All viruses included in the Flavivirus genus have some common characteristics: a size of 40-60 nanometers; an envelope that is covering an icosahedral nucleocapsid, which protects the genetic material; and a single positive sense RNA strand, which is approximately 10.000 to 11.000 bases in length. This strand has a unique open reading frame (ORD) consisting of approximately ~3400 codons, that encodes for a unique viral polyprotein (Brinton and Basu, 2015[15]). These viruses show a type I cap in the 5'-end. The presence of a methylation in 5' cap in the RNA viral genome has an important role to play in the viral translation/replication process and is involved in the evasion of the immune reactions that restrict the viral replication in the host. Presence of Poly (A) tails at 3'-termini, in the 5' and 3' terminal regions, there are short conserved sequences whose functions are still unknown. The Flavivirus genome works as a viral mRNA and as a template for the synthesis of small RNA chains. Its organization is maintained between the different genera (Lindenbach and Rice, 2003[43]; Brinton and Basu, 2015[15]).

In the family Flaviviridae, most of the members have the typical genome with a similar order of genes and some conserved non-structural proteins (Figure 1(Fig. 1)). However, they diverge notoriously in the cis-acting regulatory elements of the RNA located at their 3' and 5' ends. For example, the polyproteins translated from the genomes of Hepacivirus, Pestivirus, and Pegivirus begin with an AUG sequence that is found at the 3' end. This is the internal ribosome entry site (IRES). On the contrary the translated protein of Flavivirus is dependent on a type I cap located at the 5' end. The genome of the Flaviviridae family is organized as follows:

Flavivirus: 5'CAP (I)-5'UTR-C-prm-E-NS1-NS2A-NS2B-NS3-NS4A-NS4B-NS5-3'UTR

The ORF of the Flavivirus encodes for a polyprotein which is processed by host and viral proteases in three structural proteins, the C protein of nucleocapsid, a glycoprotein precursor of prM membrane, and a glycosylated envelope protein E, along with seven non-structural proteins (NS) viz. NS1, NS2A/B, NS3, NS4A, NS4B, and NS5, which are arranged from terminal N to C. The process of proteolysis of the polyprotein of Flavivirus is carried out through the combination of four different proteases (Bollati et al., 2010[12]).

A host signal peptidase process the anchor junctions -C/prM, prM/E, E/NS1, and NS4A-“2K”/NS4B.

Among the structural proteins, the C protein of nucleocapsid protects and encapsulates the genetic material. An envelope glycoproteins prM/M creates the superficial structure of the virion. The protein prM/M has an important role in the maintenance of the spatial structure of the E protein, which is composed of three different structures, envelope domain I, II, and III (EDI, EDII, EDIII, respectively), connected to the viral membrane (Zhang et al., 2017[102]). In YFV, the E protein is highly immunogenic and plays a central role in viral binding and fusion (Volk et al., 2009[90]).

EDI forms the central domain of the envelope protein, which establishes the protein direction and its glycosylation sites. The glycosylation domains of EDI are associated with the viral production, pH sensitivity and neuroinvasiveness in most flaviviruses. Interestingly, ZIKV has a single glycosylation site (N154) in the EDI with a longer loop (residues 145-160) compared to other flaviviruses (Kostyuchenko et al., 2013[37]). EDII plays an important role in the membrane fusion, and EDIII is the main target of the neutralizing antibodies; these proteins are essential in the biology of the Flavivirus (Shi and Gao, 2017[72]; Zhang et al., 2017[102]). The structure of the Domain III of YFV differs from the structure of other mosquito-transmitted flaviviruses, specifically in the surface exposed BC loop, since it is one amino acid shorter, than even other mosquito-borne and non-vector borne viruses. This change can modify neutralization sites and receptor interactions, and therefore, the viral pathogenesis (Volk et al., 2009[90]).

A difference in the amino acid sequence of the EDIII region has been reported in the DEN4 sylvatic and human strains, which may be related to their jump from a sylvatic host to a human one (Volk et al., 2007[89]). Therefore, this region among Flavivirus shows considerable variation.

The non-structural proteins, modulate intracellular process, such as viral replication, assembly, proteolysis, maturation, and regulation of the host immune response and are involved in the evasion of the immune response. In the case of evasion of the immune response by DENV the following are involved: NS2A, NS2B3, NS4A, NS4B and NS5 (Chen et al., 2017[19]).

