Pharmacokinetic and pharmacodynamic herb-drug interactions—part I. Herbal medicines of the central nervous system

Introduction

Medicinal plants and herbal drugs were undoubtedly among the first successful treatment modality ever used by humanity—herbal drugs are currently defined as ‘mainly whole, fragmented or broken plants or parts of plants in an unprocessed state, usually in dried form but sometimes fresh’, (European Pharmacopoeia, 2023h). Even today, the majority of the world’s population still relies on traditional herbal medicines in the broadest sense, and a large part of currently clinically used drugs is of a natural origin or has been derived from it—this fact is also being systematically followed and highlighted, e.g., in a continuing series of reviews by Newman and Cragg, at least since 1997 (Newman & Cragg, 2020).

The focus on the issue of herb–drug interactions has been initiated several decades ago by a case report concerning Citrus (grapefruit) juice interactions with cardiovascular drugs (that shall be elaborated in the contemporaneously prepared Part II of this review series). Since that, many other herb-drug interaction have been revealed or suggested. Studying herb-drug interactions is complicated by the fact, that there are not only herbal medicines but also herbal food supplements on the world market (N. B.: food supplements—EU terminology, are legally called dietary supplements in the USA) (Czigle et al., 2018). For many popular medicinal plant extracts, the same particular (standardized) extract can be found in medicinal and food products at the same time on the same local market (Czigle et al., 2018). Even that food products should be marketed with lower doses of these extracts, as they must not be intended to treat human diseases. Regardless, even if the ingredients of supplements are labelled as part of the traditional human diet by manufacturers, and their effects claimed physiological, i.e., maintaining human health, such herbal materials (herbal drugs, botanicals, and extracts) certainly contain the same biologically active constituents, no matter if they are part of medicinal or food products. These active compounds can interact with other medications, herbal or synthetic, at different levels (Cordier & Steenkamp, 2011; Singh & Zhao, 2017).

In contrast to well-investigated novel drugs which must undergo long-term preclinical and clinical investigations with the aim of defining indications, mechanism of action, as well as their undesirable effects, contraindications, and possible interactions, many phytopharmaceuticals profit from their traditional status (European Medicines Agency, 2023). This means that detailed information including data from clinical studies is mostly missing. A quite dangerous subconscious assumption has evolved, that all what comes from nature is safe (Ng, 2021). The need for new data on herbal efficacy and safety has been highlighted by recent systematic regulatory approaches in the herbal medicines segment, most prominently represented by the works of the European Medicines Agency—Committee on Herbal Medicinal Products (EMA-HMPC). European Union herbal monographs typically also contain chapters on contraindications, undesirable effects, interactions with other medicinal products, warnings, etc. (European Medicines Agency, 2023). Contrary to indications and mechanisms of action, information on less frequent undesirable effects and interactions with other medicinal products is often gained in the clinical trials phase IV, i.e., in post-market surveillance studies and/or from less systematic and less predictable sources, e.g., individual case reports. The latter usually comes from a casual overdose incident. The prevalence of use of particular herbal medicines in the general population is so well reflected in the number of clinically important herb-drug interactions known and described so far for these herbal sources (Prely et al., 2022; TRC, 2023; Williamson, Driver & Baxter, 2009, 2018; Zhang et al., 2022; Ziemann et al., 2019).

Based on recently published information, this work aims to sum up the knowledge about interactions between conventional drugs and (traditional) herbal drugs that affect the central nervous system (psychopharmaceuticals) at the level of vigilance, affectivity, and psychical integrity. Vigilance is affected by psychostimulants—herbal caffeine sources (guarana—Guaranae semen/Paulliniae semen, green/black tea—Camelliae folium, cola—Colae semen, coffee—Coffeae semen, maté leaf—Mate folium, cocoa—Cacao semen), hypnotics (valerian root—Valerianae radix) and nootropics (Ginkgo leaf—Ginkgonis folium). The affectivity is impacted by antidepressants (St. John’s wort—Hyperici herba) and anxiolytics (hop strobile—Lupuli flos), and psychical integrity by hallucinogens (cannabis aerial part/marijuana—Cannabis herba, cannabis resin/hashish—Cannabis resina).

A drug interaction is said to occur when the effect of one drug is changed by the presence of another substance, including herbal medicines, foods (supplements), beverages, and environmental agents. The outcome can be harmful if the interaction causes an increase in the side effect and/or toxicity of the drug (Cordier & Steenkamp, 2011; Singh & Zhao, 2017). A reduction in efficacy due to an interaction can sometimes be just as harmful as an increase. Patients can apparently tolerate adverse interactions remarkably well and many interactions can be accommodated, so effects may not be consciously recognized as the result of an interaction. The mechanism of herbal interaction might not be easily tracked, as herbal sources differ from conventional drug substances in that they are complex mixtures of many biologically active compounds (Nagy, Mučaji & Grančai, 2017). The resulting pharmacological action is likely to arise often on the basis of synergism. Explaining so-called herb-drug interactions is therefore often more difficult. Furthermore, many drugs can interact by more than a single mechanism. Sometimes, more mechanisms are acting in concert. For clarity, most of the mechanisms that are dealt with in this review are as though they occur in isolation (Nagy, Mučaji & Grančai, 2017). For convenience, the mechanism of interactions can be divided into those that involve the pharmacokinetics of a drug, and those that are pharmacodynamic (Cordier & Steenkamp, 2011; Singh & Zhao, 2017; Williamson, Driver & Baxter, 2009).

