A hydrolysate of poly-trans-[(2-carboxyethyl)germasesquioxane] (Ge-132) suppresses Cav3.2-dependent pain by sequestering exogenous and endogenous sulfide

Poly-trans-[(2-carboxyethyl)germasesquioxane] (Ge-132), an organogermanium, is hydrolyzed to 3-(trihydroxygermyl)propanoic acid (THGP) in aqueous solutions, and reduces inflammation, pain and cancer, whereas the underlying mechanisms remain unknown. Sulfides including H2S, a gasotransmitter, generated from l-cysteine by some enzymes including cystathionine-γ-lyase (CSE), are pro-nociceptive, since they enhance Cav3.2 T-type Ca2+ channel activity expressed in the primary afferents, most probably by canceling the channel inhibition by Zn2+ linked via coordinate bonding to His191 of Cav3.2. Given that germanium is reactive to sulfur, we tested whether THGP would directly trap sulfide, and inhibit sulfide-induced enhancement of Cav3.2 activity and sulfide-dependent pain in mice. Using mass spectrometry and 1H NMR techniques, we demonstrated that THGP directly reacted with sulfides including Na2S and NaSH, and formed a sulfur-containing reaction product, which decreased in the presence of ZnCl2. In Cav3.2-transfected HEK293 cells, THGP inhibited the sulfide-induced enhancement of T-type Ca2+ channel-dependent membrane currents. In mice, THGP, administered systemically or locally, inhibited the mechanical allodynia caused by intraplantar Na2S. In the mice with cyclophosphamide-induced cystitis and cerulein-induced pancreatitis, which exhibited upregulation of CSE in the bladder and pancreas, respectively, systemic administration of THGP as well as a selective T-type Ca2+ channel inhibitor suppressed the cystitis-related and pancreatitis-related visceral pain. These data suggest that THGP traps sulfide and inhibits sulfide-induced enhancement of Cav3.2 activity, leading to suppression of Cav3.2-dependent pain caused by sulfide applied exogenously and generated endogenously.

We have shown that sulfides including hydrogen sulfide (H 2 S), a gasotransmitter, are capable of enhancing the activity of Ca v 3.2 isoform of T-type Ca 2+ channels (T-channels), and cause somatic and visceral pain or hypersensitivity in mice and rats, an effect abolished by knockout or knockdown of Ca v 3.2 gene and by T-channel blockers [17][18][19][20][21][22][23][24]. The mechanisms by which sulfide enhances Ca v 3.2 T-channel activity have yet to be directly clarified. Nonetheless, considering the high affinity of sulfides to Zn 2+ , it is likely that, as does L-cysteine [25,26], sulfides might interact with Zn 2+ linked by coordinate bonding to a histidine residue at position 191 (His 191 ) in the second extracellular loop of domain I of Ca v 3.2, and cancel the Zn 2+ inhibition of Ca v 3.2 activity. There is also plenty of evidence that endogenous H 2 S produced from L-cysteine by cystathionine-γ-lyase (CSE), an H 2 S-forming enzyme, contributes to the Ca v 3.2-dependent pain in various animal models for inflammatory and neuropathic somatic pain [17,27,28], as well as visceral pain associated with cystitis or pancreatitis [19,[29][30][31][32]. Interestingly, there is evidence that germanium reacts with chalcogens including sulfur [33,34] and that Ge-132 relieves pain in cancer patients at a terminal stage [35]. We thus assumed that THGP, the hydrolysate of Ge-132, might directly trap sulfide, a pronociceptive molecule, resulting in pain suppression. To test this hypothesis, in the present study, we examined whether THGP would react with sulfide in vitro and inhibit sulfide-induced enhancement of T-channel-dependent currents in Ca v 3.2-transfected HEK293 cells. Further, we investigated the effects of THGP on exogenously applied sulfide-induced paw allodynia, which is dependent on Ca v 3.2 [17,18,21,23], and on visceral pain accompanying cystitis and pancreatitis, known to involve the endogenous sulfide/-Ca v 3.2 pathway [19,[29][30][31][32], in mice.

