key: cord-0983199-d41zpiyb
authors: Luo, Lu; Yang, Jun; Wang, Cheng; Wu, Jie; Li, Yafang; Zhang, Xu; Li, Hui; Zhang, Hui; Zhou, Yumei; Lu, Aiping; Chen, Shilin
title: Natural products for infectious microbes and diseases: an overview of sources, compounds, and chemical diversities
date: 2021-10-21
journal: Sci China Life Sci
DOI: 10.1007/s11427-020-1959-5
sha: 05be0aef2717155da4e64a8fb8c455a13c61ad76
doc_id: 983199
cord_uid: d41zpiyb

As coronavirus disease 2019 (COVID-19) threatens human health globally, infectious disorders have become one of the most challenging problem for the medical community. Natural products (NP) have been a prolific source of antimicrobial agents with widely divergent structures and a range vast biological activities. A dataset comprising 618 articles, including 646 NP-based compounds from 672 species of natural sources with biological activities against 21 infectious pathogens from five categories, was assembled through manual selection of published articles. These data were used to identify 268 NP-based compounds classified into ten groups, which were used for network pharmacology analysis to capture the most promising lead-compounds such as agelasine D, dicumarol, dihydroartemisinin and pyridomycin. The distribution of maximum Tanimoto scores indicated that compounds which inhibited parasites exhibited low diversity, whereas the chemistries inhibiting bacteria, fungi, and viruses showed more structural diversity. A total of 331 species of medicinal plants with compounds exhibiting antimicrobial activities were selected to classify the family sources. The family Asteraceae possesses various compounds against C. neoformans, the family Anacardiaceae has compounds against Salmonella typhi, the family Cucurbitacea against the human immunodeficiency virus (HIV), and the family Ancistrocladaceae against Plasmodium. This review summarizes currently available data on NP-based antimicrobials against refractory infections to provide information for further discovery of drugs and synthetic strategies for anti-infectious agents. SUPPORTING INFORMATION: The supporting information is available online at 10.1007/s11427-020-1959-5. The supporting materials are published as submitted, without typesetting or editing. The responsibility for scientific accuracy and content remains entirely with the authors.

Infectious diseases are a significant challenge to public health, producing the second highest number of deaths from disease globally. Natural products (NP), which have properties evolutionarily optimized for different biological functions (Atanasov et al., 2015) , have been a prolific source of antimicrobial agents with widely divergent structures and biological activities (Atanasov et al., 2015) . With the advent of combinatorial chemistry and high-throughput screening (HTS) over the last few decades (Atanasov et al., 2015) , there has been a surge in the identification of new molecular architectures and their precise mechanism of action (MOA), including an ever-increasing number of NPs (Teijaro et al., 2018) . Currently, approximately 60% of approved smallmolecule medicines are related to NPs, and 69% of all antibacterial agents are derived from NPs (Zuo et al., 2012) . Antimalarial drugs such as quinine and artemisinin are examples of widely used, effective drugs derived from NPs .

The plant kingdom serves as a unique resource for antimicrobial compounds (Atanasov et al., 2015) . The first written records of plant medicine date from 2,600 BC, in Egypt (Cragg and Newman, 2013) . Similar to traditional Chinese medicine (TCM), which had systematically been applied for over two thousand years (Winder, 1988) , the documentation of the Indian Ayurveda system dates back to the first millennium BC (Patwardhan, 2005) . The herbal medicine of Europe is largely derived from Greek and Roman medicine. The Arabs preserved many Greco-Roman medical books during the Dark and Middle Ages in the fifth to twelfth centuries (Cragg and Newman, 2013) . The search for antimicrobial agents, in particular, has largely been focused on the exploration of NPs, as almost all the antibiotic scaffolds were derived from natural sources (Newman and Cragg, 2016) . It has been estimated that 350 agents, including NPs, semi-synthetic antibiotics, and synthetic chemicals, have so far reached the world market as antimicrobials (Salam and Quave, 2018) . The successful discovery of artemisinin was due to the exploration of antique Chinese herbal medicine books.

The adaptive capability of plant eukaryotic antibiotics is still uncharted territory, although these antibiotics have historically been used against viral, bacterial, fungal, and parasitic infections (Dvorkin-Camiel and Whelan, 2008) . Microorganisms are a prolific source of structurally diverse bioactive medicines. The discovery of penicillin from Penicillium notatum by Alexander Fleming in 1928 marked a significant shift from plants to microorganisms as a source of natural antimicrobial agents (Atanasov et al., 2015) . During the following decades, various bioactive NPs, including quinine, caffeine, nicotine, codeine, atropine, colchicine, cocaine, and capsaicin, were isolated from their natural sources (Atanasov et al., 2015) . The twentieth century witnessed the development of compounds with apparently miraculous properties, mainly in the form of antibiotics (Dvorkin-Camiel and Whelan, 2008) . Many of these drugs were developed from diverse groups of fungi, in which twothirds of antibiotics are produced by members of the bacterial order actinomycetales (Cragg and Newman, 2013) . The marine environment is another rich source of bioactive compounds, many of which belong to totally novel chemical classes, not previously discovered in terrestrial sources (Akram et al., 2018) . During the decade from 1977 to 1987, about 2,500 new metabolites were reported from various marine organisms, and in 2010 alone, 1,003 new compounds were published (Blunt et al., 2012) .

NPs continue to serve as essential sources of chemical entities, supporting drug discovery for infectious diseases. A dataset comprising a relatively complete list of NPs which have been used against five categories of major infectious pathogens has been assembled through a manual selection of published articles (Pubmed, Google Scholar, Web of Science, and CNKI). In this review we use this resource to discover NP-based antimicrobials which may be active against refractory infections, attempt to predict the future for drug discovery in this arena, and endeavor to spark the curiosity of investigators who might enter this fascinating, complex landscape of NPs and their interactions with infection.

When infectious pathogens enter the human body, non-specific and specific immunity protects the body from the attack (Reller et al., 2001) . Phagocytosis, a typical non-specific mechanism of protection, induced by neutrophils and macrophages, occurs. The remaining pathogens are cleared as completely as possible by antibody-mediated humoral immunity and cell-mediated immunity, with the help of other immune molecules (Reller et al., 2001) . Immune reactions fall into two opposite yet cooperative systems, just like the Yin and Yang conception from Daoism in the basic theory of TCM: immune initiation and response (Yang), participated by activators and effectors such as T cells (Th 1 , Th 17 ), B cells, DC, Mφ, NK cells, NKG2D, and IFN-γ; and immune regulation and tolerance (Yin), managed by regulators and controllers such as IL-10, NKG2A, T reg , Th2, B reg , DC reg , and M2. The detailed strategy of the human immune system against pathogens changes with the characteristics of each pathogen, from virus to parasite. The classification of infectious pathologies in conventional medicine is pathogenoriented, and usually includes viruses, bacteria, fungus, parasites, and chlamydia and mycoplasmas. We selected and summarized potential NP candidates for severe and refractory pathologies caused by infectious pathogens from these five categories.

From the perspective of TCM, infectious pathologies occur as a result of macroclimate or physical changes, leading to imbalance of the Yin-Yang of the body, followed by the development of various symptoms. Infections are caused by an external pathogenic attack-pestilent Qi or evil Qi-due to climatic or environmental changes. A weakened immune system results in disease arising from the battle between healthy Qi (immunity) and pathogenic Qi (pathogens) (Luo et al., 2020) . The former, functioning as a defending barrier, basically consists of four layers, from the exterior to the interior: defensive Qi, Qi, nutrient Qi, and Blood. Therefore in TCM, the classification of infections is always symptomoriented. As mentioned in Volume Plain Questions, Chapter Discussion on Acupuncture, Inner Cannon of Yellow Emperor (Huang Di Nei Jing) , "all the five categories of pestilences are contagious in a similar way with similar symptoms" (Yao, 2010) . Here "five categories of pestilences" refer to cold pestilence (with symptoms manifesting cold), warm pestilence (with symptoms manifesting heat), larynx infection, infectious dysentery, and parasite infection, manifested by different syndromes. The reaction between antigen and antibody, or evil Qi and healthy Qi, induced by external attack, happens to correspond well in the two systems of medicines (Ma, 1981) . The discovery of artemisinin is one typical example from the Chinese herbal database, built based on thousands of years of clinical experience ( Figure 1 ).

