Recent Applications of Protoberberines as Privileged Starting Materials for the Development of Novel Broad-Spectrum Antiviral Agents: A Concise Review (2017–2023)

Abstract

Berberine is a well-known phytochemical with significant antiviral activity against a wide range of viruses. Due to having a unique backbone consisting of four interconnected rings, it can be used as a platform for the design and development of novel semisynthetic antiviral agents. The question here is whether novel broad-spectrum antiviral drugs with enhanced activity and toxicity potential can be obtained by attempting to modify the structure of this privileged lead compound. The present study aims to review the results of recent studies in which berberine and its close analogues (protoberberine alkaloids) have been used as starting materials for the production of new semisynthetic antiviral structures. For this purpose, relevant studies published in high-quality journals indexed in databases such as Scopus, Web of Science, PubMed, etc. in the time frame of 2017 to 2023 were collected. Our selection criterion in the current review focuses on the studies in which protoberberines were used as starting materials for the production of semisynthetic agents with antiviral activity during the indicated time period. Correspondingly, studies were identified in which semisynthetic derivatives with significant inhibitory activity against a wide range of viruses including human immunodeficiency virus (HIV), enterovirus 71 (EV71), zika virus (ZIKV), influenza A/B, cytomegalovirus (CMV), respiratory syncytial virus (RSV), and coxsackieviruses were designed and synthesized. Our conclusion is that, despite the introduction of diverse semisynthetic derivatives of berberine with improved activity profiles compared to the parent natural leads, sufficient derivatization has not been done yet and more studies are needed.

1. Introduction

Berberine (MF: C₂₀H₁₈NO₄⁺, MW: 336.4 g/mol) with the IUPAC name of 16,17-dimethoxy-5,7-dioxa-13-azoniapentacyclo[11.8.0.02,10.04,8.015,20]henicosa-1(13),2,4(8),9,14,16,18,20-octaene) is a famous bioactive alkaloid isolated from different species of the Berberis family, such as Berberis vulgaris L, Berberis aetnensis C. Presl, Berberis soulieana Schneid, Berberis poiretii Schneid, Berberis wilsoniae Hemsl, Berberis koreana Palib, Berberis thunbergii DC, Berberis aristata DC, Berberis lycium Royle, and Berberis asiatica Roxb.¹ Pharmacologically, this bioactive compound has several activities, including hepatoprotective, neuroprotective, hypolipidemic, hypoglycemic, antitumor, antioxidant, anti-inflammatory, and antimicrobial effects.² From a structural point of view, berberine is a cationic quaternary ammonium salt made by two fused scaffolds, isoquinoline and 3,4-dihydroisoquinoline. In terms of flexibility, the berberine has a rigid backbone consisting of four interconnected hexagonal rings A to D (Figure 1A).³

Figure 1.

(A) Chemical structure of berberine and substitution patterns of its analogues on the isoquinoline and 3,4-dihydroisoquinoline moieties. (B) Chemical structure of well-known protoberberine alkaloids containing 5,6-dihydrodibenzo[a,g]quinolizinium backbone.

There are diverse structural analogues of berberine (known as protoberberine alkaloids), and most of them possess a cationic quaternary ammonium group with rigid skeletons (Figure 1B). Structurally, these alkaloids belong to the class of isoquinoline alkaloids and have a 5,6-dihydrodibenzo[a,g]quinolizinium backbone. Different derivatives of protoberberines are usually distinguished by different small substituents such as hydroxyl, methoxy, and methylenedioxy in positions 2, 3 of the dihydroisoquinoline scaffold, and in positions 9, 10, and 11 in the isoquinoline scaffold. However, in the few derivatives, positions 8 and 13 (ring C) also contain alkyl substitutions. In terms of biological activity, the most famous member of the protoberberines is berberine, for which many therapeutic capabilities have been reported in the treatment of a wide range of diseases (as mentioned above). In addition, there are other well-known protoberberine alkaloids with significant biological effects such as palmatine,⁴ jatrorrhizine (C₂₀H₂₀NO₄⁺; MW: 338.4 g/mol),⁵ coptisine,⁶ etc., which are structurally very similar to berberine.

Structurally, palmatine (MF: C₂₁H₂₂NO₄⁺, MW: 352.4 g/mol) by the IUPAC name of 2,3,9,10-tetramethoxy-5,6-dihydroisoquinolino[2,1-b]isoquinolin-7-ium is one of the most similar members of protoberberines to berberine, which has two methoxy groups instead of the methylenedioxy moiety of berberine in positions 2 and 3 (Figure 1). This high structural similarity has caused these compounds to have similar biological activity and physicochemical properties. Palmatine is commonly obtained from medicinal plants Stephania yunnanensis H.S.Lo., Tinospora sagittata (Oliv.) Gagnep, Phellodendron amurense Rupr., and Tinospora cordifolia (Willd.) Miers (T. cordifolia).⁴ The pharmacology, pharmacokinetics, and toxicology of palmatine have been well investigated in previous studies. Similar to berberine, this compound also has a wide range of pharmacological effects such as anticancer, anti-inflammatory, and antiviral activities.⁴

2. Pharmacokinetics, Bioavailability, and Metabolism

Based on pharmacokinetic data, both human and animal models indicate that berberine and its derivatives (such as palmatine, jatrorrhizine, and coptisine) have poor oral bioavailability, but can be widely distributed to organs after absorption from the gastrointestinal tract.^7,8 One of the main reasons for the undesirable oral bioavailability of berberine is its insufficient dissolution rate.⁹ In addition, extensive first-pass metabolism in both the intestine and liver, as well as P-glycoprotein (Pgp)-mediated efflux are other corresponding factors for the weak bioavailability of berberine.¹⁰ To solve this shortcoming, various methods including the use of anionic surfactants, chitosan, P-gp inhibitors, and lipid microparticle delivery systems have been proposed.⁷

The question that arises is how a compound can have significant therapeutic capabilities, despite these weaknesses. In the past decades, there have been increasing efforts to understand the reasons for the therapeutic effects of berberine, despite its low plasma concentration. Since berberine and its metabolites exist simultaneously in vivo after absorption from the gastrointestinal tract, some studies state that part of the therapeutic effects of berberine are due to its metabolites.^11,12 These metabolites have been identified in different studies, all of which have structures similar to those of berberine with the same backbone. In a study conducted by Ma et al., the excretion of berberine and its metabolites in rat after oral administration (200 mg/kg) were investigated using liquid chromatography coupled with ion trap time-of-flight mass spectrometry. In this study, 16 metabolites including ten phase I metabolites and six phase II metabolites were identified (in bile, urine, and feces) after administration of berberine in vivo, among them hydroxyl-containing analogues thalifendine (M1) and berberrubine (M2) were the main metabolites (Figure 2A).¹³ Since the identification of these metabolites in the used method is based on their molecular mass, more structures can be considered as berberine metabolites. For example, in metabolites M4 (jatrorrhizine) and M6 (columbamine) with m/z = 338 the hydroxyl group can be located in any of positions 2, 3, 9, and 10. Therefore, in addition to these structures (jatrorrhizine and columbamine), the structures palmaturbine (with a hydroxyl group at position 9) and dehydrocorydalmine (with a hydroxyl group at position 10) also have the same mass, but they have been omitted. This point is also true for other metabolites, so the actual number of berberine metabolites can be considered far more than 16.

Figure 2.

