Potential antiviral effects of some native Iranian medicinal plants extracts and fractions against influenza A virus

Background Influenza A virus (IAV) infection is a continual threat to the health of animals and humans globally. Consumption of the conventional drugs has shown several side effects and drug resistance. This study was aimed to screen some Iranian medicinal plants extracts and their fractions against influenza A virus. Methods Glycyrrhiza glabra (rhizome), Myrtus  commonis (leaves), Melissa officinalis (leaves), Hypericum perforatum (aerial parts), Tilia platyphyllos (flower), Salix alba (bark), and Camellia sinensis (green and fermented leaves) were extracted with 80% methanol and fractionated with chloroform and methanol, respectively. The cytotoxicity of the compounds were determined by MTT colorimetric assay on MDCK cells. The effective concentrations (EC50) of the compounds were calculated from the MTT results compared to the negative control with no significant effects on cell viability. The effects of EC50 of the compounds on viral surface glycoproteins and viral titer were tested by HI and HA virological assays, respectively and compared with oseltamivir and amantadine. Preliminary phytochemical analysis were done for promising anti-IAV extracts and fractions. Results The most effective samples against IAV titer (P ≤ 0.05) were crude extracts of G. glabra, M. officinalis and S. alba; methanol fractions of M. communis and M. officinalis; and chloroform fractions of M. communis and C. sinensis (fermented) mostly in co- and pre-penetration combined treatments. The potential extracts and fractions were rich in flavonoids, tannins, steroids and triterpenoids. Conclusion The outcomes confirmed a scientific basis for anti-influenza A virus capacity of the extracts and fractions from the selected plants for the first time, and correlated their effects with their phytochemical constituents. It is worth focusing on elucidating pure compounds and identifying their mechanism(s) of action. Supplementary Information The online version contains supplementary material available at 10.1186/s12906-021-03423-x.

(Tamiflu ® ), as FDA-approved drug, which is a neuraminidase inhibitor molecule with the basic structure of shikimic acid was isolated and purified from the fruits of Illicium verum Hook. f. (Illiciaceae) [5]. Treatment with conventional drugs including amantadine and oseltamivir has shown side effects on the central nervous system and the gastrointestinal tract. Also, because of the genetic instability and reassortments of the virus from one side and drug resistance from the other side, the prescription of these drugs has been inefficient in some cases [6][7][8]. Due to the need for having other inhibitors of IAV, the scientists have focused on the screening medicinal plants and natural products and to find and introduce a lead compound or structure for the further preclinical trials [9].
Medicinal plants play vital roles in prevention and treatment of several diseases. The natural products and medicinal plants are popular in different parts of the world from developing to developed countries as complementary to common medicines [10,11]. Medicinal plants are cheaper, more available and culturally acceptable especially in Iran with a strong and long history of traditional medicine in comparison with some chemical drugs with undesirable adverse effects [12].
There are several plants used from ancient times till now for the treatment of common cold, flu syndrom and infections in throat and upper respiratory systems. Although there were many ethnopharmacology articles about them, a few reports highlighted their anti-influenza virus effects [13,14].
To the best of our knowledge, there are no reports about antiviral effects of G. glabra, M. commonis, S. alba, T. platyphyllos and C. sinensis (fermented) extracts against influenza A virus. Also, there are no reports about anti-influenza effects of fractions of other mentioned plants. The objective of this study was to investigate the potential effects of crude extracts and different fractions of some Iranian medicinal plants against influenza replication, and compare their activities with oseltamivir and amantadine as control drugs. For this purpose, the propagation and cytopathic effects of IAV in the presence of different extracts and fractions were determined using the hemagglutination (HA), hemagglutination inhibition (HI), and MTT cytotoxicity assays. The potent extract(s) and fraction(s) can be introduced as sources of phytochemicals with antiviral effects against influenza A virus.

Preparation of plant extracts and fractions
Rhizomes of Glycyrrhiza glabra (Leguminosae, PMP-246), leaves of Myrtus commonis (Myrtaceae, PMP-421) and Melissa officinalis (Labiatae, PMP-358), aerial parts of Hypericum perforatum (Hypericaceae, PMP-526), and bark of Salix alba (Salicaceae, PMP-924) were purchased in July 2017 from market in Tehran, Iran. Dried and fermented leaves of Camellia sinensis (Theaceae, PMP-415 and PMP-416, respectively) were obtained in November 2017 from market of Lahijan, Iran. Flowers of Tilia platyphyllos (Malvaceae, 7057-TEH) were collected from Ramsar, Mazandaran, Iran. The voucher speciments were identified by Dr. Zahra Tofighi and Dr. Yousef Ajani, and deposited in herbarium of Faculty of Pharmacy, Tehran University of Medical Sciences, Tehran, Iran. All dried plants were extracted with 80% methanol and fractionated with chloroform to give chloroform fractions and the residues were named methanol fractions.  50 ) was obtained by the hemagglutination assay using Karber method [15,16].

