Abstract
Mizoribine (MZ) inhibits the differentiation and proliferation of helper T and B cells after antigen recognition by suppressing the purine biosynthesis pathway and nucleic acid synthesis. MZ has been used in kidney transplantation, but distinct data are unavailable for islet transplantation. The present study investigated the efficacy of MZ for islet xenotransplantation. Immunosuppressive effects of MZ were determined by mixed lymphocyte reaction (MLR) assay in vitro. Toxicities for Wistar rat islets were determined by adenosine triphosphate (ATP) contents of islets during 3-day culture and stimulation index in response to glucose after culture. Immunosuppressive effects in vivo were tested in a Wistar-to-B6 islet xenotransplantation model. MZ was administered continuously for 28 days subcutaneously or intramuscularly. MZ inhibited MLR response by approximately 50% at 0.1 μg/ml. ATP contents decreased with MZ >100 μg/ml, while stimulation index was maintained. Continuous infusion of MZ at 10 mg/kg maintained blood concentrations at 0.13–0.19 μg/ml, while intramuscular injection of MZ at 100 mg/kg/day (peak 520 μg/ml at 1 h postinjection) resulted in below measurable levels (<0.03 μg/ml) within 24 h. Graft survival was significantly prolonged following continuous infusion of 10 mg/kg/day compared to controls (31.0 ± 9.5 vs. 13.2 ± 5.2 days; p = 0.002). Furthermore, animals with intramuscular injection at doses of 3.2, 10, or 100 mg/kg/day showed significantly longer graft survival (20.0 ± 7.5, 22.0 ± 7.31, and 24.5 ± 8.1 days, respectively; p < 0.05 each). Histological examination showed significant suppression of lymphocyte infiltration by MZ administration. MZ showed immunosuppressive effects in an experimental islet xenotransplantation model without adverse effects on endocrine function of islet grafts.
Introduction
Islet transplantation has spread worldwide since the successes of the Edmonton protocol (26,28,30). However, results from long-term follow-up have shown that most recipients revert to some level of insulin use, although islet transplantation can relieve glucose instability and problems with hypoglycemia by improving endogenous insulin secretion (28,31). Further progress in improving long-term results for islet transplantation will thus require immunosuppressive protocols (40,41). Sirolimus and tacrolimus were used in the Edmonton protocol, but have various adverse effects, including islet toxicity and suppression of both islet revascularization and islet regeneration (1,3,19,21,23). Immunosuppressive drugs used for islet transplantation should not only display minimal toxicity to transplanted islets, but also limited deleterious effects on renal function, since progression of renal dysfunction is one of the major complications in type 1 diabetes mellitus (1,3,9,19,21,23,28,30,31).
Mizoribine (MZ) is an imidazole class nucleoside that was isolated from culture medium of the mold Eupencillium brefeldianum M-2166 (15,18). The drug has been found to inhibit both humoral and cellular immunity by suppressing DNA synthesis through the inhibition of purine biosynthesis via antagonistic blocking of inosine monophosphate (IMP) dehydrogenase (IMPDH) (15,18). Micophenolate mofetil (MMF), in the same class of drugs, has been used for immunosuppression following islet transplantation in cases requiring discontinuation of calcineurin inhibitors or sirolimus due to adverse effects (5). However, clinical studies of MMF have shown a high incidence of leucopenia and diarrhea (2,33). In contrast, MZ has fewer adverse effects, such as hematological and gastrointestinal events, in clinical usage (36). MZ has been widely used in human kidney transplantation, particularly in Japan (37). This evidence led us to examine the effects of MZ in islet transplantation.
A shortage of donor organs represents another serious problem in islet transplantation, as each recipient currently needs two to three transplantations under present methods (28,30,31). Use of xenoislet graft is one of the best options to resolve this problem (26). Recently, modification of α-galactosyl antigen expression in donor animals and new immunosuppressive drugs including induction therapy using CD25- and CD154-specific antibodies, FTY720 and everolimus, have been shown to significantly affect some islet xenotransplantation models (10,17). These findings indicate the necessity for new immunosuppressive drugs that are effective in xenotransplantation.
The present study assessed the applicability of MZ in a rat-to-mouse islet xenotransplantation model and found that MZ has potential as an immunosuppressive drug without causing adverse effects on islet endocrine function.
