Computational approaches to drug development are rapidly growing in popularity and have been used to produce significant results. Recent developments in information science have expanded databases and chemical informatics knowledge relating to natural products. Natural products have long been well-studied, and a large number of unique structures and remarkable active substances have been reported. Analyzing accumulated natural product knowledge using emerging computational science techniques is expected to yield more new discoveries. In this article, we discuss the current state of natural product research using machine learning. The basic concepts and frameworks of machine learning are summarized. Natural product research that utilizes machine learning is described in terms of the exploration of active compounds, automatic compound design, and application to spectral data. In addition, efforts to develop drugs for intractable diseases will be addressed. Lastly, we discuss key considerations for applying machine learning in this field. This paper aims to promote progress in natural product research by presenting the current state of computational science and chemoinformatics approaches in terms of its applications, strengths, limitations, and implications for the field.

Background and Objective Immune checkpoint inhibitors (ICIs) such as programmed cell death protein 1 (PD-1) and cytotoxic T-lymphocyte antigen 4 (CTLA-4) inhibitors have greatly improved cancer treatment. However, they are associated with immune-related adverse events, including autoimmune diseases (ADs) owing to their immune enhancement effect. As there are few comprehensive studies of ADs by ICIs, it is necessary to analyze the period information of drug-induced ADs. We also assumed that the temporal information may be useful to estimate the similarity of the pathogenic mechanism between spontaneous and ICI-induced ADs. Methods A period analysis including the Weibull analysis was performed on ICI-induced ADs using the Japanese Adverse Drug Event Report (JADER) database. For evaluating the similarity of spontaneous and ICI-induced ADs, a hierarchical cluster analysis was conducted to compare the different onset-time ranges. Results Type 1 diabetes mellitus, autoimmune colitis, and pemphigoid occurred earlier with CTLA-4 inhibitors (median: 46, 29.5 and 28 days, respectively) than with PD-1 inhibitors (> 130 days). Myasthenia gravis had a median time to onset of approximately 1 month, and the risk of onset would increase over time in ipilimumab combination therapy. This result reveals ADs that require attention. Using cluster analysis, we estimated six clusters with different patterns of onset times. Based on these results and a detailed previous research survey, the possible pathogenesis of drug-induced ADs was also discussed. Conclusions This paper describes risk profiles with temporal information of ICI-induced ADs and proposes certain indicators for deciphering the mechanism of AD onset.

Natural products are an excellent source of skeletons for medicinal seeds. Triterpenes and saponins are representative natural products that exhibit anti-herpes simplex virus type 1 (HSV-1) activity. However, there has been a lack of comprehensive information on the anti-HSV-1 activity of triterpenes. Therefore, expanding information on the anti-HSV-1 activity of triterpenes and improving the efficiency of their exploration are urgently required. To improve the efficiency of the development of anti-HSV-1 active compounds, we constructed a predictive model for the anti-HSV-1 activity of triterpenes by using the information obtained from previous studies using machine learning methods. In this study, we constructed a binary classification model (i.e., active or inactive) using a logistic regression algorithm. As a result of the evaluation of predictive model, the accuracy for the test data is 0.79, and the area under the curve (AUC) is 0.86. Additionally, to enrich the information on the anti-HSV-1 activity of triterpenes, a plaque reduction assay was performed on 20 triterpenes. As a result, chikusetsusaponin IVa ( 11 : IC 50 = 13.06 μ M) was found to have potent anti-HSV-1 with three potentially anti-HSV-1 active triterpenes. The assay result was further used for external validation of predictive model. The prediction of the test compounds in the activity test showed a high accuracy (0.83) and AUC (0.81). We also found that this predictive model was found to be able to successfully narrow down the active compounds. This study provides more information on the anti-HSV-1 activity of triterpenes. Moreover, the predictive model can improve the efficiency of the development of active triterpenes by integrating many previous studies to clarify potential relationships.