Cytotoxicity assay
Two-fold serial dilutions of the extracts and fractions were exposed to the sub-cultured MDCK cells in 96-well plates (3 × 10 4 cell/well), for 48 h at 37 °C in duplicates. The test was repeated two times. The colorimetric MTT assay was performed as previously described by Mehrbod et al. [16]. Briefly, the culture medium was replaced by MTT 1X [3-(4,5-dimethyl-2-thiazolyl)-2,5-diphenyl-2H-tetrazolium bromide; Sigma, USA]. Following 3-4 h incubation at 37 °C in dark, DMSO (100 μl) was added to each well to release the purple color of formazan. The absorbance at 570 nm was measured with a microplate reader (BioTek EL 800, US). The cell viability was calculated based on the following formula: (mean Optical Density (OD) of treated cells/mean OD of control cells) × 100. The 50% cytotoxic concentration (CC 50 ), the concentration that reduced the cell viability by 50% respect to the control cells; and the 50% effective concentration (EC 50 ), the concentration required to achieve 50% protection against virus induced cytopathic effect, were also calculated by analyzing MTT data using SPSS software. The cells without any exposure which were considered as negative controls and DMSO as a vehicle control with maximum 0.5% concentration were tested in parallel. Amantadine hydrochloride and oseltamivir carboxylate, the conventional antiviral compounds, were also tested as positive control drugs.

Selectivity index
The selectivity index (SI) calculated by dividing CC 50 to EC 50 , represents the relative safety of the extracts or frections. It has been demonstrated that compounds with selectivity indices higher than 3 are potentially safe antiviral reagents [17].

Antiviral assay
In co-penetration procedure of antiviral evaluations, influenza virus (100 TCID 50 /0.1 ml) was mixed with the EC 50 of the extracts and fractions for 30 min, and then incubated with the cells at 37 °C. In pre-penetration and post-penetration procedures, the virus was added to the cells after and before the extracts and fractions. Following 1 h incubation, the unabsorbed viruses were replaced by TPCK-containing medium (1 μg/ml). Following 48 h incubation at 37 °C, viabilities of the cells were measured by MTT assay as described earlier. Concurrently, the virus titer was determined by testing the cell supernatants using the HA assay [16]. Amantadine hydrochloride (98.5 μg/ml) and oseltamivir carboxylate (394.25 μg/ml) (Sigma, Saint Louis, Missouri, USA) were considered as positive control drugs and the cells without any exposure were tested as negative controls. DMSO was tested as a vehicle control with maximum 0.5% concentration. The test was repeated two times in duplicates.

Cellular percentage of protection
The protection percentages of the samples were calculated by SPSS from the MTT data results of mockinfected and infected cells after 48 h exposure, using the following formula: where (ODT)V, (ODC)V and (ODC)M represent the absorbance of the treated sample, the virus-infected control (no compound) and the negative control (mock), respectively [18].

Hemagglutination assay (HA)
For quantification of the virus titer from the cell supernatants, the HA assay was conducted as previously described [19]. Briefly, serial dilutions of the culture media were added to 96-well U-shaped microplates in duplicates. The test was repeated two times. The HA units were measured as the reciprocal of the highest dilution giving complete agglutination with chicken red blood cells (0.5%). The precipitation and diffuse lattice formation of the RBCs demonstrate the absence and the presence of the virus, respectively.

Hemagglutination inhibition assay (HI)
For investigation of the inhibitory effect of the samples on the hemagglutinin activity, the concentration of 4 HA unit of virus particles was used. Briefly, the extracts and fractions were diluted 2-fold serially from CC 50 concentration. Then, 4 HA unit of the virus was added to each well. After pre-incubation for 45 min at room temperature, chicken red blood cells (0.5%) were added to the solution. The physical interaction between extracts/fractions and virus surface HA glycoprotein was calculated after 1 h by the agglutination inhibition pattern.

Preliminary phytochemical analysis
Crude extracts, chloroform and methanol fractions of the effective anti-influenza plants were tested for identifying the class of active metabolites such as alkaloids, cardiac glycosides, tannins, flavonoids, steroids, triterpenoids and saponins by following the standard procedures [20,21].

