How the body’s daily clock affects responses to mechanical pressure in tooth-supporting ligament cells

Orthodontic tooth movement (OTM) is induced by forces exerted by orthodontic appliances. During OTM, remodeling processes take place in the periodontal ligament (PDL) and surrounding bone. The mechanical force applied to the tooth causes compressive, but also tensile strain in various areas of the PDL. This results in biomechanical stress on periodontal tissues and especially on cells in the PDL. These PDL cells play an important role in the control of resorption and apposition of bone induced by mechanical stress [41].

Mechanical stress also results in sterile inflammation that includes an associated release of inflammatory mediators such as cytokines including interleukin 1-beta (IL-1β) [13, 19, 43, 46]. The release of these messenger substances activates both osteoclasts and osteoblasts and causes bone remodeling [3, 16]. In addition, dependent on the amount of force applied, specific factors such as runt-related transcription factor (RUNX2/Runx2) or collagen-1α (COL-1α/Col-1α) are increased or inhibited [48].

Runx2 plays an important role in bone development. It is co-decisive for osteoblast differentiation [10, 64, 65] and it has been shown previously in vitro and in vivo that RUNX2 gene expression is increased by mechanical forces in PDL cells and osteoblasts, whereas reduced exposure to mechanical strain, i.e., microgravity, decreases Runx2 expression [8, 65].

One of the most important proteins of the extracellular matrix in the periodontium is Col-1α, predominantly produced by osteoblasts and PDL cells, which is responsible for the elastic properties of the periodontium [9]. Collagen is increasingly produced in the stretched parts of the periodontium especially during tooth movement, while collagen production was shown to be reduced on the pressure side [23].

Circadian rhythms define endogenous cycle processes that occur in a 24 h period. Environmental stimuli can influence circadian rhythms and are described as “Zeitgeber” [14]. In mammals, a crucial Zeitgeber is daylight, which influences the body’s central clock in the suprachiasmatic nucleus (SCN) in the front of the hypothalamus [40]. The SCN is seen as the main pacemaker of the human body, controlling other peripheral clocks in organs and tissues via neuronal and hormonal signaling pathways, a process called ‘entrainment’ [1, 11, 59]. Prerequisite for the maintenance of a circadian rhythm in peripheral tissues is the group of clock genes. They form the peripheral oscillator at the molecular level. Different clock genes are part of activating and inhibiting feedback loops. The main feedback system comprises the transcription factors aryl hydrocarbon receptor nuclear translocator-like protein 1 (ARNTL) or brain and muscle ARNT-like 1 (BMAL1), and circadian locomotor output cycles kaput (CLOCK). They induce the formation of the period (PER; PER1/PER2/PER3) and cryptochrome genes (CRY1/CRY2). These in turn form a heterodimer together and directly inhibit the complex of CLOCK and ARNTL. Thus, the activating loop is formed by ARNTL and CLOCK, while the inhibitory feedback loop is dependent on the PER and CRY complex [26, 30, 49, 61]. This feedback loop, which is “straightforward” at the molecular level, is also described as the “core” oscillator mechanism in mammals [30].

It was shown that some of the clock genes are also involved in various other important cell functions [14]. Thus, the circadian rhythm is a crucial factor for physiological and pathophysiological processes in the human body and has an impact on diseases such as diabetes, cardiovascular disorders, sleep disorders, and also cancer [15, 28, 42, 44, 47, 53, 54].

There is evidence that circadian rhythmicity is also of relevance in oral cell homeostasis. For example, dental pulp cells, ameloblasts, odontoblasts, cementoblasts, gingival fibroblasts, and PDL cells are influenced by the circadian rhythm [17, 20–22, 60, 62, 63]. It was also shown that the proliferation of alveolar bone and tooth cementum in mice have a circadian rhythm [51]. Thereby, the expression of development- and remodeling-related genes for these tissues like OPG, RANKL, RUNX2, OCN, OPN, and COL-1α present a circadian rhythm, all of which were proven to share a correlation with the fluctuation of the expression of different clock genes [29, 56].

That the expression of clock genes follows a circadian pattern after synchronization in PDL cells was already shown [17].

The current study focused on the investigation of a possible influence of mechanical stress on the clock genes in PDL cells. Furthermore, we aimed to analyze whether tensile strain has different effects on elementary regulatory factors of PDL remodeling (IL-1β, RUNX2, COL-1α) depending on the phase of the circadian rhythm. We hypothesized that a differential influence of mechanical stress on PDL homeostasis can be assumed depending on the circadian rhythm.

In this study, the influence of mechanically induced stress on the main players of peripheral circadian rhythm of PDL fibroblasts was investigated. Static stretching altered gene expression of all clock genes investigated, suggesting an influence of mechanical stress on these peripheral messengers of the circadian rhythm. However, vice versa, the circadian rhythm also seemed to influence the effects that a mechanical force exerted on PDL cells. This was apparent by a differential regulation of three important regulatory factors, depending on the onset of mechanical strain in the course of the circadian rhythm.