基礎研究から得られた有望なシーズを迅速に実用化するためには，レギュラトリーサイエンスを推進することが必要である．独立行政法人医薬品医療機器総合機構（PMDA）は，欧米やアジアと連携した国際共同治験の推進，世界に先駆けた革新的医薬品・医療機器・再生医療等製品の実用化促進のための支援，レギュラトリーサイエンス・国際化等の推進を目標（平成26～30年度）に掲げており，レギュラトリーサイエンス研究の実施，レギュラトリーサイエンス研究に精通した人材の育成等を行っている．そこで，レギュラトリーサイエンス研究の実施状況，アカデミア等との交流・連携を通じた人材育成状況など，PMDAにおけるレギュラトリーサイエンス推進の取り組みの現状について紹介し，産官学が連携することの重要性について述べる．

The Organization for Pharmaceutical Safety and Research (OPSR), Pharmaceuticals and Medical Devices Evaluation Center (PMDEC), National Institute of Health Sciences and Ministry of Health, Labour, and Welfare (MHLW) are working together on the new pharmaceutical approval process in Japan for appropriate scientific approaches. The Consultation Division of OPSR helps the pharmaceutical companies to answer find solutions to their concerns about non-clinical and clinical trials in the new pharmaceutical development process. After a new pharmaceutical application, PMDEC evaluates the quality, efficacy, and safety of each pharmaceutical compound in the review process. Final judgment for approval is made by MHLW. Pharmacological mechanisms examined in non-clinical trials would be useful for selection of the appropriate end-point in the clinical trials. For safety assessment, International Conference of Harmonization (ICH) Guideline on Safety Pharmacology Studies was officially notified. According to pharmacological data on safety, clinical investigators can better design clinical trials to prevent patients from suffering serious adverse events. To development new epoch pharmaceuticals, continuous progress on pharmacological research will be indispensable. Further discussion among all interested parties such as pharmacological researchers, medical doctors, companies, and regulatory sites will be necessary and useful to develop more appropriate approaches in the pharmaceutical development process.

Pharmacological profiles of new orally active amide-based tachykinin NK1 receptor antagonists, N-[3,5-bis(trifluoromethyl)benzyl]-5-(4-fluorophenyl)-7,8-dihydro-N,7-dimethyl-8-oxo-1,7-naphthyridine-6-carboxamide (referred to as compound I) and two related compounds (compounds II and III), were compared with that of (+)-(2S,3S)-3-(2-methoxybenzylamino)-2-phenylpiperidine (CP-99,994), another nonpeptide tachykinin NK1 receptor antagonist. Compounds I, II, III and CP-99,994 caused parallel rightward shifts of the concentration–response curve of substance P in the guinea-pig ileum pretreated with atropine, mepyramine and indomethacin, with the pA2 values of 8.70, 7.56, 8.41 and 8.27, respectively. These antagonists did not alter the concentration–response curve of acetylcholine in the guinea-pig ileum nor that of neurokinin A in the rat vas deferens. Furthermore, contractile responses to senktide of the rat portal vein were not affected by these antagonists. In the isolated neonatal gerbil spinal cord pretreated with tetrodotoxin, substance P produced a dose-dependent depolarization of ventral roots. Compounds I, II, III and CP-99,994 caused parallel rightward shifts of the concentration–response curve of substance P in the spinal cord with the pA2 values of 7.07, 5.93, 6.40 and 7.26, respectively. In contrast, these antagonists did not affect the concentration–response curve of l-glutamate. These results suggest that compounds I, II and III are selective antagonists for tachykinin NK1 receptor both in peripheral tissues and the central nervous system.

The possible involvement of enzymatic degradation in the inactivation of enkephalins in the spinal cord of neonatal rats was investigated electrophysiologically and biochemically. In an isolated spinal cord–saphenous nerve preparation, electrical stimulation of the saphenous nerve evoked a slow depolarization lasting 20–30 s of the ipsilateral L3 ventral root. This slow depolarization was depressed by a mixture of peptidase inhibitors, consisting of actinonin (10 μM), thiorphan (0.6 μM), bestatin (10 μM), arphamenine B (10 μM) and captopril (10 μM). Naloxone (0.5 μM) not only reversed this effect of the mixture of peptidase inhibitors but also potentiated the slow depolarization beyond the pre-control level. In an isolated spinal cord preparation, electrical stimulation of a lumbar dorsal root evoked a slow depolarization of the contralateral ventral root of the same segment. This slow depolarization was depressed by application of [Met5]enkephalin in a dose dependent manner. This effect of [Met5]enkephalin was markedly potentiated by addition of the mixture of peptidase inhibitors. Among the five peptidase inhibitors, actinonin, thiorphan or bestatin alone potentiated the depressant effect of [Met5]enkephalin, whereas arphamenine B and captopril did not. Membrane fractions prepared from neonatal rat spinal cords showed degrading activities for [Met5]- and [Leu5]enkephalins and these activities were inhibited by the mixture of peptidase inhibitors. Among the five peptidase inhibitors, actinonin and thiorphan markedly inhibited the [Met5]enkephalin-degrading activity while bestatin was less effective. Arphamenine B and captopril were ineffective. The present results suggest that enzymatic degradation by peptidases plays a role in the termination of the transmitter action of enkephalins in the neonatal rat spinal cord. The present results, together with our previous results on the enzymatic degradation of tachykinins in a study in which we used the same preparations, suggest that similar but distinct combinations of peptidases are involved in the inactivation of enkephalin and tachykinin neurotransmitters.

