Databases for facilitating mechanistic investigations of traditional Chinese medicines against COVID-19

There have been debates and investigations on the clinical benefit and adverse effects of the traditional Chinese medicines used for the treatment of COVID-19 [[1], [2], [3]]. The answers to these questions require a more comprehensive mechanistic understanding and clinical evaluation of these traditional medicines. As part of the efforts for probing these questions, the possible mechanisms of these traditional Chinese medicines have been studied based on the experimental and predicted targets of the chemical ingredients, which have been derived from the liquid chromatograph-mass spectrometry [4] and obtained from literature surveys [5,6].

These studies have shown that the knowledge of the chemical ingredients of the traditional Chinese medicines are highly useful for COVID-19 investigations. In particular, the knowledge and clinical profiling of the activities of the chemical ingredients enables the multi-omics studies for finding the clinically-relevant targets [3], and the molecular structures of the chemical ingredients are needed for molecular docking and target binding studies in target assessment [5]. Additionally, the knowledge of the chemical ingredients facilitates the molecular and disease network analysis with respect to the experimental and predicted targets for understanding the network pharmacology [4], and the statistical frequency of appearance analysis of the literature-reported chemical ingredients and mechanisms for focusing on the high confidence mechanisms [6].

The mechanisms of the traditional Chinese medicines are multifaceted in general and for the treatment of COVID-19 in particular [1,2,5,6]. Therefore, more comprehensive investigations are needed for understanding the mechanisms of these traditional Chinese medicines, and for unveiling their benefits and adverse effects. A key step towards such investigations is to obtain the chemical ingredients of these traditional Chinese medicines, particularly the molecular structures and activities. In addition to the literature surveys [5,6], it is of interest to explore the rich resources of the open-access natural product (NP) and traditional medicine databases for the relevant information, particularly the molecular structures of the chemical ingredients and the activity-related information useful in the investigations of the traditional Chinese medicines against COVID-19.

Two open questions can be raised: to what extent these databases provide the molecular structures of the chemical ingredients of the traditional Chinese medicines used for the treatment of COVID-19, and what activity-related information is provided for facilitating the mechanistic investigations. These questions were probed by a database search investigation with respect to 24 traditional Chinese medicines recommended by the National Health Commission of China [1,3,6], using the names of the constituent herbs of each medicine. The searched NP databases include SuperNatural II, UNPD, ZINC, and NPB (http://www.npbdb.net:8080) databases, which provide comprehensive molecular structures, physicochemical properties, vendors, annotations (e.g. literature reported activities and mechanisms), and predicted properties (e.g. toxicity classes) for 325,000, ∼229,000, ∼80,000 and ∼444,669 NPs respectively. The searched traditional Chinese medicine databases include TCM-ID, TCM@Taiwan, TCMID, TCMSP and SymMap with ∼7,400, ∼33,000, ∼18,200, ∼13,000 and ∼19,595 NPs respectively, which provide comprehensive information on the molecular structures of the chemical ingredients, constituent herbs, prescriptions, and activity information (e.g. experimental and predicted targets or activities, NP-target relationships, and the clinical gene expression profiles of the NP targets). Moreover, a database TM-MC (∼24,000 NPs) provides comprehensive data on the traditional medicines from Northeast Asia (Korean, Chinese, and Japanese).

Our search results (Table 1 ) showed that most of the searched NP databases provide insufficient species information for searching the chemical ingredients of the traditional Chinese medicines, with the exception of the NPB database. The NPB database provides species information for 10 disciplines and the molecular structures for ∼444,669 NPs. All major traditional medicine databases support the convenient search and provide molecular structure and activity-related information for high numbers of the chemical ingredients of the 24 traditional Chinese medicines, with the exception of the TCM@Taiwan database that has been inaccessible during our search investigation. Significantly, each of these databases covers the molecular structures of 49-3,869 chemical ingredients for each traditional Chinese medicine (Table 1). In particular, the NPB database provides the molecular structures of the highest numbers of chemical ingredients for 91.7 % of the 24 traditional Chinese medicines.

Table 1.

The traditional Chinese medicines (TCMs) recommended by the National Health Commission of China, the number of chemical ingredients of each TCM searchable from the major databases of traditional medicines and natural products, and the main features of each database useful for mechanistic investigations.

