Identification and Evaluation of New Potential Inhibitors of Human Neuraminidase 1 Extracted from Olyra latifolia L.: A Preliminary Study

Sialidases, also called neuraminidases, are involved in several human pathologies such as neurodegenerative disorders, cancers, as well as infectious and cardiovascular diseases. Several studies have shown that neuraminidases, such as neuraminidase 1 (NEU-1), may be promising pharmacological targets. Therefore, the discovery of new selective inhibitors of NEU-1 are necessary to better understand the biological functions of this sialidase. In the present study, we describe the isolation and characterization of nine known compounds from Olyra latifolia L. leaves. This plant, known to have several therapeutic properties, belongs to the family of Poaceae and is found in the neotropics and in tropical Africa and Madagascar. Among the purified compounds, feddeiketone B, 2,3-dihydroxy-1-(4-hydroxy-3,5-diméthoxyphényl)-l-propanone, and syringylglycerol were shown to present structural analogy with DANA, and their effects on membrane NEU-1 sialidase activity were evaluated. Our results show that they possess inhibitory effects against NEU-1-mediated sialidase activity at the plasma membrane. In conclusion, we identified new natural bioactive molecules extracted from Olyra latifolia as inhibitors of human NEU-1 of strong interest to elucidate the biological functions of this sialidase and to target this protein involved in several pathophysiological contexts.


Introduction
Neuraminidase 1 (NEU-1) is an exoglycosidase which removes terminal sialic acid residues from glycoproteins, glycolipids, and oligosaccharides. Sialidases are widely distributed among species [1]. NEU-1 is implicated in the appearance and progress of various diseases notably through its ability to act as an elastin degradation sensor and to transmit elastin-derived peptides (EDP) signaling [2]. For example, NEU-1 activity is implicated in the development of diabetes due to an enhancement of insulin resistance through insulin receptor desialylation [3]. Interestingly, NEU-1 and the underlying activation of the PI3Kγ signaling pathway promote atherosclerosis onset in mice [4]. Otherwise, NEU-1 is associated with immune thrombocytopenia [5] and EDP contribute also to the development of nonalcoholic steatohepatitis [6]. Furthermore, this enzyme is implicated in the development of various cancers such as melanoma [7][8][9], breast [10][11][12][13], and ovarian [14] cancers.

General Procedures
NMR spectra were recorded in CD 3 OD on a Bruker Avance III 500 spectrometer (Bruker, Karlsruhe, Germany). HR-ESI-MS analysis was conducted using a Micromass Q-TOF micro instrument (Micromass, Manchester, UK). Vacuum Liquid Chromatography (VLC) was carried out on Lichroprep RP-C18 (40-63 µm) Merck using a sintered glass No. 4 and vaccum for the elution. Flash chromatography was carried out on a Grace Reveleris system equipped with dual UV and ELSD detection using Grace ® cartridges (Silica gel or RP-C 18 ) (Grace, Epernon, France). HPLC separations were performed on a Dionex apparatus equipped with an ASI-100 autosampler, an Ultimate 3000 pump, a STH 585 column oven, a diode array detector UVD 340S, and a Chromeleon software. A prepacked RP-C 18 column (Phenomenex 250 × 10 mm, Luna 5 µ) was used for semi-preparative HPLC. The eluting mobile phase consisted of H 2 O with TFA (0.0025%) and CH 3 CN with a flow rate of 5 mL/min and the chromatogram was monitored at 205 and 210 nm. Thin-layer chromatography (TLC) was carried out using silica gel 60 F 254 pre-coated aluminum plates (0.2 mm, Merck, Darmstadt, Germany). After developing with solvent systems, spots were visualized by spraying with 50% H 2 SO 4 followed by heating.

Plant Material
The leaves of Olyra latifolia L. were collected at Akoupe in the Me Region (Akoupe, Ivory Coast) in February 2017 and were identified at the National Floristic Center of the UFHB Abidjan Ivory Coast. A voucher specimen UCJ 007583 was deposited at the Hebarium of this center.

