Metabolomics and cell-based assay analysis of selected South African plants to determine anti-Rift Valley Fever Virus activity

Abstract

Rift Valley Fever Virus (RVFV) is a mosquito-borne haemorrhagic fever virus affecting both humans and animals with severe morbidity and mortality. RVFV is an emerging virus that is endemic to Africa. However, it can infect a variety of mosquito species and therefore spread to other parts of the world. Despite the efforts made to reduce or curb the RVFV infection and other tropical diseases in recent years, the re-occurrence of the RVFV may devastate the world economically, bearing in mind the shortage of therapeutic agents. There is an urgent need for novel anti-RVFV chemotherapeutic agents that can circumvent the acquired diseases. Emergence of new viruses such as the currently tormenting Coronavirus (SARS-CoV-2), which are challenging to control due to the mutative nature of the viral genome, highlights the importance for this type of research. The current study investigated the anti-RVFV, cytotoxicity, and radical scavenging activities of the 50 % aqueous-methanolic leaf extracts of twenty plant species pharmacologically proven to exhibit antiviral activities. The cytotoxicity assay was performed to determine the concentration range of extracts for the subsequent study of antiviral activity to be tested in the non-toxic range in Vero cells. A green monkey kidney (Vero) cell line was used to investigate the cytotoxicity of the extracts using the 3-[4,5-dimethyl-2-thiazol-yl]-2,5-diphenyl-2H-tetrazolium bromide (MTT) cell viability method and all plant extracts tested exhibited lethal concentrations (LC50) values > 20 μg/mL which signifies that the extracts were relatively non-toxic. Determination of the anti-RVFV activity of the plant extracts was performed using the 50% tissue culture infectious dose (TCID50) assay of tenfold serial dilutions of the virus and extracts in Vero cells. The extracts were screened in quadruplicate in 96-well tissue culture plates. After seven days, the cytopathic effect (CPE) was analysed using a light microscope and the TCID50 was calculated. This study demonstrated promising anti-RVFV activity of eight extracts (Artemisia afra, Adansonia digitata, Euclea natalensis, Elaeodendron croceum, Elaeodendron transvaalensis, Elephantorrhiza elephantina, Helichrysum aureonitens, Sutherlandia frutescens) which exerted reduction of the viral load with TCID50 < 105. Radical scavenging activity of extracts was executed on 96 well plates utilizing the 2,2-diphenyl-1-picrylhydrazyl (DPPH) and 2,2′-azino-bis-3-ethylbenzthiazoline-6-sulphonic acid (ABTS) scavenging assays. Furthermore, cell-based radical scavenging potential was done. The cell-permeant probe 2′,7′-dichlorodihydrofluorescein diacetate (H2DCF-DA) and Griess reagent assays were used to evaluate the effects of extracts on LPS-induced ROS and RNS production, respectively. The H2DCF-DA fluorescence intensity indicative of ROS production in images acquired using the Laser Scanning Confocal Microscope was observed. The EC50 values were calculated with non-linear regression analysis with a variable slope using GraphPad Prism software version 8.0 (GraphPad Software La Jolla, CA, USA). The results showed that the extracts quenched free radicals exhibiting an EC50 value range of 4.12–20.41 μg/mL and suppressed the level of pro-inflammatory mediators by 60–80 % in Vero cells. These plant extracts were shown to exhibit nontoxicity to Vero cells, but showed antiviral potency against RVFV. Moreover, extracts reduced the level of ROS which might contribute in preventing disease propagation via oxidative stress reduction. 1H-NMR-based metabolomics on samples from twenty medicinal plant species was conducted with consistent data processing and multivariate statistical analysis using MestReNova and SIMCA software, respectively. The principal component analysis (PCA), the orthogonal projections to latent structures- discriminant analysis (OPLS-DA) and hierarchical clustering analysis (HCA) plots were constructed to assess the distribution and discrimination of the samples. The variables on the far ends of the S-plot and variable importance in projection (VIP) score values > 1, were considered significant as potential biomarkers. Thorough monitoring of the performance of the model was performed using the explained variation (R2), predicted variation (Q2) coefficients, cross-validated prediction residuals (CV-ANOVA), p values < 0.05, 100 x permutation and receiver operating characteristic (Area Under the Curve) [ROC(AUC)] > 7 %. 1H-NMR spectral peaks were interpreted and annotated with the Human Metabolome Database (HMDB) and Chenomx software. Furthermore, extracts from eight plant species with potent antiviral activity were subjected to the UHPLC-qTOF-MS analysis to investigate possible metabolites responsible for the activity. The UHPLC-qTOF-MS instrument was operated in dual (positive and negative) electrospray ionisation (ESI) mode. Spectral data such as mass spectra, retention time (Rt) and ion fragments from the UPLC-qTOF-MS analysis was searched against several databases including Kyoto Encyclopedia of Genes and Genomes (KEGG: http://www.genome.jp/kegg/), PubChem (https://pubchemblog.ncbi.nlm.nih.gov/), Human Metabolome Database (HMDB: http://www.hmdb.ca/), NIST (National Institute of Standards and Technology) database, DNP (Dictionary of Natural Products: www.dnp.chemnetbase.com), MassBank (USA), mzCloud (Advanced Mass Spectral Database), and MAGMa (www.emetabolomics.org) for putative identification of metabolites. In addition, findings from the analysis were further compared to peer reviewed literature. A total of 61 putative metabolites in the samples were detected. However, of most importance is the annotation of two fatty acids, 13S-hydroxy-9Z,11E,15Z-octadecatrienoic acid and 13-hydroxy-9Z,11E-octadecadienoic acid detected in the negative ionisation mode which are present in all active samples. Our results suggest that these annotated metabolites of interest may be responsible for the antiviral activity observed. This study provided a scientific evaluation of the efficacy of medicinal plants against RVFV, provided possible avenues for mechanisms of action and demonstrated potential of using 1H-NMR-metabolomics in combination with UHPLC-qTOF-MS for rapid drug discovery.