(A) Chemical structure of phase 1 and phase 2 metabolites identified for berberine, introduced by Ma et al.¹³ Phase I metabolites are marked in blue, and phase II metabolites are marked in red. (B) Chemical structure of berberine metabolites containing sulfate functional groups with high water solubility potential, including thalifendine-10-O-sulfate (1), demethylenberberine-2-O-sulfate (2), jatrorrhizine-3-O-sulfate (3), 3,10-demethlpalmatine-10-O-sulfate (4), and 2,3,10-trihydroxyberberine-2-O-sulfate (5).¹⁴

Recently, in a study conducted by Feng and his colleagues, the pharmacokinetic and excretion profiles of berberine and its nine metabolites (including berberrubine, demethyleneberberine, jatrorrhizine, jatrorrhizine-3-O-β-d-glucuronide, jatrorrhizine-3-O-sulfate, thalfendine-10-O-β-d-glucuronide, berberrubine-9-O-β-d-glucuronide, demethyleneberberine-2-O-sulfate, and demethyleneberberine-2-O-β-d-glucuronide) were investigated in rats after a single intravenous administration (4.0 mg/kg) and oral administration (48.2, 120, or 240 mg/kg) of berberine in rats.¹² The results of this study showed that berberine (with absolute bioavailability of 0.37 ± 0.11%) is rapidly metabolized after entering the body, and all nine metabolites (mentioned above) are simultaneously present in vivo. Another important result of this study was that phase II metabolites were much higher than phase I metabolites (according to AUC_0–48h values), indicating that phase II metabolites are the main metabolites in blood circulation. In total, phase II metabolism of berberine is often proceeds through glucuronidation and sulfation of hydroxyl groups created in the first phase of metabolism. These secondary metabolites, which are much more water-soluble than the primary metabolites, are often excreted through the kidneys and urine. Figure 2B shows the chemical structure of some important berberine metabolites containing sulfate functional groups including thalifendine-10-O-sulfate (1), demethylenberberine-2-O-sulfate (2), jatrorrhizine-3-O-sulfate (3), 3,10-demethlpalmatine-10-O-sulfate (4), and 2,3,10-trihydroxyberberine-2-O-sulfate (5) (Figure 2).¹⁴ In addition, the results showed that the major metabolites of phase I/II metabolism are berberrubine and berberrubine-9-O-β-d-glucuronide, respectively.¹² In a review study reported by Wang et al., the metabolism of berberine was evaluated in terms of its metabolic pathways/organs based on the identified metabolites as well as the pharmacological activities of its active metabolites.¹⁵ After reviewing the results of a number of studies, the authors identified the dominant metabolic pathways of berberine as demethylation, demethylenation, reduction, hydroxylation, and subsequent conjugation in vivo. They also state that the pharmacological effects of berberine are not only caused by this compound alone, and berberine together with its active metabolites such as columbine and berberrubine exert a wide range of effects such as anti-inflammatory and antimicrobial effects in the body.

In conclusion, the most important metabolites of berberine are berberrubine, palmatine, jatrorrhizine, columbamine, and thalifendine, all of which play a significant role in the therapeutic effects of berberine against diseases.^13,14,16 Like berberine, other protoberberines are also metabolized through demethylation, demethylenation, hydroxylation, glucuronidation, and sulfation in vivo.¹⁷

3. Synthesis and Biosynthesis

In past studies, various synthetic methods have been reported for berberine, some of which did not have high overall yields.¹⁸ Here we review two of the recent and most important yields with remarkable overall yields. In 2014, an efficient synthetic method for the production of protoberberine alkaloids was reported by Gatland et al., using palladium-catalyzed enolate arylation (Figure 3A).²⁰ As shown, this synthetic route begins with the preparation of 2-(6-bromo-2,3-dimethoxyphenyl)-1,3-dioxolane (intermediate 7) by methylation and acetal protection of the benzaldehyde compound 6-bromo-2-hydroxy-3-methoxybenzaldehyde (starting material 6) in high yield (98%) over two steps. In a parallel reaction, intermediate 9 (2-(6-acetylbenzo[d][1,3]dioxol-5-yl)ethyl pivalate) was synthesized through a three-step reaction by BH₃ reduction of compound 8 (2-(benzo[d][1,3]dioxol-5-yl)acetic acid) and subsequent protection of the free alcohol moiety as the pivalate ester. in the last step for the production of intermediate 9, the Friedel–Crafts acylation in the presence of Ac₂O and ZnCl₂ obtains the target intermediate in remarkable yield (73%). In the next step, through the reaction of intermediates 7 and 9 in the presence of Cs₂CO₃ and 5 mol % [(Amphos)₂PdCl₂], the coupling reaction was performed well and the ketone intermediate 10 was obtained with a remarkable yield (84%). In the last step of the reaction of intermediate 5 with NH₄Cl in EtOH/H₂O solvent and at a temperature of 90 °C, aromatization was first performed to provide the desired isoquinoline-containing intermediate 11, and then by increasing the reaction temperature from 90 to 110 °C, berberine produced in high yield (82%). In total, this synthesis method was carried out in 5 steps with an overall yield of 50%, which is remarkable.²⁰

Figure 3.

Demonstration of two efficient synthesis routes for the production of berberine in high yields, conducted by Gatland et al. (route A, overall yield: 50%)²⁰ and Mori-Quiroz et al. (route B, overall yield: 54%).²¹

Another high-yield synthetic method for berberine production was reported in a 2018 study by Mori-Quiroz et al. (Figure 3B).²¹ Since this method provides the final product in only four steps with a good overall yield of 54%, the researchers of this study claim that this method is the most concise and efficient synthesis route of berberine to date. The first step of this method begins through a Pictet–Spengler reaction by the interaction of 3,4-(methylenedioxy)phenethylamine 12 with the acetal compound 2,2-dimethoxyacetaldehyde in the presence of trifluoroacetic acid and yields intermediate 5-(dimethoxymethyl)-5,6,7,8-tetrahydro-[1,3]dioxolo[4,5-g]isoquinoline 13 with a yield of 69%. In the next step, through the reductive amination of intermediate 13 with the reagent of 2,3-dimethoxybenzaldehyde 14, intermediate 15 was obtained with a yield of 79%. The interaction of intermediate 15 with triflic acid in dichloromethane solvent and room temperature through a Friedel–Crafts alkoxyalkylation leads to the closure of the saturated ring in the middle part of structure 16, which then gives the compound lambertine (17) through the elimination of the methoxy moiety. In the final stage, lambertine oxidation using iodine in the presence of potassium acetate produces the target compound berberine with a very high yield of 99%.²¹

In addition to the difficult/costly synthetic laboratory methods, berberine is also extracted from natural sources. Today, many plants such as Berberis vulgaris have been identified that can biosynthesize berberine as a metabolite in significant quantities; therefore, this important alkaloid can be obtained through extraction and isolation processes from medicinal plants. Recently, Prajwala and his colleagues in a brief review study presented a list of plants that contain berberine along with its diverse biological activities.²² The path of berberine biosynthesis in plants is a multistep and almost long course that starts from smaller molecules such as tyrosine, tyramine, and dopamine and finally reaches the final isoquinoline quaternary molecule berberine (Figure 4). Recently, some studies have shown/established this path.²³ Since endogenous small molecules such as tyrosine and dopamine are factors with wide practical and structural applications in living organisms, (S)-norcoclaurine (18) should be considered as an important primary precursor of berberine, which is constructed by the enzyme (S)-norcoclaurine synthase from dopamine and 4-hydroxyphenylacetaldehyde. This molecule turns into the intermediate (S)-reticuline (22) in four consecutive steps of O-methylation (biosynthesis of compound 19, by the use of enzyme (S)-norcoclaurine-6-O-methyltransferase), N-methylation (biosynthesis of compound 20, by the use of enzyme (S)-coclaurine N-methyltransferase), hydroxylation (biosynthesis of compound 21, by the use of (S)-N-methylcoclaurine 3′-hydroxylase), and O-methylation (biosynthesis of compound 22, by the use of 4′-methyltransferase), respectively (Figure 4). The conversion of (S)-reticuline to (S)-secoulerine (23) can be considered the most important step in the biosynthesis of berberine (and similar alkaloids), where the berberine bridge enzyme (BBE) closes a saturated ring in the middle part of the reticuline structure.²⁴ The next two steps are the conversion of scoulerine to (S)-tetrahydrocolumbamine (24) and then the construction of (S)-canadine (25), which are catalyzed by the enzymes scoulerine 9-O-methyltransferase and sanadine synthase CYP719A21, respectively.

Figure 4.

Berberine biosynthesis pathway starts with the small biomolecule tyrosine. As highlighted in blue in the figure, (S)-norcoclaurine is the first specific precursor of protoberberines, which is constructed through the coupling of dopamine and 4-hydroxyphenyl, catalyzed by (S)-norcoclaurine synthase. The conversion of (S)-reticuline to (S)-secoulerine is the most key step of berberine biosynthesis because, in this step, reticuline oxidase makes the main backbone of the molecule by closing a saturated ring in the middle of the structure. Considering the key role of this enzyme in berberine biosynthesis, the gene encoding reticuline oxidase is an important target for metabolic engineering studies in medicinal plants.