About the non-structural proteins, they change in functions and structure between different Flaviviruses. For example, NS1 is a highly conserved protein that is encoded exclusively by the members of genus Flavivirus and is considered as the major antigenic marker during infection. This non-structural protein is a dimeric protein with molecular weight between 46-55 kDa depending on the degree of glycosylation, which is important for the efficiency of the viral secretion, replication, and virulence. Thus, deletion of NS1 glycosylation sites of DENV and YFV, significantly reduce their viral replication. NS1 can exist as a monomer, dimer (protein connected to the mNS1 membrane) and also as a hexamer (secreted protein, sNS1) (Rastogi et al., 2016[66]; Chen et al., 2017[19]). A recent study shows that the overall structure of NS1 is similar among YFV, ZIKV, DENV, and WNV (Wang et al., 2017[94]). NS1 of ZIKV has a diversity of electrostatic surface features in the host-virus interaction context as well as unique dimeric assembly (Song et al., 2016[80]).

The NS1 function is to activate the Toll-like receptors (TLRs) and inhibit the complement system in order to interact with the NS4 protein. The activation of the TLRs is important as it leads to the activation of transcription factors and the production of cytokines as a part of the innate immune response. The main receptor for Flaviviridae is TLR3, which in turn can modulate the activation of dendritic cells, macrophages, and B cells as well as triggers the IFN mRNA transcription, thus inducing the innate antiviral responses (Alayli and Scholle, 2016[3]; Chen et al., 2017[19]). Spontaneous mutation resulting in a single amino acid substitution in ZIKV NS1 protein increased its infectivity in Aedes aegypti mosquitoes, which most likely had promoted its transmission in the recent epidemic (Liu et al., 2017[45]).

In the case of WNV, NS1 reduces the regulation of the IFN-β and IL-6 transcription mediated by TLR3; nonetheless, this inhibition can be dependent on cell type (Chen et al., 2017[19]). It has been reported that the expression of the WNV, YFV or DENV-2 NS1 protein, does not alter the TLR3 translation levels in two cell lines (HeLa and HEK-293) (Baronti et al., 2010[6]). This could indicate that the TLR3 pathway is not exclusive in the control of flaviviruses replication and they possess other mechanisms for evading the host immune response.

On the other hand, in the dengue virus, NS1 anchors the replication complex to the host endoplasmic reticulum membrane, and interacts directly with NS4B, As a result, it plays an important role in the viral replication process. Furthermore, YFV and DENV are associated with hemorrhagic fever, this may be due to the increased vascular permeability. However it can be mediated by the NS1 protein, because it has been shown that its binding to heparin sulfate of the endothelial glycocalyx layer may alter the capillary permeability (Puerta-Guardo et al., 2016[64]). This is also relevant in the severe forms of dengue since anti-NS1 antibodies join to the surface of the endothelial cells and platelets producing cell damage (Chen et al., 2018[18]). Besides, ZIKV NS1, from Asian lineage strains, can be involved in the developmental complications such as microcephaly in fetuses and also severe neurological manifestations in adults. However, currently it is not clear how the virus can penetrate the hematoencephalic barrier (Wang et al., 2017[93]; Xia et al., 2018[98]). However, an alternative protein, the NS1', has been reported in ZIKV (specifically Asian lineage) and other neuroinvasive Flavivirus. This suggests that this protein may be involved with the viral neurotropism and therefore with its neuroinvasiveness (Tambonis et al., 2017[82]).

As a consequence about the previously reviewed information, it is possible to establish that the NS1 protein of the Flavivirus, induces a protective immune response against infection; it is an important immunogen in the viral infection course (Chen et al., 2017[19]), and a biomarker for the diagnostics of dengue, zika and chikungunya infections (Cecchetto et al., 2017[17]).