Materials and Methods
Animals
Male C57BL/6 (B6; H-2b) mice and Wistar rats, approximately 8 weeks old, were purchased from Clea Japan (Osaka, Japan). Wistar rats were used as donors and B6 mice as recipients. Diabetes was induced by intraperitoneal injection of streptozotocin (STZ) (Sigma Chemical, St. Louis, MO) at 250 mg/kg body weight, and was diagnosed when nonfasting blood glucose levels increased to >400 mg/dl on two consecutive measurements. This experimental protocol was approved by the Ethics Review Committee for Animal Experimentation at Fukushima Medical University.
Islet Isolation and Transplantation
Islet isolation and transplantation were performed as described previously (7). Briefly, we injected Hank's balanced salt solution (Nissui, Tokyo, Japan) containing 2 mg/ml of collagenase (collagenase S-1; Nitta Gelatin, Osaka, Japan) into the common bile duct. The distended pancreas was removed and incubated at 37°C. Collagenase-digested islets were purified by centrifugation on gradients composed of three different Ficoll (type 400, Sigma) densities (1.120, 1.090, and 1.050). After centrifugation, the distinct layer of islets was collected and washed. Diabetic B6 mice acting as recipients were anesthetized with diethyl ether (Wako Pure Chemicals, Osaka, Japan). The left kidney was exposed through a lumbar incision under sterile conditions, and 350–400 islets were transplanted into the left renal subcapsular space of recipients.
Mixed Lymphocyte Reaction (MLR)
MLR was performed as previously described (38). Briefly, irradiated (2000 rads, cesium source) donor strain (Wistar rat) splenocytes were used as stimulators. A total of 8 × 105 cells isolated from the spleen of the recipient C57BL/6 were used as responders. Naive C57BL/6 splenocytes were used as controls. Responder cells and an equal number of irradiated stimulator cells were cultured with 0, 0.1, 0.5, 1, 2, or 5 μg/ml of MZ in 96-well round-bottom plates for 4 days at 37°C in 5% CO2 and humidified air. All experiments were performed in triplicate. Cells were pulsed with [3H]thymidime for the last 16 h of culture, then harvested on days 3, 4, and 5. Proliferation activity was calculated in triplicate microwells by incorporation of 1 μCi/well of [3H]thymidime for 6 h. Cells were harvested and counted with a Betaplate (Pharmacia LKB Biotechnology, Uppsala, Sweden). Responses are reported as the mean cpm of triplicate measurements.
Measurement of Blood MZ Concentration
B6 mice were used as recipients for measurement of blood MZ concentrations after intramuscular or continuous subcutaneous injection of MZ. Blood samples were collected from the right ventricle of the heart at 1, 2, and 4 h after injection of the drug intramuscularly at doses of 1, 10, or 100 mg/kg/day, or at 3, 7, and 14 days after starting continuous administration of the drug subcutaneously at doses of 1, 10, 20, 30, or 60 mg/kg/day. Blood MZ concentration was measured by high-performance liquid chromatography (12).
Assessment of Islet Graft Function In Vitro
The insulin secretory capacities in response to low (50 mg/dl) and high (300 mg/dl) glucose, as glucose stimulation index, were evaluated using a modification of the method reported by Ricordi et al. after culture for 3 days following isolation with or without MZ (27). Doses of MZ were 0, 0.5, 1.0, 10, or 100 μg/ml. Briefly, after 6 h of culture in RPMI-1640 solution at 37°C under 5% CO2 and 95% air in a humidified atmosphere, five sets of 20 islets of 150–200 μm in diameter were handpicked. These islets were placed in a 12-well transwell microplate (Corning Transwell 3403, pore size 12 μm; Corning, Lowell, MA, USA) with 50 mg/dl of glucose RPMI-1640 and 0.1% fetal calf serum (FCS) at 37°C for 60 min for stabilization. After preincubation, the transwell was placed in the second well cluster containing 3.3 mM glucose RPMI-1640 and 0.1% FCS and incubated for 60 min. The transwell was then placed in the third well cluster containing 300 mg/dl glucose for 60 min. Supernatants of the second and third well clusters were immediately collected and frozen at −20°C until assessment. Insulin content was measured using ultrasensitive rat insulin enzyme-linked immunosorbent assay kits (Morinaga, Kanagawa, Japan). Stimulation index was calculated by dividing insulin secretion at high glucose by that at low glucose.