Pharmacokinetic interactions are those that can affect the processes by which drugs are absorbed, distributed, metabolized, and eliminated (the so-called ADME interactions) (Cordier & Steenkamp, 2011; Singh & Zhao, 2017; Williamson, Driver & Baxter, 2009, 2013). The classical example of a target is the most well-investigated drug transporter efflux transporter P-glycoprotein (P-gp), also known as multidrug resistance protein 1 (MDR1) or ATP-binding cassette sub-family B member 1 (ABCB1) (Williamson, Driver & Baxter, 2009). Some well-known examples include ginkgo (maidenhair tree)—Ginkgo biloba (with active ginkgolides) is an inhibitor of P-gp and St. John’s wort—Hypericum perforatum is an inductor of P-gp (Czigle & Tóth, 2016). Other transporters that are involved in some drug interactions are organic cation transporters (OCTs) and organic anion-transporting polypeptides (OATPs) (Ali et al., 2020; Rodrigues et al., 2019). Although all of these mechanisms are undoubtedly relevant for interactions with herbal medicines, this review will focus mainly on cytochrome P₄₅₀ and the P-gp drug transporter protein, as information on other transporters is currently limited or of unknown biological relevance. Some of the secondary metabolites are enzyme inductors or inhibitors (CYP isoenzymes) (Amaeze et al., 2021; Manikandan & Nagini, 2018; Pandian et al., 2020; Wanwimolruk & Prachayasittikul, 2014; Williamson, Driver & Baxter, 2009; Zhou et al., 2003). With the exception of prodrugs, inhibitors slow down drug metabolism, which often leads to a substantial increase in serum drug levels with significant risk of adverse or even toxic effects (Cordier & Steenkamp, 2011; Singh & Zhao, 2017). Inductors cause the opposite, increased drug metabolism. Again, with the exception of prodrugs, this consequently often leads to lower-than-expected serum drug levels with potential treatment failure (Williamson, Driver & Baxter, 2009). St. John’s wort—Hypericum perforatum is a well-known inductor of CYP isoenzymes (Czigle & Tóth, 2016; Williamson, Driver & Baxter, 2009), inhibitors include ginkgo—Ginkgo biloba (Czigle & Tóth, 2016; Williamson, Driver & Baxter, 2009), valerian—Valeriana officinalis (Czigle & Tóth, 2016; Williamson, Driver & Baxter, 2009), hop—Humulus lupulus (Czigle & Tóth, 2016; Williamson, Driver & Baxter, 2009), among others (Czigle & Tóth, 2016; Williamson, Driver & Baxter, 2009). Finally, several other enzymes have been shown to play a role in herb-drug interactions, such as UDP-glucuronyltransferases (UGT) (Williamson, Driver & Baxter, 2009, 2018).

Pharmacodynamic interactions are those where the effects of one drug are changed by the presence of another drug at its site of action. Sometimes the drugs and the biologically active plant metabolites directly compete for a particular target (e.g., receptor, enzyme), but often the reaction is more indirect and involves interference with physiological mechanisms. If two drugs that have the same or similar pharmacological effect are given together, the effects can be additive and, in some cases, even potentiating (Nagy, Mučaji & Grančai, 2017). In contrast, there are some pairs of drugs with activities that are opposed to each other (antagonist) (Nagy, Mučaji & Grančai, 2017). For example, a particular additive effect is well known to occur between ginkgo—Ginkgo biloba and antithrombotic agents (Czigle & Tóth, 2016; European Medicines Agency, 2012d) and similarly between St. John’s wort—Hypericum perforatum and antidepressant drugs (Czigle & Tóth, 2016; European Medicines Agency, 2021). The exposures to drugs, for which knowledge of the potential adverse herb-drug interactions is lacking, might be a risk to patients’ safety (Czigle & Tóth, 2016; Williamson, Driver & Baxter, 2009, 2018). On the contrary, some interactions may be therapeutically beneficial (Czigle & Tóth, 2016; Williamson, Driver & Baxter, 2009, 2018) and might be used to develop new therapeutic strategies in the future.

Survey methodology

This work aims to summarize the knowledge about interactions between conventional drugs and (traditional) herbal drugs; and, as part of an intended series on this topic, concentrates on principal herbal drugs that affect psychopharmaceuticals. Subheads include herbal drugs that affect the central nervous system at the level of vigilance, affectivity, and psychical integrity. The given topic has been comprehensively summarized from the current scientific view on herbal medicines and has been currently (re-)evaluated by the Evidence Based Medicine approach (factor—risk/benefit ratio; the beneficial effects of an herbal medicine can outweigh the side effect). The minimum acceptance criteria were similar to the guidelines for authors of reviews by Munn et al. (2018). The survey databases used were SciFinder, PubMed, which identified primary scientific publications (e.g., case reports), and several secondary databases on interactions were used later on as well (Therapeutic Research Center—Natural Medicines database (TRC, 2023), Stockley’s Herbal Medicines Interactions (Williamson, Driver & Baxter, 2009, 2013, 2018), and other databases). The search was focused on officinal herbal drugs (European Pharmacopoeia, 2023h), herbal drug preparations (European Pharmacopoeia, 2023g), and herbal drug extracts (European Pharmacopoeia, 2023f). Keywords for search are listed here: herbal medicine-drug interaction, pharmacokinetic and pharmacodynamic interaction, selected botanical genera: Camellia, Coffea, Paullinia, Ilex, Cola, Theobroma, Valeriana, Ginkgo, Hypericum, Humulus, Cannabis. Subsequently, the names of the particular herbal drugs were included in the search, as Camelliae folium non fermentatum/fermentatum, Coffeae semen, Guaranae semen, Mate folium, Colae semen, Cacao semen, Valerianae radix, Ginkgonis folium, Hyperici herba, Humuli flos, Cannabis flos/herba/resina. Some botanical synonyms and related terms. We compiled information about pharmacokinetic and pharmacodynamic interactions with biologically active metabolites of these plants or herbal drugs. The criteria for the inclusion or exclusion of specific literature sources were interaction results, the conclusions of case report evaluations, clinical relevance of particular interactions or their absence, and the exhaustiveness of the description of the interaction. The same herbal medicines are often used in multiple indications, in different individual diagnoses; on the other hand, the mechanism of action of particular secondary metabolites is related to several diagnoses by EU herbal monographs (European Medicines Agency, 2023). Our objective was to explain the biological mechanisms of pharmacological activities for particular secondary metabolites in individual indications. The resulting pharmacological effect often is due to the synergism of the individual secondary metabolites. As for legislative aspects, the European Medicines Agency (EMA) applies the same criteria on all herbal drugs and on all diseases/health problems in its given methodological procedures. We summarize, synthesize, paraphrase data, and give an overview of the main points out of each source and combine them into a coherent whole. We analyze and interpret (clinically relevant) herb-drug interactions in the tables in the next sections with clarification, notes after analyses of theoretical or practical information. The Anatomical Therapeutic Chemical (ATC) classification system of the World Health Organization (WHO) Collaborating Centre for Drug Statistics Methodology (World Health Organization, 2023) is used to order and organize drug molecules in tables to highlight links in therapeutic and pharmacological/chemical subgroups. Further criteria reflected in this article include the availability of information about the potential of interaction of the particular drug, about interactions prevention measures, as well as practical recommendations after the manifestation of the interaction.