Tests for alkaloids
A) Dragendorff 's test: The apperance of reddish brown turbidity after addition of Dragendorff reagent was indicator for the presence of alkaloids. B) Hager's test: By addition of Hager reagent to the samples, yellow turbidity was seen in the presence of alkaloids.

Tests for cardiac glycosides
Keller-killiani test: The mixture of ferric chloride and glacial acetic acid were added to the sample solutions. Apperance of bluish green in the upper layer and reddish color in the lower layer of the solutions, after addition of concentrated sulphuric acid confirmed the existence of cardiac glycosides in the samples.

Test for tannins
FeCl 3 test: After adding a few drops of 5% w/v solution of ferric chloride III in 90% alcohol to the sample solutions, the appearance of deep blue or dark green color indicated the tannin existence in the samples.

Test for flavonoids (Shinoda test)
After adding few drops of concentrated HCl and magnesium metal to the samples, the appearance of pink, red, crimson or magenta color was the sign of flavonoids presentation.

Tests for steroids and triterpenoids
Libermann-Buchard test: The appearance of brown ring in the middle of green upper layer and deep red lower layer after addition of sulfuric acid containing few drops of acetic anhydride were indicator for the presence of steroids and triterpenoids. Salkowski test: By addition of chloroform and concentrated H 2 SO 4 to the extract and shaking well, the color of chloroform layer changed to red, and acid layer showed greenish yellow fluorescent if sterols and/or triterpenes existed.

Test for saponins (foam test)
The volume of 50 mg of samples in water was shaken vigorously in a test tube. If froth characteristic was obtained, the presence of saponins was confirmed.

Statistical analysis
The data were expressed as mean ± SD, and analyzed by one-way analysis of variance (ANOVA) and General Linear Model (GLM) (SPSS 18.0), Tukey and Duncan post-hoc tests. Sample values with P ≤ 0.05 and P ≤ 0.01 were considered statistically significant and highly significant, respectively.

Results
In this study, the efficacy of the crude extracts, chloroform and methanol fractions of 8 selected plants with history of usages in traditinal and folklore Iranian medicine were tested against IAV. The cytotoxicity of the extracts were evaluated, and non-toxic concentrations were defined prior to antiviral assay. The ability of the samples against viral titer and viral cytopathic effects varied with different extracts and fractions of the same plant as determined by the HA and MTT assays. The profile of the extracts and fractions of selected medicinal plants used in this study is listed in Table 1.

Cytotoxicity and selectivity indices of the compounds
The results for CC 50 were largely different for various samples ( Table 2). Among the crude extracts, the highest CC 50 value belonged to S. alba (3647.45 ± 52.52 μg/ ml) and the lowest CC 50 value (10.41 ± 0.00 μg/ml) was obtained for M. communis and H. perforatum equally. Among the chloroform fractions, S. alba and M. communis showed the highest (1755.28 ± 0.89 μg/ ml) and the lowest (2.60 ± 0.00 μg/ml) CC 50 values, respectively. Amongst methanol fractions the highest and the lowest CC 50 values were obtained for M. officinalis (4413.80 ± 3.69 μg/ml) and T. platyphyllos (277.56 ± 3.19 μg/ml), respectively. The EC 50 values which were the same as non-cytotoxic concentrations (NCTC) with no significant effects on the cell viability were calculated using data obtained from MTT and one-way ANOVA analysis as compared to the negative control ( Table 2). The selectivity index values were obtained with the highest SI value of 32 with M. officinalis (chloroform fraction), H. perforatum (methanol fraction), S. alba (crude extract and methanol fraction) and C. sinensis unfermented (methanol fraction) and the lowest value of 2 with G. glabra (chloroform fraction).

Inhibitory effect of extracts and fractions on influenza a virus
The samples were tested in an in vitro screening assay to define the antiviral activity against IAV. The antiviral activity of the plants extracts/fractions was analyzed based on the Log 10 HA titer (Table 3) and Log HA decrement (Table 4). Among them, in co-penetration treatments, G. glabra (crude extract; 5.00 ± 4.24), G. glabra (methanol fraction; 5.50 ± 3.54), M. communis (chloroform fraction; 5.50 ± 3.54) and M. officinalis (methanol fraction;6.00 ± 2.83); in pre-penetration treatment, M. communis (methanol fraction; 5.50 ± 0.71); and in postpenetration treatment: M. communis (chloroform fraction; 5.50 ± 3.54) showed highly significant (P ≤ 0.01) decrease in HA titer compared to the virus inoculation. These results were in accordance with the Log HA decrement (Table 5) which showed 5-6 logs decrement in Log HA titer for all samples.