Since the extraction of biologically active compounds from plants relies on laborious and time-consuming agricultural processes and, on the other hand, the chemical synthesis of natural molecules is expensive and challenging due to the complexity of these structures, scientists are always looking for cheaper and easier ways to achieve these molecules. One of the attractive and developing methods for this purpose is the microbial manufacturing of these molecules, which is generally a more efficient, cost-effective, and environmentally friendly method. In recent years, several studies have been conducted to improve the efficiency of the microbial production of berberine and its derivatives, which have generally produced significant results. For example, in a recent study by Han and his colleague Li, they developed complete berberine biosynthesis in Saccharomyces cerevisiae by engineering 19 genes including 12 heterologous genes from plants and bacteria.²⁵ In this study, they increased the berberine titer 643-fold to 1.08 mg/L by inducing the overexpression of bottleneck enzymes, increasing the fermentation scale, and postfermentation heat treatment, which can strengthen the supply chain of berberine and advance the development of the production of such compounds.

Another strategy that scientists are interested in to obtain more valuable bioactive compounds such as berberine is the manipulation of some genes encoding plant enzymes through metabolic engineering strategies.²⁶ For example, in a study by Huang et al., by modulating the genes encoding the BBE, they created changes in benzylisoquinoline alkaloid biosynthesis, resulting in much higher amounts of berberine (and some other alkaloids) in plant Macleaya cordata.²⁷ This enzyme plays a key role in closing one of the saturated rings in the intermediates of the berberine production biosynthesis path (converting (S)-reticuline to (S)-secoulerine). Since the main hydrocarbon backbone of berberine is created at this stage, today the gene encoding BBE is considered by scientists as an important target for metabolic engineering works.

4. Antiviral Activity

Isoquinoline alkaloids make up a well-known class of natural bioactive compounds, which are mostly considered for their impressive antiviral and anti-inflammatory activities. In our recent studies, we have focused on the description of antiviral activities and therapeutic capabilities of different derivatives of these alkaloids against severe acute respiratory syndrome coronavirus (SARS-CoV)-2 infectivity.^3,28 Berberine is the most famous member of this category, which is mostly considered for their impressive antiviral activities. Since a review study has recently been published regarding the antiviral activity of berberine,²⁹ we only mention here the results of some studies published on the broad-spectrum antiviral activity of berberine in the past decade. Furthermore, we summarize the in vitro data of the past decade on the antiviral activity and cytotoxicity of berberine in Table 1.

Table 1. Summary of the Last Decade’s In Vitro Data (2010-2023) on the broad-spectrum antiviral activity of the berberine Alkaloid Including 50% Maximal Effective Concentration (EC₅₀), 50% Maximal Inhibitory Concentration (IC₅₀), 50% Cytotoxicity Concentrations (CC₅₀), Selectivity Index (SI), along with the Mechanism of Action and Cell Lines Used (Studies Are Listed in Chronological Order).

Virus type	Antiviral mechanism of action	IC50 or EC50	CC50	SIa	Cell line	ref
Herpes simplex virus-1 (HSV-1)	Inhibition of viral DNA synthesis	IC50 = 0.082 (mg/mL)	13.2 (mg/mL)	161	Vero E6	(Chin et al., 2010)30
HSV-2	Inhibition of viral DNA synthesis	IC50 = 0.090 (mg/mL)	13.2 (mg/mL)	147	Vero E6	(Chin et al., 2010)30
Infuenza A H1N1, (strains PR/8/34 and WS/33)	Inducting post-translationally effects in the virus life cycle to inhibit virus protein trafficking/maturation.	IC50 (strain PR/8/34) = 0.01 μM	-	-	RAW 264.7	(Cecil, Davis, Cech, Laster, 2011)48
		IC50 (strain WS/33) = 0.44 μM	-	-	RAW 264.7
Infuenza A/FM/1/47(H1N1)	Suppression of virus-induced cytopathogenic effects (CPE) and reduction of viral neuraminidase activity	IC50 (g/L) = 0.025	0.242 (g/L)	9.69	MDCK	(Wu et al., 2011)49
HIV-1 NL4.3	Inhibiting reverse transcriptase (RT) viral enzyme	EC50 = 0.13 μM	2.09 μM	16.07	CEM-GFP	(Bodiwala, Sabde, Mitra, Bhutani, Singh, 2011)50
HSV-1	Suppressing HSV-induced host cell JNK and NF-kB activation, as well as disrupting early stage of the HSV-1/2 replication cycle (between viral attachment/entry and genomic DNA replication).	EC50 = 6.77 μM	>400 μM	>59.08	HEC-1-A	(Song et al., 2014)51
		EC50 = 6.77 μM	165.7 μM	24.48	HEK293T
HSV-2	Suppressing HSV-induced host cell JNK and NF-kB activation, as well as disrupting early stage of the HSV-1/2 replication cycle.	EC50 = 5.04 μM	>400 μM	>79.37	HEC-1-A	(Song et al., 2014)51
		EC50 = 5.04 μM	165.7 μM	32.88	HEK293T
Chikungunya virus (CHIKV)	Inhibition of viral replication in a dose-dependent manner by reducing viral protein expression.	EC50 = 1.8 μM	>100 μM	>55.6	BHK-21	(Varghese, Kaukinen et al., 2016)52
		EC50 = 1.9 μM	>100 μM	>52.6	Huh-7.5
CHIKV	Decreased viral RNA synthesis and protein expression in a dose-dependent manner.	EC50 (strain LR2006 OPY1) = 4.5 μM	202.6 μM	45	HEK-293T	(Varghese, Thaa et al., 2016)53
		EC50 (strain LR2006 OPY1) = 12.2 μM	429.5 μM	35	HOS
		EC50 (strain LR2006 OPY1) = 35.3 μM	-	-	CRL-2522
		EC50 (strain CHIKV-SGP11) = 44.2 μM			CRL-2522
		EC50 (strain CNR20235) = 50.9 μM			CRL-2522
O’nyong nyong virus (ONNV)	Decreased viral RNA synthesis and protein expression in a dose-dependent manner.	EC50 = 29.2 μM	-	-	CRL-2522	(Varghese, Thaa et al., 2016)53
Enterovirus 71 (EV71)	Inhibiting EV71 replication (via reduction of viral RNA and protein synthesis as well as inhibiting MEK/ERK pathway), and suppressing virus-induced autophagy (by activating AKT protein and inhibiting JNK and PI3KIII phosphorylation).	IC50 (strain H) = 8.55 μM	73.10 μM	8.55	Vero E6	(Wang et al., 2017)35
		IC50 (strain SHZH98) = 10.25 μM	73.10 μM	7.13	Vero E6
		IC50 (strain JS-52) = 7.43 μM	73.10 μM	9.84	Vero E6
		IC50 (strain BrCr) = 7.43 μM	73.10 μM	9.84	Vero E6
EV71	Inhibition of activation of MEK/ERK signaling pathway and suppression of EV71-induced autophagy by activation of AKT and inhibition of phosphorylation of JNK and PI3KIII proteins.	IC50 (strain S) = 21.2 μM	146.9 μM	6.93	Vero E6	(Wang et al., 2018)54
		IC50 (strain JS-52) = 14.0 μM	147.0 μM	10.5	Vero E6
		IC50 (strain SHZH98) = 28.2 μM	147.2 μM	5.22	Vero E6
		IC50 (strain BrCr) = 9.45 μM	147.4 μM	15.6	Vero E6
Hepatitis C Virus (HCV)	Inhibiting HCV binding and entry/fusion steps without inactivating free virus particles or affecting expression of host cell entry factors and postentry viral replication	EC50 = 7.87 μM	82.75 μM	10.51	Huh-7.5	(Hung et al., 2019)36
Human cytomegalovirus (HCMV); Different strains	Inhibiting transactivating functions of the viral transcription factor immediate-early 2 (IE2), thus impairing efficient E gene expression and the progression of the viral replication cycle.	EC50 (strain AD169) = 2.65 μM	390 μM	147	HFF	(Luganini et al., 2019)37
		EC50 (strain TB40-UL32-EGFP) = 2.70 μM	390 μM	144	HFF
		EC50 (strain VR1814) = 4.00 μM	390 μM	98	HFF
		EC50 (strain 388438U) = 1.30 μM	390 μM	300	HFF
Murine cytomegalovirus (MCMV)	Inhibiting the activity of the MCMV transcription factor IE3.	EC50 = 1.95 μM	192 μM	98	NIH 3T3	(Luganini et al., 2019)37
Influenza A H3N2 (strain A/Hong Kong/4801/2014)	Upregulating the MAPK/ERK pathway and hijacking this pathway for nucleolar export of the viral ribonucleoprotein.	IC50 = 17 μM	107 μM	6	A549	(Botwina et al., 2020)38
		IC50 = 52 μM	1053 μM	20	MDCK
		IC50 = 4 μM	521 μM	123	LET1
		IC50 = 16 μM	-	-	HAE
HIV-1 (Different HIV-1 pseudotyped viruses)	Disruption of HIV-1 envelope glycoprotein function and inhibition of virus-mediated cell–cell fusion (inhibiting the fusion process of viral entry into host cells).	IC50 (strain IIIB) = 6.356 (μg/mL)	107 (μg/mL)	-	-	(Shao, Zeng, Tian, Liu, Fu, 2020)55
		IC50 (strain NL4–3) = 5.574 (μg/mL)		-	-
		IC50 (strain Bal) = 5.586 (μg/mL)		-	-
		IC50 (strain JRFL) = 10.275 (μg/mL)		-	-
		IC50 (strain JRCSF) = 7.559 (μg/mL)		-	-
		IC50 (strain AD8) = 7.532 (μg/mL)		-	-
SARS-CoV-2	-	IC50 = 10.58 μM	>400 μM	>37.84	Vero E6	(Pizzorno et al., 2020)43
SARS-CoV-2	Inhibition of viral replication by disrupting the late viral infection cycle and by targeting MAPK pathways (ERK, JNK, and p38 MAPK).	EC50 = 9.1 μM	149.2 μM	16.4	Vero E6	(Varghese et al., 2021)44
Dengue Virus (DENV)	Two distinct mechanisms of action include direct virucidal activity, as well as inhibiting infectious virus production through an effect on adenosine-monophosphate activated kinase (AMPK) activation and lipid metabolism.	IC50 = 42.87 μM	-	-	BHK-21	(Ratanakomol, Roytrakul, Wikan, Smith, 2021)39
		EC50 = 3.66 μM	115 μM	31.42	BHK-21
		-	79.54 μM	-	A549
		-	8.24 μM	-	HEK293T/17
		-	70.98 μM	-	Huh-7
		-	244.1 μM	-	HepG2
Zika virus (ZIKV)	-	IC50 = 11.42 μM	-	-	BHK-21	(Ratanakomol et al., 2021)39
CHIKV	-	IC50 = 14.21 μM	-	-	BHK-21	(Ratanakomol et al., 2021)39
Cyprinid herpesvirus 2 (CyHV-2)	Inhibiting the viral gene transcription and suppressing the viral replication in a dose-dependent manner	EC50 = 9.74 μg/mL	>25 μg/mL	>2.57	RyuF-2	(Su, Tang, Wang, Lu, 2021)40
Herpes simplex virus 1 (HSV-1)	Inhibition of transcription and translation of genes related to HSV-1 activity (gD, ICP-4, ICP-5, and ICP-8) and suppression of phosphorylation of MAPK proteins (JNK and p38).	IC50 = 45.6 μM	405.11 μM	8.9	HEK293T	(Cui, Zhang, Hu, Yang, 2022)56
EV71	Preventing the viral replication stage by regulating the Keap1- Nuclear factor erythroid 2-related factor 2 (Nrf2) axis, resulting in the abolishing of virus replication and inflammation.	IC50 = 2.79 μM	>100 μM		U251	(Cui et al., 2022)57
		IC50 = 4.03 μM	>100 μM		SK-N-MC
		IC50 = 6.83 μM	>100 μM		A549
DENV (different strains)	Suppression of the formation of intracellular and extracellular infectious viral particles (without affecting the early events of the viral replication cycle or viral protein expression), as well as inhibiting the activation of ERK1/2 and p38 MAPK	IC50 (strain Hawaii) = 9.1 μM	>200 μM	>21.9	Vero E6	(Giannone et al., 2023)58
		IC50 (strain NGC) = 2.4 μM	>200 μM	>83.3	Vero E6
		IC50 (strain H87) = 11.6 μM	>200 μM	>17.3	Vero E6
		IC50 (strain 8124) = 6.2 μM	>200 μM	>32.3	Vero E6
		IC50 (strain NGC) = 2.7 μM	89.9 μM	33.3	A549
		IC50 (strain NGC) = 3.5 μM	>200 μM	>57.1	HepG2
ZIKV (different strains)	Inhibiting the activation of ERK1/2 and p38 cell signaling pathways caused by ZIKV and reducing p38 MAPK phosphorylation	IC50 (strain INEVH116141) = 5.7 μM	>200 μM	>35.1	Vero E6	(Giannone et al., 2023)58
		IC50 (strain PRVABC59) = 0.7 μM	>200 μM	>285.7	Vero E6
		IC50 (strain DAK-AR-41524) = 3.8 μM	>200 μM	>52.6	Vero E6
		IC50 (strain INEVH116141) = 5.6 μM	89.9 μM	16.1	A549
		IC50 (strain INEVH116141) = 5.6 μM	>200 μM	>35.7	HepG2