Drug interactions of herbal drugs containing caffeine (Guaranae semen, Camelliae folium, Colae semen, Coffeae semen, Mate folium, Cacao semen)

The most common herbal drugs used as sources of caffeine include green and black tea—Camelliae folium non fermentatum/fermentatum (Theae folium), coffee—Coffeae semen, guarana—Guaranae semen (Paulliniae semen, Guarana, Pasta guarana), maté leaf—Mate folium, cola—Colae semen, and cocoa—Cacao semen. Significantly, in addition to their use in herbal medicinal products, all of these herbal sources are commonly present in foods and beverages, as well (tea, coffee, cola drinks, and energy drinks) (Czigle & Tóth, 2010b). The caffeine content in the named herbal drugs decreases in the following order: Guarana (3.6–6%) > Camelliae folium (2–4%) > Colae semen (1–2.5%) > Coffeae semen (1–2%) > Mate folium (0.5–0.8%) > Cacao semen (0.07–0.4%) (Czigle & Tóth, 2010b). Camelliae folium and Colae semen additionally contain appreciable amounts of tannins, for Colae semen and Coffeae semen the presence of chlorogenic acids, and for Cacao semen the content of fatty acids is substantial (Czigle & Tóth, 2010b; Nagy, Mučaji & Grančai, 2017).

The European Pharmacopoeia lists officinal monographs for: green tea (Camelliae sinensis non fermentata folia) (European Pharmacopoeia, 2023d), guarana (Guaranae semen) (European Pharmacopoeia, 2023e), cola (Colae semen) (European Pharmacopoeia, 2023a), and maté leaf (Mate folium) (European Pharmacopoeia, 2023j). Coffee (Coffeae semen, Coffea L. spp., Rubiaceae) and cocoa (Cacao semen, Theobroma cacao L., Malvaceae) are not officinal in the European Pharmacopoeia.

The plant species Camellia sinensis (L.) Kuntze (Theaceae), with the botanical synonym Thea sinensis L., is known as the source of green tea (Camelliae folium non fermentatum), black tea (Camelliae folium fermentatum) and oolong (Camelliae folium semifermentatum) (The World Flora Online, 2022). There is a wide array of indications, well supported by in-depth studies on a number of defined chemical constituents of Camellia sinensis (L.) Kuntze. The main biologically active compounds are xanthine derivatives, such as caffeine (2–4%), theobromine (0.05%) and theophylline (traces); polyphenols (up to 30%), including gallotannins, (-)-epigallocatechin-3-O-gallate (EGCG, up to 12%), (-)-epicatechin-3-O-gallate (ECG), catechin tannins, other flavonoids, ellagitannins, phenolic carboxylic acids; theanine (amino acid), ascorbic acid, triterpene saponins, and minerals (Nagy, Mučaji & Grančai, 2017).

The dried seed of Paullinia cupana Kunth (syn. Paullinia sorbilis Mart.) (Sapindaceae), guarana, is also known as a food product, often available in the form of candy lozenges with Pasta guarana (guarana extract) (Czigle & Tóth, 2010b).

The dried leaf of Ilex paraguariensis A.St.-Hil. (Aquifoliaceae) is known as maté tea. After harvesting, the leaf is rapidly desiccated by heating, and cut (Czigle & Tóth, 2016).

Cola is produced from the plant species Cola nitida (Vent.) Schott et Endl. (syn. C. vera K. Schum.) and its varieties, as well as from Cola acuminata (P. Beauv.) Schott et Endl. (syn. Sterculia acuminata P. Beauv.) (Malvaceae); the herbal drug consists of whole or fragmented dried seeds, which were freed from the testa. Cola drinks contain extracts of this herbal drug (Czigle & Tóth, 2016).

Caffeine-containing herbal drugs with psychostimulant effects are used medicinally ‘for the relief of symptoms of (temporary) fatigue and sensation of weakness’, (European Medicines Agency, 2023). Based on a large amount of verified scientific data, several of these herbal drugs have been classified as traditional herbal medicinal products with the above cited indication: Camelliae folium non fermentatum (whole dried leaf, comminuted and powdered herbal substance) (European Medicines Agency, 2012c), Guaranae semen (guarana seed in solid dosage forms for oral use) (European Medicines Agency, 2011b), Colae semen (as a powdered herbal substance in solid dosage forms or as a herbal tea for oral use, as liquid extract or tincture) (European Medicines Agency, 2010b), Mate folium (as a comminuted herbal substance such as herbal tea for oral use) (European Medicines Agency, 2008b). In addition to this general therapeutic indication, maté leaf was also classified as a traditional herbal medicinal product ‘to increase the amount of urine to achieve flushing of the urinary tract as an adjuvant in minor urinary complaints’, (European Medicines Agency, 2008b).

Caffeine itself is a CYP1A2 inhibitor (Guo et al., 2021; Kot & Daniel, 2008; Williamson, Driver & Baxter, 2013). Caffeine metabolites are paraxanthine (84%), theobromine (12%), and theophylline (4%) (Faber, Jetter & Fuhr, 2005). In commonly achievable plasma concentrations, caffeine is an antagonist at adenosine A₁ and A₂-receptors (Cauli & Morelli, 2005; Nehlig, 2018). In higher concentrations, it acts as well as a phosphodiesterase inhibitor (PDE1, PDE4, PDE5) (Boswell-Smith, Spina & Page, 2006); γ-aminobutyric acid receptor (GABA_A) antagonist (Isokawa, 2016; Lopez et al., 1989); ryanodine receptor (RYR) agonist (Kong et al., 2008); and it also affects the nervus vagus (Hibino et al., 1997). Caffeine and paraxanthine have similar sympathomimetic actions (Benowitz et al., 1995).

This review presents and summarizes relevant and more recent data on the interaction potential of caffeine and caffeine-containing herbal drugs that are, notably, not included in the European Union herbal monographs and their related assessment reports (European Medicines Agency, 2008a, 2010a, 2011a, 2012a, 2020a, 2022).

