and IL-1 stimulate and PGE (2) inhibits its activity.
MMP-9 associated with degenerative processes   [16] Cyclooxygenase-2 Human disk and synovial membranes There were obvious distinction of COX-2 immunoreactivity between the control specimens and internal derangement cases, in the region of posterior and/or anterior loose connective tissues. intensive COX-2 expression was detected in the synovial membrane of internal derangement cases.

COX-2 expression
and TMD  [17] MMP and aggrecanase Human synovial fluid MMP-2, -9, and aggrecanase expression in the ID group were significantly higher than those in the normal group. Those with anterior disc displacement without reduction and severe OA showed significantly high expression of MMP-9 compared with other disease subgroups.

Yoshida et al. (2005) [18]
Aggrecanase Human synovial fluid Aggrecanase expression in TMJD group were significantly higher than that in the normal control group. MMP-2 and MMP-9 Rat trigeminal ganglion MMP expression in the trigeminal ganglion was shown to vary during the phases of the inflammatory process. MMP-9 regulated the early phase and MMP-2 participated in the late phase of this process. Furthermore, increases in plasma extravasation in periarticular tissue and myeloperoxidase activity in the joint tissue, which occurred throughout the inflammation process, were diminished by treatment with DOX, a nonspecific MMP inhibitor.
MMP-2 and MMP-9 associated with process of TMJ inflammation

MMP-13
Human discs MMP-13 is not significantly involved in collagen degradation.

Cytokines and Inflammatory Markers
Our review of the literature found several different cytokines and inflammatory markers to be in association with TMD, including interleukins, CD antigens, TNF alpha, TGF beta, VEGF, IGF, NFkB, IFN-y, capsaicin, and bradykinin. These markers are involved in many pathophysiological functions, including initiating and mediating inflammatory responses, activating immune cells, regulating apoptosis and angiogenesis, and regulating pain and fever.
As previously mentioned, the interleukins are a major family of cytokines that are extensively involved in the immune response. They are secreted by macrophages and T cells in order to modulate inflammation and fever, activate and maintain immune cells and bone marrow, produce immunoglobulins and acute phase proteins, and maintain granulomas. Various studies in our review found IL-1, IL-2, IL-6, IL-8, and IL-10 to be associated with TMD. Kim (2012) did not detect IL-4 and IL-5 in either the TMD group or the control group during a study of human synovial fluid [23].  2017) found no significant difference when comparing TMD groups of ADDwR and ADDwoR, and with and without OA [24][25][26][27]. (Table 2).

IHC Marker Study Design and Tissue Expression Results Conclusions
Fu et al. (1995) [24] IL-6 Human synovial fluid Interleukin-6 level was greater than 100 U/mL in 13 of 18 patients with degenerative joint disease and in 5 of 12 patients with disc displacement. However, the interleukin-6 level was less than 100 U/mL (range, 20 to 75 U/mL) in all patients with masticatory muscle disorder.

TGF-beta and Tenascin
Human synovial membranes TGF-beta and tenascin were distributed in the affected synovial membrane of TMJ with internal derangement. These findings suggested that TGF-beta and tenascin might have a close relationship with synovitis, followed by tissue repair. The up-regulation of CD44H observed in some dysfunctional TMJ discs seems to indicate a prevention of apoptosis in fibroblast-like cells and an important role in phenotypical change of fibrochondrocytes into chondroblastlike cells, enabling the aggregation of chondroid tissue pericellular matrix components.  [25] IL-1 Beta and IL-6

Prevention of apoptosis in TMJ discs
Human synovial cells IL-1 beta increased IL-6 production in synovial cells. Enhanced production of IL-6, which is associated with bone resorption and inflammatory response, seems to be related to the progression of TMD. The concentrations of TNF-alpha, IL-6, IL-1beta, sTNFR-I, and sTNFR-II were significantly higher in the synovial fluid of patients than in controls (p < 0.05). TNF-alpha level was positively correlated with those of IL-6, sTNFR-I, and sTNFR-II. In particular, there was a highly significant positive correlation between sTNFR-I and sTNFR-II.
All associated with TMD Hypoxia and interleukin-1beta and MMPs

Rabbit discs
The results showed that the combination of hypoxia and IL-1beta caused a significant increase in MMP-1, MMP-3, MMP-9 and MMP-13 mRNA.
Hypoxia and IL-1beta increases MMPs  Hypoxia and interleukin-1beta Human synovial fluid and discs The combination of hypoxia and interleukin-1beta caused a significant increase in tenascin-C protein and mRNA of synovial fibroblasts, but not in disc cells.
Hypoxia and interleukin-1beta increase tenascin-C protein  Certain cytokines (noted left) associated with TMD Camejo (2013) [49] Fas Ligand (FasL) Human discs A higher area of in situ immunostaining of FasL was found in temporomandibular discs with ADDwR 1 .

Apoptosis and TMD
Sicurezza (2013) [50] β-Defensin Human discs The presence of β-defensin-4 in human TMJ discs affected by ADDwoR 2 was found, hypothesizing its possible role in articular bone disruption. Antiinflammatory effects of IL-37. 1 Anterior Disc Displacement with Reduction; 2 Anterior Disc Displacement without Reduction.