Cell viability and cellular percentage of protection
The optical densities (ODs) in MTT assay for the antiviral activity against IAV revealed highly significant increments in the majority of the combined treatments compared to the virus-inoculated cells (P ≤ 0.01). Amantadine and oseltamivir combined exposures also resulted in high cell viability (P ≤ 0.01) ( Table 5). The ODs in combined treatments were analyzed to obtain the percentage of compounds protection on the cell viability against the virus infectivity. Data are presented in Table 6.

Dose-dependent response
The test was repeated twice. Some of the extracts and fractions showed RBC precipitation (HI+) until a certain dilution which showed dose-dependent responses for their HA physical interaction. But some other samples showed RBC precipitation (HI+) in all dilutions which is indicative of HA physical interaction in all dilutions (Table 7).

Preliminary phytochemical analysis results
The existence of secondary metabolites was investigated by different preliminary analyses for the crude extracts, and chloroform and methanol fractions of five potent anti-influenza virus plants including G. glabra, M. communis, M. officinalis, S. alba and C. sinensis (fermented). Phytochemical analysis data (
In the current research, based on the antiviral properties of medicinal plants against IAV and other viruses, and their traditional and folklore usage in Iran, the antiviral efficacy of the crude extracts, and chloroform and methanol fractions of some Iranian native medicinal plants including G. glabra, M. communis, M. officinalis, H. perforatum, T. platyphyllos, S. alba, and C. sinensis (fermented and non-fermented) were evaluated against IAV with more details. Previous investigation revealed that glycyrrhizin derived from the rhizomes of G. glabra has protective effects against IAV by induction of interferon [26]. In Table 3 Log 10 HA titer from HA assay in combined treatments with virus as compared to virus control group Data presented as mean ± SD are averages of two independent HA titration which were tested in duplicates. *, **: Significantly and highly significantly different from values obtained for compound-treated samples compared to untreated sample (P ≤ 0.05 & P ≤ 0.01) analyzed by SPSS, Tukey post-hoc test

Plant
Extracts or Fractions Log HA (mean ± SD) addition, other compounds of licorice showed significant inhibition on influenza A neuraminidase in a computerbased approach [27]. The effects of polyherbal formula containing licorice were confirmed for the prevention and treatment of influenza-like syndrome, clinically [28,29]. M. officinalis essential oil could inhibit avian influenza virus (H9N2) through various replication cycle steps especially direct interaction with the virus particles [30]. Also, its extract demonstrated a significant anti-influenza effect against H1N1 strain of influenza virus [31].