Clinically significant pharmacokinetic and pharmacodynamic interactions are summarized in Table 1. In relation of pharmacodynamic interactions, it is important to emphasize that preparations containing caffeine reduce the sedative action and increase the side effects caused by sympathomimetic drugs (European Medicines Agency, 2008b, 2010b, 2011b, 2012c). Patients taking monoamine oxidase (MAO) inhibitors should be cautious when concomitantly using Paulliniae semen (European Medicines Agency, 2011b), Mate folium (European Medicines Agency, 2008b), and/or Colae semen (European Medicines Agency, 2010b). However, there are, also some important pharmacokinetic interactions based either on the inhibition of CYP1A2 metabolism of other drugs by caffeine (e.g., melatonin) (European Medicines Agency, 2008b, 2010b, 2011b, 2012c) or by inhibition of the same cytochrome by other drugs resulting in potential caffein toxicity (e.g., fluvoxamine, ciprofloxacine) (European Medicines Agency, 2008b, 2010b, 2011b, 2012c).

Drug interactions of Ginkgonis folium

Ginkgo biloba L. (Ginkgoaceae), commonly also known as ginkgo or the maidenhair tree, is a species of dioecious tree native to China (Bensky, Gamble & Bensky, 1986; Czigle et al., 2018; van Beek, 2000).

The European Pharmacopoeia lists a monograph for the dried leaf of Ginkgo biloba L. (Ginkgonis folium) (European Pharmacopoeia, 2023c) and requires the content of not less than 0.5% flavonoids, expressed as flavone glycosides (M_r 757). The pharmacopoeial monograph for the refined and quantified dry extract produced from Ginkgo leaf (Ginkgonis extractum siccum raffinatum et quantificatum) (European Pharmacopoeia, 2023b) contains three positive criteria on active constituents: 22.0–27.0% flavone glycosides (M_r 756.7), 2.8–3.4% ginkgolides (A, B and C), and 2.6–3.2% bilobalide. The fourth criterion requires that the content of ginkgolic acids is less than 5 ppm (European Pharmacopoeia, 2023b).

Products prepared from Ginkgo leaf and from its refined and quantified dry extract, often marketed as EGb 761, are top-selling herbal medicinal products, especially in Europe, as well as important major botanical food supplements on the world market. The quality of these food supplements on the European market is often dubious and prone to adulteration (Czigle et al., 2018). In western allopathic medicine, Ginkgo biloba medications underwent strict evaluations of their clinical therapeutic use, for example, by the European Medicines Agency (EMA), with the following results: the Ginkgo leaf herbal medicinal dry extract is indicated in the well-established medicinal use category ‘for the improvement of (age-associated) cognitive impairment and of quality of life in mild dementia’, (European Medicines Agency, 2012d), ATC code: N06DX02 (World Health Organization, 2023), and the powdered Ginkgo leaf herbal substance is beneficial as a traditional herbal medicinal product ‘for the relief of heaviness of legs and the sensation of cold hands and feet associated with minor circulatory disorders’ (European Medicines Agency, 2012d). The European Union herbal monograph is accompanied by a detailed assessment report (European Medicines Agency, 2012b).

The indications for G. biloba leaf are well supported by in-depth studies on a number of defined chemical constituents. The best known are flavonoids (flavonols, e.g., quercetin, kaempferol, isorhamnetin and their mono-, di-, and triglycosides; biflavonoids, e.g., amentoflavone, ginkgetin, bilobetin, sciadopitysin) (Hasler et al., 1992; Hasler & Meier, 1993; Meier, Hasler & Sticher, 1992) and terpene lactones (several diterpenes, e.g., ginkgolides A, B, C, D, K, L, M, N, P, Q, and the pentanorditerpene bilobalide) (van Beek, 2005; van Beek & Lelyveld, 1992; van Beek & Montoro, 2009), but they also include hypersensitive (ginkgolic acid) (Fuzzati, Pace & Villa, 2003; van Beek & Wintermans, 2001) and neurotoxic compounds (ginkgotoxin) (Leistner & Drewke, 2010).

Ginkgolides are inhibitors of CYP1A2, CYP2C9, CYP2C19, CYP2D6, CYP2E1, CYP3A4 (Czigle & Tóth, 2009a, 2016; Williamson, Driver & Baxter, 2009, 2013) and of the P-gp (causing decreased absorption and increased elimination) (Czigle & Tóth, 2009a, 2016; Williamson, Driver & Baxter, 2009, 2013). Ginkgolides are also antagonists of the platelet-activating factor (PAF); (Czigle & Tóth, 2009a, 2016; Czigle et al., 2018; Oberpichler et al., 1990; Smith, Maclennan & Darlington, 1996), Ginkgo flavonoids affect hemorheology (Czigle & Tóth, 2016; Chen et al., 2019a; Witte, Anadere & Walitza, 1992) and vasopressin level (Czigle & Tóth, 2016; Kubota et al., 2006). The EGb 761 extract increases the synaptosomal uptake of 5-hydroxytryptamine (Czigle & Tóth, 2016; Ramassamy et al., 1992), acts as an GABA_A antagonist (Czigle & Tóth, 2016; Ivic et al., 2003), and has a neuroprotective effect (Ahlemeyer & Krieglstein, 2003). Biflavones (ginkgetin, isoginkgetin, bilobetin, and amentoflavone), and some flavonoids (apigenin, quercetin, isorhamnetin, luteolin, and kaempferol) inhibit thrombin activity (Chen et al., 2019c).

This review presents and summarizes relevant and more recent data on the drug interaction potential of Ginkgo leaf and its active metabolites. Interactions can occur at both the pharmacokinetic and pharmacodynamic levels (Table 2). Ginkgo biloba can increase the effects of anticoagulants, such as coumarin derivatives (European Medicines Agency, 2012d); patients with concomitant anticoagulant (e.g., warfarin, phenprocoumon) and antiplatelet treatment (e.g., clopidogrel, acetylsalicylic acid, and other non-steroidal anti-inflammatory drugs (NSAIDs)) should only use Ginkgo medicinal products after consultation with a physician (European Medicines Agency, 2012d). Concomitant use of Ginkgo biloba preparations and efavirenz (a non-nucleoside reverse transcriptase inhibitor) is not recommended, as plasma concentrations of efavirenz can be decreased due to induction of CYP3A4 (European Medicines Agency, 2012d). An interaction study with talinolol (a selective beta-blocking agent) indicates that Ginkgo biloba may inhibit P-gp (European Medicines Agency, 2012d). This can cause increased exposure of drugs markedly affected by the P-gp in the intestine, such as dabigatran etexilate (a direct thrombin inhibitor); caution is advised for the combination of Ginkgo biloba with dabigatran (European Medicines Agency, 2012d). A study of interaction has indicated that the C_max of nifedipine (a calcium channel blocker) increased by up to 100% in some individuals that resulted in dizziness and an increased severity of hot flashes (European Medicines Agency, 2012d).