Evaluation of MZ Toxicity
Islets were cultured for 3 days in the presence of MZ at concentrations of 0, 1.0, 10.0, 100, or 1,000 μl/ml, and adenosine triphosphate (ATP) levels of 10 islet equivalent (IEQ) islets were measured. ATP contents of islets were measured using the bioluminescent enzymatic cycling assay as previously reported (14). Briefly, 33 μl of 0.5 N HClO4 was added to the well containing islets with 300 μl of culture medium, and then this solution was frozen at −80°C until assayed. Next, 20 μl of sample and 80 μl of Tris buffer were mixed, then 100 μl of each bioluminescence reagent [firefly luciferase alone; firefly luciferase with pyruvate kinase (PK); or PK with pyruvate orthophosphate dikinase] was added. Bioluminescence intensity was measured within 90 s using a Lumitester C-100N (Kikkoman, Tokyo, Japan), MLX Microtiter plate luminometer (DYNEX, Chantilly, VA, USA), and Mithras LB940 (Berthold, Bad Wildbad, Germany).
Survival Study of Islet Xenografts
In the Wistar rat to B6 mice combination, recipient mice were divided into groups with MZ (MZ group) or without MZ (control group). In the MZ group, MZ was administered continuously at a dose of 10 mg/kg for 28 days subcutaneously using an osmotic pump (0.2 ml, Alzet 2001; Durect, Cupertino, CA, USA) implanted subcutaneously in the back of the mouse at the time of transplantation or intramuscularly at doses of 1.0, 3.2, 10, or 100 mg/kg/day on the medial aspect of the thigh once daily (29). The pump was exchanged using the same procedure after 14 days and was removed on day 28.
Nonfasting blood glucose levels of each recipient were measured daily to monitor islet graft survival during the first 3 weeks and then every other day. Graft rejection was defined by the day that blood glucose level exceeded 300 mg/dl on 2 successive days.
Histological Examination of Grafted Tissue
Grafts were removed at 7 days after transplantation for histological studies in the control group and continuous infusion of MZ at a dose of 10 mg/kg/day. Specimens containing the transplantation site were fixed in 10% buffered formalin, embedded in paraffin, and cut into 4.5-μm-thick sections. Sections were stained with hematoxylin and eosin (H&E) or stained immunohistologically with rabbit anti-human insulin polyclonal antibody (Santa Cruz Biotechnology, Santa Cruz, CA, USA) and rabbit anti-human glucagon polyclonal antibody (Progen Bitechnik, Heidelberg, Germany). Cross-reac-tivity to mice of these antibodies has already been established by the supplying companies.
To detect CD4 and CD8 cells, cryostat sections were air dried, fixed in acetone for 10 min, and hydrated in Tris-buffed saline (pH 7.4) for 10 min. After fixation in formol calcium solution for 1 min, sections were washed three times with phosphate-buffered saline (PBS) and incubated overnight with biotin-conjugated rat anti-mouse CD8a monoclonal antibody (BD PharMingen, San Diego, CA, USA) or biotin-conjugated rat anti-mouse CD4 monoclonal antibody (BD PharMingen) at 4°C. After washing with PBS, slides were incubated with peroxidase-conjugated streptavidin for 30 min. Slides were washed with PBS and fixed with 1% glutaraldehyde solution for 3 min before staining for peroxidase activity with 3,3′-diaminobenzidine tetrahydrochloride (DAB; Wako Pure Chemicals) in 500 mM Tris-HCl buffer (pH 7.6) containing 0.01% H2O2.
Statistical Analysis
Data are expressed as mean ± SD. Graft survival in different experimental groups was compared using the log-rank test. All statistical calculations were performed using Statview-J5.0 system software (SAS Institute, Cary, NC, USA). Graft survival-prolonging effects were analyzed using the Kaplan-Meier method and log-rank test. Differences between groups in ATP levels and stimulation index (SI) were examined for statistical significance using Student's t-test with Bonferroni correction. All values of p < 0.05 were considered indicative of a significant statistical difference.
Results
In Vitro MLR Assay
MLR assay was used to determine the immunosuppressive dose of MZ in vitro (Fig. 1). Maximum MLR responses were obtained on days 3 and 4 in this xenocombination. The lymphoproliferative reaction was inhibited by MZ in a concentration-dependent manner. The concentration of MZ at which the reaction was inhibited by approximately 50% was 0.1 μg/ml, while concentrations >0.5 μg/ml completely inhibited the reaction.
