A one sample SEM picture from each group can be shown in Figure 1. SEM of dental elements showed the exposed tubules in E0 and E1 groups while no tubules exposure were observed in the control groups (Figure 1). There was higher dentin hypersensivity scores in rats with dentin erosion exposed to cold water (0.538±0.144) than warmish (0.154±0.087; Z= 2.169, p = 0.030, the Mann-Whitney test), confirming thatcold stimulation increased the pain response Figure 1.

Figure 2 shows pain modulation of neuronal plasticity induced by low-frequency stimulation and high-frequency stimulation.

The mean fEPSP slope and mean PS amplitude recorded 0-5 minutes after induction were used to assess short-term synaptic and somatic plasticity, respectively (Figure 3). LFS to the perforant pathway induced comparable early depression of synaptic strength in the control (as shown by fEPSP slope of 88.2±6.0% of baseline), E0 (81.6±2.7%) and E1 groups (79.4±2.8%, F_2,15=1.227; p =0.321). The slope of EPSP following HFS, on the other hand, increased by 112.3±10.6% of baseline in the control group, 119.2±7.9% of baseline in the painless group, and 131.6±5.8% of baseline in the painful group (F_2,15=1.378; p=0.277). Despite of the lack of a significant difference among groups these results indicate that both short-term synaptic depression induced by LFS and short-term synaptic potentiation induced by HFS are amplified by pain sensation.

Contrary to expected, an early potentiation of the PS amplitude was observed following both LFS (121.2±25.2% of baseline) and HFS (193.9±20.2% of baseline) in control rats. The PS amplitude decreased by 67.8±8.1% of baseline after LFS but increased by 448.0±70.1% of baseline after HFS in the painful group. The PS amplitude in the painless group did not show any significant difference to pre-induction after LFS (94.3±3.2 of baseline), but potentiated by 195.0±23.4% of baseline after HFS. After a one-way ANOVA, the PS amplitude was revealed to have a significant group difference (F_2,18=10.929; p=0.001), however there was a tendency for significance after LFS (F_2,15=3.006; p=0.080). Tamhane's post-hoc test confirms that painful stimulation facilitates post-tetanic potentiation of the PS amplitude. (vs. control group, p=0.031; vs. painless group, p=0.031). These results indicate that short-term somatic potentiation induced by HFS is amplified by pain sensation.

The mean fEPSP slope and mean PS amplitude recorded 55-60 minutes after induction were used to assess long-term synaptic and somatic plasticity, respectively (Figure. 4). In the control group, both LFS (198.8±35.9% of baseline) and HFS (154.0±16.5% of baseline) to the perforant pathway induce a long-term increase in PS amplitude although no change in fEPSP slope was observed (100.1±10.7 and 99.9±4.5 of baseline, respectively).In contrast to the control group, LFS to the perforant pathway resulted in synaptic LTD in painless group (82.5±6.0%, t₅ = 0.034) and painful group (79.0±4.9, t₆ = 0.008). LFS-induced long term potentiation of PS amplitude was lower in painless group (130.6±13.6%) and painful group (94.0±12.6%), but higher in painless group (197.5±31.9%) and in painful group (337.3±53.2%). One-way ANOVAs yielded significant group difference for the PS amplitude after LFS (F_2,15=5.200, p=0.019) and HFS (F_2,18=6.671, p=0.007).There was no significant group difference in fEPSP slope after LFS (F_2,15=2.13, p=0.146) and after HFS (F_2,18=1.369, p=0.281). Tamhane's post-hoc test confirms that painful stimulation facilitates long-term potentiation of the PS amplitude after HFS (p=0.038). Nevertheless, the difference between control and E1 rats showed a trend toward statistical significance after LFS (p=0.093).

The biochemical conversion from short-term plasticity to LTP requires activation of intracellular kinase pathways and gene expression induced with high-frequency stimulation ^{14; 10}. MAPK control of activity-dependent gene expression is one of the critical events for long-lasting changes at the synapse. Therefore, we looked at how HFS-induced MAPK activation and gene expression differed between rats exposed to painful and painless stimuli (Figure 4). We found that the p-ERK1/2 (ꭓ²=7.692; p=0.021) levels, but not total-ERK1/2 (ꭓ²=3.978; p=0.137), in the hippocampus differed among groups 60 minutes after HFS. Post-hoc Mann Whitney test revealed that p-ERK1/2 levels in the painless group was considerably lower than in the control group (Z=2.727, p=0.006), suggesting that pain sensation prevents down regulation of ERK1/2 phosphorylation by dentin erosion. There was no significant difference among three groups for JNK and p38-MAPK (data not shown).

In addition, we found that PSEN2-mRNA levels significantly elevated in the rats exposed to painful stimuli (n=5; 1.60±0.13 fold relative to a control rat), compared to the rats exposed to painless stimuli (n=5; 0.97±0.18) in response to HFS (Z=2.104; p=0.030, Mann-Whitney U test). There was no significant difference in MAPT-mRNA (painful:1.06±0.07; painless:0.84±0.15) and JNK-mRNA (painful:1.57±0.16; painless:1.05±0.21), but the significance of difference remained at 0.052 level for BACE1-mRNA (painful:1.60±0.13; painless:0.74±0.14). Figure 5

LTP and LTD are synaptic plasticity types that characterize a potentiation or depression of the PS amplitude with or without fEPSP slope. The present study investigated whether pain sensation modulates neuronal plasticity. A bulk of finding from our study indicates that chronic and intermittent pain which is induced by dentin hypersensitivity facilitates the induction of plasticity. In control rats, neither LFS nor HFS resulted in a lasting change in synapse strength; however both stimulation methods induced somatic LTP. Interestingly, LFS was able to induce synaptic depression in the rats with dentin erosion. Moreover HFS-induced LTP increased in magnitude if the rats with dentin erosion were exposed to painful stimuli. The current findings suggest that pain sensation increases glutamatergic granule cell somatic plasticity after HFS but decreases it after LFS. In agreement with this finding, pain significantly increased LTP induced by theta burst stimulation in the DG and CA1 area of hippocampal slices of rats suffering from persistent nociception 50.