Proteoglycans
The review of the literature found several different proteoglycans that are involved with TMD, such as tenascin, vimentin, fibronectin, decorin, biglycan, veriscan, and elastin. Proteoglycans are proteins located in the extracellular matrix, cell surface, or intracellular granules. They are involved in several processes, such as cell signaling and organization of the extracellular matrix [53]. Yoshida and Leonardi found that tenascin was associated with degenerated tissue, specifically in the portion of the TMJ synovial membrane affected by internal derangement [54]. Additionally, Toriva et al. (2006) [55] found that versican is associated with TMD and causes growth-related changes and regional differences in the TMJ discs of rats. The full table of proteoglycans and their respective results may be seen in the table above. (Table 3).  [57] Decorin and Biglycan Rat discs Regional differences in staining for decorin became prominent at 4, 8 and 16 weeks; decorin was more abundant in the peripheral area of the band than in the central area. In contrast, staining for biglycan was evenly distributed throughout the disc until 4 weeks, and after that became intense in the anterior and posterior bands. Human synovial fluid and discs Tenascin was produced specifically in synovial cells and vascular endothelial cells and. Fibroblasts were affected in the portion of TMJ with internal derangement.  [60] Type II Collagen Human discs

Tenascin expression in TMD
The percentage of type II collagen in immunoreactive disc cells was significantly higher in the outer part (the articular surfaces) than in the inner part (the deep central areas) of the disc.
Increased collagen synthesis in TMJ discs Yoshida et al. (2004) [54] Vimentin Human discs and synovial membrane There was an obvious distinction of vimentin immunoreactivity between the control specimens and internal derangement cases, in the posterior and/or anterior loose connective tissues. In particular, intensive vimentin expression was detected in the hypertrophic synovial membrane of internal derangement cases.
Vimentin present in synovial membrane of TMD  Chondromodulin-1 (ChM-1) Rabbit discs ChM-1 may play a role in the regulation of TMJ remodeling by preventing blood vessel invasion of the cartilage.
IL-1 β induces a significant increase in lumican mRNA, but not in fibromodulin mRNA.

Leonardi et al. (2011) [70]
Lubricin Human discs A longstanding TMJ disc injury, affects lubricin expression in the TMJ disc tissue and not its surfaces; moreover, lubricin immunostaining is not correlated to TMJ disc histopathological changes.
No correlation between TMD and lubricin expression

Hormones
Our search of the literature resulted in seven studies looking at the involvement of three hormones in the pathogenesis of TMD. These studies examined relaxin, estrogen and progesterone in tissues from rabbit, mice and baboon discal tissues. Relaxin is a hormone that has been shown to be involved in matrix remodeling during pregnancy, and its effects are potentiated in the presence of estradiol [75]. A common finding among the studies conducted by Naqvi, Kapila, Hashem, and Park showed that an increased level of relaxin in disc samples was associated with increased levels of MMPs, such as collagenase and stromelysis, and a decrease in the levels of collagen and glycosaminoglycans (GAGs) [4,71,75]. Hashem, however, also found that the hormone progesterone prevented the effects of relaxin and estrogen, limiting the amount of matrix loss by preventing induction of MMPs [76].
An alternative methodology of looking at the hormonal involvement in the pathogenesis was utilized in some studies by examining the estrogen and/or relaxin receptor. Wang's study examined the level of estrogen receptors α and β, relaxin receptors LGR7 and LGR8, and progesterone receptor in various joints, and they found a higher number of hormone receptors in the TMJ of mice when compared to the knee joint [76]. In a study conducted by Puri, it was found that the level of estrogen-α receptors in the TMJ, however, could be influenced by inflammation-an increase in inflammation was correlated with a decrease in estrogen-α receptors [77]. Together, these help illustrate the multifactorial etiology that makes understanding this disorder so difficult.
These findings support the idea that hormonal influences are a key factor in the gender disparity associated with TMD. With studies showing a large number of hormonal receptors located in the joint, and with studies showing an increase in relaxin and estrogen associated with increased MMP activity and decreased collagen and GAGs, evidence points toward hormones playing a key role in this disparity. Although these studies suggest this, however, it is important to note that all studies were carried out on different species, and more research should be conducted with human samples for stronger support. Similarly, future studies should focus on specific receptors and hormones to identify which hormones have the largest impact on the pathogenesis, as well as look at the role of inflammation. (Table 4).  [77] Relaxin, β-estradiol, and progesterone alone or in various combinations.

Rabbit discs
Relaxin produces a dose-dependent induction of tissue-degrading enzymes of the matrix metalloproteinase family, specifically MMP-1, MMP-3, MMP-9, and MMP-13 in cell isolates and tissue explants from TMJ fibrocartilage. The induction of these MMPs is accompanied by loss of collagen and glycosaminoglycans, which was blocked by a pan-MMP inhibitor.
Progesterone attenuated the induction of MMPs.