Both English databases, including PubMed (Canese and Weis, 2013) , Google Scholar (Vine, 2006) , and Web of Science (Analytics, 2017) , and Chinese databases, including CNKI (Xia et al., 2008) and Wanfang (Wang and Shi, 2012) were searched to collect currently available literature. We conducted activity-based selection on NPs against 21 infectious pathogens covering five categories: viruses (hepatitis B virus (HBV), influenza virus (IAV), herpes simplex virus (HSV), human immunodeficiency virus (HIV), severe acute respiratory syndrome coronavirus (SARS-CoV), Middle East respiratory syndrome coronavirus (MERS-CoV), and severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2); bacteria (Salmonella typhi, Shigella, Mycobacterium tuberculosis, methicillin-resistant staphylococcus aureus (MRSA), Streptococcus pneumonia, and Helicobacter pylori); fungi (Cryptococcus neoformans and Malassezia furfur); parasites (Plasmodium, Schistosoma, Trichomonas vaginalis); and other categories (Leptospira, Treponema Pallidum, and Prion). All included articles were classified into five levels of evidence regarding the robustness of data and experiments conducted. Articles containing only in vitro experiments, conducted or primarily screened by molecular docking, were considered Level One evidence; those with in vitro experiments plus animal model testing or the identification of specific molecular interactions were considered Level Two; those involving signal pathways were considered Level Three; those involving examination of molecular effects by gene knockdown were considered Level Four; and those including molecular docking or other experiments to test the direct binding of the target based on pharmacodynamics and mechanism studies, were considered Level Five.

A total of 618 articles were collected. These articles included 646 NP-based compounds, from 672 natural sources, with biological activities against 21 pathogenic organisms, from five different categories. Of the articles, 91.59% (566) reported in vitro experiments, 2.75% (17) reported a combination of in vivo and in vitro research, 2.10% (13) were in vivo studies, 1.29% (8) were in silico studies, and 0.16% (1) was an in vitro study with in silico screening. With respect to level of evidence, 85.11% (526) of the studies available were categorized as providing level one evidence, 11.33% (70) provided level two evidence, 2.75% (17) had level-three evidence, 0.16% (1) had level-four evidence, and 0.64% (4) provided level-five evidence, indicating that most of the current studies into NPs are still at the preliminary stage, and the MOA and specific targets should be a focus of further exploration (Supplementary materials in Supporting Information).

We selected data from some of the most representative pathogens from each category for further discussion.

There are numerous, potentially useful, NPs to be evaluated and exploited for therapeutic applications against different virus families (Kitazato et al., 2007) . Three major mechanisms of inhibition of HBV by compounds from natural sources have been reported: the inhibition of HBV covalently closed circular DNA (cccDNA) (Wang, 2008) , regulation of HBV replication, and inhabitation of HBV replication by modifying JNK/MIAPK signal pathway. Studies have shown that some effective components from medicinal plants can inhibit the replication of HBV DNA as well as cccDNA (Su et al., 2017) . Epigallocatechin gallate, a catechin compound extracted from green tea, has an antiviral effect on the HBV tolerant cell line Hep G2.117, and has an inhibitory effect on pre-core mRNA, cccDNA, and replication intermediate DNA, which are involved in HBV expression (Wang, 2008) .

Treatments for acquired immune deficiency syndrome (AIDS) generally fall into the categories of HIV suppression or HIV reactivation (Cary and Peterlin, 2018) . A promising area of research involves a deeper understanding of the identification of drug targets which prevent HIV transcription is the goal to achieve a functional cure Deeks, 2012) . These HIV reverse transcriptase inhibitors include biflavonoids from Rhus succedanea (Lin et al., 1997) , michellamines from Ancistrocladus korupensis (Boyd et al., 1994) , and lanostane-type triterpenes from Polyalthia suberosa (Li et al., 1993) . Three of the active compounds identified are known to be HIV integrase inhibitors: flavonoid gallate ester from Acer okamotaanum (Kim et al., 1998) , and dicaffeoylquinic acids from Achyrocline satureioides (Zhu et al., 1999) . Some active compounds were found to be HIV protease inhibitors: watersoluble lignins from Inonotus obliquus (Ichimura et al., 1998) , uvaol and ursolic acid from Crataegus pinatifida (Min et al., 1999) , and maslinic acid from Geum japonicum (Xu et al., 1996; Chisembu and Hedimbi, 2009) .

At present, no preventive vaccines or established antiviral therapies are available for coronaviruses (Sohrabi et al., 2020) . However, several synthetic compounds have shown promise, including hydroxychloroquine and choloroquine phosphate (Cortegiani et al., 2020) , which act through several mechanisms, including alkalinization of the host cell phagolysosomes. Despite the short time since the emergence of SARS-CoV-2 several studies have reported on the use of computer modeling for screening for the inhibition of this virus (Liu and Zhou, 2005) . Typically, these models determine the free binding of energy between a ligand and a receptor, with a lower free binding energy indicating a stronger bond between the ligand and the receptor (Lung et al., 2020) . Several researchers have utilized virtual computer docking models to screen for potential compounds that could bind to and inhibit key proteins present in SARS-CoV (Liu and Zhou, 2005) , highlighting the potential antiviral activity of compounds such as sabadinine and aurantiamide acetate. Only a handful of studies have investigated the potential of NPs as therapeutic agents against MERS-CoV (Richardson et al., 2020) . Silvestrol, a phytochemical from Aglaia sp., has been found to be a potent inhibitor of MERS-CoV replication (EC 50 of 1.3 nmol L −1 ) . Some promising compounds for the treatments of coronavirus include scutellarein, Silvestro L, tryptanthrin, saikosaponin B 2 , griffithsin, lycorine, and polyphenolics .

NPs have played a vital role in the development of antimicrobial drugs, such as discovering of rifamycins (1957), quinolone (1962), trimethoprim (1968), oxazolidinone (2000) , and antibacterial lipopeptides (2003), polyketides, non-ribosomal peptides, and aminoglycosides (Butler, 2005; Pham et al., 2019) . Staphylococcus aureus infection is the leading cause of hospital-acquired pneumonia and infection of surgical wounds (Ambé et al., 2015) . MRSA tends to be multi-drug resistant (MDR), exhibiting resistance not only to β-lactam antibiotics but also to a variety of antibiotic classes. Most of the active extracts contained tannins, (poly)phenols (including flavonoids, lignans and coumarins), terpenoids or alkaloids, which have been previously reported to be active compounds against MRSA. The considerable antibacterial activity against MRSA of grape seed extract (Vitis vinifera L; Vitaceae), and immune regulation and tolerance (Yin), managed by regulators and controllers such as IL-10, NKG2A, T reg , Th2, B reg , DC reg , and M2. Right panel, From the perspective of traditional medicine, the classification of infections is always symptom-oriented. The diseases occur due to imbalance between healthy Qi (immunity) and pathogenic Qi (pathogen). The Inner Cannon of Yellow Emperor (Huang Di Nei Jing) classified five categories of pestilences, cold pestilence, warm pestilence, larynx infection, infectious dysentery, and parasite infection, manifested by different syndromes. The recovery from diseases depends on the battle between healthy Qi (immunity) and pathogenic Qi (pathogen). The former, functioning as a defending barrier, basically consists of four layers from the exterior to the interior: defensive Qi, Qi, nutrient Qi, and Blood. which is rich in potent antioxidant polyphenolics, signified a major advancement in the treatment of MRSA diseases (Al-Mousawi et al., 2020) . Aqueous extract of Enantia polycarpa, a tropical plant of the Annonaceae family (Ambé et al., 2015) , had bactericidal activity against 75% of MRSA strains tested. It had minimum inhibitory concentration (MIC) and minimum bactericidal concentration (MBC) values of 3.125 and 12.5 mg mL −1 , respectively. Phytochemical analysis revealed the presence of saponins, alkaloids, and tannins in an aqueous extract (Ambé et al., 2015) .

Many of the antibiotics currently used for the treatment of infections come from fungi. Fungi imperfecti, or Penicillium, and Cephalosporium were the two of the first fungi identified as having antibiotic activity (Dvorkin-Camiel and Whelan, 2008) . There are five groups of antifungal agents approved for human consumption: the azoles, the polyenes, the echinocandins, the allylamines, and the antimetabolites (Howard et al., 2020) .