Hydrogen sulfide (H2S), which is well known as a toxic gas, is produced endogenously in mammalian tissues from L-cysteine mainly by two pyridoxal-5′-phosphate-dependent enzymes, cystathionine β-synthetase and cystathionine γ-lyase. Recently, we showed that cystathionine β-synthetase in the brain produces H2S, and that H2S facilitates the induction of hippocampal long-term potentiation by enhancing NMDA receptor activity. Here we show that mRNA for another H2S producing enzyme, cystathionine γ-lyase, is expressed in the ileum, portal vein, and thoracic aorta. The ileum also expresses cystathionine β-synthetase mRNA. These tissues produce H2S, and this production is blocked by cystathionine β-synthetase and cystathionine γ-lyase specific inhibitors. Although exogenously applied H2S alone relaxed these smooth muscles, much lower concentrations of H2S greatly enhanced the smooth muscle relaxation induced by NO in the thoracic aorta. These observations suggest that the endogenous H2S may regulate smooth muscle tone in synergy with NO.

As a step to clarify the profiles of tachykinin receptors in the mammalian central nervous system, we examined the effects of various tachykinin receptor agonists and antagonists on motoneurons in isolated spinal cord preparations from rats, gerbils and hamsters. After treatment with tetrodotoxin, potential changes were recorded extracellularly from lumbar ventral roots at 27 degrees C. Bath-application of tachykinin NK1, NK3 receptor agonists produced depolarizing responses of ventral roots. In contrast, selective NK2 agonists exerted no or only marginal depolarizing action. Neurokinin A (NKA), however, exerted a distinct depolarizing action on motoneurons. Tachykinin NK1 receptor antagonists antagonized the actions of SPOMe and NKA in a competitive manner. The present results suggest that tachykinin NK1 and NK3 receptors are present on spinal motoneurons of newborn rats, gerbils and hamsters, and that NKA acts on the NK1 receptors.

The longitudinal muscle of guinea-pig ileum was used as a test tissue. Progressive (up to 70 min) treatment with phenoxybenzamine (10(-6) M) inhibited progressively the response to histamine and progressively decreased the density of histamine H1-receptors. However, the 90 min treatment has no further significant inhibitory effect on the contractile response and did not decrease the density of the receptors. These results suggest that phenoxybenzamine discriminates between two distinct populations of H1-receptors. Furthermore, it was found that the presence of GTP gamma-S made some populations of H1-receptors resistant to phenoxybenzamine and facilitated an interaction of mepyramine with H1-receptors.

The pharmacological properties of a newly synthesized 3-acetoxy-6 beta-acetylthio-10-oxo-N-cyclopropylmethyl-dihydronormorphine (KT-95) were examined. This compound, as well as (-)-3-acetyl-6 beta-acetylthio-N-cyclopropylmethylnormorphine (KT-90) and morphine, inhibited the twitch response to electrical stimulation of the guinea-pig ileal preparation that contains mu- and kappa-receptors. The inhibitory effect of KT-95 was about 17 times more potent than morphine, and about 4 times more potent than KT-90. In the guinea-pig ileal preparation, KT-95 behaved as a mu-antagonist against morphine in the presence of norbinaltorphimine (3 x 10(-8) M). In the rabbit vas deferens, containing kappa-opioid receptors, KT-95 inhibited the twitch response to electrical stimulation in a concentration-dependent manner. Norbinaltorphimine concentration-dependently caused parallel rightward shifts of the concentration-response curves to KT-95 in the guinea-pig ileum and in the rabbit vas deferens after electrical stimulation, suggesting that KT-95 behaved as an agonist for the kappa-opioid receptor. In the mouse vas deferens, that contains delta-receptors. KT-95 behaved also as a delta-antagonist against Leu-enkephalin in the presence of norbinaltorphimine. KT-95, KT-90 and morphine were examined for their potencies in displacing the specific binding of [3H]naloxone (mu-selective ligand), [3H]U69593 (kappa-selective ligand), and [3H]D-Ala2-D-Leu5-enkephalin (delta-selective ligand) to synaptosomal fractions from rat brain. Although KT-95 had a higher nonselective affinity to mu-receptors than KT-90 and morphine, the affinity of KT-95 to kappa-receptors was about 18 times higher than that of morphine, and about 5 times higher than that of KT-90. In the acetic acid-induced writhing test, subcutaneously injected KT-95 was more potent than morphine. Furthermore, the analgesic effect, induced by KT-95 (0.062 mumol/kg, s.c.), was abolished by simultaneous administration of norbinaltorphimine (0.020 mumol/mouse, s.c.), suggesting that the analgesic action of KT-95 is mediated through the kappa-opioid receptor. In the pressure test, KT-95 was 20.17 times more potent than morphine. The analgesic action, induced by KT-95 (2.05 mumol/kg, s.c.), was also in this test abolished by simultaneous administration of norbinaltorphimine (0.14 mumol/rat, s.c.), suggesting that this action of KT-95 is mediated through the kappa-opioid receptor. These results indicate that KT-95 behaves as a kappa-agonist with mu- and delta-antagonistic activities, and suggest that analgesia, induced by KT-95, is mainly mediated through kappa-receptors.