^aThe SI is a ratio that measures the window between cytotoxicity and antiviral activity by dividing the given CC₅₀ value by the EC₅₀ value (EC₅₀/CC₅₀).

In a study, Chin and co-workers have shown the antiherpes simplex virus (HSV) effects of berberine. They illustrated that berberine is active against both HSV-1 and HSV-2 and inhibits viral DNA synthesis by intercalating into DNA.³⁰ In a study conducted by Zha et al., it was shown that berberine can remarkably inhibit human immunodeficiency virus (HIV) protease inhibitor (PI)-induced tumor necrosis factor-alpha (TNF-α) and IL6 expression by modulating endoplasmic reticulum (ER) stress signaling pathways in murine macrophages.³¹ In an experiment conducted by Mahata et al., it was shown that berberine is a potentially promising compound for the treatment of cervical cancer infected with HPV, by the mechanism of targeting both the host and viral factors responsible for the development of cervical cancer by blocking activator protein-1 (AP-1) and repressing viral oncoproteins E6 and E7 expression.³² A study by Shin et al. showed that berberine dramatically decreased the replication of the respiratory syncytial virus (RSV) by decreasing viral protein and mRNA syntheses, and induction of inhibitory effect on Toll-like receptor 4 (TLR4) activation which leads to suppression of p38 mitogen-activated protein kinases (p38 MAPK) activation.³³ A study conducted by Kim et al. showed that the major compound of Cortex Phellodendri extract, berberine, has promising antiviral properties against a wide range of viruses including influenza A subtypes (H1N1, H5N2, H7N3, and H9N2), Newcastle Disease Virus (NDV), Vesicular Stomatitis Virus (VSV), HSV, EV-71, and Coxsackie Virus (H3-GFP). The researchers stated that berberine induces mRNA expression of antiviral genes and cytokine secretion, modulates the production of pro-inflammatory cytokines, and stimulates the antiviral state in infected host cells via type I interferon (IFN) induction.³⁴ Wang and colleagues reported that berberine can suppress the EV-71 replication through the mechanisms of downregulating the phosphorylation of the mitogen-activated protein kinase kinase (MEK)/extracellular signal-regulated kinase signaling pathway, inhibiting enterovirus 71 (EV71)-induced autophagy, and blocking the phosphorylation of c-Jun N-terminal Kinase (JNK) and phosphoinositide 3-kinase III (PI3KIII).³⁵