Mixed lymphocyte reaction (MLR). Responder cells (splenocytes of B6 mice) and an equal number of irradiated stimulator cells (splenocytes of Wistar rats) were cultured with 0, 0.1, 0.5, 1, 2, or 5 μg/ml concentrations of mizoribine (MZ; open square, filled triangle, closed square, filled circle, multiplication sign, and open circle, respectively).
Pharmacokinetic Profile of MZ
MZ concentrations in blood after intramuscular administration at doses of 1, 10, and 100 mg/kg/day were 0.2, 1.5, and 500 μg/ml at 1 h, respectively (n = 3 in each group). In animals given 100 mg/kg/day of MZ, these concentrations rapidly decreased to 2.3 μg/ml after 4 h and were below the measurable level after 24 h. Blood concentrations in animals given intramuscular MZ at 1.0 or 10 mg/kg/day decreased more rapidly below the measurable level within 4 h after administration (Fig. 2A).
Blood MZ concentration after intramuscular (A) or continuous subcutaneous (B) administration. Blood MZ concentrations after intramuscular administration at doses of 1, 10, or 100 mg/kg/day rapidly decreased and were below the measurable level after 24 h (A). Blood concentrations of MZ, after subcutaneous administration by continuous pump infusion at doses of 1, 10, 20, 30, or 60 mg/kg/day, remained at measurable levels until 14 days after starting MZ (B).
In animals continuously infused with 1.0, 10, 20, 30, or 60 mg/kg/day, blood MZ concentrations on day 3 after starting administration were <0.03 (below the measurable limit): 0.17 ± 0.01, 0.20 ± 0.04, 1.00 ± 0.48, and 1.20 ± 0.42 μg/ml, respectively (n = 3 in each group).
These concentrations remained at measurable levels until 14 days after starting MZ, with the exception of 1.0 mg/kg/day. However, mice that received MZ at doses of 20, 30, and 60 mg/kg/day developed signs of severe anorexia, dehydration, and hypoglycemia, and died at 7–13 days after starting MZ (Fig. 2B).
Evaluation of MZ Toxicity on Islet ATP Content
ATP contents of islets cultured for 3 days with 0, 1.0, 10.0, 100, or 1,000 μl/ml of MZ were 25.0 ± 1.2, 28.2 ± 2.6, 22.3 ± 1.3, 15.2 ± 1.5, and 13.8 ± 1.0 pmol/10 IEQ, respectively (Fig. 3A). Groups of islets cultured with 100 or 1,000 μl/ml of MZ showed significantly lower ATP levels than islets without MZ (p < 0.001). These concentrations were thus considered as toxic levels.
Evaluation of MZ toxicity by measuring of ATP content in islets cultured for 3 days (A) and glucose stimulation index at 3 days after isolation (B). Adenosine triphosphate (ATP) content of islets cultured with 100 and 1,000 μl/ml of MZ showed significantly lower levels than islets without MZ (p < 0.001) (A). Data represent mean ± SD. *p < 0.001 between groups. Glucose stimulation index showed no significant difference between control and MZ groups (B).
Assessment of Islet Graft Function In Vitro
Glucose stimulation index after 3 days culture in the presence of MZ at doses of 0, 0.5, 1.0, 10, and 100 μg/ml were 2.3 ± 0.3, 2.4 ± 0.6, 2.4 ± 0.1, 2.7 ± 0.8, and 2.0 ± 0.4, respectively (Fig. 3B). No significant differences were noted between MZ-treated islets compared with untreated islets, indicating no immediate adverse effects of this drug with the doses and duration examined here on insulin secretory capacity in response to glucose.