COVID-19 clinical cases		Traditional medicine for COVID-19 treatment	Database
			TCM-ID	TCMID	TCMSP	TM-MC	SYMMAP	NPB
			Database main features useful for COVID-19 investigations
			Integrated TCM data, exp targets, clinical target profile	Integrated TCM data, exp and predicted targets, activities, diseases	TCM systems pharmacology data, exp and predicted targets, target and disease networks	North eastern Asia traditional medicines	Integrated TCM data, symptoms, exp and predicted targets, networks of herbs, symptoms, targets and diseases	Largest natural product database for 10 disciplines, predicted targets or activities
			Number of chemical ingredients in database
Medical observation period
Fatigue with gastro-intestinal discomfort		Huoxiang Zhengqi capsules 藿香正气胶囊	778	907	1045	1350	1745	2,347
Fatigue with fever		Lianhua Qingwen capsules 连花清瘟胶囊	733	790	1312	1605	1627	3084
		Jinhuaqinggan granula 金花清感颗粒	496	819	834	1097	1180	2215
		Shufengjiedu capsules 疏风解毒胶囊	473	571	1004	1322	1041	1762
Clinical treatment period
Mild cases	Hanshi Yufei Formula 寒湿郁肺证处方		743	1055	1256	1724	1413	2042
	Shire Yunfei Formula 湿热蕴肺证处方		668	930	1274	1454	1541	2699
General cases	Shidu Yufei Formula 湿毒郁肺证处方		549	628	1021	1386	1247	2185
	Hanshi Zufei Formula 寒湿阻肺证处方		561	836	840	1071	1066	1170
Severe cases	Huashi Baidu Formula 化湿败毒方		738	863	1213	1425	1481	2572
	Qiying Liangfan Formula气营两燔证处方		463	671	770	961	1050	1688
	Xiyanping Injection 喜炎平注射液		55	76	49	95	89	179
Critical cases	Suhexiang pill 苏合香丸		457	796	693	1027	873	1596
	Angongniuhuang pill 安宫牛黄丸		258	456	505	585	609	1102
	Neibi Waituo Formula 内闭外脱证处方		347	451	463	574	423	627
	Shenfu Injection 参附注射液		76	113	137	174	491	597
	Shengmai Injection 生脉注射液		219	237	200	240	644	590
	Shenmai Injection 参麦注射液		85	62	74	106	249	349
Recovery period	Feipi Qixu Formula 肺脾气虚证处方		555	745	833	1091	1440	1956
	Qiyin Liangxu Formula 气阴两虚证处方		705	855	1064	1358	1502	2570
Mild, general, severe and critical cases	Qingfeipaidu decoction 清肺排毒汤		1001	1242	1850	2221	2104	3869
Severe and critical cases	Xuebijing Injection 血必净注射液		416	622	728	792	1064	868
	Xingnaojing Injection 醒脑静注射液		146	303	328	346	361	451
	Reduning Injection 热毒宁注射液		227	342	421	516	688	1029
	Tanreqing Injection 痰热清注射液		236	275	482	621	523	1201

These databases also provide a variety of the activity-related information useful for mechanistic investigations (Table 1). Specifically, TCM-ID provides experimental targets and the clinical gene expression profiles of the targets. TCMID gives experimental targets and predicted targets (by chemical structure similarity), activity values, and the profiles of the related drugs and diseases. TCMSP includes the experimental targets and predicted targets (by SysDT software), and the derived ingredient-target networks and associated target-disease networks. SymMap gives the experimental targets and predicted targets (by SysDT software), and the ring networks of herbs, TCM symptoms, modern medicine (MM) symptoms, ingredients, targets and diseases. NPB provides the predicted targets or activities (by chemical structure similarity).

Collectively, these traditional medicine and natural product databases are highly useful sources of the chemical ingredient data and activity information for facilitating the mechanistic investigations of the traditional medicines against COVID-19.

Author contributions

Qiuji Cui, Bingwei Ni, Yingying Chen developed NPB database, Sida Jiang and Yin Tan analyzed traditional medicine data, Yu Zong Chen directed the research, Yu Zong Chen and Weiping Chen designed this work and wrote the manuscript.

Declaration of Competing Interest

The authors state no conflict of interest and have received no payment in preparation of this manuscript.