The remarkable antiviral activities of berberine have also been of great interest in recent studies (2019–2023). In a study by Hung et al., it was reported that berberine blocks hepatitis C virus (HCV) attachment and inhibits N-terminal kinase its entrance to the cell through the interaction with the HCV E2 glycoprotein which in turn can suppress HCV replication.³⁶ A 2019 study by Luganini et al. showed that berberine inhibits the replication of human cytomegalovirus (HCMV) by interfering with the activation function of the HCMV immediate-early 2 (IE2) protein, which plays an important role in the progression of viral replication and viral pathogenesis. In this study, it was shown that low micromolar concentrations of berberine are effective in suppressing the proliferation of different strains of HCMV. Mechanistic evaluations also showed that berberine compromised viral cycle progression at a stage prior to viral DNA replication and Early (E) gene expression but after IE protein expression.³⁷ Another recent study by Botwina demonstrated the significant anti-influenza A activity of berberine at nontoxic concentrations through the mechanism of inhibition of the MAPK/ERK1 pathway, which is required for the transport of viral ribonucleoproteins to the cytoplasm. Moreover, it was illustrated that berberine did not interfere with the virus itself, virus entry, trafficking, and genome replication while it affects virus assembly, maturation, or egress suggesting that berberine blocks the influenza A virus replication at late stages of infection.³⁸ In a 2021 study by Ratanakomol et al., berberine was reported to have a direct virucidal effect as well as reducing new virion production against dengue virus (DENV) through a postreplication mechanism. The results demonstrated that berberine has cellular effects that lead to an increase in cellular DENV E protein without concomitant impact on DENV nonstructural proteins (NSPs), indicating an effect on viral particle formation or egress.³⁹ Another 2021 study by Su et al. reported a significant suppressive effect of berberine on the replication of cyprinid herpesvirus 2 (CyHV-2) in vitro and in vivo. The results of the evaluations stated that berberine systematically prevents viral gene transcription and suppresses the replication of CyHV-2, causing the suppression of viral replication, reducing viral pathogenesis, and increasing survival rates in tested animals.⁴⁰ Recently, in a 2023 study conducted by Seteyen et al., it was reported that berberine has valuable effects in fighting O’nyong–nyong virus (ONNV) by inhibiting the replication and regulating the type I IFN antiviral signaling pathway and inflammatory pathways and mediators.⁴¹

Because of the strong antiviral effects of berberine against a wide range of viruses, some researchers have decided to test this alkaloid against the newly emerging virus SARS-CoV-2. In recent years, several review/research/trial studies have been published on the capabilities of berberine in the fight against SARS-CoV-2 and the treatment of coronavirus disease of 2019 (COVID-19);^42,43 here we will have a brief overview of the results of some of the studies. In an in vitro study conducted by Varghese et al., the antiviral activity of berberine against SARS-CoV-2 was assessed using Vero E6 and nasal epithelial cells. They illustrated that this alkaloid is effective against two different SARS-CoV-2 isolates from Bavaria and Nijmegen1 at low micromolar concentrations. Conducting a dose–response assay in infectious titers showed that berberine has anti-SARS-CoV-2 effects with 50% effective concentration (EC₅₀) values of 9.1 μM and 2.1 μM for Bavaria and Nijmegen1 isolate form, respectively. Time-of-addition evaluations also showed that berberine acts in the late stage of the viral life cycle.⁴⁴ In a 2021 study done by Rodriguez-Rodriguez et al., it was shown that berberine chloride could reduce SARS-CoV-2 replication dose-dependently in Vero E6 cells suggesting an impact on the late phase of the viral life cycle.⁴⁵ By reviewing the various studies by Babalghith and his colleagues, it was concluded that berberine can be used as a possible anti-SARS-CoV-2 drug due to its strong anti-inflammatory, antioxidant, and antiviral effects.⁴⁶ We recently argued that despite the extensive efforts made in the fight against COVID-19, pharmacotherapy and drug discovery of this disease have not yet been successful, and efforts need to be pursued seriously.⁴⁷ Considering the significant antiviral and anti-inflammatory effects of protoberberines, structural modification of these antiviral alkaloids is one of the research fields with high potential to find a treatment for COVID-19, which can be given more attention by researchers.

The studies reviewed here show that the natural molecule berberine has abundant antiviral capabilities against a wide range of viruses. In terms of the mechanism of action (as shown in the second column of Table 1), berberine counteracts the infectivity of different viruses by exerting direct antiviral effects and also by modulating some host-based targets, such as NF-kB and MAPK signaling pathways. In addition, through the exertion of strong anti-inflammatory effects that are often manifested by inhibiting the release of inflammatory factors, this compound shows a high potential for clinical applications against viral diseases (especially in acute cases). As a result, this molecule can be a suitable structural platform for use in structural optimization studies in medicinal chemistry with the aim of obtaining more effective compounds with a more optimal activity and toxicity profile.

5. Development of Semisynthetic Analogues with Antiviral Activities

With the understanding of the dangers caused by viruses to human life in recent years, the design and development of new antiviral compounds have recently become more important.⁵⁹ With the aim of finding new antiviral molecules, medicinal chemists continuously conducted various studies, some of which used semisynthetic strategies for the derivatization of the berberine platform. Here, an attempt is made to describe the semisynthetic methods used for the production of antiviral berberine-based structures along with the reported pharmacological results by describing the most important studies conducted in this field.

In a study conducted by Bodiwala and co-workers, a new semisynthetic series of berberine derivatives (general structure 29, Figure 5) was prepared and then evaluated as anti-HIV agents.⁵⁰ The Structural modification was done at the C-9 position of the berberine by converting the berberine chloride to berberrubine, which in the next step converted to berberine-9-ester derivatives by reacting with different acid chlorides in acetonitrile solvent. All newly synthesized compounds first were tested for their cytotoxicity in the CEM-GFP cell line by an MTT assay. In complementary assessments, three of the most active compounds with more than 70% inhibition of HIV replication in CEM-GFP cells were selected for further cell-based anti-HIV activity. Berberine chloride with EC₅₀ = 0.13 μM and berberrubine (which is known as the most important in vivo metabolite of berberine^13,15) with EC₅₀ = 2.8 μM showed promising results compared to azidothymidine (AZT) as a positive control with EC₅₀ = 1.05 μM. The structural modifications on the C-9 position of berberine core revealed that attachment of unsubstituted phenyl, napthoyl, and adamantoyl rings decrease the anti-HIV activity of novel compounds while phenyl ring with electron-withdrawing groups or compounds with less bulky heterocyclic substituents had more potent anti-HIV results. Among tested compounds, two molecules with thiophene (compound 27) and 2,4-difluorophenyl substitutions (compound 28) showed promising results with EC₅₀ = 1.82 and 0.95 μM respectively and therapeutic index (TI) more than 50. To further explore the mechanism of action, a colorimetric assay was performed to analyze the anti-HIV reverse transcriptase (RT) activity of the most active compounds, berberine, 26, 27, and 28 (Figure 5) at a concentration of 20 μg. Results of enzymatic tests showed 99% inhibition of HIV-1 RT for three compounds 26–28, berberine showed 94% inhibition at the same concentration of 20 μg, but the standard drug nevirapine showed 95.5% inhibition at an even lower concentration of 0.1 μg. Since berberine (EC₅₀ = 0.13 μM) and its active derivatives 26–28 strongly inhibit the virus replication compared to the standard drug azidothymidine (EC₅₀ = 1.05 μM), but it inhibits the RT enzyme with less potency in the enzymatic test, the researchers concluded that RT inhibition is not the only/main mechanism of anti-HIV activity of the tested compounds.⁵⁰ Obtaining more information in this regard requires additional mechanistic experiments that were not performed in this study.

Figure 5.

General synthesis route for the preparation of 9-substituted berberine analogs 29 as novel semisynthetic anti-HIV agents, as well as the chemical structures and cytotoxicity/anti-HIV activity of the most active compounds 26, 27, and 28 (cytotoxicity concentration 50% (CC₅₀) and EC₅₀ in μM) in the presence of lead compound berberine and standard drug azidothymidine. (A) Highest noncytotoxic concentration; (B) EC₅₀ = concentration of compound to achieve 50% inhibition of infected cells; (C) CC₅₀ = concentration of compound indicating 50% cytotoxicity in uninfected cells; (D) TI = CC₅₀/EC₅₀.⁵⁰

The synthesis method for the production of target compounds is shown in Figure 5. This route was started with vacuum pyrolysis of the starting material berberine at 190 °C and under reduced pressure (20–30 mmHg), which resulted in the production of 26 with a desirable yield. In the next step, the acylation of intermediate 26 in a polar aprotic solvent with various acid chlorides led to the production of final semisynthetic compounds with quantitative yield (92–96%).

In a study conducted by Wang et al., three series of semisynthetic 9-substituted ester and ether derivatives of quaternary benzylisoquinoline alkaloids were designed and synthesized, and then their anticytopathic effect (CPE) was evaluated against four different genotype EV71 strains including H, JS-52, SHZH98 and BrCr.⁵⁴ Natural alkaloids berberine, palmatine, and jatrorrhizine were used as starting materials in the synthesis of these categories (Figure 6). Based on the biological results, berberine-based ether-containing compounds (general structure 30) with different substituents on the phenyl ring had better antiviral activity, cytotoxicity effects, and selectivity index (SI) than ester-containing derivatives. In the case of palmatine- and jatrorrhizine-based derivatives with an opened methylenedioxy ring (structures 33–36), the same results were seen and repeated. Figure 7 shows the chemical structure of the most active compounds identified in this study (structures 37–41). Taking together with IC₅₀ and SI values, 9-ether containing derivative 39 showed broad-spectrum anti-EV71 activity with IC₅₀ ranging from 7.12 to 14.8 μM against all tested strains, lower cytotoxicity, and better SI than berberine hydrochloride as standard drug. Further analysis via Western blot assay revealed that the mentioned compound could inhibit the activation of the MEK/ERK signaling pathway and suppress the EV71-induced autophagy by activating AKT protein and inhibiting the phosphorylation of JNK and PI3KIII proteins.