Co-pen Pre-pen Post-pen
The extract of H. perforatum showed anti-IAV effect both in vitro and in vivo. The EC 50 of the extract was 40 μg/ml against IAV while its CC 50 in MDCK cell line was 1.5 mg/ml [32]. In an experiment, it was observed that H. perforatum extract had significant efficacy for the treatment of mice infected with IAV [32]. In another study an opposite response occurred. The consumption of oral H. perforatum extract in the mice infected with influenza A virus, enhanced transcription of the suppressor of cytokine signaling 3 (SOCS3) and led to the impaired immune defense and higher mortality [33].
The anti-influenza activity of green tea (Camellia sinensis) against H1N1 virus was equivalent to green tea by-products (EC 50 equal to 6.72 and 6.36 μg/ml, respectively). Also, hexane-soluble and ethyl acetate-soluble fractions of green tea by-products possessed strong anti-influenza activity in chickens [34]. The other studies demonstrated that dimeric polyphenol molecules in green tea display more potent antiviral effects against both influenza A and B viruses than monomers. In addition, the existance of C-4′ hydroxyl group in the B ring of planar flavonols is necessary for the anti-influenza B virus activity [35,36]. It was confirmed clinically that formulations containing C. sinensis or green tea metabolites including catechines and theanine could prevent influenza infection [37,38]. According to the results, in terms of the selectivity index (SI), the extracts and fractions of all tested herbs were considered safe for the antiviral treatments except chloroform fraction of G. glabra. Its SI value was the same as conventional drugs. In addition, C. sinensis, S. alba, H. perforatum and M. officinalis were categorized as the safest plants in cellular studies.
In HI test, the extracts of T. platyphyllos, S. alba and C. sinensis showed RBC precipitation in all tested dilutions which indicates the strong physical interaction of these compounds with the HA surface glycoprotein of the virus. However, amongst the others, methanol fraction of M. officinalis showed the weakest interaction (3rd dilution), and on the opposite side M. communis and H. perforatum showed stronger interaction (5th dilution). The results of this study confirmed dose-dependent response for most of the extracts and fractions. In a previous research, the formation of complexes between tannins and proteins was confirmed [39]. Also it was demonstrated that antiviral inhibitory effects of hydrolyzable tannins were related to the intractions blocking between viral glycoproteins and cell surface glycosaminoglycans (GAGs) [40]. It is interesting that in our experiment, M. officinalis, unlike the others, did not contain any tannin which justificated the weakest intraction.
With a thorough scrutiny on the Tables 3, 4, 5 and 6 and comparing their results with the SI values in Table 2 and the fact that SI values higher than 3 are safe compounds, it was concluded that except one almost all tested samples were safe but not effective against IAV titer.
Moreover, General Linear Model (GLM) analysis which estimated marginal means of all the respective values for different exposure ways (combined treatments) confirmed all the outcomes. The data are shown in the Supplementary Figs. 1, 2, 3 and 4. Therefore, most of the above mentioned plant species might be promising Table 5 Cell viabilities from MTT assay in combined treatments with virus as compared to virus control group Data presented as mean ± SD are averages of two independent MTT assays which were tested in duplicates. **: highly significantly different from values obtained for drugs-treated samples compared to untreated sample (P ≤ 0.01) analyzed by SPSS, Tukey post-hoc test alternatives to decrease flu unfavorable effects by affecting the viral and cellular receptors. The data were considerable because the conventional antiviral drugs; amantadine and oseltamivir showed promising effects against virus infection, however, growing drug resistance has caused a significant challenge [41,42]. The phytochemical analysis of the potent anti-IAV extracts/fractions demonstrated that they were rich in flavonoids, tannins, triterpenoids and steroids. Natural products have different mechanisms against viral infections from interfering with entry, transcription, replication and translation of the virus, nuclear export of the virus, viral assembly, packing and budding to enhance the host responses [43].
There were many reports that showed flavonids act as anti-IAV compounds with various mechanisms [44,45]. Flavonoids are natural phenolic compounds of plants with potent antioxidant and antiviral properties. They can help viral-infected cells to fix their biochemical imbalance resulting from oxidative stress [46]. Also they have shown potential inhibition on the neuraminidase active site of influenza virus. The potency of NA blocking reduced from aurones to flavon(ol)es, isoflavones, flavanon(ol)es and flavan(ol)es, respectively. The structure activity relationship (SAR) studies of flavonoids against influenza virus demonstrated that the presence of 7-OH, 4′-OH, C 2 = C 3 and C 4 = O functionalities were necessary, but the existance of a sugar group reduced the effects [47,48].
Triterpenoids and steroids are natural components elucidated from plants and other organisms which have various biological activities including antiviral activities. Mechanistic studies revealed triterpenoids bind tightly to the viral hemagglutinin (HA) and disrupt the attachment of viruses to the cell receptors [49][50][51].
Influenza virus has two envelope glycoproteins named hemagglutinin (HA) and neuraminidase (NA). The binding of hemagglutinin to sialic acid residues of the host cells is a key step for initiating the IAV infection. The role of NA is facilitating the movement of virus from infection  21:246 sites to the respiratory tract [52,53]. The presence of flavonoids, tannins, steroids and triterpenoids together in the potent anti-influenza extracts and fractions of this study, covers the inhibitory effects on both the HA and NA may be a reason for such significant effects. It is good to note that the virus strain used in this study, A/PR/8/34, is not pathogenic to humans and may not be the best model for these types of studies. However, this strain is generally used because it provides acceptable oucomes comparable to the pathogenic strains and is also safe for the users.

Conclusions
The emergence of new strains of the influenza A virus makes us think about innovative strategies for the development of new drugs with improved antiviral effects, higher safety and better tolerability. Research focusing on traditional herbs as complementary therapies or preventive medicine is becoming more attractive. Medicinal plants have been used for various ailments, particularly infectious diseases. The current study indicated that treatment of IAV with the selected extracts and fractions reduced the hemagglutination activity of the virus, which may result from the physical interaction of the samples with virus hemagglutinin. Based on this scientific confirmation, the selected plants may have the capacity to ease the symptoms and burden of flu. The next focus of this study will be the purification of pure compounds responsible for the bioactivity against IAV infection as well as their mode of action. Recommendations are proposed for strategizing the future role and place of medicinal plants in prevention and treatment of influenza and other infectious