Tandem mass spectrometry analysis
NaSH at 1 mM and THGP at 1 mM were mixed and incubated for 10 min at room temperature, and the structure of the reaction products was analyzed by tandem mass spectrometer, LCMS-8060 (Shimadzu, Kyoto, Japan), equipped with heated ESI probe. The pH of the mixture was adjusted to 7.0 with NaOH. Mass spectrometry (MS) and tandem mass spectrometry (MS/MS) were carried out in the negative ion mode. The samples were analyzed by flow injection analysis. The mobile phase consisted of water/methanol (1:2, v/v) and it was pumped at a flow rate of 0.2 mL/min. The overall run time was 1 min. The injection volume was 2 μL. Probe voltage was set at − 3000 V, DL temperature was 150 • C and 250 • C, heat block temperature was 400 • C, and interface temperature was 100 • C. Nebulizer gas flow was 3 L/min, drying gas flow was 10 L/min, and heating gas flow was 10 L/min. ESI mass spectra for reaction products were measured at a scan range from m/z 100 to 250. The ion at m/z 195 in the negative ion full-scan mass spectra obtained by MS analysis of the mixture of THGP and NaSH was selected as a precursor for MS/MS analysis, since the relative signal at m/z 195 in the mixture of THGP and NaSH was greater than that in THGP alone. Product ion spectra were measured at a scan range from m/z 50 to 250 with the collision energy of − 20 and − 35 V. Data were acquired and analyzed using LabSolutions LCMS software (Shimadzu). The values of calculated m/z were determined using ChemBioDraw Ultra 11.0 (PerkinElmer, Inc., Waltham, MA, USA).

Proton nuclear magnetic resonance analyses of the interaction between THGP and sulfides
In proton nuclear magnetic resonance ( 1 H NMR) analysis, a phosphate buffer solution (20 mM, pH 7.4) was prepared with D 2 O and used for preparing samples. The solution of THGP at a final concentration of 10 mM was mixed with NaSH at 10 mM or Na 2 S at 1.25-10 mM, 10 min (or 5 min in some experiments) before beginning the NMR analysis. In the experiments to study the reaction between THGP and Na 2 S, each at 10 mM, in the presence of ZnCl 2 at 0.1-10 mM, a Tris-HCl buffer solution (1 M, pH 7.4) instead of the phosphate buffer were prepared using D 2 O and deuterium chloride solution, in order to avoid the formation of Zn 3 (PO 4 ) 2 precipitation. The deposited ZnS in the reaction mixture was filtered off immediately before NMR analysis. 1 H NMR spectrum of the reaction mixtures or its filtrates was measured at 25 • C by a JEOL JNM-ECA 800 (800 MHz, 1 H).

Measurement of T-currents by the whole-cell patch-clamp recording
T-channel-dependent Ba 2+ currents (T-currents) were measured by the whole cell-patch clamp recording in hCa v 3.2-HEK293 cells, as described elsewhere [23,24,37]. The composition of the extracellular solution (mM) was: 152 tetraethylammonium (TEA)-Cl, 10 BaCl 2 and 10 HEPES, adjusted to pH 7.4 with TEA-OH. The composition of the intracellular solution (mM) was: 110 Cs-MeSO 4 , 14 creatine phosphate, 10 HEPES, 9 EGTA, 5 Mg-ATP, and 0.3 Tris-GTP, adjusted to pH 7.2 with CsOH. The resistance of the patch electrodes was a range of 3-5 MΩ. Series-resistance was compensated by 80%, and current recordings were low-pass filtered (<5 kHz). The cell membrane voltage was held at − 80 mV, and T-currents were elicited every 12 s by a test pulse of 200 ms duration at − 20 mV. The cells were superfused at a rate of 3 mL/min with the extracellular solution, and after stabilized, exposed to the solution containing THGP (1-10 mM) for 5 min, and then to the solution containing Na 2 S at 10 μM in addition to THGP for 10 min.

Animals
Female (for the cystitis model) or male (other experiments) ddY mice (18-35 g) were purchased from Kiwa Laboratory Animals Co. Ltd. (Wakayama, Japan). The mice were housed in a temperature-controlled room at 24 • C under a 12-h day/night cycle, and had free access to food and water. All animals were used with approval by Kindai University's Committee for the Care and Use of Laboratory Animals, and all procedures employed in the present study were in accordance with the guidelines of the Committee for Research and Ethical Issues of IASP (www.iasp-pain.org/Education/Content.aspx?ItemNumber=1217).