Figure 6.

General synthesis route, reagents, and conditions for the preparation of target berberine-based derivatives 30, 31, and 33–36 as novel anti-EV71 agents.⁵⁴

Figure 7.

Chemical structures and antiviral activity (IC₅₀ in μM) of most active semisynthetic compounds 37–41 and well-known antiviral alkaloids berberine and jatrorrhizine against four different genotype EV71 strains including H, JS-52, SHZH98 and BrCr, in the in vitro situations. (A) TC₅₀ = median toxic concentration; (B) SI = TC₅₀/IC₅₀.⁵⁴

The synthesis path for the preparation of the target structures is shown in Figure 6. Based on previous reports, demethylation of the starting materials berberine and palmatine under vacuum pressure (30–40 mmHg) and 195 °C temperature for 1h gave hydroxyl-containing compounds 29 and 32 in 80–86% yield. Subsequently, alkylation of these intermediates (along with jatrorrhizine) using corresponding bromides and K₂CO₃ in DMF as a solvent at 70 °C to yield final etheric compounds 30, 33, and 35 with yields ranging from 20 to 58%. Similarly, the esterification of intermediates 29, 32, and jatrorrhizine with corresponding chlorides and K₂CO₃ in CH₃CN solvent at 70 °C led to the preparation of final compounds 31, 34, and 36 in 21–38% yields.⁵⁴

In a study by Enkhtaivan et al., a novel series of berberine-piperazine derivatives conjugated through a pentyloxy side chain were designed and synthesized, and their antiviral activities were examined against influenza A and B strains including A/PR/8/34 (H1N1), A/Vic/3/75 (H3N2), B/Lee/40, and B/Maryland/1/59, in the presence of oseltamivir as standard drug.⁶⁰ To evaluate the toxicity profile, the cytotoxicity of synthesized derivatives was also screened toward MDCK noncancer cell lines. The title compounds with different piperazine substitutions showed significant anti-influenza activity with IC₅₀ ranging from 35 to 89 μM, respectively. From the structure activity relationship (SAR) finding in this study, it could be concluded that a change in the functionalities of groups connected to the piperazine ring resulted in the alternation of the antiviral efficacies of synthesized molecules. Biological results revealed that structures with disubstitution electron donating groups present on the aryl ring of piperazine have better antiviral effects compared to other structures with mono- or disubstituted electron-withdrawing groups. Figure 8B depicts the chemical structures of the most active compounds identified in this study (compounds 44–48). Among them, 1-(2,4-dimethoxyphenyl)piperazine-containing derivative (compound 45) demonstrates the highest anti-influenza activity with IC₅₀ ranging from 29.17 to 35.16 μM against different viral strains with the best results against B/Maryland/1/59, and high TI of 110.65. In addition, the mentioned compound had promising inhibitory results against neuraminidase enzyme in a dose-dependent manner as 44.55% at 0.1 μg/mL, 21% at 1 μg/mL, and 18.5% at 10 μg/mL, respectively (which are comparable to the antineuraminidase activity of oseltamivir as a selective neuraminidase inhibitor, including 31.95% at 0.1 μg/mL, 19.9% at 1 μg/mL, and 13.65% at 10 μg/mL). The molecular docking studies indicated more than 20 interactions at the neuraminidase catalytic site with a higher binding energy (−8.2 kcal/mol) compared to oseltamivir (- 6.1 kcal/mol).⁶⁰

Figure 8.

(A) General synthesis route, reagents, and conditions for the preparation of semisynthetic berberine-based derivatives 43 as novel anti-influenza agents; (B) antiviral activity (IC₅₀ in μM) and chemical structures of the most active semisynthetic compounds 44–48 and standard antiviral drug oseltamivir against four different influenza strains including A/PR/8/34 (H1N1), A/Vic/3/75 (H3N2), B/Lee/40 and B/Maryland/1/59 in MDCK cells.⁶⁰

As shown in Figure 8A, for the synthesis of the target compounds of this study, berberine as the starting material was heated in a vacuum oven at 190 °C and 20–30 mmHg pressure for 40 min under a method described in the literature for producing a critical intermediate berberrubine, with a 85% yield.⁶¹ In the next step, when the 9-hydroxy group was alkylated under refluxing with dibromopentane in dry acetonitrile for 6h achieved intermediate 42. The synthesis process was followed by a nucleophilic substitution reaction under coupling with various substituted piperazines and heterocyclic cores such as morpholine, piperidine, and piperidine carbazole in the presence of potassium carbonate and DMF as base and solvent (at 80 °C for 6–8 h) to produce target compounds 43 in satisfactory yields.⁶²

In a more recent study, Wang’s research team designed and produced three novel series of semisynthetic molecules using berberine, palmatine, and jatrorrhizine as starting materials (similar to their previous study⁵⁴) with different substitutions on 3, 9, and 10 positions of 5,6-dihydroisoquinolino[3,2-a]isoquinolin-7-ium backbone (general structures 49–50 (Figure 9A), 58–59 (Figure 9C), and 61 (Figure 9D)), and analyzed their antiviral activities against six different genotypes Coxsackievirus B (CVB1–6) strains.⁶³ In addition to the three semisynthetic series, the researchers produced a fully synthetic series (Figure 9B, general structures 56–57) in order to evaluate the effect of the substitution on the 10 positions of the berberine on the antiviral activity. The structural modifications were done by attachment of different carbamates for the achievement of the target primary amines. In the case of C-9 and C-10 substituted derivatives, SAR findings indicate that those with primary amines had promising antiviral activities against all tested CVB (1–6) strains and a better SI compared with carbamate derivatives, but this difference is not so significant. In addition, the introduction of a primary amine by a linker at position 3 might have a beneficial impact on both antiviral activity and safety. In general, it can be said that the anti-CVB activity of the compounds in all four series is almost at the same level (with single-digit average IC₅₀ values) and there is not much difference between them. Figure 10 shows the chemical structure of the most active compounds (62–68) from each structural series, along with their antiviral activities. Among tested compounds, the best results were seen by 67 with IC₅₀ values ranging from 3.08–9.94 μM against all tested CVBs 2–6 strains and a satisfactory SI value of 34.3 on CVB3, better than that of berberine as the positive control. In order to explore the mechanism of anti-CVB3 inhibitory activity, further analysis was performed in both RNA and protein levels. The results confirmed that this compound (67) inhibits virus replication by down-regulating the expression of VP1 protein and VP1 RNA and also suppresses the phosphorylation of ERK, JNK, and p38 MAPK, which is important for CVB3 replication. Analyzing the results, the researchers stated that these berberine-based semisynthetic compounds can be considered a new class of anti-CVB agents with the advantage of broad-spectrum anti-CVB potency.⁶³

Figure 9.

General synthesis route, reagents, and conditions for the synthesis of target berberine-based derivatives 49–50, 56–57, 58–59, and 60–61 as novel anticoxsackievirus B agents.⁶³

Figure 10.

Chemical structures and cytotoxicity, antiviral activity (against six different genotypes coxsackievirus B (CVB1–6) strains) and SI of most active synthesized compounds 62–68 in the in vitro situations.