Cryptococcal meningitis (CM) is a vital mycosis caused by Cryptococcus neoformans and Cryptococcus gattii, which leads to over one million cases and 600,000 deaths per year (Sloan and Parris, 2014; Liu et al., 2012; Park et al., 2009) . Cryptococcosis primarily manifests after C. neoformans enters and colonizes the lungs (Saag et al., 2000; Samie et al., 2019) . The successful colonization of the host by C. neoformans is attributed to its approximately 150 reported virulence factors (Malachowski et al., 2016) . Pelargonium sidoides extract has been found to retain its antifungal activity for up to four years after preparation. The yeast cell wall and membrane are its first line of defense against its extracellular environment (Bahn et al., 2005) . C. neoformans enlarges its capsule through distal growth by releasing and attaching capsular components stored within secreted vesicles to the outer edge of the capsule (Zaragoza et al., 2006) . Preliminary investigation of the antifungal mechanism revealed that the active compounds induced apoptosis of C. neoformans cells and arrested the cell cycle at the G1/S phase. Berberine, a protoberberine-type isoquinoline alkaloid isolated from Berberis aquifolium, Berberis vulgaris, Berberis aristata, and Hydrastis canadensis, has been shown to have broad antibacterial and antifungal activity. Berberine causes damage to the cell membrane, and the possible impairment of cell function. Berberine may affect mitochondrial respiratory function, causing the breakdown of ΔΨm and a lack of accumulation of rhodamine 123 in the mitochondria (Ludovico et al., 2001) .

Schistosomiasis, caused by flatworms of the genus Schisto-soma, is one of the most neglected tropical diseases (Veras et al., 2012) . Schistosoma mansoni is the major etiological agent of human schistosomiasis, and praziquantel (PZQ) is the only drug available to treat this neglected disease. There is therefore an urgent need for new drugs. Recent studies have indicated that extracts from Piper aduncum (Piperaceae) are active against adult worms of S. mansoni. Extracts of cardamonin from P. aduncum caused tegumental alterations, and reduction of oviposition and motor activity in worms of S. mansoni. Another two compounds, luteolin and (3R,6R)-linalool oxide acetate, showed anthelmintic activity against S. mansoni (IC 50 range from 5.8 to 36.9 μg mL −1 ). Luteolin induced tegumental damage in S. mansoni (Wangchuk et al., 2016) . A bioguided phytochemical study identified 2-methoxy-6-pentyl-benzoquinone, also known as Primin, as the major bioactive metabolite producing toxicity against adult S. mansoni worms (Viegas et al., 2017 ). An in vivo study showed that the methanolic fraction of Clerodendrum umbellatum reduced the hepatosplenomegaly induced by infection with S. mansoni, and prevented the elevated malondialdehyde (MDA) level induced by the infection, while producing a significant increase in catalase activity and glutathione levels. Phytochemical analysis of an aqueous fraction from C. umbellatum leaf extract revealed the presence of alkaloids, flavonoids, cardiac glycosides, phenols, saponins, tannins, and terpenoids (Jatsa et al., 2015) . Extract of Ramalina aspera shows a cercaricidal effect on S. mansoni at a concentration of 5.0 μg mL −1 , with effective molluscicidal activity against the embryos and adult snails of the species B. glabrata and cercariae of S. mansoni. It could be a promising molluscicidal and cercaricidal agent (Silva et al., 2019) .

Prions are the infectious agents that cause neurodegenerative diseases, known as prion diseases or transmissible spongiform encephalopathies (TSEs). These diseases include Creutzfeldt-Jakob disease (CJD), kuru, fatal familial insomnia, scrapie, and bovine spongiform encephalopathy (Aguzzi and Liu, 2017) . A variety of anti-prion compounds, including pentosane polysulfate, polyamines, amantadine, astemizole, dextran sulfate, Congo red, suramin, rapamycin, and quinacrine have been reported to bind to the prion protein PrP C , acting as chemical chaperones to reduce the accumulation of PrP Sc in cell culture (Ishibashi et al., 2016; Imberdis et al., 2016) , with specific binding to the hotspot region of PrP C (Ferreira et al., 2014) . However, most of the compounds which were active in vitro have failed in vivo (Li and Weng, 2017) . Two novel and potent NPs, BNP-03 ((1s,3R,4r,5S)-3,5-bis(((E)-3-(3,4-dihydroxyphenyl)acryloyl)oxy)-1,4-dihydroxycyclohexane-1-carboxylic acid) and BNP-08

yphenyl)acryloyl)oxy)-4,5-dihydroxycyclohexane-1-carboxylic acid) selected from self-constructed dabase by Choi et al. (2018) , showed inhibitory effects, reducing PrP Sc signals in a standard scrapie cell assay (Klöhn et al., 2003) . Several naturally occurring polyphenols, phenothiazines, antihistamines, statins, and antimalarial compounds are potent prion inhibitors. Several of the new PrP Sc inhibitors cross the blood-brain barrier, and thus have the potential to be active after TSE infection reaches the brain (Kocisko et al., 2003) . In one study, a total of 500 marine invertebrate extracts were searched for yeast-prion inhibiting extracts. Five compounds exhibiting detectable anti-prion activities were isolated: aplysamine-1, aplysamine-2, purealidine-Q, 3,5dibromoverongiaquinol, and 3,5-dibromoverongiaquinol dimethyl ketal (Jennings et al., 2018) . All active compounds identified in the extract of the sponge Suberea ianthelliformis contained an ethylaminodibromophenyl (EADP) moiety, and were identified as potent inhibitors of yeast prions (Hamann et al., 1993; Tsukamoto et al., 1996) . The EADP structure class may serve as a useful lead for the future development and design of novel and improved anti-prion therapeutics (Tilvi et al., 2004; Ishibashi et al., 1991) . The H 3 receptor, ApoE and AChE have all been identified as potential targets for treating various neurodegenerative diseases (Morisset et al., 2000; Torrent et al., 2015) . From the data collected, several different mechanisms appear to be involved in treating infectious diseases by natural products. Anti-microorganism mechanisms include enzyme inhibition and the disruption of the permeability or integrity of the membranes of microorganisms. The former acts through: (i) inhibition of integrase, reverse transcriptase, or RNA polymerase, such as treatments for HIV and IAV infection Panthong et al., 2015; Huerta-Reyes et al., 2004) , which inhibit nucleic acid (DNA/RNA) replication of viruses in host cells; (ii) inhibition of essential enzymes for viral propagation like ATPase/MERS-3CLpro enzyme/SARS-3CLpro (Yu et al., 2012; Jo et al., 2019; Su et al., 2020) ; (iii) inhibition of the functional enzymes of microbes to reduce their pathogenicity, like GlmU/InhA/iNOS and COX-2 (Han et al., 2019; Hartkoorn et al., 2012; Alves-Silva et al., 2020) . Additionally, NPs could affect the viability or activity of microorganisms (cell wall/cell membrane/mitochondrial membranes) (Silva et al., 2019 Zuzarte et al., 2011 . Secondly, NPs act through the modulation of host cells by activating or inhibiting the inflammation-associated signaling pathways of host cells infected by microorganisms, such as NF-κB/JNK/MAPK/p53, to adjust the potential of enemy-destroying and self-destructive process to a better balance (Zhou et al., 2017; Han and Guo, 2012; Pang et al., 2011) . Thirdly, NPs exert protective effects on cells through antioxidant mechanisms (Alam et al., 2012; Fankem et al., 2019; Reddy et al., 2010) . Lastly, NPs can induce DNA damage (Singh et al., 2018) , interfere with substance meta-bolism (Agarwal et al., 2008) , exert immunomodulation (Basso et al., 2020) or autophagy modulation (Kaneda et al., 1991; Laconi et al., 2014) .

Novel active skeleton compounds, such as Asperflavipine A, have become a topic of interest in NP research. Terpenoids, polyketones, alkaloids, and heterozygous KS-NRPS NPs have attracted considerable interest. Most of these NPs have anti-tumor, antibacterial, anti-inflammatory, and other biological activities. Novel NPs also have great potential because of their unique characteristics . The novel skeletons of heteroterpenes, such as Rhodomyrtusials derived from terpenoid biosynthesis, have shown great potential in studies of antibacterial activities. Due to their novel and diverse structural skeletons, they are widely distributed in microorganisms, including fungi and actinomycetes, sponges, Myrtaceae, Ericaceae, Lamiaceae, and some large fungi such as Ganoderma lucidum (Matsuda and Abe, 2016; Peng and Qiu, 2018; Zhao et al., 2020) .

