DNMT1 involved in the analgesic effect of folic acid on gastric hypersensitivity through downregulating ASIC1 in adult offspring rats with prenatal maternal stress

Abstract Aims Gastric hypersensitivity (GHS) is a characteristic pathogenesis of functional dyspepsia (FD). DNA methyltransferase 1 (DNMT1) and acid‐sensing ion channel 1 (ASIC1) are associated with GHS induced by prenatal maternal stress (PMS). The aim of this study was to investigate the mechanism of DNMT1 mediating the analgesic effect of folic acid (FA) on PMS‐induced GHS. Methods GHS was quantified by electromyogram recordings. The expression of DNMT1, DNMT3a, DNMT3b, and ASIC1 were detected by western blot, RT‐PCR, and double‐immunofluorescence. Neuronal excitability and proton‐elicited currents of dorsal root ganglion (DRG) neurons were determined by whole‐cell patch clamp recordings. Results The expression of DNMT1, but not DNMT3a or DNMT3b, was decreased in DRGs of PMS rats. FA alleviated PMS‐induced GHS and hyperexcitability of DRG neurons. FA also increased DNMT1 and decreased ASIC1 expression and sensitivity. Intrathecal injection of DNMT1 inhibitor DC‐517 attenuated the effect of FA on GHS alleviation and ASIC1 downregulation. Overexpression of DNMT1 with lentivirus not only rescued ASIC1 upregulation and hypersensitivity, but also alleviated GHS and hyperexcitability of DRG neurons induced by PMS. Conclusions These results indicate that increased DNMT1 contributes to the analgesic effect of FA on PMS‐induced GHS by reducing ASIC1 expression and sensitivity.

development of functional bowel disorders in adulthood. 6,7 Our previous study demonstrated that prenatal maternal stress (PMS) induced GHS through enhanced acid-sensing ion channel 1 (ASIC1) demethylation. 8 Others also reported that the pathogenesis of visceral pain involves dynamic changes in the epigenetic markers and gene expression. [9][10][11] As a result, epigenetic regulation may represent a novel mechanism of FD, and deciphering the mechanisms underlying is crucial for the treatment of visceral pain.
DNA methylation is a type of epigenetic modification and is catalyzed primarily by a family of DNA methyltransferases (DNMTs), including DNMT1, DNMT3a, and DNMT3b. 12,13 Epigenetic regulation is not static as it can be altered by environmental stimuli. 14,15 The contribution of DNMT1 and DNMT3a-triggered DNA methylation to neuropathic pain has been reported. [16][17][18] The mRNA of Dnmt1, but not Dnmt3a and Dnmt3b, is downregulated in dorsal root ganglion (DRG) of PMS rats, indicating an involvement of DNMT1 in the development of FD. 8 Nevertheless, the role of DNMT1 in the pathogenesis of PMS-induced GHS has not yet been validated.
Folic acid (FA) is a major component of one-carbon metabolism that plays a critical role in the nervous system. 19,20 Accumulating evidence links disorders of one-carbon metabolism to neurodegenerative and neuropsychiatric diseases, including Alzheimer's dementia and depression. 21,22 An adequate amount of FA is required during pregnancy for the rapid rate of tissue growth. 23 FA supplementation during the periconceptional period reduces the prevalence of neural tube defects. 24,25 Moreover, systemic administration of FA improves axonal regeneration through DNA methylation. 26 FA also acts through the elevation of DNMT activity to increase neuronal differentiation in neural stem cells. 27 The mandatory fortification with FA might alter the expression of DNMT1 in cervical carcinogenesis. 28 Therefore, FA may be useful to promote healing in neural tissues. However, whether FA supplementation alleviates PMS-induced GHS has not yet been explored.
We have established a PMS-induced FD model and confirmed the decrease of DNMT1 and increase of ASIC1 in PMS rats. 8 In the present study, we aimed to further investigate the mechanism of DNMT1 in the analgesic effect of FA on GHS. Our findings indicated that FA is required to relieve GHS, possibly through elevating DNMT1, then inhibiting ASIC1 and neuronal excitability. Treatments with DNMT1 inhibitor blocked the analgesic effect of FA on GHS.
Moreover, overexpression of DNMT1 reduces neuronal hyperexcitability and GHS. Therefore, FA and DNMT1 provide a potential treatment strategy for visceral pain.