Also, the molecular concept of a circadian rhythm in PDL fibroblasts could be observed and significant changes of the clock genes ARNTL, PER1, and PER2 over the course of 24 h after synchronization became evident. As already demonstrated by Hilbert et al., the peripheral circadian rhythm is an important process that is controlled in PDL fibroblasts by a cyclic regulation of the clock genes [17]. Also in this study, regulation occurred after synchronization of PDL fibroblasts with the help of dexamethasone. An interesting finding was that the gene expression of ARNTL and Per2 showed an opposite course. This opposite regulation is a feature of the peripheral molecular clock, as ARNTL is a so-called positive factor of the molecular feedback loop of the circadian clock, whereas the Period genes belong to the corresponding negative factors [50]. Also, in the paper of Hilbert et al., regulations of ARNTL and PER2 were found in particular, whereas CLOCK and PER1 were influenced to lesser extents. This could suggest that ARNTL and PER2 are profoundly important for regulation of the molecular clock in PDL fibroblasts, but this assumption should be further investigated, because the experiments of Hilbert et al. and the experiments of the present study were performed with the same PDL cells from Lonza in the same laboratory.

Mechanical strain increased the gene expression for the negative factors of the circadian clock PER1 and PER2, whereas the positive factor ARNTL was reduced by mechanical stimulation at every time point measured from the 8 h time point onward. CLOCK gene expression was also affected by mechanical stimulation. Furthermore, 4 h after synchronization, an increase was detected, whereas a reduction was observed after 12 h. The influence of mechanical stimulation on molecular clock genes which our results suggest are consistent with the existing literature. Qin et al. examined rat PDL cells after orthodontic tooth movement. In their in vivo study, an increase in the expression of PER1 was demonstrated to be induced by mechanical stimulation [38].

In addition to the question of whether a mechanical force affects the peripheral circadian clock, we tried to determine whether the circadian clock alters the effects that mechanical forces induce in PDL fibroblasts. Therefore, mechanical strain was initiated at different time points after synchronization and the effects of the mechanical force on key regulatory factors were examined. We picked a second time interval 12 h after synchronization for the stimulation, because ARNTL and PER2 had demonstrated an opposite regulation starting at this time point in the control cells. Furthermore, Hilbert et al. already showed that the time point 12 h after synchronization was well suited to monitor gene expression changes of regulatory factors in PDL cells in the course of the circadian rhythm [17].

Starting mechanical strain at t0 (directly following synchronization, group SI1) after 2 and 4 h, there was an increase in protein expression of Col-1α compared to t0, whereas the same mechanical stimulation starting at t12 induced a reduced gene expression (after 2 and 4 h) and protein expression (after 4 h) compared to t12. Hilbert et al. also discovered reduced gene expression of COL-1α 12 h after synchronization of the cells. These results provide evidence for an influence of the circadian rhythm on Col-1α synthesis. Qin et al. also investigated the regulation of Col-1α synthesis in relation to the circadian rhythm, but in a rat model. The authors found a reduction in COL-1α gene expression at each measured time point in the 12–18 h time interval, although this reduction did not reach significance. As mentioned in the introduction, Col-1α is critical in cell homeostasis and connective tissue turnover [12, 18, 33]. Interestingly, at the transcriptional level, the mechanical stress had an inhibitory effect on COL-1α gene expression independent of the time of its onset within the circadian cycle. On the other hand, at the protein level, an increase in protein expression was observed after mechanical stimulation in group SI2. This is in accordance with findings of another study which also showed that stretching of PDL cells increased the protein biosynthesis of Col-1α [18]. Qin et al. observed an upregulation of COL-1α gene expression after 3 and 21 h following mechanical load in vivo [38]. Of course, in that in vivo experiment, it was not possible to locally separate the true types of strain triggered by the force, being it tension or compression, as in our in vitro approach. Moreover, in their study continuous force was applied, while in our experimental setup, we picked the force onset at different time points after synchronization. Thus, our results are difficult to compare with those of Qin et al. [38].

Runx2 is a critical factor in osteoblast differentiation and it is important for bone remodeling during orthodontic tooth movement [64]. Our results showed an upregulation of RUNX2 gene expression 4 h after synchronization in group SI1, whereas in group SI2, gene expression was decreased after 2 h as well as after 4 h of mechanical strain. Again, parallels can be drawn to the results of Hilbert et al., as they found a decreased gene expression for RUNX2 12 h after synchronization compared to t0 [17]. Other studies also found Runx2, which is also a transcriptional factor crucial for the differentiation of ameloblasts, to be upregulated in dependence of other clock genes such as Nr1d1 [4, 52]. Mechanical strain, like tension, has repeatedly been shown to lead to increased gene expression of RUNX2 alongside cell proliferation. This has been demonstrated in murine stem cells and in human PDL cells [2, 48]. However, the magnitude of the tensile strain plays a crucial role. Sun et al. found that excessive elongation of the cell decreased the gene expression of RUNX2 [48]. We used a strain of the cells leading to an elongation of about 20%, which is considered to belong the higher spectrum of tissue strain induced by orthodontic forces [24]. It was noticeable that for Runx2 no significant differences, neither at the gene nor the protein level, resulted from mechanical stimulation. Hence, further studies with variable stretch intensities and time intervals are necessary to gain further insights into the effects of mechanical strain on RUNX2 in dependence of the circadian rhythm.