Nonfasting Blood Glucose Levels and Graft Survival After Islet Transplantation
Nonfasting blood glucose levels in animals given islet xenograft without MZ treatment (Fig. 4A) or animals with MZ treatments, continuously at a dose of 10 mg/kg for 28 days subcutaneously (Fig. 4B) or intramuscularly at doses of 1.0 (Fig. 4C), 3.2 (Fig. 4D), 10 (Fig. 4E), or 100 (Fig. 4F) mg/kg/day were restored to normal within a couple of days following transplantation, indicating no toxic effects of MZ on glucose metabolism. Animals given continuous infusion of 10 mg/kg/day of MZ showed significantly longer graft survival compared to controls (control, 12.0 ± 0.9 days; continuous infusion of 10 mg/kg/day MZ, 31.0 ± 3.4 days; p = 0.002) (Fig. 5B). Furthermore, animals with intramuscular injection at doses of 3.2, 10, and 100 mg/kg/day showed significantly longer graft survival at 20.0 ± 3.4, 22.0 ± 3.1, and 24.5 ± 4.5 days, respectively (p < 0.01) (Fig. 5D-F). However, intramuscular administration at a dose of 1.0 mg/kg/day failed to prolong graft survival (Fig. 5C).
Nonfasting blood glucose levels after islet transplantation. Nonfasting blood glucose levels of mice undergoing islet transplantation with untreated (A: n = 36), treated with continuous infusion of MZ at 10 mg/kg/day (B: n = 7), and treated with an intramuscular once daily dose of 1 mg/kg/day of MZ (C: n = 8), 3.2 mg/kg/day of MZ (D: n = 7), 10 mg/kg/day of MZ (E: n = 6), or 100 mg/kg/day of MZ (F: n = 4).
Graft survival after islet transplantation with or without MZ. Graft survival of animals given continuous infusion of 10 mg/kg/day of MZ was compared with that of control groups (B). Graft survival of animals with intramuscular injection of 1.0 (C), 3.2 (D), 10 (E), or 100 (F) mg/kg/day of MZ was also compared with that of controls, respectively.
Histological Examination
Histologically, marked infiltration of inflammatory cells was observed at transplanted sites of grafts in the control group, whereas relatively fewer inflammatory cells were noted in MZ groups at 7 days after xenotransplantation. The microscopic structure of islets was maintained relatively well in MZ groups (Fig. 6B). Immunohistochemical study using anti-insulin antibody and anti-glucagon antibody revealed that many of the respective positive cells were observed in the MZ group compared with the control group (Fig. 6C-F).
Histological findings 7 days after islet transplantation and immunohistochemical findings using anti-CD4 and anti-CD8 antibodies. Histology of the transplanted site in controls is expressed in upper rows, while findings from the MZ group are expressed in lower rows including H&E staining (A, B), immunohistochemistry with anti-insulin antibody (C, D), and anti-glucagon antibody (E, F) (100x magnification). Histology of the transplanted site in the control group is expressed in upper rows, while findings from the MZ group are expressed in lower rows including immunohistochemistry with anti-CD4 antibody (G, H) and anti-CD8 antibody (I, J) (100x magnification).
Immunostaining of anti-CD4 and anti-CD8 revealed that marked CD4- and CD8+ lymphocyte infiltration around transplanted islets was suppressed in the MZ groups in xenorecipients compared to controls (Fig. 6G–J).
Discussion
MZ is an immunosuppressive drug showing antiproliferative effects by the blockade of IMPDH activity (15,18). This drug is widely used in clinical renal transplantation in Japan with a significant effect in preventing acute rejection and reducing the incidence of adverse effects, including leukopenia and diarrhea (13,36,37). However, MZ has not been applied for islet transplantation experimentally or clinically. To the best of our knowledge, this is the first investigation of the effects of MZ using in vitro assays and an islet xenotransplantation model. Our results from MLR showed that immunosuppressive effects of MZ were evident at concentrations ≥0.1 μl/ml. The immunosuppressive effects of MZ as evaluated by MLR in humans have been reported to be obtained with the same concentration used in our study (34). Furthermore, insulin secretory capacities were not suppressed with these concentrations. However, ATP content in islets significantly decreased with 100 and 1,000 μl/ml of MZ, suggesting some toxic effects at these doses. For this study, we therefore selected a target concentration for MZ of >0.1 μl/ml but <100 μl/ml.