Drug interactions of Valerianae radix

Valerian is an old favorite in European phytotherapy. The monograph of valerian root (Valerianae radix) in the European Pharmacopoeia describes the dried, whole or fragmented underground parts of Valeriana officinalis L. s. l. (Valerianaceae), including the rhizome surrounded by the roots and stolons (European Pharmacopoeia, 2023o). The pharmacopoeia also lists further monographs with pharmaceutical quality used in human medicine: valerian root, cut for the purpose of being used in herbal teas (Valerianae radix minutata) (European Pharmacopoeia, 2023p), valerian dry hydroalcoholic extract (Valerianae extractum hydroalcoholicum siccum) (European Pharmacopoeia, 2023n), valerian dry aqueous extract (Valerianae extractum aquosum siccum) (European Pharmacopoeia, 2023m), and valerian tincture (Valerianae tinctura) (European Pharmacopoeia, 2023q).

Recently, Valeriana officinalis medications underwent evaluations of their clinical therapeutic use, e.g., by the European Medicines Agency, with the following results: the dry extract of Valerianae radix is indicated in the category of well-established medicinal use for the ‘relief of mild nervous tension and sleep disorders’, (European Medicines Agency, 2015c), ATC code: N05CM09 (World Health Organization, 2023), and as a traditional medicinal product Valerianae radix and its dry extracts, tinctures, and expressed juice (from fresh root), are beneficial for the ‘relief of mild symptoms of mental stress and to aid sleep’, (European Medicines Agency, 2015c). The valerian essential oil (Valerianae aetheroleum), as a traditional herbal medicinal product is beneficial for the ‘relief of mild symptoms of mental stress and to aid sleep’, (European Medicines Agency, 2015b). This review article only presents relevant and more recent data that are not part of these herbal monographs or related assessment reports (European Medicines Agency, 2015a).

The combination of valerian root and hop strobile (Valerianae radix and Lupuli flos) is indicated in the category of well-established medicinal use for the ‘relief of sleep disorders’, and in the category of traditional use for the ‘relief of mild symptoms of mental stress’, and ‘to aid sleep’, (European Medicines Agency, 2017b).

Sedative herbal tea combinations (Species sedativae) shall contain 2, 3 or 4 herbal substances out of hop strobile—Lupuli flos (Humulus lupulus L., Cannabaceae), lavender flower—Lavandulae flos (Lavandula angustifolia Mill., Lamiaceae), Melissa leaf—Melissae folium (Melissa officinalis L., Lamiaceae), passionflower herb—Passiflorae herba (Passiflora incarnata L., Passifloraceae), and valerian root—Valerianae radix (Valeriana officinalis L., Valerianaceae). Depending on the chosen herbal substances, the percentage of Valerianae radix in combination is 15–57% or 33–57%. This herbal tea is beneficial as a traditional herbal medicinal product for the ‘relief of mild symptoms of mental stress and to aid sleep’, (European Medicines Agency, 2017a).

The major biologically active compounds of Valerianae radix are flavonoids, iridoids, lignans, volatile compounds, such as monoterpenes (borneol, bornyl acetate), and sesquiterpenes (valerenone, valerenic and acetoxyvalerenic acids) (Czigle & Tóth, 2010a, 2016; Orhan, 2021).

A possible mechanism, by which Valerianae radix may cause sedation, is by increasing the synaptic amount of γ-aminobutyric acid (GABA) (Johnston et al., 2006; Orhan, 2021; Savage et al., 2018). Flavonoids (Wasowski & Marder, 2012) and monoterpenes (bornyl acetate) (Wang & Heinbockel, 2018) have GABA agonistic properties, and flavonoids also act as MAO inhibitors (Dhiman et al., 2019). Valerenic acid could attenuate increases in serum corticosterone levels in a mouse model of stress, the extract could modulate serotonin (5-HT) and norepinephrine (NE) turnover in the hippocampus and amygdala region (Jung et al., 2015). Valerenic acid is a selective positive allosteric modulator of the GABA_A receptor (Khom et al., 2007), an agonist at the metabotropic glutamate receptor mGlu I, but an antagonist at the mGlu II (Del Valle-Mojica & Ortíz, 2012; Feinberg et al., 2005); a partial agonist at the 5-HT_5A receptor (Dietz et al., 2005), it also has antidepressant potential. Isovaltrate (a valepotriate) is an inverse agonist at the adenosine A₁ receptor (Sichardt et al., 2007), and it presents antidepressant-like activity, mediated by dopaminergic, noradrenergic, serotonergic neurotransmissions (Dietz et al., 2005; Jayaraj et al., 2020; Orhan, 2021; Ortiz, Nieves-Natal & Chavez, 1999). It also has inhibitory activity towards the Na⁺/K⁺-ATPase (Müller et al., 2015). Some lignans, for example 1-hydroxypinoresinol, are GABA_A receptor agonists; and 4′-O-β-d-glucosyl-9-O-(6″-deoxysaccharosyl)olivil is a partial agonist at adenosine A₁ receptors (Cavadas et al., 1995; Savage et al., 2018; Schumacher et al., 2002).

Interactions occur at the pharmacokinetic and pharmacodynamic levels (Table 3). Different preparations of valerian root have been proven to inhibit the cytochrome P₄₅₀ isoenzyme CYP3A4. In contrast, Valeriana does not affect other cytochrome P₄₅₀ isoenzymes CYP1A2, CYP2C9, CYP2C19, CYP2D6 and CYP2E1 (Bogacz et al., 2014; Czigle & Tóth, 2010a, 2016; Williamson, Driver & Baxter, 2009, 2013, 2018). Pharmacodynamic interactions are related to the mechanism of action of secondary metabolites. Valeriana does not affect the pharmacokinetics of caffeine, alprazolam, or midazolam to a clinically relevant extent (European Medicines Agency, 2015c). The stimulant effect of caffeine may oppose the hypnotic effects of valerian (European Medicines Agency, 2015c). However, additive anxiolytic effects are possible with benzodiazepines (European Medicines Agency, 2015c).