Assessment of mechanical allodynia induced by intraplantar injection of Na 2 S in mice
Mice were placed on a risen wire mesh floor, covered with a clear plastic box (10 × 10 × 10 cm) and habituated to the experimental environment. Then, the mid-plantar surface of the right hindpaw was stimulated with von Frey filaments (0.008, 0.02, 0.04, 0.07, 0.16, 0.4, 0.6 and 1.0 g), and 50% paw withdrawal threshold was determined according to the up-down method [23,38]. After measurements of baseline thresholds, mice received intraplantar (i.pl.) injection of Na 2 S at 10 pmol/paw in a volume of 10 μL in the right hindpaw, and thereafter, the nociceptive threshold was measured every 15 min. THGP at 0.02, 0.2 and 2 μg/paw (0.1, 1 and 10 nmol/paw, respectively) was co-injected i.pl. with Na 2 S 10 pmol/paw, and THGP at 30 and 100 mg/kg (150 and 500 μmol/kg) was administered i.p. 30 min before i.pl. Na 2 S at the same dose.

Creation of a cyclophosphamide-induced cystitis model and assessment of bladder pain-like nociceptive behavior, referred hyperalgesia, bladder swelling and micturition in mice
The model of cyclophosphamide (CPA)-induced cystitis in mice was prepared according to the previously reported method [39][40][41] with minor modifications, in which the increased dose, 400 mg/kg, but not 300 mg/kg, of CPA was used to induce highly reproducible bladder pain-like behavior and referred hyperalgesia in mice [30,42]. The bladder pain-like nociceptive behavior, such as licking the skin region of the bladder and pressing the lower abdomen against the floor, was counted for a 30-min period starting 3.5 h after i.p. CPA, and subsequently, the referred hyperalgesia was evaluated by stimulating the skin region between the anus and urethral opening with each of four distinct von Frey filaments (0.008, 0.07, 0.4 and 1.0 g) [41]. Thereafter, micturition frequency was measured by a voiding spot method for 2 h [32,43]. Finally, i.e. approximately 6.5 h after i.p. CPA, the mice were killed by cervical dislocation, and the bladder was isolated for determination of the wet tissue weight as a marker of bladder swelling and for Western blot analysis of CSE protein levels. THGP at 30 and 100 mg/kg or TTA-A2, a selective T-channel blocker, at 1 mg/kg was administered i. p. 3 h after i.p. CPA.

Creation of a cerulein-induced pancreatitis model and assessment of referred hyperalgesia and plasma amylase activity in mice
The pancreatitis model was created in male mice by repetitive i.p.
injection of cerulein at 50 μg/kg at 1-h intervals, 6 times in total, as described previously [19,44]. Referred hyperalgesia was assessed by using four distinct von Frey filaments (0.02, 0.07, 0.16 and 1.0 g), 5.5, 6.0 and 6.5 h after the onset of cerulein injection. Thereafter, i.e. 7 h after the onset of cerulein injection, citrated blood was withdrawn from the aorta of mice anesthetized with i.p. injection of midazolam at 4 mg/kg, medetomidine 0.3 mg/kg and pentobarbital 10 mg/kg, to determine the plasma amylase activity as an indicator of the severity of the evoked pancreatitis, and the pancreas was excised afterwards for determination of tissue weight as a marker of pancreatic edema and for Western blot analysis of CSE protein levels in the pancreatic tissue. THGP at 100 mg/kg or TTA-A2 at 1 mg/kg was administered i.p. 5 min after the final (6th) injection of cerulein.

Western blot analysis
Protein expression levels of CSE was analyzed by Western blotting in the isolated bladder and pancreas of mice, as described previously [19,29]. The primary antibodies employed were an anti-CSE rabbit antibody (sc-135203, Santa Cruz Biotechnol., Santa Cruz, CA, USA) and an anti-GAPDH rabbit antibody (sc-25778, Santa Cruz Biotechnol.). A HRP-conjugated anti-rabbit IgG (Cell Signaling Technol., Beverly, MA, USA) was used as a secondary antibody. Immunopositive bands were developed by Chemi-Lumi One Super (Nacalai Tesque, Kyoto, Japan), and quantified using densitometric software (ImageJ 1.44p, http: //imagej.nih.gov/ij).

Statistics
Data are represented as the mean ± S.E.M. Statistical significance for parametric data were analyzed by an analysis of variance followed by the Tukey's test for multiple comparisons or Student's t-test for twogroup data. For non-parametric analyses, Kruskal-Wallis H-test followed by a least significant difference-type test was employed for multiple comparisons. Significance was set at a level of p < 0.05.