The present study used dentin hypersensivity model induced by dentin erosion mediated by acidic solution. Dentin hypersensitivity can be defined as a sharp, short pain arising from exposed dentin in response to thermal, tactile, osmotic, or chemical stimuli. According to the Hydrodynamic Theory, the pain of dentin hypersensitivity is caused by the quick movement of dentinal fluid which, in turn, excites the mechanoreceptors in the periphery of the pulp 5.In a functional magnetic resonance research, higher BOLD (blood-oxygen-level dependent) signals were identified in cortical areas involved in emotion and pain control when painful electrical stimulation was applied to maxillary canine teeth versus painless stimulation 7. As a result, it is possible to deduce that exposing teeth with compromised tissue integrity to cold stimulation causes a pain response. Such conclusions were supported by the analysis of hypersensitivity scores that shows cold stimulation caused increased motor response in the rats with dentin erosion.

The present study also investigated pain modulation of neuronal plasticity-related MAPK activity in experimental model of dental erosion. The present finding shows that peripheral tissue injury reduces ERK1/2 phosphorylation in the dentate gyrus, which is avoided by pain perception. It has been demonstrated that intra-plantar saline or bee venom injection, which simulate temporary or chronic pain, may cause a strong and long-lasting activation of ERKs in the hippocampus and primary somatosensory cortex21. Elk mRNA isoforms, a major target of ERKs within the nucleus, are individually up-regulated 2.3-fold in the dorsal root ganglia following peripheral nerve damage 25. Activation of the ERK isoforms of MAPK has been demonstrated to be necessary for the induction of NMDA receptor-dependent LTP in the dentate gyrus 9. Previous studies have provided evidence that chronic or persistent pain can cause malfunction in numerous brain structures involved in amnesia, insomnia, and depression 171330. These findings support to the functional importance of ERKs-mediated signaling pathways in the processing of consequences of pain associated with cognitive dimensions.

We show that painful stimuli increase the expression of certain neurodegeneration-related genes, particularly PSEN2. PSEN1 and PSEN2 genes encode presenilins (presenilin 1 and presenilin 2) which constitute the catalytic subunits of the γ-secretase intramembrane protease protein complex. Mutations in presenilins cause autosomal dominant, early-onset familial AD (FAD) and promote cerebral Amyloid β accumulation 28. Transgenic mouse studies have revealed that FAD-linked presenilin 1 is associated with a greater degree of LTP induction in the hippocampus 3548383412. To the best of our knowledge, no study has addressed a direct interaction between LTP, pain and presenilin 2. Our findings support the idea that a greater induction of LTP can be related with pain-induced up-regulation of PSEN2.

Previous studies have reported that many of patients with neurodegenerative disease such as Alzheimer s disease 24, Parkinson s disease 3 and Huntington s disease 1 complain of painful symptoms though their origin is variable. The presented study draws attention to the increase in the transcription of some neurodegenerative proteins in the hippocampus tissue, which is involved in a number of neurodegenerative diseases, due to exposure to painful stimuli. Among these proteins, the pain-induced upregulation of PSEN2, which forms the catalytic subunit of the γ-secretase complex, during LTP induction seems to be important according to our study results. Other possible functions of presenilins are on important neuronal processes such as regulation of Ca homeostasis 8, modulation of apoptosis 11 and intracellular trafficking 33. It was also shown that presenilin expression is increased in the brains of some AD patients194214. Therefore, our study highlights that exposure to painful stimulation has the potential to increase susceptibility to neurodegeneration accompanied by impaired regulation of plasticity.

The main limitation of the research is that it did not focus on the mechanism of which substance or substances mediate the modulation of plasticity by pain. A candidate mediator may be Orexin because the participation of Orexin receptors within the DG has been demonstrated in the anti nociceptive effects of lateral hypothalamic stimulation 6, a region that responds to both acute heat stimuli and a chronic inflammatory irritant 39. The DG region has the Orexin-containing axon terminals that come from neurons in the lateral hypothalamus 44. Orexin, when administered directly to the dentate gyrus, increased LTP, which was prevented by pretreatment with SB-334867, a selective Orexin 1 receptor antagonist 44. Exogenous orexin exacerbated hippocampal LTP depression, and increased Aβ and tau pathologies in APP/PS1/tau triple-transgenic AD model mice by affecting BACE1 29.

As a result, pain perception may influence synaptic and somatic neural plasticity, as well as ERK phosphorylation and PSEN2 expression in the dentate gyrus. Further investigations are needed to understand the molecular targets that mediate the pain modulation mechanism of plasticity. These studies will contribute to both a better understanding of the function of the hippocampus in the formation of memory traces of painful stimuli and to revealing the role of pathological pain processes in neurodegeneration.