| Animals
Experiments were performed on pregnant Sprague-Dawley (SD) rats and their adult male offspring. The onset of pregnancy was confirmed by the presence of a vaginal plug, which was defined as gestational day 1 as described previously. 29 All experiments were approved by the Institutional Animal Care and Use Committee of Soochow University. The care and handling of all rats followed the guidelines of the International Association for the Study of Pain.

| PMS procedure
Prenatal maternal stress pregnant dams were subjected to a heterotypic intermittent stress protocol, as described previously. 30,31 Agematched control (CON) pregnant dams underwent sham stress, and their male offspring were designated control rats.

| Drugs and administration
Folic acid (80 μg/kg, Sigma) or normal saline (NS) was intraperitoneally injected from gestational day 7 to delivery according to the changed body weight in PMS rats. To determine the role of DNMT1 in the analgesic effect of FA, DC-517, solubilized in dimethyl sulfoxide (DMSO), was used to inhibit DNMT1. The intrathecal injection of DC-517 or DMSO was adopted once a day for 7 days consecutively in FA-treated PMS rats at the age of 6 weeks. On the other hand, lentivirus (LV-NC) or lentivirus expressing full-length Dnmt1 (LV-DNMT1) was used as the negative control or to overexpress DNMT1 (Shanghai Gene Chem Co. Ltd). The estimated titer of the concentrated LV-NC or LV-DNMT1 was 1.25 × 10 8 transducing units per milliliter (TU/mL). A volume of 10 μL of LV-NC or LV-DNMT1 was intrathecally injected in PMS rats at the age of 5 weeks. After 2 weeks, EMG recordings were applied to determine GHS, and whole-cell patch-clamp recordings were conducted to detect the excitability of DRG neurons.

| EMG recordings
The procedure for the insertion of the gastric distension (GD) balloon and recording EMG activities were conducted as described previously. 6 Gastric hypersensitivity (GHS) to balloon distension was measured by an observer in a blinded manner. Rats received a series of 20 s graded GD: 20, 40, 60, 80, and 100 mmHg, with 2 min intervals between distensions. A Biopac electromyogram (EMG)-150C amplifier (Biopac Systems, Inc) continuously recorded EMG. The traces were analyzed using Acknowledge (Biopac Systems, Inc). The response to GD was defined as an increase in EMG activity above baseline during the 20 s GD period. Data were reported as the area under the curve (AUC) of the integrated EMG after baseline subtraction.

| Whole-cell patch clamp recordings
The changes in neuronal excitability and proton-elicited currents were determined as described previously. 33 Briefly, T7-T10 DRGs were dissected out and transferred to an ice-old and oxygenated fresh dissecting solution Then, the DRG neurons were dissociated as described previously. 34

| Real-time polymerase chain reaction (RT-PCR)
Total RNA was extracted from T7-T10 DRGs using TRIzol reagent (Invitrogen), and cDNA was synthesized using EasyScript First-Strand cDNA Synthesis SuperMix kit (Transgen Biotech), according to the manufacturer's instructions. The primer sequences used in this study are listed in Table 1. Negative control reactions did not contain the cDNA temple. The relative expression levels of the target were calculated using the comparative 2 −∆∆Ct method and normalized to that of GAPDH.

| Western blot
Western blot analyses with T7-T10 DRGs were performed as described previously. 32  (1:4000, Sigma) at room temperature for 1 h. The immunoreactive bands were detected with ECL reagents using Image J software on a gel imaging system (Bio-Rad).

| Immunofluorescence
To detect the co-expression of DNMT1 and ASIC1 in gastric DRG neurons, animals were deeply anesthetized and then perfused transcardially with 4% paraformaldehyde (PFA). The cryosections of T7-

| Rotarod test
Rotarod test was performed as described previously. 35,36 The rats were placed on a rotating rod with a non-slippery surface motionless for 5 min to adapt to the rotarod apparatus. The time for which an animal could remain on the rotating rod was recorded using an accelerating rotarod. During the procedure on 3 consecutive days, the speed of the rod rotation increased gradually from 5 to 25 rpm over the course of 5 min. The rats stayed on the rod until falling off, which was measured using a timing device.

| Statistics
Statistical analysis was conducted using OriginPro 8 (Origin Lab) and Prism 6 (GraphPad) software. Normal distribution of data was analyzed using the Kolmogorov-Smirnov test. Differences in groups with normally distributed data were analyzed using the two-sample t-test, two-way repeated-measures analysis of variance (ANOVA); data that do not exhibit a normal distribution were analyzed using TA B L E 1 Oligonucleotides used in the study.