In this rat-to-mouse xenotransplantation model, prolongation of graft survival was achieved by continuous subcutaneous administration of MZ at a dose of 10 mg/kg/day, with blood concentrations of MZ maintained within target levels during treatment. Furthermore, animals with intramuscular injection at doses of 3.2, 10, or 100 mg/kg/day showed significantly longer graft survival than untreated animals. Shimmura et al. reported that prolongation of graft survival in allocombination using a heart transplantation model was achieved when the drug was administered intramuscularly at >80 mg/kg/day (32). Our results indicate that prolongation of graft survival for islet xenotransplantation is achieved using a lower dose, unlike solid organ allotransplantation. Of note, prolongation was possible with both continuous infusion and intramuscular injection with an immediate peak concentration, although toxicity was more profound when given continuously. Murase et al. reported that MZ monotherapy at a dose of 7.5 mg/kg/day in hamster-to-rat heart transplantation had only marginal effects on prolongation of graft survival (22). MZ was administered by gastric instillation in that study, but no data were available for blood concentration. Our results demonstrate that MZ is effective either by continuous infusion or daily intramuscular injection for survival benefit in islet xenotransplantation.
Histological study of the xenografting model using Wistar rats to B6 mice revealed suppression of infiltration by CD4+ and CD8+ cells at transplanted islet sites in the subcapsular space of the kidney. We have previously reported that islet xenografts in the same combination using Wistar-to-B6 were rejected in a CD4-dependent pathway, since administration of anti-CD4 antibody significantly prolonged islet graft survival in this combination (8). MZ reportedly has an immunosuppressive effect in suppressing IL-2 production by CD4+ T cells (35). The CD4 pathway could thus be dependently suppressed by MZ treatment, which might be a leading cause of prolonged survival of islet xenografts.
Histological study revealed that infiltration of CD4+ and CD8+ cells around transplanted islets was suppressed by MZ administration. The mechanism underlying immunosuppression by MZ is the inhibition of DNA synthesis in proliferative lymphocytes following antigen presentation (4,15,18). A similar mechanism has been reported for MMF, which is already used in clinical islet transplantation in combination with other immunosuppressive drugs (5). Recently, details have been reported of the mechanism underlying inhibition of IMPDH activity by MZ and mycophenolic acid (MPA), the active form of MMF (6). MPA inhibits the conversion of IMP to xanthine monophosphate (XMP) by binding to the nicotinamide adenine dinucleotide (NAD)-binding site of IMPDH and trapping E-XMP, an intermediate of IMP conversion to XMP. In contrast, MZ monophosphate inhibits IMPDH activity by binding to the IMP-binding site of IMPDH. Also, the strain of Candida albicans that shows MPA-resistant IMPDH is reportedly sensitive to MZ, since MPA and MZ have different mechanisms of IMPDH inhibition (16). MZ also shows other immunosuppressive activities, including inhibition of B-cell function and antibody production (11,25). Our results thus indicate that MZ could be one candidate for islet xenotransplantation in combination with other immunosuppressive drugs.
Although the Edmonton protocol has been spread worldwide, discontinuation of the immunosuppressive drugs, tacrolimus or rapamycin, represents one of the main problems, since these drugs have many adverse effects, such as β-cell toxicity, aphtha, and renal dysfunction (1,3,19,21,23). In clinical situations, MMF has been used instead of these drugs, but clinical studies of MMF have shown a 12.1–37.8% incidence of leukopenia and a 13.9–38.4% incidence of diarrhea (2,5,33). Reports have described 18.1–36.3% of patients who received organ transplantation under MMF requiring reduction or discontinuation of MMF administration due to adverse events (2,33). MZ might be one of the good candidates for resolving these problems. The low incidence of adverse effects and the relative safety of MZ are evident, because MZ has already been used clinically for kidney transplantation and treatment of relapsing steroid-dependent nephritic syndrome (24,36,37). In addition, high-dose MZ therapy was reported as experimentally effective for ongoing acute humoral rejection (20). In this study, the MZ dose was limited due to toxicities to mice, but human cells are reportedly more resistant than murine cells to both MZ and the aglycone (39). The present study shows that the function of isolated islet cells tolerates high-dose MZ, encouraging us to use higher doses of MZ in large animals (including nonhuman primates) before clinical trials in the future.
In conclusion, administration of MZ for islet xenotransplantation in mice is capable of suppressing rejection without affecting islet function. MZ has potential as an immunosuppressive drug for islet xenotransplantation.
Footnotes
Acknowledgments
This work was supported in part by grants from the Japanese Ministry of Education, Culture, Sports, Science and Technology and in part by Grant-in-Aid for Research on Human Genome, Tissue Engineering Food Biotechnology, Health Sciences Research Grants, Ministry of Health, Labour and Welfare of Japan. The reagents of adenosine nucleotides and luciferase were provided by Kikkoman Corp. The authors declare no conflict of interest.