Miscellaneous
Many of the studies that resulted from our search of the current literature included biomarkers that do not fit the preceding categories. Leonardi et. al. looked at the levels of heat shock protein 27  in human disc tissues and found that an increased level of the protein was associated with TMD, while the level of HSP-27 was virtually undetectable in samples from fetal tissue and normal joint tissue [79]. Heat shock proteins act within cells to protect against protein denaturation; however, they have been suggested to be involved in the pathogenesis of many diseases with elevated levels for prolonged periods of time.
Huang et al. looked at the presence of BCL-2 and BAX in disc tissue from rabbit samples [80]. BCL-2 is a protooncogene that is anti-apoptotic in nature, while BAX is a proapoptotic gene. Immunohistochemical staining of the disc samples revealed the presence of receptors for both BCL-2 and BAX, which suggests the possibility of their involvement in the regulation of the apoptotic process of TMD [80]. With respect to tumor suppressors, Castorina et. al. looked at the presence of P53 expression (along with VEGF) in human disc samples from subjects with TMD and found a positive correlation between pathological changes of the tissue and expression of the tumor suppressor protein P53, as well as levels of VEGF [81]. These studies are added evidence of the complex nature of the regulation of the apoptotic pathway in TMD.
Aquaporins are membrane channel proteins that allow water to pass through cell membranes to help regulate osmotic hemostasis [82]. They are found as part of normal physiology, with their presence identified in processes in the kidney, gastrointestinal tract, among other tissues, but have also been identified as part of various pathologic processes.
In two studies carried out by Loreto, human disc samples were utilized to examine the presence of aquaporin-1 channels [53,83]. In the first study, it was found that these channels are part of normal homeostatic regulation of the TMJ; however, in the second study, it was found that the channel protein was upregulated in disc samples from subjects with TMD when compared with samples without TMD [82,84].
It has been previously shown that angiogenesis plays an important role in the progression of TMD, and, with that, the involvement of VEGF. Dickkopf-related protein (DKK-1) is a protein that inhibits the Wnt/beta-cantenin pathway and has been previously shown to play a role in physiologic and pathologic processes thorugh its role in promoting angiogenesis and recruitment of endothelial cells [85]. In a study conducted by Jiang, human synovial fibroblasts were examined for the presence of DKK-1 and found a high level of expression in the synovial fluid in samples obtained from subjects with TMD [84].
Elastin-derived proteins (EDPs) are bioactive peptides that arise with the degradation of elastin, a major component of the extracellular matrix [86]. An increase in elastin degradation and turnover has been shown to be a part of many pathological processes and through induction of inflammatory markers. Kobayachi used immunohistochemical staining to examine levels of EDPs in human synovial fluid and found an association between EDP, IL-6 and MMP-12, suggesting a role in EDPs in regulation of the inflammatory cascade in the TMJ [86].
Finally, in a study conducted by Fujimura, human synovial fluid was analyzed using electrophoresis in an attempt to look at synovial protein patterns [87]. The group found a significantly higher concentration of total protein in the synovial fluid of samples of patients with various stages of TMD when compared to healthy samples, especially in the amount of proteins with a higher molecular weight. Even when comparing the differences of protein concentration within the stages, a more advanced (osteoarthritis) stages were associated with a higher concentration of protein than earlier stages. These results support the idea that there is an increase in the level of inflammation as TMD progresses to higher stages. (Table 5). Approximately 22 different protein bands with molecular weights ranging from 14 to 700 kd were clearly discernible on electrophoresis. The relative amounts of specific proteins in the SF of the TMD group were also different from those in the AS group (p < 0.05). The major difference in total protein concentration appeared to be due to the increased abundance of relatively high molecular weight proteins (>140 kd) in the TMD patients as compared to the AS group.
Increased concentration of synovial proteins associated with increased stage of TMD.
Loreto (2012) [85] Aquaporin Human discs AQP1 is normally expressed in the TMJ disc and confirm a role for it in the maintenance of TMJ homeostasis.
Aquaporin acts on TMJ homeostasis Loreto (2012) [85] Aquaporin Human discs Aquaporin-1 is expressed and upregulated in temporomandibular joint with anterior disc displacement (both with and without reduction).

Conclusions
The pathogenesis underlying the development and progression temporomandibular joint disorder is a very complex biochemical process. As evidenced by this review, much research has been conducted to identify the different underlying involvement of biological markers. Through this review, recurring evidence provides confidence in suggesting the involvement of matrix metalloproteinases that may be involved in basement membrane and disc pathogenesis, inflammatory markers that may be modulating pain and TMJ tissue breakdown, as well as point to differences in gender prevalence of TMD. Moreover, proteoglycans and several proteins were found to be involved in the inflammatory and apoptotic cascade contributing to TMD progression and severity. Evidence in the current literature has allowed us to increase the comprehension of the underlying pathophysiology of TMD; however, as mentioned throughout this paper, more organized research on larger human samples needs to be conducted in order to delve deeper into the understanding of how this disease develops and progresses.