Plasmids and Transfection Reagent
Plasmid encoding human PPCA protein was provided by Pr. Alessandra d'Azzo and has been described previously [32]. Plasmid encoding human NEU-1 was purchased from ImaGenes GmbH (Berlin, Germany). JetPEI DNA transfection reagent used for cell transfections was purchased from Polyplus transfection.
Sialidase activity at the plasma membrane of macrophages was performed as described previously [33]. K-elastin (KE) harboring the GxxPG bioactive motif was produced by chemical hydrolysis of insoluble elastin coming from bovine neck ligaments. Differentiated THP-1 cells, seeded in 12-well culture dishes (5 × 10 5 cells/well), were washed with PBS and pre-incubated 15 or 30 min with vegetal compounds or DANA, a reaction buffer containing 20 mM of CH 3 COONa (pH = 6.5) and 400 µM of Muf-NANA, with or without K-elastin (50 µg/mL). After the pre-incubation step, cells were put 2 h 30 min at 37 • C in the dark. After incubation, the reaction was stopped by adding 0.4 M of glycine buffer (pH = 10.4) and the fluorescent 4-methylumbeliferone product released in the medium was measured using the Infinite F200 Pro (TECAN) hardware and Magellan software.

Western Blot
Protein samples in appropriate buffers according to the experiments were diluted in Laemmli buffer (62.5 mM Tris, 2% SDS, 10% glycerol, 0.05% bromophenol blue, pH 6.8) and heated 10 min at 100 • C. After electrophoresis in a 10% acrylamide SDS-PAGE gel, proteins were transferred onto a nitrocellulose membrane at 100 V for 1 h in a Tris/glycine buffer supplemented with 10% ethanol. After blocking of the nitrocellulose membrane with 0.05% TBS Tween-20 (TBS-T) supplemented with 5% milk for 1 h at room temperature, membrane was probed with primary antibodies diluted at 1:200 for NEU-1 (NEU-1 F8 Santa Cruz) and at 1:750 for β actin (Santa Cruz Biotechnology, Heidelberg, Germany) in TBS-T with 3% BSA overnight at 4 • C. Membrane was then washed in TBS-T and incubated with HRP-linked secondary antibodies diluted at 1/10,000 in TBS-T with 5% milk at room temperature. Anti-mouse HRP-linked antibodies (Cell Signaling, Danvers, MA, USA) were used for protein detections. Chemiluminescent protein detection was done using ECL Prime and ODYSSEY Fc (LI-COR Biosciences-GmbH, Bad Homburg, Germany) hardware and the Image Studio software.

Statistical Analysis
Results are expressed as mean ±SEM. Statistical significance was evaluated using Student t-test or ANOVA followed by a Dunnett's multiple comparison test.

Structure Identification
The purification of CH 2 Cl 2 and 80% MeOH extracts obtained on defatted O. latifolia leaves gave nine compounds.