The synthetic route of target compounds is shown in Figure 9A–D. For the semisynthetic derivatives of berberine and palmatine (Figure 9A and D), first, the starting natural substance is subjected to dimethylation at the 9-position using a vacuum oven under reduced pressure of 30–40 mmHg and 195 °C for 1 h until it is converted to the corresponding hydroxylated intermediates 29 and 32. Alkylation of these intermediates with substituted p-toluene sulfonates (TsO-R-NHBoc) or Br-R-NHBoc in the presence of K₂CO₃ in acetonitrile solvent resulted in carbamate compounds (49 and 60, respectively) with a different yield. In the last step, the acidification of carbamate compounds using a mixture of CH₃OH/HCl gave the final compounds containing primary amines (general structures 50 and 61). The production of semisynthetic jatrorrhizine derivatives 58 and 59 is shorter compared to berberine and palmatine derivatives due to the lack of the first step of demethylation (Figure 9C). The next two steps of the synthesis of these compounds are completely similar to the synthesis methods of the two series of berberine and palmatine derivatives. As shown in Figure 9B, a fully synthetic method was used to produce berberine derivatives substituted at the 10-position (general structures 56 and 57), because the demethylation of berberine under vacuum/heat only results in the demethylation of the 9-position. For this purpose, commercially available compounds 3-hydroxy-2-methoxybenzaldehyde (A) and 2-(benzo[d][1,3]dioxol-5-yl)ethan-1-amine (B) were used as starting materials. After connecting the amine and aldehyde parts of these reagents by heating at 100 °C for 8 h, the first intermediate containing the imine group (51) is obtained. Reduction of the imine part by NaBH₄ in methanol under reflux conditions and then use of oxalaldehyde (glyoxal) and formic acid reagents in the presence of CuSO₄ and HCl causes the middle ring to close and the main scaffold of berberine to be formed (intermediates 52 and 53). The final steps of the synthesis of the target compounds 56 and 57 are similar to the methods described in the synthesis of semisynthetic derivatives of this study.⁶²

In a study by Nguyen et al., a few 13-benzylberberine derivatives containing aromatic substituents at the 13-position (Figure 11, general structures 70 and 71) were prepared and evaluated as anti-Zika virus (ZIKV) infections.⁶⁴ Based on the experimental cell-based assay, three compounds, 73–75 with IC₅₀ ranging from 5.3 to 9.8 μM, showed better inhibitory than berberine as the positive control (Figure 11). The best anti-ZIKA results were explored by 2,6-difluorobenzyl substituted compound 74 with the highest Anti-ZIKV activity and selectivity (IC₅₀ = 5.3 μM and SI > 15) compared to berberine as a positive control (IC₅₀ = 15.5 μM, and SI > 4.1). Also, a molecular docking study between the mentioned compounds in the active site of the ZIKV virus (NS2B-NS3 protease) recorded a docking score of −7.31 kcal/mol with a number of H-bounding and hydrophobic interactions which confirmed the highest binding affinity to this protease and exhibited vital role in the inhibition activity of the compound.

Figure 11.

General synthesis rout for the preparation of target compounds 70- 71 along with the chemical structures of the most active derivatives 72–75, presenting their antiviral and cytotoxic activities.⁶⁴

As illustrated in Figure 11, the preparation of target berberine derivatives was started by treating berberine chloride in a mixture of 5.0 M NaOH and acetone at 0 °C to produce the desired intermediate 69. In the next step, the reaction of compound 69 with various benzyl/heteroaromatic benzyl halides in the presence of CH₃CN solvent and NaI as the catalyst (at 80 °C) afford intermediates 70. Finally, target compounds 71 were synthesized by treating intermediates 70 with aqueous KOH (20%) in the air at 115 °C temperature.⁶⁴

In a recent study conducted by Kumar et al., several benzothiazole-containing of derivatives berberine (Figure 12, general structure 77) were designed and synthesized with the aim of potential effects as anti-influenza activity.⁶⁵ The synthesized compounds were evaluated against different strains of influenza A (A/PR/8/34, A/Vic/3/75) and influenza B (B/Lee/40, B/Maryland/1/59) viruses by applying an in vitro sulforhodamine B (SRB) bioassay. All title compounds revealed more significant antiviral results and bearable cytotoxicity against MDCK cell lines and therefore provided outstanding therapeutic indices that were comparable to the control drug oseltamivir and much better than the parent berberine molecule. SAR findings indicated that the existence of different electron-donating or -withdrawing functional groups on the benzothiazole ring resulted in variable degrees of antiviral effects. Based on the cell-based antiviral effects, unsubstituted benzothiazole derivative (compound 78), two compounds with electron-donating groups including methyl and methoxy (compounds 79 and 80), and two compounds with electron-withdrawing functional groups including mono- and difluoro (compounds 81 and 82) showed remarkable anti-influenza activity compared to the control drug oseltamivir. In evaluating the mechanism of action, the neuraminidase inhibitor activity of the target benzothiazole-berberine derivatives was evaluated, and the mentioned derivatives exhibited comparable neuraminidase inhibition to control oseltamivir drugs. These results confirmed well by in silico research using a molecular docking study between benzothiazole-berberine derivatives and neuraminidase protein. The obtained results suggest that these ligands act selectively on the active site of neuraminidase and assist in tighter binding and improved activity.

Figure 12.

(A) General synthesis route, reagents, and conditions for the preparation of semisynthetic berberine-based derivatives 77 as novel anti-influenza A/B agents; (B) antiviral activity (IC₅₀ in μM) and chemical structures of the most active semisynthetic compounds 78–82 and standard antiviral drug oseltamivir.⁶⁵

The synthesis method of the target compounds is shown in Figure 12. As shown, in the first step, the hydroxylated intermediate 29 was obtained similar to the previously described methods via demethylation of starting material berberine under reduced pressure (20–30 mmHg) and 190 °C temperature for 40 min with an 85% yield. Subsequently, compound 29 was refluxed with 1,5-dibromopentane in acetonitrile as a solvent to afford intermediate 76. In the final step, nucleophilic replacement in intermediate 76 with various benzo[d]thiazol-2-amine compounds under a reflux situation in the presence of K₂CO₃ in DMF solvent led to forming target compounds 77.⁶⁵

In a study by Enkhtaivan et al., twenty-three semisynthetic berberine derivatives containing piperazinyl substituents at the 12-position were synthesized as the novel anti-influenza agents (Figure 13, general structure 83), and their antiviral activities were evaluated against four different strains of influenza A and B virus (A/PR/8/34 (H1N1), A/Vic/3/75 (H3N2), B/Lee/40, and B/Maryland/1/59).⁶⁶ For this purpose, all newly synthesized compounds were screened first against the A/PR/8/34 strain of influenza virus in vitro using CPE reduction assay with the SRB colorimetric method. The results showed that the title compounds can significantly inhibit the A/PR/34/8 strain with remarkable IC₅₀ values ranging from 0.87 to 60.1 μg/mL, while their cytotoxicity level was observed in the desired range of 63.16 to1639 μg/mL. Among the tested compounds, seven synthesized derivatives (including compounds 84–87, Figure 13) showed better results than berberine and oseltamivir as positive controls with IC₅₀ less than 3.5 μM against the A/PR/8/34 strain. For further evaluations, most active compounds 84–87 were subjected to tests against the A/Vic/3/75 strain of influenza A, and two influenza B strains including the B/Lee/40 and B/Maryland/1/59. Interestingly, the results showed that all 4 compounds have stronger antiviral activity against all strains tested compared to the parent molecule berberine and standard drug oseltamivir (Figure 13). The structural activity relationship study confirmed that the attachment of electron-withdrawing groups such as fluoro and chloro on phenylpiperazine had beneficial effects on the antiviral activity of designed derivatives. The most promising results were recorded by 4-fluorophenyl-piperazine derivative (compounds 86) against all tested influenza strains which were 3 to 10 times more potent than oseltamivir as a standard drug. These promising results were further confirmed by a neuraminidase inhibitor activity assay and molecular docking study. The mentioned compound showed better neuraminidase inhibition activity (with 43.1% inhibition at 0.1 μg/mL concentration) and greater binding ability on viral neuraminidase subunit compared to commercial neuraminidase inhibitor drug oseltamivir.

Figure 13.