The hurdle remains in the researches of natural antimicrobial agent discoveries is that only the verification of their activity in vitro is effective, but insufficient research has been carried out in vivo, and the mechanisms of activity are difficult to explain, limiting the practical application of some effective NPs. The anti-microorganism mechanisms of NPs is the result of more than one mode of action, so the specific mechanism of an NP could be explored based on the integrated analysis of its effects on both the microorganism and the host cells, including studies of morphology, molecular mechanisms, and function. This is why the mechanisms of numerous conventional drugs have not been explored in depth. However, given the considerable potential of NPs for anti-microorganism activity, it is important to study their specific mechanisms of action (MOA).

In this section we provide a comprehensive analysis of the currently collected data using three approaches. To analyze the major classification of compounds selected and their chief MOAs, we used network pharmacology. To explore the structural relationships between NPs targeting different infectious pathogens, we calculated the maximum Tanimoto similarity scores between all molecule pairs of selected compounds. We also explored the diversity of their natural sources.

From the collected data, we selected a total of 268 available NP-based compounds which have activity against 19 infectious pathogens, including HBV, IAV, HSV, HIV, SARS-CoV, SARS-CoV-2, MERS-CoV, Salmonella typhi, Shigella, Mycobacterium tuberculosis, MRSA, Helicobacter pylori, Cryptococcus neoformans, Malassezia furfur, Plasmodium, Schistosoma, Trichomonas vaginalis, Leptospira, and prions. Information about the MOAs of the compounds against microorganisms was obtained from the literature, and a network diagram was constructed using Cytoscape 3.8.0 (Shannon et al., 2003) (Figure 2 ). Target genes of compounds with a minimum required interaction score (indicating the prediction of compound targets) greater than 0.7 (high confidence) were obtained from the STITCH website (Szklarczyk et al., 2016) . Functional enrichment analysis of target genes and the graphical display of the results of the analysis were conducted using the ClueGO v2.5.7 plug-in of Cytoscape 3.8.0 software (Bindea et al., 2009) (Figure 3 ).

The 268 NP-based compounds were classified into ten categories: sugars and glycosides, quinonoids, penylpropanoids, flavonoids, terpenoids, steroids, alkaloids, tannins, polyphenols, fatty acids, amino acids, and sulfur compounds. Based on our analysis, the compounds the most credible evidence of activity were agelasine D and dicumarol against Mycobacterium tuberculosis; dihydroartemisinin against Plasmodium; and pyridomycin against Mycobacterium tuberculosis. Other promising interactions, which would repay further investigation, are compounds 1,8-cineole, borneol, camphene, berberine, andα-pinene against Cryptococcus neoformans; artemisinin against Plasmodium; Baicalin, emodin, myricetin, and scutellarein against SARS-CoV; and methyl jasmonate against Trichomonas vaginalis (Figure 2 ).

The possible MOAs of four compounds baicalin, emodin, berberine, and quercetin against the above infectious pathogens were illustrated in detail, as they have shown the most abundant inhibition effects. The most plausible pathways are those related to immunoregulation, anti-inflammatory activities, and modification of the replication of DNA and translation of the proteins of infectious pathogens. For instance, emodin negatively regulates DNA metabolism and silencing by miRNA (Figure 3 ). Emodin can protect infected cells by inhibiting the replication and maturation of the EV71 virus (Zhong et al., 2017) . This observation also suggests that emodin may play a role against pathogens such as HIV and HSV by inhibiting their DNA replication.

These four compounds regulate similar signaling pathways, including (i) inflammation-immune-response-related signaling pathways: after pathogenic microorganisms infect body cells, they induce a series of inflammatory and immune responses to resist the invasion of pathogenic microorganisms (Häcker, 2018) , which helps to regulate the response to infection. For example, emodin can regulate the production of chemokines and TNF-α; berberine can regulate IL-1 β; and quercetin can regulate the production of IL-8 and chemokines. (ii) Apoptosis-related signaling pathways, infec-tion, and the subsequent excessive inflammation and immune response may induce apoptosis and cell damages (Zhang and Wang, 2014) . These four compounds can regulate apoptosis-related signaling pathways, and may reverse cell damage through this common mechanism.

To explore the structural relationships of NPs targeting different infectious pathogens, we calculated the maximum Tanimoto similarity scores between all molecule pairs of the selected compounds. A total of 268 NP-based compounds targeting 19 pathogenic organisms from five categories were collected: virus (HBV, IAV, HSV, HIV, SARS-CoV, SARS-CoV-2, and MERS-CoV); bacteria (Salmonella typhi, Shigella, Mycobacterium tuberculosis, MRSA, and Helicobacter pylori); fungus (Cryptococcus neoformans, and Malassezia furfur); parasites (Plasmodium, Schistosoma, and Trichomonas vaginalis) and others (Leptospira and prions). The molfiles of the compounds in the Canonical SMILES format were downloaded from the ChEMBL online database (Gaulton et al., 2016) . The Tanimoto similarity was calculated using a Perl program rewritten based on MolPrint 2D (Kong et al., 2011) . To examine the relationship between the targeted pathogen types and the chemical diversity we subdivided the dataset into five subgroups (virus, bacterium, fungus, parasite, and others) (Pye et al., 2017) . Figure 4 shows that compounds, especially those inhibiting parasites, had high similarity, with a relatively large "hammerhead" distribution. Compared with the other subgroups, particularly the bacteria, fungi, and viruses, the distribution of the maximum Tanimoto scores is centered at relatively low values, suggesting that the chemistries of the compounds inhibiting these pathogenic organisms show high structural diversity. The targets and mechanisms of inhibition of parasites might be relatively less diversified comparing with those against bacteria and viruses, which contain a variety of targets and use different approaches. However, except for the "others" subgroup, the maximum Tanimoto scores of all infectious pathogens are considered to be high (T>0.4). These results indicate that, in the absence of significant innovation in discovery approaches, there are diminishing returns in terms of the discovery of fundamentally new chemical diversity from continued investigations into the same classes of natural sources. These conclusions agree with those of Pye et al. (2017) . Therefore, more efforts should be paid to exploring novel targets, deciphering mechanisms, and discovering compounds. Recently, there have been some efforts to describe NP chemical space and to use this space as a boundary for designing NP-like synthetic screening libraries (Rosén et al., 2009; Pascolutti et al., 2015) . For a given biosynthetic class of NPs, a huge number Figure 2 Compound-pathogen network of natural products based on a pharmacology network. Compounds are divided into ten groups and pathogens into five groups. Green circles and red rhombuses correspond to compounds and infectious pathogens, respectively. The size of each node is proportional to its degree. The thickness and distance to the red line signifies the credibility of the supporting evidence. , 190 Oxymatrine, 191 Palmatine, 192 Plakinidine A, 193 Pseudopalmatine, 194 Quinine, 195 Roemerine, 196 Siamenol, 197 Stephanine, 198 Tetrahydroberberines, 199 of theoretical molecules can be created from primary building blocks such as amino acids, sugars, acetate, and propionate. Using these diverse and often chiral components, NP libraries should, therefore, exceed synthetic libraries in terms of the structural diversity of chemical scaffolds (Pye et al., 2017) .

Using the most reliable data, we collected information about compounds targeting four pathogenic organisms from four subgroups: C. neoformans (fungus), Salmonella typhi (bacterium), HIV (virus), and Plasmodium (parasite), to explore the diversity of their natural sources. A subset of 331 species of medicinal plants known to produce compounds with biological activities against the above four pathogenic microorganisms was collected from the literature. Data about 129 species producing NPs were collected for C. neoformans, 62 species for HIV, 26 for Plasmodium, and 114 for Salmonella typhi. In order to visualize the lineages selected above, four species trees with the different subgroup of compounds based on the chloroplast genome ( Figure 5 ) were constructed in R 3.4.2 (The R Core Team, 2013) with the "phytools" package, using the function S.PhyloMaker (Qian and Jin, 2016) . The backbone of the tree and nodes was obtained together with the S.PhyloMaker script (Qian and Jin, 2016) . The tree was visualized and annotated in iTOL . NP-based compounds were relatively more abundant in the Asteraceae, Celastraceae, Lamiaceae, and Myrtaceae families for C. neoformans; the Anacardiaceae, Asteraceae, Euphorbiaceae, Lamiaceae, Leguminosae, and Polygonaceae families for Salmonella typhi; the Cucurbitaceae, Euphorbiaceae, Lamiaceae, Rutaceae, and Schisandraceae families for HIV; and the Ancistrocladaceae, Annonaceae, Apocynaceae, Lauraceae, and Zingiberaceae families for Plasmodium ( Figure 5 ). We concluded that, though the discovery of unexplored and unusual sources provides opportunities for finding novel NPs, NP-based compounds from the same family, genera or species may share similar secondary metabolic pathways, mechanisms, and targets, in which specific synthetic pathways could be further explored.