Effects of Selected Compounds 1-3 on Sialidase Activity in COS-7 Cells
Overexpressing NEU-1 We next evaluated the ability of compounds 1-3 to decrease membrane sialidase activity of COS-7 cells overexpressing NEU-1. COS-7 cells were transfected with both human NEU-1 and PPCA encoding plasmids, and the capacity of the compounds 1-3 to inhibit Biomedicines 2021, 9, 411 9 of 15 sialidase activity was assessed 48 h post-transfection in crude membrane preparations and compared to DANA at the same concentrations. A four-fold increase of membrane sialidase activity, from 100 to 437 ± 37%, was observed between untransfected cells and COS-7 cells overexpressing NEU-1 (Figure 3a). After 15 min of incubation at 0.1 and 1 µM, both DANA and the vegetal compounds are able to decrease NEU-1 sialidase activity in a similar way (Figure 3b). Compounds 1-3 and DANA at 0.1 µM decrease NEU-1 sialidase activity by 10.4 ± 3.1, 9.4 ± 2.1, 13.1 ± 1.8, and 9.3 ± 1.9%, respectively, in comparison with the control (untreated cells) (Figure 3b). A treatment by compounds 1-3 and DANA at 1 µM decreases NEU-1 sialidase activity by 14.3 ± 1.3, 14.0 ± 1.5, 10.3 ± 0.9, and 12.2 ± 1.5%, respectively, in comparison with the control (Figure 3b). Similar effects were observed after 30 min of incubation at both 0.1 and 1 µM (Figure 3c). Given that all the experiments were performed in presence of 0.024% of DMSO (final concentration), effects of DMSO alone at the same concentration were evaluated. No effect of DMSO was observed compared to the control condition (untreated cells) (Figure 3d). Moreover, DANA and the vegetal compounds do not alter significantly NEU-1 membrane expression level in transfected COS-7 cells (Figure 3e). Altogether, these results suggest that these three vegetal compounds are able to decrease significantly NEU-1 mediated sialidase activity in an overexpression cell model, in similar extent to the DANA molecule.
In both cell lines, compounds 1-3 act quickly at low concentrations to inhibit NEU-1 sialidase activity. These inhibitory effects of vegetal compounds were more important in THP-1-derived macrophages under K-elastin stimulation than in COS-7 cells overexpressing NEU-1. All vegetal compounds tested show comparable sialidase activity inhibition ranges. Their inhibition potentials were also similar to those observed with DANA at same concentrations.    In both cell lines, compounds 1-3 act quickly at low concentrations to inhibit NEU-1 sialidase activity. These inhibitory effects of vegetal compounds were more important in