General synthesis route for the production of Mannich base berberine derivatives 83, along with the chemical structures, cytotoxicity, and antiviral activities (IC₅₀ in μg/mL) of the most potent berberine derivatives 84–87 against the different strains of the influenza viruses A and B.⁶⁶

The general synthesis pathway for preparing piperazine-based N-Mannich bases of berberine derivatives is described in the previous studies.⁶⁷ In the first step, hydroxyl-containing structure berberrubine could be obtained via heating commercial berberine chloride at 190 °C in a vacuum oven at reduced pressure (20–30 mmHg) for 40 min with 85% yield under selective demethylation at position 9. Then, a Mannich reaction happened by with formaldehyde and a broad range of heterocyclic cores such as substituted piperazines, morpholine, piperidine, and piperidine carbazole in ethanol as solvent at 80 °C or room temperature for 24h to give final compounds 83.⁶⁶

In a recent study conducted by Das and Srinivas, several semisynthetic berberine derivatives were synthesized by modification of the proto-BBR backbone by hydrogenation and/or insertion of substituents at positions 9 or 12, and tested against RSV strain A2. Cytotoxicity and antiviral properties of novel compounds were assessed using the HEp-2 cell line (HeLa derivative cell culture) in the presence of ribavirin as a standard drug. The structural modification was accomplished at various positions of the berberine by using different substituents. Figure 14 illustrates the chemical structures of the most active compounds recognized in this study (compounds 88 and 92–95). The results demonstrated that berberine and its 12-nitro analogue (88), as well as 9-O derivatives (92–95), have significantly stronger antiviral activities with IC₅₀ ranging from 3.4 to 6.7 μM compared to ribavirin (IC₅₀ = 31.1 αM) as the reference drug (Figure 14).⁶⁸

Figure 14.

(A) General synthesis route for the preparation of target compounds 88 and 91; (B) chemical structures, cytotoxicity, and anti-RSV activity of the most active derivatives 88 and 92–96, berbeine and its derivatives.⁶⁸

In terms of cytotoxicity, berberine and its hydrogenated derivatives (such as compound 96) in which a quaternary amine has been converted to a tertiary amine have the least cytotoxicity. Of course, when the middle rings of berberine are hydrogenated, the antiviral activity of berberine is significantly reduced or even lost. Compound 96 is the only analogue of berberine that despite the saturation of two middle rings (rings B and C) has a better situation in terms of total activity and toxicity. The important point is that the cytotoxicity (CC₅₀) of compounds 95–92 is significantly high and is in the range 10 αM, while the CC₅₀ of compound 88 with a −NO₂ group in position 12 is equal to 187.5, which is desirable. Among berberine derivatives, this compound has the best selectivity index (SI = 39) after berberine with SI = 69.6. Although compounds 92–95 have strong antiviral activity, they are not promising compounds for clinical use due to having a single-digit SI index. By summarizing the results, researchers introduced only compounds 88 and 96 as promising lead compounds for clinical applications due to having favorable SI indexes.⁶⁸

As illustrated in Figure 14A, berberine was utilized as a starting material for synthesizing the target compounds. Compound 88 was produced in 40% yield by treatment of concentrated nitric acid and berberine in glacial acetic acid at 10 °C temperature. Then, after completing the addition of nitric acid, the reaction temperature increased to room temperature and continued at this temperature for 8 h. To synthesize the final compounds 92–95, three consecutive steps were performed. First, berberrubine (89) was produced due to selective demethylation of the C-9 methoxy group under microwave irradiation for 5 min. In the next step, treatment of berberrubine 89 and 1,2-dibromoethane in DMF at 40 °C under a nucleophilic substitution reaction led to the generation of 9-O derivative 90 with 72% yield, which had a terminal halogen atom in the aliphatic moiety. In the final step, the thioether derivatives 91 were produced when intermediate 90 reacted with commercially available heteroaromatic thiols in ethanol solvent using sodium methylate as a base under stirring for 1 day.⁶⁷

6. Conclusion

Due to having simultaneous antiviral and anti-inflammatory effects, protoberberins are suitable candidates for use in the treatment of acute viral diseases. Berberine is the most well-known member of the protoberberines, which has been subjected to dozens of clinical trials in the past to evaluate its therapeutic capabilities in various diseases (https://clinicaltrials.gov/). Despite having significant biological activity, berberine does not have favorable pharmacokinetics or bioavailability, which can negatively affect its therapeutic index. On the other hand, due to its moderate cytotoxic profile, it may cause side effects in high doses. Therefore, it is important to find ways to reduce its therapeutic weaknesses and strengthen its capabilities.

Semisynthetic strategies have long been used in medicinal chemistry with the aim of improving the activity profile of natural compounds through reducing toxicity and increasing potency. In the last two decades (especially the past decade), berberine and its analogues have attracted the attention of medicinal chemists as starting materials for the design and development of semisynthetic antiviral agents. In most of the studies (as reviewed), the modification is done on position 9 of substrates berberine or palmatine, where the methoxy group is demethylated under reduced pressure, and then the active hydroxyl group becomes available for the next steps. Other structural positions that are targeted for modification in the backbone of protoberberines (5,6-dihydrodibenzo[a,g]quinolizinium) include positions 2, 3, 10, and 12. The structure–activity relationships reported in the reviewed studies indicate that in most cases, the placement of the same substitutions in the mentioned positions (especially 2, 3, 9, and 10) does not create a significant difference in the biological activity of the target semisynthetic compounds. These findings indicate the importance of the 5,6-dihydrodibenzo[a,g]quinolizinium structural platform as a bulky pharmacophoric scaffold in the occurrence of antiviral activities. Of course, it should not be overlooked that the structural modifications carried out in some studies have led to a significant improvement in the antiviral activity and safety profile of the target structures compared to the parent protoberberine alkaloids.

In some studies, jatrorrhizine alkaloid, which has a hydroxyl group in position 2, has also been used as a starting material, which shortens the synthesis path by one step. The use of other protoberberine structures containing a hydroxyl group as starting materials (such as berberrubine, thalifendine, palmatrubine, columbamine, and stepharanine) can be further considered in future studies, because it can reduce cost and time by eliminating the demethylation step. In addition, the use of such analogues can provide new information about the pharmacological potential of these compounds, which have received less attention in previous studies.

In general, the findings of the reviewed studies show that new semisynthetic compounds with more optimal biological activity and toxicity profile can be obtained through structural modification/optimization on the protoberberine substrates. Although the modifications made on protoberberines (especially berberine) have promising results in obtaining better antiviral agents, the studies done in this field have not been sufficient/extensive. In addition, in most of the studies only in vitro evaluation has been done, and the antiviral potential of new semisynthetic compounds has not been investigated in vivo (which should be further considered in future studies). In conclusion, we emphasize that there is a significant potential to obtain semisynthetic berberine derivatives with more optimal antiviral effects, which should be given more attention in future studies.

Glossary

Abbreviations

AP-1

Activator Protein-1

AZT

Azidothymidine

AMPK

Adenosine-monophosphate activated kinase

AUC

Area under curve

BBE

Berberine Bridge Enzyme

CC₅₀

50% Cytotoxicity Concentration

CMV

Cytomegalovirus

COVID-19

Coronavirus Disease of 2019

CHIKV

Chikungunya virus

CPE

Cytopathic Effect

CyHV-2

Cyprinid Herpesvirus 2

CVB1–6

Coxsackievirus B1–6

DENV

Dengue virus

EC₅₀

50% Effective Concentration

EV71

Enterovirus 71

ER

Endoplasmic Reticulum

ERK

Extracellular Signal-Regulated Kinase

JNK

Jun N-Terminal Kinase

HCV

Hepatitis C Virus

HCMV

Human Cytomegalovirus

HIV

Human Immunodeficiency Virus

HSV

Herpes Simplex Virus

IE2

Immediate-Early 2

IC₅₀

50% Inhibitory Concentration

IFN

Interferon

MEK

Mitogen-Activated Protein Kinase

MCMV

Murine cytomegalovirus

NDV

Newcastle Disease Virus

NSPs

Nonstructural Proteins

NF-kB

Nuclear factor kappa B

Nrf2

Nuclear factor erythroid 2-related factor 2

ONNV

O’nyong-Nyong Virus

PI

Protease Inhibitor

PI3KIII

Phosphoinositide 3-Kinase III

P38 MAPK

P38 Mitogen-Activated Protein Kinases

Pgp

P-glycoprotein

RT

Reverse Transcriptase

RSV

Respiratory Syncytial Virus

SAR

Structure–Activity Relationship

SARS-CoV-2

Severe acute respiratory syndrome coronavirus 2

SI

Selectivity Index

SRB

Sulforhodamine B

TNF-α

Tumor Necrosis Factor-alpha

TLR4

Toll-Like Receptor 4

45 TI

Therapeutic Index

46 VSV

Vesicular Stomatitis Virus

47 ZIKV

Zika Virus

Author Contributions

M.V. contributed in the conceptualization, supervision, writing, and editing of the original draft. Z.Z. and A.A. contributed in writing a part of the synthesis and medicinal chemistry topics, respectively. E.A. contributed in writing and editing the manuscript.

The authors declare no competing financial interest.