Drug interactions of Hyperici herba

The herbal drug Hyperici herba consists of whole or fragmented, dried flowering tops of St. John’s wort (Hypericum perforatum L., Hypericaceae), harvested during flowering time (European Pharmacopoeia, 2023l). The Pharmacopoeia also lists a monograph for the St. John’s wort dry extract, quantified (Hyperici herbae extractum siccum quantificatum) (European Pharmacopoeia, 2023k) with pharmaceutical quality used in human medicine. In Western allopathic medicine, Hypericum perforatum medications were subjected to strict evaluations of their clinical therapeutic use, e.g., by the European Medicines Agency, with the following results: the Hypericum dry extract is indicated in the category of well-established medicinal use for the ‘treatment of mild to moderate depressive episodes, and for the short term treatment of symptoms in mild depressive disorders’, (European Medicines Agency, 2021), ATC code: N06AX25 (World Health Organization, 2023), and as a traditional medicinal product the Hyperici herba dry and liquid extracts, tinctures, expressed juice (from the fresh herb), and the comminuted herbal substance are beneficial for the ‘relief of temporary mental exhaustion, for the symptomatic treatment of minor inflammations of the skin (such as sunburn) and as an aid in healing of minor wounds, as well as for the symptomatic relief of mild gastrointestinal discomfort’, (European Medicines Agency, 2021). This review article presents relevant and more recent data that are not part of the European Union herbal monograph or the related assessment report (European Medicines Agency, 2016a).

The final antidepressant effect arises on the basis of the synergy of different biologically active compounds (Czigle & Tóth, 2009b; Nagy, Mučaji & Grančai, 2017). The phloroglucinol derivatives hyperforin and adhyperforin are re-uptake inhibitors of serotonin, dopamine, and norepinephrine (Tian et al., 2014). Hyperforin is a transient receptor potential channels (TRPC6) activator as well, and it induces the release of calcium and zinc from mitochondria (Tu, Gibon & Bouron, 2010). Recent data showed that the increase in the intracellular concentration of calcium ions is mediated by the activation of the mentioned TRPC6, a diacylglycerol-sensitive C-class of transient receptor potential channels, and without the activation of other isoforms (TRPC1, TRPC3, TRPC4, TRPC5, and TRPC7) (Nagy, Mučaji & Grančai, 2017; Tu, Gibon & Bouron, 2010). Hyperforin is also a GABA reuptake inhibitor (Zanoli, 2004). Flavonoids and proanthocyanins have a GABA agonistic effect (Hanrahan, Chebib & Johnston, 2011). The naphthodianthrone derivative hypericin is a MAO-A and MAO-B inhibitor (Butterweck, 2003), and dopamine β-hydroxylase inhibitor (Kleber et al., 1999), leading to increased dopamine levels, although thus possibly decreasing norepinephrine and epinephrine. It also affects 5-HT_1A (Teufel-Mayer & Gleitz, 1997) and D₃ receptors (McClatchey et al., 2009), and is a β-sympathomimetic (Nagy, Mučaji & Grančai, 2017).

Interactions occur at both the pharmacokinetic and pharmacodynamic levels. The Hypericum dry extract induces the expression of CYP3A4, CYP2C9, CYP2C19 (Czigle & Tóth, 2009b, 2016; Williamson, Driver & Baxter, 2009, 2013, 2018), which results in increased drug metabolism; it also inhibits the P-glycoprotein (causing decreased oral absorption and increased elimination) (Czigle & Tóth, 2009b, 2016; Dürr et al., 2000; Williamson, Driver & Baxter, 2013). Special care should be taken in the case of concomitant use of all drug substances whose kinetics are influenced by CYP3A4, CYP2C9, CYP2C19, or P-gp (e.g., digoxin, simvastatin, methadone, benzodiazepine derivatives, amitriptyline, finasteride, fexofenadine), because a reduction in plasma concentrations of these drugs is possible (Table 4). Concomitant use of immunosuppressive drugs for systemic use (cyclosporine/ciclosporin, tacrolimus) (Alscher & Klotz, 2003; Czigle & Tóth, 2016), antivirals for systemic use (protease inhibitors, such as amprenavir, indinavir, and others), several antineoplastic agents, such as topoisomerase 1 inhibitors (irinotecan), and anticoagulants, vitamin K antagonists (warfarin) is contraindicated (Table 4). A reduction in plasma concentrations of hormonal contraceptives for systemic use can lead to increased intermenstrual (breakthrough) bleeding and to a failure of the contraceptive effect. Women who use oral contraceptives should take alternate or additional contraceptive measures (Table 4) (Czigle & Tóth, 2016; European Medicines Agency, 2021; Hall et al., 2003).

The outcome of the pharmacodynamic interaction is the risk of serotonin syndrome (Czigle & Tóth, 2016; Spadaro et al., 2022; Volpi-Abadie, Kaye & Kaye, 2013) or Call-Fleming syndrome (Czigle & Tóth, 2016; Skandhan, Ramakrishnan & Anand, 2013). The reason is that the Hypericum dry extract increases serotonergic effects when combined with antidepressants affecting serotonin, such as selective serotonin reuptake inhibitors (e.g., sertraline, paroxetine, nefazodone), tricyclic antidepressants, selective serotonin (5-HT_1A) agonists employed in anxiety treatment (buspirone), or selective serotonin (5-HT_1BD) agonists used for acute migraine attacks (triptans) (Table 4). The serotonin syndrome (Volpi-Abadie, Kaye & Kaye, 2013) is a potentially life-threatening condition based on the overactivation of both the central postsynaptic and peripheral 5HT_1A, and most notably the 5HT_2A receptors. Symptoms include a combination of altered cognitive functions (cephalgia, agitation, hypomania, confusion, hallucination, coma), autonomic hyperactivity (tremor, diaphoresis, hypertension, tachycardia, nausea, vomitus), neuromuscular hyperactivity (rigidity, tremor, myoclonus, hyperreflexia) and hyperthermia (Spadaro et al., 2022; Volpi-Abadie, Kaye & Kaye, 2013). Symptoms usually begin within 24 h of an increased dose of a serotonergic drug/agent, the addition of another serotonergic agent to a drug regimen, or overdosing. Most patients will seek help in a hospital in 6 h; however, patients with mild symptoms may have a more subacute or chronic presentation (Volpi-Abadie, Kaye & Kaye, 2013). The Call-Fleming syndrome (Skandhan, Ramakrishnan & Anand, 2013), also called reversible cerebral vasoconstriction syndrome (RCVS), is characterized by thunderclap headache and reversible vasoconstriction of the cerebral arteries.