Discussion
The key role of NEU-1 in the degradation of sialyloconjugates in lysosomes has been known for many years [43], but its implication in cellular regulatory mechanisms has been identified only recently. These new findings are associated with the discovery of NEU-1 localization at the cell plasma membrane. As described previously, human NEU-1, the ERC catalytic subunit, is known, after activation of its sialidase activity by EDP, to be involved in several pathophysiological contexts as cancers [7][8][9][10][11]13,14,44], diabetes [3], atherosclerosis [4], and nonalcoholic steatohepatitis [6]. That is why several strategies have been envisaged to disturb the activation of NEU-1 to prevent the progression of various diseases. Among these, chemical compounds [22,45] or transmembrane peptides as inhibitors of NEU-1 dimerization/activation [24] are currently under development. During the last years, different selective inhibitors of human NEU-1 have been notably developed based on the DANA chemical structure [22,23]. Indeed, the C9-amido analog of DANA (C9-BA-DANA) shows micromolar IC50 against human NEU-1 [23] and the C5-hexanamido-C9-acetamido analog has a K i of 53 ± 5 nM and 340-fold selectivity over other isoenzymes [22]. In fact, C9-BA-DANA not only blocks NEU-1 sialidase activity but also its bioactivities in human lung in vitro and murine lung in vivo [45]. Furthermore, a recent study has also shown that interfering peptides targeting the transmembrane domain of NEU-1 are able to decrease both NEU-1 dimerization and sialidase activity [24]. The aim of this study was to identify plant-derived natural products that share analogy with the DANA molecule and to evaluate their bioactivity against plasma membrane NEU-1 sialidase activity. We have selected Olyra latifolia because in Ivory Coast, their leaves were used for treating diabetes and the diabetic wound [28] and this activity could be related to NEU-1 inhibitory activity [3]. Nine known compounds were isolated from the leaves of Olyra latifolia and three compounds (1-3) were shown to possess structural analogy with DANA: the feddeiketone B, the 2,3-dihydroxy-1-(4-hydroxy-3,5-dimethoxyphenyl)-1-propanone, and the syringylglycerol. Two cell models have been used in the present work, COS-7 cells which overexpressed NEU-1 and THP-1-derived macrophages which endogenously express this target protein. For all these compounds and DANA, no significant effect on cell toxicity was observed in the two cell lines after 2 or 3 h of incubation at 0.1 and 1 µM. Therefore, we studied the effects of these compounds on NEU-1 sialidase activity in COS-7 cells and in THP-1-derived macrophages at this concentration range. In all experiments, final percentages of DMSO used were under 0.03% in order to avoid cell toxicity. Our results suggest that compounds 1-3 are able to decrease significantly NEU-1 mediated sialidase activity in both cell models, but the inhibitory effects were much greater in THP-1 cells than in COS-7 for all the tested concentrations (0.1 and 1 µM). These results might be explained by the fact that in COS-7 cells, NEU-1 is overexpressed and, therefore, the ratio between the NEU-1 expression level and the compound concentrations is higher in these cells compared to THP-1 cells. Kawecki et al. have previously shown that incubation with K-elastin significantly increased sialidase activity at the plasma membrane and that silencing NEU-1 by siRNA completely blocked the effects of K-elastin: these data definitely demonstrated involvement of NEU1 in K-elastin-induced sialidase activity at the plasma membrane of THP-1 cells [33]. Our results obtained with the same cell model showed that the natural compounds 1, 2 and 3 strongly inhibit NEU-1 sialidase activity triggered by K-elastin: these experiments suggest that these 3 compounds are able to directly inhibit human NEU-1, the ERC catalytic subunit, after activation of its sialidase activity by K-elastin. In addition to these results, we observed that the inhibitory effects of the three tested compounds extracted from Olyra latifolia are very similar to those obtained with DANA in the same experimental conditions which indicate that these natural molecules are potent inhibitor of NEU-1 as well as DANA. The glycerol lateral chain in syringylglycerol (3) and DANA seems to be important for the activity or at least a ketone and a primary alcohol group on a C 3 -substituant. In this preliminary study, we tested the effects of these three compounds at very low concentrations (0.1 and 1 µM). Indeed, these compounds are solubilized in DMSO which cannot be used at higher concentrations due to its toxicity. However, the effects of these molecules obtained at low concentrations are very promising and suggest that at higher concentrations (which requires less toxic solvents for solubilization), similar to those classically used for DANA, these inhibitory effects could be more important. Consequently, these vegetal compounds could be used as natural putative inhibitors of NEU-1 and could serve as models to design more potent synthetic inhibitors. Furthermore, as these compounds appear to be effective in inhibiting membrane NEU-1 in vitro, they could be used in in vivo models to inhibit this protein involved in several diseases such as atherosclerosis, thrombosis, insulin resistance, non-alcoholic steatohepatitis, and cancer.
In this preliminary study, we report the isolation and characterization of nine compounds (1-9) isolated for the first time from O. latifolia leaves. Among them, compounds 1-3-having structural analogy with DANA-showed inhibitory effects against human NEU-1-mediated sialidase activity at the plasma membrane. The effects of these molecules obtained at low concentrations are very promising and suggest that at higher concentrations, similar to those classically used for DANA, these inhibitory effects could be more important. Thus, these compounds are considered as new natural bioactive inhibitors of human NEU-1 which may be further exploited as potential leads to elucidate the biological functions of this sialidase and to target this protein involved in several pathophysiological contexts. Consequently, this preliminary study predicted an inhibitory potential of these vegetal compounds that could be a starting point for the development of new natural putative inhibitors of NEU-1 and for the design of more potent synthetic inhibitors (identified by determining their IC50 and K i values and comparing with those of other inhibitors). Furthermore, in order to better characterize compounds 1-3 and their possible use as tools for a better understanding of sialic acid biology, it could be interesting to evaluate their effect on another plasma membrane associated sialidase such as NEU-3.but also on two other members of sialidase family, NEU-2 and NEU-4. Last, as these compounds appear to be effective in inhibiting membrane NEU-1 in vitro, they could be used in in vivo models to down-regulate this protein involved in several diseases such as atherosclerosis, thrombosis, insulin resistance, non-alcoholic steatohepatitis, and cancer.
Author Contributions: All authors listed have made a substantial, direct, and intellectual contribution to the work and approved it for publication. All authors have read and agreed to the published version of the manuscript.

Conflicts of Interest:
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.