Based on the above analysis, sugars and glycosides, quinonoids, penylpropanoids, flavonoids, terpenoids, steroids, alkaloids, tannins, polyphenols, fatty acids, amino acids, and sulfur compounds from NP sources are the ten most promising compounds against infectious pathogens (Table 1) . Inflammation and immune response-related signaling pathways and apoptosis-related signaling pathways are the two major pathways through which these compounds act against pathogenic microorganisms. The NP library is in need of further exploration, especially for the identification of novel targets and the deciphering of mechanisms, as the structural diversity of chemical scaffolds is of great abundance. The families Asteraceae, Celastraceae, Anacardiaceae, Cucurbitaceae, Ancistrocladaceae and associated genera and species are found to contain abundant compounds which act against specific pathogenic microorganisms. Synthetic versions of these pathways should be further explored, to discover effective compounds. The targets and mechanisms for inhibiting bacteria and viruses are diverse. More efforts should be put into exploring novel targets, deciphering mechanisms of action, and discovering compounds in NP-like synthetic screening libraries.

NP drug discovery is a multidimensional problem, requiring the consideration of several factors for both natural and synthetic compounds, including safety, pharmacokinetics, and efficacy, to be evaluated during the selection of drug candidates (Chi et al., 2006) . NP discovery is achieved using two main approaches. As the conventional "top-down" approach has already been largely abandoned (Jensen et al., 2014) , the advent of latest technologies such as artificial intelligence (AI), "organ-on-chip", and microfluidics technologies (Chi et al., 2006) has bolstered the emergent "bottom-up" approach, offering the capacity to access the unexpressed genetic potential of microorganisms (Bachmann et al., 2014) . Recent advances in analytical and computational techniques have made possible new approaches to the analysis of complex NPs, to design synthetic versions of compounds and to derive new and innovative drugs (Chi et al., 2006) . Metabolic pathway engineering and synthetic biology have transformed NP discovery, production, and engineering (Pye et al., 2017) . The area of NP discovery in silico is becoming increasingly active, mainly due to the reduced time, risks, and resources taken by this approach compared to traditional experimental approaches (Luo et al., 2020) (Figure 6 ). Computational models can be used to identify inadequate knowledge about the targets of NP, undesirable pharmacokinetic expressions upon target interaction, or off-target effects. Therefore, it is necessary to implement in silico methods and powerful data resources to facilitate the design and redesign of NP-like molecules with desired bioactivity, and to predict and validate NP targets. There is an urgent need to find therapeutic NPs when infectious diseases are prevalent (Trosset and Cavé, 2019) . NPs have a large quantity of lead-like molecules, which could be used as scaffolds to expand the chemical library (Gu et al., 2013) . Chemical similarity search depends heavily on the existence of reference compounds for known targets, and the availability of data about existing NPs or preclinical NP candidates (Mujawar et al., 2018) . We collected databases from the literature (Table 2) . Virtual NP databases can be categorized into (i) encyclopedic and general NP databases, (ii) databases enriched with NPs used in traditional medicines, and (iii) specialized databases focused on specific habitats, geographical regions, organisms, biological activities, or even specific NP classes. However, the availability of materials for experimental evaluation produces a bottleneck in NP-based drug discovery, as many of those resources do not offer free downloads (Kinghorn et al., 2019) .

In silico methods have strongly impacted the way in which new targets are identified for old ligands (Keiser et al., 2009; Cameron et al., 2013) , the prediction of side effects (Lounkine et al., 2012) , and the development of anatomical therapeutic indicators of approved ligands (Wu et al., 2013) . Over the past few decades, several strategies have become important, including network-based analysis; molecular docking; ligand-based and structure-based approaches, including chemical structures and reactions; protein structure; protein-protein interactions; signal transduction; genetic interactions; and metabolic networks (Agamah et al., 2020) .

These strategies have been widely used to predict candidate targets and ligand-target interactions. Molecular docking is used to predict the geometry and score the interactions of proteins in complex with small-molecule ligands (Kitchen et al., 2004) . These methods can be used to predict whether a given ligand is potentially able to bind other targets. Ligandbased analysis indicates that similar compounds tend to have similar biological properties, and have been extensively used to analyze and predict the activity of ligands for new targets (March-Vila et al., 2017) . Structure-based analysis is based upon the idea that proteins with similar structures are likely to have the same functions, and to recognize related ligands. In the field of drug repurposing, protein comparison is used as a method to identify secondary targets of an approved ligand (Ehrt et al., 2016) (Figure 6A and B) .

Machine learning (ML) or data mining (DM) approaches have permeated in silico methods, and play critical roles in uncovering significant patterns in chemical and pharmacological property space, which are essential for compound discovery. Advanced machine learning models and algorithms such as support vector machines (Burbidge et al., Workflow of strategies for drug discovery of natural products against infectious diseases. A, The lead-like molecules from NP sources are used as scaffolds to expand the chemical library, by identifying new targets for old ligands and predicting side effects and anatomical therapeutic indicators. Network-based molecular docking, ligand-based and structure-based approaches, together with analysis of chemical structures and reactions, protein structures, protein-protein interactions, signal transduction, genetic interactions, and metabolic networks, have been widely used to predict candidate targets and ligandtarget interactions. B, The "top-down" (bioassay-or chemical signature-guided isolation) vs. the "bottom-up" approaches (genetic information-driven natural product isolation) for natural product diversity. C, Advanced machine learning or DM approaches have contributed to recognizing patterns underlying the relationships between compounds and calculated molecular descriptors or experimental measurements within a large chemo-genomic space that correlates specific activities or classifications for a set of compounds with their features. D, The development of novel targeted therapies by exploiting the polypharmacology of NP with deep learning and multi-task learning. 2001), neural networks (Lo et al., 2018) , logistic regression (Wu and Zhao, 2019) , naive Bayesian classification (Flach and Lachiche, 2004) , binary kernel discrimination (Wu and Zhao, 2019) , partial least squares (Mazandu et al., 2011) and random forests (Lo et al., 2018) , have contributed to recognizing patterns underlying the relationship between compounds and calculated molecular descriptors or experimental measurements within a large chemo-genomic space (Weaver, 2004) . In this space specific activities or classifications for a set of compounds are correlated with their features, thus enabling clustering of similarities among NPlike compounds in multidimensional space (Weaver, 2004 Fang et al. (2017) developed approaches to the development of novel targeted therapies by exploiting the poly-pharmacology of NP. In the new chemical entities (NCE), the initial resource value of substructure associations, drug-target interactions, and drug-substructure associations can be defined as A CS (i, j), A DS (i, j) and A DT (i, j), respectively, The initial resource matrix can be built via these three values for the substructure-drug (or NCE)-target network, and transformed into a final resource matrix by appropriate parameterization. Deep learning and multi-task learning emerge as important in silico methods ( Figure 6C and D). For example, Francesco introduced Deep Docking (DD), a novel deep learning platform suitable for docking billions of molecular structures in a rapid yet accurate fashion (Gentile et al., 2020) . Using a biomedical knowledge graph, a new potential medication, baricitinib, was identified by BenevolentAI for inhibiting SARS-CoV-2. It has shown promising effects on controlling infection and reducing inflammatory injury, and clinical trials have now started.