Tachykinin NK1 receptor antagonists were used to explore the physiological functions of substance P (SP) and neurokinin A (NKA). Pharmacological profiles of three NK1 receptor antagonists, GR71251, GR82334, and RP 67580, were examined in the isolated spinal cord preparation of the neonatal rat. These tachykinin receptor antagonists exhibited considerable specificities and antagonized the actions of both SP and NKA to induce the depolarization of ventral roots. Electrical stimulation of the saphenous nerve with C-fiber strength evoked a depolarization lasting about 30 s of the ipsilateral L3 ventral root. This response, which is referred to as saphenous-nerve-evoked slow ventral root potential (VRP), was depressed by these NK1 receptor antagonists. In contrast, the saphenous-nerve-evoked slow VRP was potentiated by application of a mixture of peptidase inhibitors, including thiorphan, actinonin, and captopril in the presence of naloxone, but not after further addition of GR71251. Likewise, in the isolated coeliac ganglion of the guinea pig, electrical stimulation of the mesenteric nerves evoked in some ganglionic cells slow excitatory postsynaptic potentials (EPSPs), which were depressed by GR71251 and potentiated by peptidase inhibitors. These results further support the notion that SP and NKA serve as neurotransmitters producing slow EPSPs in the neonatal rat spinal cord and guinea pig prevertebral ganglia.

Pharmacological characteristics of [D-Pro9,[spiro-gamma-lactam]Leu10,Trp11]physalaemin-(1-11) (GR82334), a tachykinin NK1 receptor antagonist, and its effects on slow depolarizing responses of lumbar ventral roots evoked by primary afferent stimulation were examined in isolated spinal cord preparations of neonatal rats. GR82334 (1-3 microM) caused dose-dependent rightward shifts of the concentration-response curves for substance P, substance P methyl ester, delta-aminovaleryl [Pro9,N-Me-Leu10]substance P-(7-11) (GR73632) and neurokinin A in normal artificial cerebrospinal fluid and those for substance P methyl ester, GR73632 and neurokinin A in the presence of tetrodotoxin. GR82334 (10 microM) did not evoke gamma-aminobutyric acid (GABA) release from spinal cords of neonatal rats, whereas [D-Pro9,[spiro-gamma-lactam] Leu10,Trp11]substance P (GR71251), another tachykinin NK1 receptor antagonist, induced a significant increase in GABA release. GR82334 (1-3 microM) markedly depressed the slow depolarizing response of ventral roots, referred to as slow ventral root potential, evoked by the stimulation of the contralateral dorsal root or the ipsilateral saphenous nerve. In contrast, cyclo[Gln,Trp,Phe,Gly,Leu,Met] (L-659,877, 1 microM), a selective tachykinin NK2 receptor antagonist, did not depress the saphenous nerve-evoked slow ventral root potential and did not antagonize the action of neurokinin A to induce ventral root depolarization. The present results provide further evidence for the involvement of substance P, neurokinin A and tachykinin NK1 receptors in the primary afferent-evoked slow ventral root potentials.