THGP directly interacts with sulfide in vitro
To test the possible direct interaction of THGP, the hydrolysate of Ge-132 (Fig. 1A), with sulfide, we analyzed the structure of THGP alone and the mixture of THGP and NaSH, using the MS/MS system. In MS analysis of THGP alone at 1 mM, the negative ion full-scan mass spectra showed five major signals at m/z 193, 195, 196, 197 and 199, corresponding to the calculated m/z values of THGP containing five different major natural isotopes of germanium, 70 Ge, 72 Ge, 73 Ge, 74 Ge and 76 Ge, respectively (Fig. 1B). The relative abundance of the signal at m/z 197 for 74 Gecontaining THGP (calculated m/z: 196.9505) was the greatest (Fig. 1B), in agreement with the previous report from our group [45]. It is to be noted that additional five minor signals were detected at m/z 207, 209, 210, 211 and 213 (Fig. 1B), as seen in the previous report [45], which might indicate the existence of methylated products of THGP (Fig. 1B) possibly due to the reaction with methanol present in the mobile-phase. Next, in MS analysis of the mixture of THGP at 1 mM and NaSH at 1 mM, the relative abundance of the signal at m/z 195 was the greatest in the negative ion full-scan mass spectra (Fig. 1C), differing from the results of the MS analysis of THGP alone (Fig. 1B), thereby indicating that the increased relative signal at m/z 195 reflects generation of a reactant between THGP and NaSH, in addition to 72 Ge-containing THGP itself (see Fig. 1B (Fig. S1B). Collectively, the mixing of THGP and NaSH appears to generate a chemical in which two hydroxyl groups on a germanium (Ge) atom of THGP are replaced with a sulfur atom (Fig. 1F).
Next, we analyzed the structure of compounds present in the mixture of THGP and NaSH using a 1 H NMR technique. In the NMR spectrum of THGP at 10 mM alone, two proton triplet signals were observed at δ H 2.48 (a) and 1.57 (b), indicating two methylene groups adjoining a carbonyl group and a Ge atom, respectively ( Fig. 1F and G). On the other hand, in the NMR spectrum of the mixture of THGP and NaSH, each at a final concentration of 10 mM, as assessed after 10-min reaction at room temperature, the two proton triplet signals, (a) and (b), of THGP shifted to the higher magnetic fields, i.e. δ H 2.40 (a') and 1.42 (b'), respectively, indicating that a large proportion (83%) of THGP changed to the sulfurcontaining reaction product through the reaction with sulfide (Fig. 1F,  H). It is noteworthy that NMR analysis performed even 5 min after mixing THGP and NaSH provided almost the same results, indicating that the reactions between THGP and NaSH were very fast and quickly reached an equilibrium. We also confirmed such interaction between THGP and Na 2 S, another sulfide compound. THGP at 10 mM was mixed with Na 2 S at 1.25, 2.5, 5 or 10 mM and incubated at room temperature for 10 min. The NMR spectrum of the mixture of THGP and Na 2 S showed that two proton triplet signals, (a) and (b), shifted to the higher magnetic fields, (a') and (b'), respectively, in a manner depending on concentrations of Na 2 S (Fig. 2A). The % peak areas of (a') and (b') signals increased depending on Na 2 S concentrations, and conversely, (a) and (b) signals decreased (Fig. 2B and C). For instance, it is roughly estimated that, when 10 mM THGP is mixed with 5 and 10 mM sulfide, approximately 50% and 95% of THGP change to the sulfur-containing reaction product, respectively ( Fig. 2B and C). We also confirmed that the above- mentioned NMR profiles remained constant even when the incubation time of the mixture of THGP and Na 2 S was prolonged overnight, indicating that the reaction between THGP and Na 2 S is reversible and quickly reaches an equilibrium. Collectively, THGP may be able to trap sulfide that possibly interact with Zn 2+ linked via coordinate bonding to His 191 of Ca v 3.2, which inhibits the channel activity [46].

The reaction sensitivity of THGP to sulfide in the presence of Zn 2+
It is essential to compare the relative affinities of THGP and Zn 2+ to sulfide, if we use THGP to inhibit the interaction of sulfide with Zn 2+ linked to His 191 of Ca v 3.2. Thus, we examined the reaction sensitivity of THGP to Na 2 S in the presence of Zn 2+ at different concentrations. It is to be noted that a Tris-HCl buffer instead of the phosphate buffer was used to avoid the formation of Zn 3 (PO 4 ) 2 precipitation in this experiment. Under this condition, 82-83% of 10 mM THGP, once mixed with 10 mM Na 2 S in the absence of ZnCl 2 , changed to the sulfur-containing reaction product, as estimated from changes in % peak area of triplet signals from (a) and (b) to (a') and (b'), respectively (Fig. 3A, top center, and Fig. 3B and C). This reaction between THGP and sulfide decreased by addition of ZnCl 2 in a concentration-dependent manner (Fig. 3), e.g. ZnCl 2 at 3 and 10 mM decreased the % peak areas of (a') and (b') signals of the 10 mM THGP/10 mM Na 2 S mixture from 82-83% to 46-47% and 0%, respectively, and conversely, (a) and (b) signals increased (Fig. 3).