There are many difficulties, related to the availability of efficient and rapid bioassay systems, in screening and identifying conventional NP-based drugs with auspicious antimicrobial activities (Akram et al., 2018) . The production of NPs requires reliable supplies of the source material, which may vary by seasonal or through environmental change, or may be lost through extinction or legislation (Tanrikulu et al., 2009) . The transition of a natural compound from a "screening hit" through a "drug lead" to a "marketed drug" is associated with demands for increasing amounts of the compound (Atanasov et al., 2015) . The traditional approach for NP discovery once relied heavily upon luck (Jensen et al., 2014) . The complex chemical structures of bioactive NPs hamper the development of methods for total synthesis or derivation that might be needed for the property optimization of lead candidates (Atanasov et al., 2015) . Another major challenge for NP drug discovery is the incompatibility of NPs with HTS (Koehn and Carter, 2005) . Adaption and changes in sample preparation and assay designs are necessary to use HTS for the detection of bioactivity in plant extracts and to identify potent pure compounds (Torres et al., 2017) . A further challenge is to clarify the precise molecular mechanisms and signaling pathways involved in the bioactivity (Corson and Crews, 2007) . Due to high costs, clinical trials of NPs are rarely supported by industry, while the pharmaceutical companies also worry about the patentability of NPs (Atanasov et al., 2015) . These issues have all limited the development of NP medicines.

Antibiotic abuse, especially in developing countries, has resulted in the emergence of resistant microbes due to the evolutionary selection pressure imposed by antibiotics (Gwynn et al., 2010) . Around 90%-95% of Staphylococcus aureus strains worldwide are resistant to penicillin, and 70%-80% of the same strains are methicillin-resistant in many Asian countries (Hemaiswarya et al., 2008) . The number of effective therapeutic measures against life-threatening infectious pathogens has fallen dramatically because of the emerging MDR pathogens (Jensen et al., 2014) . Therefore, it is extremely urgent for the discovery of novel antibiotics to keep pace with the growing threat of drug resistance (Silber et al., 2016) .

The lack of effective therapies and vaccines for various viral infections, and the rapid emergence of new drug-resistant viruses, have produced a growing need for the development of effective new natural-product-based agents to treat viral diseases. New viruses have emerged throughout human history, causing tens of millions of deaths. At present, climate change and globalization have created favorable conditions for the spread of viruses, and in the future, new virus outbreaks may be even more frequent. Therefore, the development of effective antiviral drugs, especially broadspectrum antiviral drugs for conservative targets of different viruses, to fight potential outbreaks of new and re-emerging viruses in the future is a topic of general concern in the academic and industrial communities.

In viruses, the sequences and structures of proteins performing the same functions are often highly similar, such as the spike proteins of SARS-CoV and SARS-CoV-2. These proteins can be used as common targets in the development of antiviral drugs. Among them, representative drugs are Rukobia (fostemsavir, bms-663068), which went on the market in 2020, and the SARS-CoV-2 antibody developed by Eli Lilly and Regeneron Pharmaceuticals, which has been approved for emergency use in the COVID-19 outbreak. However, virus SFTSV is usually under high mutation pressure, so it is difficult to develop broad-spectrum antiviral inhibitors against this target.

At present, only nucleoside analogs targeting viral DNA/ RNA polymerase, such as ribavirin and favipiravir, and compounds targeting host proteins, can be used as broadspectrum antiviral drugs against a large class of viruses. Viruses from the same genus usually contain several highly conserved epitopes, such as TmS and viral protease, which can be further developed as common targets for broadspectrum antiviral drugs. For example, EK1, a virus inhibitor developed for coronavirus before the outbreak of COVID-19, also showed antiviral activity against SARS-CoV-2, indicating the feasibility of designing a broad-spectrum inhibitor for the viruses of the same genus in advance. Therefore, cocktail therapy composed of broad-spectrum viral drugs targeting DNA/RNA polymerase combined with two to three kinds of drugs targeting SP transmembrane subunits or viral proteases can be used as a reserve scheme for the treatment of new viruses from the same genus.

Meanwhile, unlike bacteria, viruses have no characteristics in common, such as cell walls. For example, coronavirus and orthomyxovirus have few common components. Therefore, common targets can only be found for certain families of viruses. The natural immune system of the human body plays an important role in antiviral immunity. Different viral infections lead to changes in the expression of various genes in the host, leading to the host presenting different physiological states. After the virus invades the body, it can induce type I and type III interferons (IFNs) to stimulate the body cells to produce hundreds of interferon stimulated genes (ISGs) to combat the virus. The proteins expressed by ISGs can not only directly inhibit the replication of the virus, but also regulate the expression of IFNs, which play indirect antiviral roles. A large number of studies have reported the antiviral effects and related mechanisms of a variety of ISGS proteins and ISGS family proteins. Different ISGs can inhibit the virus at different stages of infection, from invasion to release. Therefore, an herbal medicine targeting human ISG genes can play a broad-spectrum antiviral role by regulating the release of specific ISG genes.

New NP screening technologies are currently being exploited to improve hit rates for antibacterial discovery. Lead optimization could be enhanced by the identification of new antibiotic classes with improved tractability, and by expanding the predictability of in vitro safety assays. Implementing multiple screening and target identification strategies is recommended for improving the likelihood of discovering new antibacterial compounds that address unmet needs (Jensen et al., 2014) . The intrinsic complexity of NPbased drug discovery necessitates highly integrated interdisciplinary approaches (Atanasov et al., 2015) . NP isolation is laborious and time-consuming, and careful selections of environmental sources of microbes and experimental methods are required to improve the outcomes of isolation procedures (Jensen et al., 2014) . The recent rapid increase in genetic information technology has led to novel screenings, and genetic techniques have facilitated the implementation of combinatorial biosynthetic technology and genome mining. The knowledge gained has allowed large numbers of previously unknown molecules to be identified (Akram et al., 2018) . Omics analysis, including genomics, transcriptomics, proteomics, metabolomics, and metabolomics, results in the generation of complex multivariate datasets that require computational and chemometric tools for interpretation (Tanrikulu et al., 2009 ). Genomics has been applied in the identification of NPs and biomarkers, proteomics has been applied to NP validation and biomarker identification, metabolomics and metabonomics have been used in identification studies, and big data has been applied to NP-based drug development and precision medicine as well as target identification and reverse pharmacology (Prachayasittikul et al., 2015) .

Continued development of technologies such as genome sequencing, bioinformatics, microbial genomics, metabolic pathway engineering, and synthetic biology are poised to usher in a new age of NP discovery and drug development (Jensen et al., 2014) . This is a highly informed approach, in which strains can be prioritized based on a bioinformatic assessment of their biosynthetic potential (Teijaro et al., 2018) . The advent of high-throughput genome sequencing techniques has made accessible increasing numbers of microbial genomes which can be evaluated in silico for their capacity to produce secondary metabolites (Hoffmann et al., 2018) . Advancements in microbial genomics have shown that, for instance, in Actinobacteria, approximately 90% of biosynthetic capacity has not been realized (Teijaro et al., 2018; Harvey et al., 2015) .

The NPs from various sources, including terrestrial and marine microorganisms, fungi, invertebrates, and plants, each possess unique structural features. These features result from the production of NPs via enzymatic reactions catalyzed by proteins programmed by DNA sequences that differ greatly between the source organisms (Naman et al., 2017) . Taxonomic distribution analysis of candidate signals can be especially useful to support the discovery of new NPs (Hoffmann et al., 2018) . A study conducted by Pye et al. (2017) demonstrated that most NPs share structural similarity to previously published compounds. Combining the chemotaxonomic method with molecular phylogenetic data would help to select natural species from genera or families known to produce compounds or compound classes associated with certain bioactivities or therapeutic effects in a targeted manner (Ramos Barbosa et al., 2012) . Phylogenetic and phytochemical studies have shown that there is a strong phylogenetic signal in the distribution of secondary metabolites in NPs (Saslis-Lagoudakis et al., 2011; Saslis-Lagoudakis et al., 2012) . In particular, the exploration of crosscultural ethnomedical patterns within a phylogenetic framework is regarded as a very powerful tool for the identification of highly promising plant groups, when phylogenetically related plant species from very distant regions are found to be used for medical conditions with the same therapeutic aims (Saslis-Lagoudakis et al., 2011; Saslis-Lagoudakis et al., 2012; Atanasov et al., 2015) .