Along with other interleukins, IL1‑β is an important inflammatory marker detectable in the remodeling processes of mechanically induced tooth movement in the periodontium [13]. Similar to Hilbert et al., in our control group an increase in gene expression of IL1‑β was seen 2 and 4 h after synchronization and then a decrease in the time slot from t12 to t16a [17]. IL1‑β has already been linked to the circadian rhythm by other studies. It is an important factor in sleep rhythm and can interfere with spontaneous sleep, when inhibited [25]. Alterations of IL-1β during the course of the circadian rhythm have also been demonstrated in oral saliva [39]. The mechanical force triggered an increase in gene expression in group SI2 from starting point t12 onwards with a maximum value measured 2 h after mechanical strain. It is already known from several studies that intermittent forces promote IL-1β production, which would be consistent with our results [34, 35].

For comparable measurements, we followed the protocol of Hilbert et al. regarding the synchronization protocol [17]. Since our cell culture lacks the central Zeitgeber, we used an artificial pacemaker to induce synchronization of our cells. Balsalobre et al. showed that dexamethasone as a serum component causes high reproducibility of circadian synchronization [6, 7]. With dexamethasone stimulations during different time points in the course of a 24 h phase, a change in circadian rhythm has to be expected. This may favor but also disrupt the circadian rhythm [57]. Other stimuli such as Ca²⁺, protein kinase C, and glucocorticoid hormones are also able to induce fibroblast synchronization, but these are already components of the used serum [5, 6]. Additionally, the possible implications of the supplements in our DMEM have to be discussed. We added 1% serum, 1% Pepstrep 100 µg/ml, Plasmocin, and vitamin C. We know from previous studies that serum, for example, can influence the circadian rhythm [6]. In addition, central and peripheral oscillators can be influenced differently by factors such as nutrients and feeding cycles [36]. The other supplements, such as vitamin C as an antioxidant, are also likely to influence the circadian rhythm [55]. Detailed studies investigating the causal relationship between circadian rhythm and oxidative stress are still lacking. Therefore, we cannot exclude that the administration of different supplements in our paper and the work of Hilbert et al. influenced the circadian rhythm of PDL cells [17].

Furthermore, the use of the right cell passage is important, as cell senescence has an impact on circadian expression of clock genes in vitro [27]. Kunieda et al. reported an attenuated circadian expression of clock genes in senescent passages compared to young cell passages in cultured human aortic vascular smooth muscle cells. Shimizu et al. investigated cell senescence in human periodontal ligament cells during mechanical loading [45]. They characterized passages 5–6 as young passages. From passage 15 onwards, the cell doubling population slows down. Passages over 22 were identified as senescent passages. Therefore, in regard to cell senescence, there should not be attenuated effects on the clock gene expression in our experiments as we used passage 6, a comparable passage to Hilbert et al., who used passage 4 [17].

Additionally, there are also disadvantages of our chosen in vitro approach. The in vivo study performed by Qin et al. certainly offers a more comprehensive picture of the overall influences of the circadian rhythm on the periodontal tissues. However, especially for such a scarcely researched topic as the interaction between the effects of the circadian rhythms and those of a mechanical force, in vitro studies have the special advantage that different individual and biasing factors can be controlled to a higher degree. In our study, we applied a tensile strain of 20% which, according to Kanzaki et al., is rather strong, to induce tooth movement clinically [24]. A study applying a force of lower magnitude would certainly be of interest, as would be studies focusing on compressive strain. Furthermore, our measurements were limited to a period of 24 h. The reason for this were the findings of Hilbert et al. which suggested that without further synchronization the recognizability for the circadian rhythm of cells in vitro seems to weaken during 48 h [17].

It seems that mechanical stress affects the expression of clock genes of the circadian rhythm in periodontal ligament (PDL) cells, which are at the same time involved in various other important cell functions. Furthermore, it appears that the cell response to the mechanical stimulus varies in the different phases of the circadian rhythm. Thus, our results could open the way to more efficient timing for orthodontic treatment devices in the future, especially for those appliances used on an hourly basis.

Open Access funding enabled and organized by Projekt DEAL.

L.I. Peters, J. Marciniak, E. Kutschera, C. Luiz, E. Calvano Küchler, C. Kirschneck, A. Jäger and S. Beisel-Memmert declare that they have no competing interests.