Inhibitory effect of THGP on the Na 2 S-induced enhancement of Tcurrents in Ca v 3.2-transfected HEK293 cells
The effect of THGP on sulfide-induced enhancement of Ca v 3.2 Tchannel activity was examined in HEK293 cells that stably express human Ca v 3.2 channels, using a whole-cell patch-clamp technique. After the confirmation of stable T-channel-dependent currents (T-currents), the cells were stimulated with Na 2 S at 10 μM almost doubled T-currents, an effect persisting for at least 10 min (Fig. 4A and B), as reported previously [23]. THGP at 1-10 mM prevented the Na 2 S-induced enhancement of T-currents in a concentration-dependent manner (Fig. 4C-F), whereas THGP alone in the same concentration range did not alter the baseline T-currents (Fig. 4C-E, G).

Inhibitory effect of THGP on the Ca v 3.2-dependent pain induced by exogenously applied sulfide in mice
Given the direct interaction between THGP and sulfide in vitro, as shown above, we tested whether THGP could suppress the sulfideinduced pain, in which Ca v 3.2 plays an essential role [18,20,23]. Intraplantar (i.pl.) injection of Na 2 S at 10 pmol/paw caused mechanical allodynia in mice, an effect suppressed by co-injection of THGP at 0.02-2 μg/paw (0.1-10 nmol/paw) in a dose-dependent manner   (150 or 500 μmol/kg, respectively) also reduced the i.pl. Na 2 S-induced allodynia (Fig. 5B).

Inhibitory effect of THGP on Ca v 3.2-dependent visceral pain caused by endogenous sulfide in mice
Endogenous sulfide/H 2 S generated from L-cysteine by the upregulated CSE participates in Ca v 3.2-dependent visceral pain associated with CPA-induced cystitis and cerulein-induced pancreatitis in mice [19,[29][30][31]. A single systemic (i.p.) administration of CPA induced upregulation of CSE protein in the bladder tissue (Fig. 6A), bladder pain-like nociceptive behavior (Fig. 6B), referred hyperalgesia in the skin region between the anus and urethral opening (Fig. 6C), bladder swelling (increased bladder weight) (Fig. 6D) and increased urinary frequency (increased voiding spots) (Fig. 6E), as reported elsewhere [29,30]. THGP at 100 mg/kg as well as TTA-T2, a selective T-channel blocker, at 1 mg/kg, when administered i.p. 3 h after CPA administration, significantly reduced the nociceptive behavior and referred hyperalgesia ( Fig. 6B and C), but not bladder swelling or increased urinary frequency ( Fig. 6D and E). Cerulein at 50 μg/kg, repeatedly administered i.p. at 1-h intervals, 6 times in total, caused upregulation of CSE protein in the pancreatic tissue (Fig. 7A), referred hyperalgesia in the upper abdomen (Fig. 7B), pancreatic swelling (increased pancreatic weight) (Fig. 7C) and increased plasma amylase levels ( Fig. 7D), in agreement with our previous study [19]. THGP at 100 mg/kg as well as TTA-A2 at 1 mg/kg, administered i.p. 5 min after the final injection of cerulein, significantly suppressed the referred hyperalgesia (Fig. 7B), but not pancreatic swelling (Fig. 7C) or increased plasma amylase activity (Fig. 7D). Thus, THGP capable of trapping sulfide and the T-channel inhibitor suppressed visceral pain accompanying cystitis and pancreatitis, without affecting tissue swelling or damage, being consistent with the previous findings that CSE inhibitors prevented visceral pain in the same cystitis and pancreatitis models [19,29].