Another strategy employed to overcome mechanisms of resistance is the use of a combination of drugs producing synergistic activity against microorganisms (Hemaiswarya et al., 2008) . Antibiotics generally target the microbial cell wall, but using the approaches described in this review, new therapeutic targets are being identified. The auto-inducer molecules produced by quorum sensing systems are currently targeted by novel compounds to control antibiotic resistance and biofilm formation, quorum sensing, and quorum quenching, and are clearly important as a potential target for natural substances (Mulat et al., 2019) . The future development of new antimicrobial agents will rest with those who employ synthetic and semi-synthetic methodology and a further understanding of microbial cell architecture and drug resistance mechanisms (Thomford et al., 2018) .

Although this review presents a thorough evaluation of publicly available data, there may be some NP-derived compounds that have been overlooked. Emerging technologies, such as genomics and synthetic biology, are enabling new ways of discovering and utilizing NP-based products. We are entering an exciting era in which the ancient wisdom distilled into the world's traditional herbal medicines can be reinterpreted and exploited through the lens of modern science, demystifying conventional medicines with modern approaches .

Relevant information about 268 NP-based compounds classified into ten groups against 19 selected infectious pathogens was collected for network analysis. The distribution of maximum Tanimoto scores indicated that compounds which inhibit parasites exhibited a low diversity, whereas the chemistries inhibiting bacteria, fungi, and viruses showed more structural diversity. There is a lack of innovation in the methods for discovery of NPs, and more efforts should be paid to exploring novel targets, deciphering mechanisms of action, and discovering bioactive compounds. A total of 331 species of medicinal plants with compounds exhibiting pharmacological activities against four specific infectious pathogens were selected to classify the family sources. NP libraries offer a diversity of chemical components for the discovery of drugs against infectious pathogens. This review summarizes the currently available data on NP-based antimicrobials against refractory infections, to provide information for further discovery and synthesis of drugs acting against infectious pathogens.

The author(s) declare that they have no conflict of interest.

Computational/in silico methods in drug target and lead prediction

Role of heme in the antifungal activity of the azaoxoaporphine alkaloid sampangine

A role for astroglia in prion diseases

Antiviral potential of medicinal plants against HIV, HSV, influenza, hepatitis, and coxsackievirus: A systematic review

Methanolic extracts of Withania somnifera leaves, fruits and roots possess antioxidant properties and antibacterial activities

Effect of black grape seed extract (Vitis vinifera) on biofilm formation of methicillin-resistant Staphylococcus aureus and Staphylococcus haemolyticus

Chemical characterization and bioactivity of the essential oil from Santolina insularis, a Sardinian endemism

Diversity of medicinal plants used in the traditional treatment of the diarrhea on the markets of Abidjan

Web of Science. Trust the Difference. Web of Science Fact Book

Reproducible molecular networking of untargeted mass spectrometry data using GNPS

Discovery and resupply of pharmacologically active plant-derived natural products: a review

Microbial genome mining for accelerated natural products discovery: is a renaissance in the making?

Specialization of the HOG pathway and its impact on differentiation and virulence of Cryptococcus neoformans

Immunomodulatory activity of βglucan-containing exopolysaccharides from Auricularia auricular in phagocytes and mice infected with Cryptococcus neoformans

The ChEMBL bioactivity database: an update

ClueGO: a Cytoscape plug-in to decipher functionally grouped gene ontology and pathway annotation networks

Marine natural products

Anti-HIV michellamines from Ancistrocladus korupensis

A standard database for drug repositioning

Drug design by machine learning: support vector machines for pharmaceutical data analysis

Natural products to drugs: natural product derived compounds in clinical trials

Chemical informatics uncovers a new role for moexipril as a novel inhibitor of cAMP phosphodiesterase-4 (PDE4)

PubMed: the bibliographic database

Natural products and HIV/AIDS

iSMART: an integrated cloud computing web server for traditional Chinese medicine for online virtual screening, de novo evolution and drug design

TCM Database@Taiwan: the world's largest traditional Chinese medicine database for drug screening in silico

TTD: therapeutic target database

Highlights on the progress of traditional Chinese medicine and natural drugs during

Natural products in drug discovery and development

A survey of plants with anti-HIV active compounds and their modes of action

Application of the fragment molecular orbital method to discover novel natural products for prion disease

The Drug Discovery Portal: a resource to enhance drug discovery from academia

UniProt: a hub for protein information

The Drug Repurposing Hub: a next-generation drug library and information resource

Molecular understanding and modern application of traditional medicines: triumphs and trials

A systematic review on the efficacy and safety of chloroquine for the treatment of COVID-19

Natural products: A continuing source of novel drug leads

The Comparative Toxicogenomics Database: update 2017

HIV: shock and kill

Tropical American plants in the treatment of infectious diseases

Impact of binding site comparisons on medicinal chemistry and rational molecular design

Quantitative and systems pharmacology. 1. In silico prediction of drugtarget interactions of natural products enables new targeted cancer therapy

Antioxidant activity of dichloromethane fraction of Dichrocephala integrifolia in Salmonella typhi-infected rats

Anti-prion activity of a panel of aromatic chemical compounds: in vitro and in silico approaches

Naive Bayesian classification of structured data

The ChEMBL database in 2017

Deep Docking: a deep learning platform for augmentation of structure based drug discovery

Use of natural products as chemical library for drug discovery and network pharmacology

Challenges of antibacterial discovery revisited

Biogenetically diverse, bioactive constituents of a sponge, order Verongida: bromotyramines and sesquiterpene-shikimate derived metabolites

Antibacterial and anti-inflammatory activity of traditional Chinese herb pairs, Angelica sinensis and Sophora flavescens

Action of dicumarol on glucosamine-1-phosphate acetyltransferase of GlmU and Mycobacterium tuberculosis

Towards a new tuberculosis drug: pyridomycin-nature's isoniazid

The reemergence of natural products for drug discovery in the genomics era

Traditional Chinese medicine database and application on the web

SuperTarget goes quantitative: update on drug-target interactions

Synergism between natural products and antibiotics against infectious diseases

Systematic integration of biomedical knowledge prioritizes drugs for repurposing

The Transformer database: biotransformation of xenobiotics

Correlating chemical diversity with taxonomic distance for discovery of natural products in myxobacteria

A comprehensive overview of the medicinal chemistry of antifungal drugs: perspectives and promise

The NCGC pharmaceutical collection: a comprehensive resource of clinically approved drugs enabling repurposing and chemical genomics

HIV-1 inhibitory compounds from Calophyllum brasiliense leaves

Inhibition of HIV-1 protease by water-soluble lignin-like substance from an edible mushroom, Fuscoporia obliqua

Identification of anti-prion compounds using a novel cellular assay

Structure-based drug discovery for prion disease using a novel binding simulation

Purealidin A, a new cytotoxic bromotyrosine-derived alkaloid from the Okinawan marine spongePsammaplysilla purea

Schistosomicidal, hepatoprotective and antioxidant activities of the methanolic fraction from Clerodendrum umbellatum Poir leaves aqueous extract in Schistosoma mansoni infection in mice

Yeastbased screening of natural product extracts results in the identification of prion inhibitors from a marine sponge

Challenges and triumphs to genomics-based natural product discovery

Characteristics of flavonoids as potent MERS-CoV 3C-like protease inhibitors

In vitro effects of berberine sulphate on the growth and structure of Entamoeba histolytica, Giardia lamblia and Trichomonas vaginalis

Progress in the Chemistry of Organic Natural Products-Cheminformatics in Natural Product-Based Drug Discovery

KEGG: Kyoto Encyclopedia of Genes and Genomes

The Library of Integrated Network-based Cellular Signatures NIH Program: system-level cataloging of human cells response to perturbations

Predicting new molecular targets for known drugs

A new flavonol glycoside gallate ester from Acer okamotoanum and its inhibitory activity against human immunodeficiency virus-1 (HIV-1) integrase

PubChem substance and compound databases

Predicting drug metabolism: experiment and/or computation?