The pharmacological characteristics of RP 67580, a non‐peptide tachykinin NK 1 receptor antagonist, and its effects on a reflex response evoked by stimulation of primary afferent fibres, were examined in isolated neonatal spinal cord preparations of the rat. Potentials were recorded extracellularly from a lumbar ventral root and drugs were bath‐applied in normal artificial cerebrospinal fluid (CSF) or in the presence of tetrodotoxin (TTX). In normal artificial CSF, RP 67580 (0.1‐0.3 μ m ) caused rightward shifts of the concentration‐response curves for substance P (SP), neurokinin A (NKA) and substance P methyl ester (SPOMe), an NK 1 selective agonist, with pA 2 values of 7.25, 7.47 and 7.49, respectively. In the presence of TTX (0.3 μ m ), RP 67580 also caused rightward shifts of the concentration‐response curves for SPOMe and NKA. The pA 2 value of RP 67580 against SPOMe (6.75) was significantly lower than that against NKA (7.22). RP 67580 (0.3‐1 μ m ) did not cause a clear parallel shift of the concentration‐reponse curves for SP, and it depressed the depolarizations induced by low concentrations of SP, but slightly potentiated those induced by high concentrations of SP. RP 67580 (1 μ m ) did not depress the depolarizing responses to bombesin, 1 –glutamate, γ‐aminobutyric acid (GABA), thyrotropin‐releasing hormone and muscarine. RP 67580 (1 μ m ), however, depressed the response to acetylcholine in the presence of atropine and the response to nicotine. RP 68651 (1 μ m ), the enantiomer of RP 67580 devoid of activity at tachykinin NK 1 receptors, also depressed the response to acetylcholine in the presence of atropine. RP 67580 (1 μ m ) did not induce GABA release from the rat spinal cord. In the neonatal gerbil spinal cord, the antagonist effects of RP 67580 (0.3‐1 μ m ) against SPOMe were much less potent than in the neonatal rat spinal cord. In the rat spinal cord‐saphenous nerve preparation, electrical stimulation of the saphenous nerve at C‐fibre strength evoked a prolonged depolarization of the ipsilateral L3 ventral root (slow VRP). RP 67580 (0.1‐1 μ m ) depressed the saphenous nerve‐evoked slow VRP. In contrast, RP 68651 (0.3 μ m ) had no effect on the slow VRP . The results of the present study indicate that RP 67580 acts as a high affinity NK 1 receptor antagonist in the neonatal rat spinal cord, although it also possesses an antinicotinic action. This study further suggests the existence of a subpopulation of tachykinin NK 1 receptors that are activated by NKA and SPOMe, as well as by low concentrations of SP, and are sensitive to the antagonist action of RP 67580 in the neonatal rat spinal cord. This study also provides further evidence for the involvement of SP and NKA in the slow VRP evoked by C‐fibre stimulation in the neonatal rat spinal cord.

The possible involvement of enzymatic degradation in the inactivation of tachykinin neurotransmitters was examined in the spinal cord of the neonatal rat. The magnitude of substance P (SP)‐ or neurokinin A (NKA)‐evoked depolarization of a lumbar ventral root in the isolated spinal cord preparation was increased by a mixture of peptidase inhibitors, consisting of actinonin (6 μm), arphamenine B (6 μm), bestatin (10 μm), captopril (10 μm) and thiorphan (0.3 μm). The mixture augmented the response to NKA more markedly than that to SP. In the isolated spinal cord‐cutaneous nerve preparation, the saphenous nerve‐evoked slow depolarization of the L3 ventral root was augmented by the mixture of peptidase inhibitors in the presence of naloxone (0.5 μm) but not in the presence of both naloxone and a tachykinin receptor antagonist, GR71251 (5 μm). Application of capsaicin (0.5 μm) for 6 min to the spinal cord evoked an increase in the release of SP from the spinal cord. The amount of SP released was significantly augmented by the mixture of peptidase inhibitors. Synaptic membrane fractions were prepared from neonatal rat spinal cords. These fractions showed degrading activities for SP and NKA and the activities were inhibited by the mixture of peptidase inhibitors. The degrading activity for NKA was higher than that for SP and the inhibitory effect of the mixture for NKA was more marked than that for SP. Although some other fractions obtained from homogenates of spinal cords showed higher degrading activities for SP, these activities were insensitive to the mixture of peptidase inhibitors. Effects of individual peptidase inhibitors on the enzymatic degradation of SP and NKA by synaptic membrane fractions were examined. Thiorphan, actinonin and captopril inhibited SP degradation, while thiorphan and actinonin, but not captopril, inhibited NKA degradation. The potency of the inhibition of each peptidase inhibitor was lower than that of the mixture. The present results suggest that enzymatic degradation is involved in the inactivation of tachykinin neurotransmitters in the spinal cord of the neonatal rat.