Discussion
Under physiological conditions, the activity of Ca v 3.2 T-channels is attenuated in part by Zn 2+ linked via coordinate bonding to a nitrogen atom in the imidazole ring of His 191 present in the second extracellular loop of domain I of Ca v 3.2 (Fig. 8A), and the Zn 2+ inhibition of Ca v 3.2 can be reversed by Zn 2+ chelators or other Zn 2+ -sensitive compounds including thiols such as L-cysteine [25,26]. Considering that sulfides, but not polysulfides, easily react with Zn 2+ and produce ZnS [47,48], HS − derived from endogenous H 2 S generated by CSE or other enzymes and from exogenously applied H 2 S donors is considered to interact with Zn 2+ linked by coordinate bonding to His 191 of Ca v 3.2 and cancel the Zn 2+ inhibition of Ca v 3.2 activity (Fig. 8B). The present study demonstrated that THGP, the hydrolysate of Ge-132, directly reacted with HS − and formed a reaction product containing a sulfur atom binding to the germanium atom of THGP (Fig. 1A), thereby interfering the interaction of HS − with Zn 2+ linked via coordinate bonding to His 191 of Ca v 3.2 and consequently inhibiting the sulfide-induced enhancement of Ca v 3.2 activity (Fig. 8C).
The acceleration of Ca v 3.2 activity by sulfide/H 2 S, particularly generated by CSE, is involved in a number of pathological pain including visceral pain accompanying cystitis or pancreatitis [19,[29][30][31], as shown in the present study, and also somatic inflammatory and neuropathic pain [17,27,28]. In this context, THGP might be useful to treat a wide variety of intractable somatic pain, in addition to bladder and pancreatic pain. Our findings from 1 H NMR analysis that the reaction of cerulein at 50 μg/kg at 1-h intervals, 6 times in total. THGP at 100 mg/kg or TTA-A2 at 1 mg/kg was administered i.p. 5 min after the final (6th) cerulein injection, and the referred hyperalgesia was repeatedly evaluated 5.5, 6 and 6.5 h after the onset of repeated cerulein injections (i.e. 0.5, 1 and 1.5 h after the final cerulein injection). Seven hours after the onset of repeated cerulein injections, the blood was withdrawn from the anesthetized mice for the assessment of plasma amylase activity, and the pancreatic tissue was excised from the sacrificed mice afterwards, to perform pancreatic weight measurement and Western blotting. between 10 mM THGP and 10 mM Na 2 S decreased by nearly half in the presence of 3 mM ZnCl 2 (see Fig. 3) suggest that about 3-fold higher concentrations of THGP than Zn 2+ might be required to halve the interaction between Zn 2+ and sulfide. Since serum zinc concentrations are around 15 μM in the mammalian body [49], it is estimated that the effective blood concentrations of THGP to halve the sulfide-Zn 2+ binding are around 45 μM. It has been reported that the serum THGP concentration reaches around 20 μM (4 μg/mL) at a peak time, 3-6 h after oral administration of Ge-132 at 100 mg/kg in rats [45]. Therefore, the estimated effective blood concentration, 45 μM, of THGP is considered achievable after i.p. administration of THGP at 100 mg/kg that inhibited sulfide-dependent somatic and visceral pain in the present study (see Figs. 5-7). We consider that Zn 2+ and THGP would react with sulfide in a competitive manner at least in a solution state, which is also associated with solubility equilibrium of ZnS in a biological fluid. H 2 S is now known as a multifunctional gasotransmitter along with NO and CO, and exhibits a variety of biological activity through its interaction with metals binding to proteins including heme proteins, antioxidant activity, sulfhydration of cysteine residues present in proteins, and others [50]. It is likely that these biological activities of H 2 S is also inhibited by THGP. Sulfide/H 2 S interacts with several metals, such as zinc, iron, copper and nickel [50], whereas the effects of THGP on the interaction of sulfide with metals other than zinc in the mammalian body have yet to be examined.

Conclusions
Our study demonstrates that THGP, a hydrolysate of Ge-132, directly reacts with sulfide, resulting in suppression of Ca v 3.2-dependent pain caused by sulfide applied exogenously or generated endogenously. Therefore, Ge-132 may serve as a medicine for treatment of pathological pain. H 2 S, a toxic gas, at higher concentrations binds to cytochrome c oxidase in the electron transport chain, leading to cell death [50], and needs pharmacological interventions [51]. Such a toxic outcome of endogenous and inhaled H 2 S may also be protected by THGP. Endogenous sulfide/H 2 S appears to play a dual role, i.e. being pro-inflammatory and anti-inflammatory (protective), and long-term complete inhibition of H 2 S-generating enzymes may not necessarily be beneficial, particularly in the cardiovascular system [50,52]. In this context, Ge-132 may be useful to trap the excessive sulfide/H 2 S generated in pathological conditions and maintain appropriate blood sulfide/H 2 S levels.

Declaration of competing interest
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Data availability
Data will be made available on request.