Viral infectious disease and natural products with antiviral activity

Docking and scoring in virtual screening for drug discovery: methods and applications

A quantitative, highly sensitive cell-based infectivity assay for mouse scrapie prions

New inhibitors of scrapie-associated prion protein formation in a library of 2,000 drugs and natural products

The evolving role of natural products in drug discovery

Historical variation of structural novelty in a natural product library

The SIDER database of drugs and side effects

Autophagy activation and antiviral activity by a licorice triterpene

DrugBank 4.0: shedding new light on drug metabolism

Interactive tree of life (iTOL) v3: an online tool for the display and annotation of phylogenetic and other trees

YaTCM: Yet another Traditional Chinese Medicine Database for Drug Discovery

Demystifying traditional herbal medicine with modern approach

Anti-AIDS Agents, 9. Suberosol, a New C 31 lanostane-type triterpene and anti-HIV principle from Polyalthia suberosa

Discovery of small molecules binding to the normal conformation of prion by combining virtual screening and multiple biological activity evaluation methods

In vitro anti-HIV activity of biflavonoids isolated from Rhus succedanea and Garcinia multiflora

SARS-CoV protease inhibitors design using virtual screening method from natural products libraries

Molecular mechanisms of cryptococcal meningitis

Machine learning in chemoinformatics and drug discovery

Large-scale prediction and testing of drug activity on side-effect targets

Assessment of mitochondrial membrane potential in yeast cell populations by flow cytometry

The potential chemical structure of anti-SARS-CoV-2 RNA-dependent RNA polymerase

Analysis on herbal medicines utilized for treatment of COVID-19

Conception on infectious diseases from the perspective of traditional medicine

Systemic approach to virulence gene network analysis for gaining new insight into cryptococcal virulence

On the integration of in silico drug design methods for drug repurposing

Biosynthesis of fungal meroterpenoids

Contribution of microarray data to the advancement of knowledge on the Mycobacterium tuberculosis interactome: use of the random partial least squares approach

Inhibitory effect of triterpenes from Crataegus pinatifida on HIV-I protease

High constitutive activity of native H3 receptors regulates histamine neurons in brain

Paradigm shift in drug repurposing from phenalenone to phenaleno-furanone to combat multidrug resistant Salmonella enterica serovar typhi

Medicinal plant compounds for combating the multi-drug resistant pathogenic bacteria: a review

Broad-spectrum antiviral activity of the eIF4A inhibitor silvestrol against corona-and picornaviruses

KNApSAcK-3D: a three-dimensional structure database of plant metabolites

Chapter 5. Modern natural products drug discovery and its relevance to biodiversity conservation

Natural products as sources of new drugs from 1981 to 2014

Pharos: collating protein information to shed light on the druggable genome

AfroDb: a select highly potent and diverse natural product library from African medicinal plants

Ethanol extract from ampelopsis sinica root exerts anti-hepatitis B virus activity via inhibition of p53 pathway in vitro

Anti-HIV-1 integrase activity and molecular docking of compounds from Albizia procera bark

Estimation of the current global burden of cryptococcal meningitis among persons living with HIV/AIDS

Capturing nature's diversity

Ethnopharmacology and drug discovery

Clinical practice guidelines for rare diseases: the orphanet database

Meroterpenoids from Ganoderma species: a review of last five years

Pharmacophore-driven identification of PPARγ agonists from natural sources

A review of the microbial production of bioactive natural products and biologics

Retrospective analysis of natural products provides insights for future discovery trends

An updated megaphylogeny of plants, a tool for generating plant phylogenies and an analysis of phylogenetic community structure

Selecting medicinal plants for development of phytomedicine and use in primary health care

A comparative in vitro study on antifungal and antioxidant activities of Nervilia aragoana and Atlantia monophylla

Infectious disease pathology

Baricitinib as potential treatment for 2019-nCoV acute respiratory disease

Glimmers in illuminating the druggable genome

Novel chemical space exploration via natural products

TCMSP: a database of systems pharmacology for drug discovery from herbal medicines

Practice guidelines for the management of cryptococcal disease

Opportunities for plant natural products in infection control

The antifungal and Cryptococcus neoformans virulence attenuating activity of Pelargonium sidoides extracts

The use of phylogeny to interpret cross-cultural patterns in plant use and guide medicinal plant discovery: an example from Pterocarpus (Leguminosae)

Phylogenies reveal predictive power of traditional medicine in bioprospecting

Applications of integrated data mining methods to exploring natural product space for acetylcholinesterase inhibitors

Cytoscape: a software environment for integrated models of biomolecular interaction networks

Virtual screening on natural products for discovering active compounds and target information

From discovery to production: biotechnology of marine fungi for the production of new antibiotics

Toxicological effects of Ramalina aspera (lichen) on Biomphalaria glabrata snails and Schistosoma mansoni cercariae

Hydroxychavicol, a key ingredient of Piper betle induces bacterial cell death by DNA damage and inhibition of cell division

Cryptococcal meningitis: epidemiology and therapeutic options

World Health Organization declares global emergency: a review of the

Cnidimonins A-C, Three types of hybrid dimer from Cnidium monnieri: structural elucidation and semisynthesis

Anti-SARS-CoV-2 activities in vitro of Shuanghuanglian preparations and bioactive ingredients

STITCH 5: augmenting protein-chemical interaction networks with tissue and affinity data

Structure-based pharmacophore screening for Natural-Product-Derived PPARγ agonists

Challenges and opportunities for natural product discovery, production, and engineering in native producers versus heterologous hosts

R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing

Natural products for drug discovery in the 21st century: innovations for novel drug discovery

New bromotyrosine alkaloids from the marine sponge Psammaplysilla purpurea

Interaction of prion protein with acetylcholinesterase: potential pathobiological implications in prion diseases

Antibacterial synergism of extracts from climbers belonging to Bignoniaceae family and commercial antibiotics against multi-resistant bacteria

Target Identification and Validation in Drug Discovery

iScreen: world's first cloud-computing web server for virtual screening and de novo drug design based on TCM database@Taiwan

Ceratinamine: an unprecedented antifouling cyanoformamide from the marine sponge Pseudoceratina purpurea

Development of a natural products database from the biodiversity of Brazil

Activity of epiisopiloturine against Schistosoma mansoni

In vitro schistosomicidal activity of the crude extract, fractions and Primin, the major active benzoquinone constituent from the leaves of Miconia willdenowii (Melastomaceae)

Google scholar

DGIdb 2.0: mining clinically relevant drug-gene interactions

New target and development direction of chronic hepatitis B treatment (in Chinese)

Spirostaphylotrichin X from a marine-derived fungus as an anti-influenza agent targeting RNA polymerase PB2

Comparison among the search platforms of Wanfang Database, CNKI database and VIP database

Compounds derived from the bhutanese daisy, ajania nubigena, demonstrate dual anthelmintic activity against schistosoma mansoni and trichuris muris

Applying data mining techniques to library design, lead generation and lead optimization

Medicine in China. A history of pharmaceutics. Berkeley, Los Angeles and London

T3DB: the toxic exposome database

Machine learning technology in the application of genome analysis: A systematic review

Relating anatomical therapeutic indications by the ensemble similarity of drug sets

Five large Chinese biomedical bibliographic databases: accessibility and coverage

Anti-HIV triterpene acids from Geum japonicum

TCMID: Traditional Chinese Medicine integrative database for herb molecular mechanism analysis

Identification of myricetin and scutellarein as novel chemical inhibitors of the SARS coronavirus helicase, nsP13

Inner Cannon of Yellow Emperor (Huang Di Nei Jing) (in Chinese)

The polysaccharide capsule of the pathogenic fungus Cryptococcus neoformans enlarges by distal growth and is rearranged during budding

Npass: natural product activity and species source database for natural product research, discovery and tool development

Inflammatory response of macrophages in infection

Rhodomyrtus tomentosa (Aiton.): A review of phytochemistry, pharmacology and industrial applications research progress

Rheum emodin inhibits enterovirus 71 viral replication and affects the host cell cycle environment

Saxifragifolin D attenuates phagosome maturation arrest in Mycobacterium tuberculosisinfected macrophages via an AMPK and VPS34-dependent pathway

Irreversible inhibition of human immunodeficiency virus type 1 integrase by dicaffeoylquinic acids

Evaluation of traditional Chinese medicinal plants for anti-MRSA activity with reference to the treatment record of infectious diseases

Chemical composition and antifungal activity of the essential oils of Lavandula viridis L'Hér

 Jiqing Zhang, Qin Huang, Qianwen Liu, Dr.Naishi Li, Prof.Haoyu Hu, Dr. Jimeng Zhao, and Dr. Lei Fan for their contributions to this article.

The supporting information is available online at https://doi.org/10.1007/s11427-020-1959-5. The supporting materials are published as submitted, without typesetting or editing. The responsibility for scientific accuracy and content remains entirely with the authors.