Gapdh-F TGGAG TCT ACT GGC GTCTT
Gapdh-R TGTCA TAT TTC TCG TGG TTCA Mann-Whitney test. All data were expressed as means ± standard error of the mean (SEM). A p-value <0.05 indicated a statistically significant difference.

| PMS induces GHS and decreases DNMT1 expression in DRGs
First, we confirmed that PMS could induce GHS, which was consistent with our previous study. GHS was assessed by measuring the viscerosomatic responses to graded GD using EMG. The gastric balloon and electrodes were implanted at the age of 6 weeks. One week after recovery from surgery, GD-induced painful responses were recorded ( Figure 1A). The adult offspring with PMS were more sensitive to GD, as evident by AUC. The results showed significant effects at the distension pressures of 80 and 100 mmHg ( Figure 1B, n = 6 rats for each group, ***p < 0.001 compared to CON, two-way repeated-measures ANOVA). The motor function detected by rotarod was not affected ( Figure 1C, n = 6 rats for each group, p > 0.05 To determine whether an epigenetic mechanism, such as DNA methylation contributes to PMS-induced GHS, we detected the potential changes of DNMTs. There was no change in the mRNA level of Dnmt1, Dnmt3a, and Dnmt3b ( Figure 1D-F, n = 4 rats for each group, p > 0.05 compared to CON, Mann-Whitney test). However, the protein level of DNMT1 was decreased ( Figure 1G, n = 5 and 6 rats for CON and PMS group, respectively, *p < 0.05 compared to CON, two-sample t-test), while no change was detected in DNMT3a (n = 5 and 6 rats for CON and PMS group, respectively) and DNMT3b (n = 4) ( Figure 1H,I, p > 0.05 compared to CON, two-sample t-test or Mann-Whitney test). These data suggested that DNMT1 was significantly downregulated at the protein levels in PMS rats. Therefore, the present study mainly focused on the role of DNMT1 in PMSinduced GHS.

| FA treatment reverses GHS and enhances DNMT1 expression in T7-T10 DRGs of PMS rats
As a cofactor in a multitude of single-carbon transfer reactions, FA has a direct influence on epigenetic maintenance. 19 (Figure 2A,B, n = 6 rats for each group, *p < 0.05, **p < 0.01, ***p < 0.001 compared to PMS + NS, two-way repeated-measures ANOVA). These data suggested that FA alleviates GHS. Moreover, FA treatment reversed DNMT1 expression that was reduced by PMS ( Figure 2C, n = 4 and 7 rats for NS and FA group, respectively, ***p < 0.001, compared to PMS + NS, two-sample t-test).

| FA treatment reverses hyperexcitability of DRG neurons induced by PMS
Since FA inhibited PMS-induced EMG responses, we investi-

| FA suppresses ASICs hypersensitivity of DRG neurons induced by PMS
ASICs are the main H + receptors in the nervous system and ASIC1, the major subunit responsible for acid-activated current, plays pivotal roles in diverse functions including synaptic transmission and plasticity. 37 We have shown that ASIC1 contributes to PMS-induced GHS. 8 In the present study, PMS increased ASIC1 expression ( Figure 3A, n = 9 and 11 rats for CON and PMS group, respectively, *p < 0.05 compared to CON, two-sample t-test) and FA treatment significantly decreased ASIC1 in PMS rats ( Figure 3B compared to NS, two-sample t-test). Therefore, FA could suppress the ASIC sensitivity activated by PMS. Furthermore, DiI was injected into the stomach wall to label gastric DRG neurons, followed by double immunofluorescence using ASIC1 and DMNT1 antibodies. The results showed that ASIC1 and DNMT1 were co-localized in gastric DRG neurons ( Figure 3D). This suggested a possibility of the interaction between DNMT1 and ASIC1 in PMS-induced GHS.