During the treatment with Hypericum extracts, intense UV exposure should be avoided. Hypericin (a red pigment) ( $G=(V,E)$ 595 nm) (Skalkos et al., 2006) is phototoxic (Schempp et al., 2000, 2001, 2003). A synergistic photosensitivity reaction has been observed between aminolevulinic acid (photodynamic therapy) (Boiy, Roelandts & de Witte, 2011; Ritz et al., 2012; Schneider-Yin et al., 2009) and hypericin (Table 4).

Drug interactions of Lupuli flos

The pharmacopoeial herbal drug hop strobile (Lupuli flos), having pharmaceutical quality and used in human medicine, is described as the dried, generally whole, female inflorescence (strobile) of hop (Humulus lupulus L., Cannabaceae) (European Pharmacopoeia, 2023i). Hops (hop strobiles) are also commonly used culinarily as a bittering and flavoring agent in the production of beer. In allopathic medicine, Humulus lupulus medications underwent strict evaluations of their clinical therapeutic use, e.g., by the EMA with the following results: Lupuli flos dry and liquid extracts (ethanolic or sweet wine extract), tincture, as well as the comminuted or powdered herbal substance are indicated in the category of traditional herbal medicinal products that are beneficial for the ‘relief of mild symptoms of mental stress and to aid sleep’, (European Medicines Agency, 2013). This review article presents relevant and more recent data that are not included in the European Union herbal monograph on hop strobile and the related assessment report (European Medicines Agency, 2005).

The medicinal use of valerian root and hop strobile as a combination (Valerianae radix et Lupuli flos) is indicated in the category of well-established medicinal use for the ‘relief of sleep disorders’, and in the category of traditional use for the ‘relief of mild symptoms of mental stress and to aid sleep’, (European Medicines Agency, 2017b).

Sedative herbal tea combinations (Species sedativae) shall contain 2, 3 or 4 herbal substances out of hop strobile—Lupuli flos (Humulus lupulus L., Cannabaceae), lavender flower—Lavandulae flos (Lavandula angustifolia Mill., Lamiaceae), Melissa leaf—Melissa folium (Melissa officinalis L., Lamiaceae), passionflower herb—Passiflorae herba (Passiflora incarnata L., Passifloraceae), and valerian root—Valerianae radix (Valeriana officinalis L., Valerianaceae). Depending on the chosen herbal substances, the percentage of Lupuli flos in combination is 15–40%, 15–43%, or 21–36%. This herbal tea is beneficial as a traditional herbal medicinal product for the ‘relief of mild symptoms of mental stress and to aid sleep’, (European Medicines Agency, 2017a).

The final anxiolytic/hypnotic effect of hop strobile is based on the synergy of different biologically active compounds. Lupuli flos contains resins with phloroglucinols (α-acids (humulone and its derivates), β-acids (lupulone and its derivates), and their degradation product 2-methyl-3-buten-2-ol), flavonoids (xanthohumol), and terpenes (β-myrcene, β-caryophyllene) (Nagy, Mučaji & Grančai, 2017).

Some Lupuli flos components, including those contained in beer (non-alcoholic), are believed to act in the CNS by influencing GABA, adenosine, serotonin, and melatonin neurotransmission with an effective sedative action that both modulates the circadian rhythms and the sleep-wake cycle and is beneficial for the induction of sleep (Abourashed, Koetter & Brattström, 2004; Brattström, 2007; Butterweck et al., 2007; Dietz et al., 2005; Franco et al., 2012; Savage et al., 2018). Humulone is a positive allosteric modulator of GABA_A receptors (Benkherouf et al., 2020). Some components of hop resin (including the degradation product 2-methyl-3-buten-2-ol) increase GABA activity by modulating GABA_A receptors (Franco et al., 2012). The Lupuli flos extract is an agonist of the melatonin receptor MT₁ but α- and β-acids do not affect melatonin receptors (Grundmann et al., 2006). Melatonin secretion has a circadian character, its receptors (MT₁ and MT₂) are localized in the suprachiasmatic nucleus; melatonin regulates circadian rhythms and hence the sleep-wake cycle in humans. Humulus active compounds have an α₂-agonistic effect (β-myrcene) (Surendran et al., 2021), they affect the cannabinoid CB₂ receptor (β-caryophyllene) (Alberti et al., 2017; Klauke et al., 2014) and opioid receptors (Park et al., 2012). Lupuli flos also contains several compounds with estrogenic activity, such as the prenylated flavonoid 6-prenylnaringenin (Carbone & Gervasi, 2022; Dietz et al., 2017; Hemachandra et al., 2012; Tan et al., 2014; Yoshimaru et al., 2014). Prenylflavonoids (xanthohumol, isoxanthohumol and 8-prenylnaringenin) are capable of modulating aromatase activity, decreasing estrogen synthesis (Monteiro et al., 2006).

Interactions occur at both the pharmacokinetic and pharmacodynamic levels. Lupuli extract is a CYP1A2, CYP2C8, CYP2C9, CYP2C19 inhibitor (Czigle & Tóth, 2016; Foster et al., 2011; TRC, 2023; Williamson, Driver & Baxter, 2013; Yuan et al., 2014). 8-Prenylnaringenin is an inhibitor of multidrug resistance-associated transporters, P-glycoprotein and MRP1 (Wesołowska et al., 2010). Abundant Phase II conjugates of the prenylated flavonoids were observed, including monoglucuronides, diglucuronides, monosulfates and sulfate-glucuronic acid deconjugates (van Breemen et al., 2020). No clinically significant interactions are reported in the European Union herbal monograph (European Medicines Agency, 2013). The interactions between Humulus and other drugs are based mainly on experimental evidence (European Medicines Agency, 2013; Williamson, Driver & Baxter, 2009, 2013, 2018) (Table 5).