F I G U R E 1 PMS induces GHS and decreases DNMT1 expression. (A) GHS induced by PMS at the age of 7 weeks. EMG recordings
were applied to grade GD at 20, 40, 60, 80, and 100 mmHg in control and PMS rats. (B) The AUC from PMS rats was significantly higher than controls at distension pressures of 80 and 100 mmHg (n = 6 rats for each group, ***p < 0.001 vs. CON, two-way repeated-measures ANOVA). (C) PMS treatment did not affect the time for rats to stay on the rod (n = 6 rats for each group, p > 0.05 vs. CON, two-sample t-test). (D-F) RT-PCR assays showed that the mRNA expression of Dnmt1, Dnmt3a, and Dnmt3b was not altered (n = 4 rats for each group, p > 0.05 vs. CON, Mann-Whitney test). (G) Western blot assays demonstrated that DNMT1 protein expression was significantly decreased in T7-T10 DRGs of PMS rats (n = 5 and 6 rats for CON and PMS group, respectively, *p < 0.05 vs. CON, two-sample t-test). (H and I) DNMT3a (n = 5 and 6 rats for CON and PMS group, respectively) and DNMT3b (n = 4 rats for each group) protein expression was not changed in T7-T10 DRGs of PMS rats (p > 0.05 vs. CON, two-sample t-test).

| DNMT1 overexpression suppresses GHS and ASIC1 hypersensitivity
To examine whether DNMT1 contributes to PMS-induced GHS and ASIC1 hypersensitivity, LV-NC or LV-DNMT1 was intrathecally injected in PMS rats at the age of 5 weeks. Two weeks after injection, EMG was recorded. LV-NC treatment had no significant effect on EMG amplitude. However, LV-DNMT1 resulted in a dramatic reduction in EMG responses at 60, 80, and 100 mmHg GD pressures, as evident by the AUC (Figure 5A,B, n = 6 rats for each group, **p < 0.01, ***p < 0.001 compared to LV-NC, two-way repeated-measures ANOVA). These data suggested that DNMT1 was involved in the development of PMS-induced GHS. To further determine the correlation between DNMT1 and ASIC1, we examined whether LV-DNMT1 suppresses ASIC1 expression and sensitivity. Western blot analysis showed that LV-DNMT1 treatment significantly reversed DNMT1 expression that was downregulated by PMS ( Figure 5C, n = 5 rats for each group, ***p < 0.001, compared to LV-NC, two-sample t-test). Furthermore, LV-DNMT1, significantly blocked the increase of ASIC1 induced by PMS ( Figure 5C, n = 5 rats for each group, **p < 0.01, compared to LV-NC, two-sample t-test).
In addition, currents evoked by H + in DRG neurons obtained from LV-NC-or LV-DNMT1-treated PMS rats were measured. The average peak current density was −168.8 ± 13.8 and −93.1 ± 18.5 pA/ pF for the LV-NC and LV-DNMT1 groups, respectively ( Figure 5D, n = 23 and 20 cells from LV-NC and LV-DNMT1 groups, respectively, **p < 0.01 compared to LV-NC, two-sample t-test). Together, these findings indicated that DNMT1 suppressed ASIC1 expression and hypersensitivity induced by PMS.

| DNMT1 overexpression decreases neuronal excitability in PMS rats
Since LV-DNMT1 inhibited EMG responses, we investigated

| DISCUSS ION
We have reported that PMS induced GHS in adult offspring, accompanied by DNMT1 decrease and ASIC1 increase. 8 The present study showed that FA ameliorates PMS-induced GHS via increasing DNMT1 and reducing ASIC1 sensitivity and neuronal excitability. report that DNMTs were downregulated under diabetic neuropathic pain and bone cancer pain. 32,42,43 However, it was different from another study that showed an active mechanism in inflammatory pain. 44 Although the detailed mechanism for this difference is unclear, the role of DNMT1 in PMS-induced GHS may differ from that of inflammatory and neuropathic pain.
According to the clinical phenotype, pregnant women need to supplement FA as it can prevent various malformations of neural tubes. [45][46][47] FA also induces spinal axon regeneration via DNA methylation 26,48,49 and alleviates neuropathic pain in spinal cord injury rats. 50 However, the role of FA in PMS-induced GHS was unclear.
The present study showed that FA alleviated GHS, accompanied by increased DNMT1 expression and decreased neuronal excitability.
Moreover, PMS upregulated ASIC1 expression that was in accordance with our previous report. 8   data collection and analysis, decision to publish, or preparation of the manuscript.

CO N FLI C T O F I NTE R E S T S TATE M E NT
No conflicts of interest, financial or otherwise, are declared by the authors.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.