Drug interactions of cannabinoids-containing herbal drugs (Cannabis flos, Cannabis herba, Cannabis resina)

Observing the impact of marijuana (Cannabis herba) or hashish (Cannabis resina) on the human body opened the way for the medical use of some cannabinoid receptor substrates in the therapy of certain diseases (e.g., as analgesics, antiemetics, antiepileptics) (Czigle & Tóth, 2016; Hill et al., 2017; NIH, 2023; Stasiłowicz et al., 2021). It took a long time to finally discover human cannabinoid receptors for in the 1980s; there are two subtypes: CB₁ and CB₂ (Kendall & Yudowski, 2017; Stasiłowicz et al., 2021; Svízenská, Dubový & Sulcová, 2008). Subsequently, the importance of the endocannabinoid system was clarified and endogenous ligands of the cannabinoid receptors—metabolites of arachidonic acid—were discovered. They include anandamide (AEA), 2-arachidonoylglycerol (2-AG), and palmitoylethanolamide. Less well-established endocannabinoids are virodhamin, noladin, and N-arachidonoyl dopamine (Mouslech & Valla, 2009). The main substances of marijuana or hashish are phytocannabinoids, e.g., Δ⁹-tetrahydrocannabinol (Δ⁹-THC, dronabinol, (-)-THC), Δ⁸-tetrahydrocannabinol (Δ⁸-THC), cannabidiol (CBD) and cannabinol (CBN). A total of almost 70 isolated cannabinoids are known (European Medicines Agency, 2020b; Gülck & Lindberg Møller, 2020). Currently, there is information on the therapeutic application of the above-mentioned herbal drugs, their extracts and/or individual active metabolites, as well as on some analogues with slight chemical structural modifications. Cannabis flos and Cannabis herba (European Medicines Agency, 2020b) of pharmaceutical quality (Cannabis sativa L., Cannabaceae) are used as analgesics in oncology and to treat neuropathic pain (oral use in the form of a water infusion, i.e., herbal tea, or as inhalation therapy) (Fitzcharles et al., 2021; Giacoppo, Bramanti & Mazzon, 2017; Hill et al., 2017; NIH, 2023; Stasiłowicz et al., 2021; Vulfsons et al., 2020). Synthetic agonists at cannabinoid receptors encompass dronabinol (Δ⁹-THC) and nabilone (both are used as antiemetic agents during cancer treatment and in the treatment of anorexia of patients with HIV) (Beal et al., 1995; DAC, 2001; Ng & Gupta, 2022). Cannabidiol (CBD) is used as an orphan drug in the therapy of specific forms of epilepsy: the Dravet syndrome (European Medicines Agency, 2014; Chen, Borgelt & Blackmer, 2019b) and the Lennox-Gastaut syndrome (European Medicines Agency, 2017c; Chen, Borgelt & Blackmer, 2019b). EMA granted an orphan designation for cannabidiol for the treatment of tuberous sclerosis (European Medicines Agency, 2018) and glioma (European Medicines Agency, 2016b). An inverse agonist at cannabinoid receptors, rimonabant, was tested in the treatment of addiction and obesity, but was withdrawn from the market due to side effects (2% incidence of depression) and suicidal risk (Bielawiec, Harasim-Symbor & Chabowski, 2020; Leite et al., 2009; European Pharmacopoeia, 2023h). Cannabis sativae oleum, the fatty oil obtained from Cannabis semen is important for the treatment of eczema and intertrigo (Sklenář, 2021).

Pharmacokinetic interactions occur mainly at the level of metabolism (Table 6). Phytocannabinoids are extensively metabolized by the cytochrome P₄₅₀ (Czigle & Tóth, 2011, 2016). Cannabinoids inhibit CYP2C9 and CYP3A4 (Czigle & Tóth, 2011; Williamson, Driver & Baxter, 2009, 2013, 2018), while marijuana smoking induces CYP1A2 (Czigle & Tóth, 2011, 2016; Williamson, Driver & Baxter, 2009, 2013, 2018). THC and CBD inhibit CYP2D6, CYP2C19, CYP2B6 and CYP2J2 (Qian, Gurley & Markowitz, 2019). CYP3A4/5/7 is potentially inhibited by CBD (Qian, Gurley & Markowitz, 2019). Δ⁹-THC induces CYP1A1 also activates CYP2C9 (Czigle & Tóth, 2011, 2016; Qian, Gurley & Markowitz, 2019; Williamson, Driver & Baxter, 2009, 2013, 2018), the cannabinoids mixture also induces the expression of CYP2E1 (Czigle & Tóth, 2011, 2016; Williamson, Driver & Baxter, 2013) and CYP2D6 (Czigle & Tóth, 2011, 2016; Williamson, Driver & Baxter, 2009, 2013, 2018). UGT1A9 is inhibited by CBN and CBD, UGT2B7 is activated by CBN but inhibited by CBD (Qian, Gurley & Markowitz, 2019). Carboxylesterase 1 (CES1) is potentially inhibited by CBD and THC (Qian, Gurley & Markowitz, 2019). Pharmacodynamic interactions are related to the mechanism of action of cannabinoids (Table 6).

Preventing drug interactions/practical recommendations

Conclusions

The possibility of drug interactions between conventional drugs and herbal medicines (or herbal food supplements) is an emerging concern for therapeutic safety, as subsequent side effects and direct toxicities constitute a possible hidden threat to successful clinical therapy. Simultaneous use of herbal products and other synthetic drugs can modify, increase, or decrease the therapeutic effects of a drug; in addition, new side effects can arise. Plausible cases of herb-drug interactions include: herbal medicines or beverages containing caffeine reduce the activity of sedatives and increase the side effects caused by sympathomimetic drugs; the stimulant effect of caffeine can oppose the hypnotic effects of Valeriana officinalis; additive anxiolytic effects are possible when benzodiazepines are combined with Valeriana officinalis or Humulus lupulus; bleeding may occur when anticoagulants/antiplatelet drugs are combined with Ginkgo biloba; serotonin syndrome can threaten patients who combine Hypericum perforatum with serotonin-affecting medications such as serotonin reuptake inhibitors. For medicinal Cannabis, antagonistic effects were observed for combinations with some NSAIDs, as well as a potentiation of the antinociceptive effect of anodynes. Patient compliance is essential. The medical doctor and pharmacist should be informed about self-medication.