The NS2 protein, is encoded by the NS2 region of the Flaviviridae genome. The NS2 protein can be also processed into the matured proteins such as NS2A and NS2B (Chen et al., 2017[19]). The NS2A is a 226 amino acid long hydrophobic protein, that is inserted in the membrane and interacts specifically with negatively charged phospholipids. This protein contains some transmembrane domains and is associated with the membrane of host endoplasmic reticulum (RE). It is involved in the evasion of the immune response since it suppresses Interferon β transcription (IFN-β), and as a result, the interferon response, which is one of the host's first lines of defense. The NS2A is required for the processing of NS1 containing specific recognition motifs targeted by specific proteases. It is also involved in the virion assembly and the synthesis of viral RNA. For example, in the case of YFV, a Lys-to-Ser mutation at position 190 in NS2A blocked the production of virus particles (Leung et al., 2008[39]). Additionally, it has been reported that in DENV there are two different sets of NS2A molecules, which could be interesting for the development of control strategies (Xie et al., 2015[99]). The functions of NS2A are related to the membrane structure and orientation, and mainly in both viral RNA synthesis and virion assembly (Murray et al., 2008[50]; Xie et al., 2015[99]). NS2B is an integral protein of 14 kDa and is composed of three domains, two transmembrane segments located at the N and C terminal regions, and a central region of 47 amino acids that acts as cofactor for the protease NS3. The complex NS2B-NS3 serves as a place for the assembly of the replication complex of the flaviviruses and modulates both the viral pathogenesis and the immune response of the host (Murray et al., 2008[50]; Chen et al., 2017[19]). Although the functions of NS2A and NS2B proteins are not well understood, it is known that in DENV, apoptosis is induced by NS2B-NS3 protease precursor and NS3 protease, most likely through the caspase-8 pathway or NF-κB pathway (Uno and Ross, 2018[88]).

NS3 is an enzyme that shows approximately 65 % sequence identity between DENV, ZIKV, and YFV (Brand et al., 2017[13]), therefore is interesting for the development of control strategies. The NS3 protein is the second largest protein after NS5, and is composed by two regions: the N-terminal region that contains a serine protease domain and a chymotrypsin protease domain. The C-terminal region that contains both a nucleoside triphosphatase (NTPase) and a RNA helicase domain. Serine protease is involved in the proteolytic processing of the viral polyprotein in three single proteins, as mentioned above, and thus plays a central role in the viral replication cycle. The chymotrypsin protease cleaves the viral precursor polyprotein to release each individual NS protein and the NTPase/dependent RNA helicase, which are involved in the replication of the genome and synthesis of viral RNA (Natarajan, 2010[54]). Likewise, DENV-NS2B is required as a cofactor for NS3 protease activity, for example, it has been described that in DENV is responsible for the cleavage between NS2A/2B, NS2B3, NS3/4A and NS4B/5 (Chen et al., 2017[19]).

DENV NS4 protein is cleaved by the NS3 host signal protease and peptidase, producing mature NS4A and NS4B proteins. This process is necessary for the inhibition of the IFN signaling pathway, as well as for the viral replication process (Zou et al., 2015[104]; Chen et al., 2017[19]). NS4A is a small integral membrane protein and it contains four transmembrane segments that act as a scaffold for the replication complex. It has been proposed that NS4A leads to membrane modifications and plays an important role in the formation of structures derived from the host's membrane. The N-terminal region of NS4A interacts specifically with curved membrane regions. It is an essential component of the viral replication cycle because it interacts directly with the vimentin protein of the cytoskeleton, an event that is necessary for the correct localization of the replication complex in the perinuclear region (Teo and Chu, 2014[84]). On the other hand, the protein NS4A is involved in the induction of autophagy as a mechanism for infection (McLean et al., 2011[48]). NS4B is formed by three C-terminal trans-membrane segments and is involved in blocking the signaling transduction of the IFN-α/β, as well as modulating the function of NS3 helicase. In YFV and WNV, NS4B and STAT1 show cytoplasmic localization, and can inhibit the 'Interferon Stimulated Response Element' activation mediated by IFNβ (Nemésio et al., 2012[55]; Uno and Ross, 2018[88]). Even though the NS4B protein is highly conserved among the DENV serotypes, mutations in the NS4B of YFV can modulate its neuro-invasion capacity, as well as its neuro-virulence (Zmurko et al., 2015[103]). During Zika infection, NS4A and B are involved in the induction of autophagy in the fetal neural stem cells leading to defective neurogenesis (Liang et al., 2016[41]).

Finally, NS5, a 104 kDa protein is encoded by the C terminus of Flavivirus genome. It is highly conserved among the Flavivirus, with a homology of about 68 % between DENV, ZIKV and YFV (Potisopon et al., 2014[61]; Brand et al., 2017[13]). NS5 is composed of two domains: the methyltransferase domain, indispensable for the formation of the CAP structure of the viral genome, and the RNA-dependent RNA polymerase domain, which is necessary for the RNA genome replication process. The NS5 protein of the Flavivirus can inhibit the IFN signals during the infections caused by DENV, YFV, ZIKV, and HCV, in order to evade innate immunity; this is achieved through the interaction with STAT2 protein or by modulating RNA splicing within the host cell (Murray et al., 2008[50]). DENV infected cells show localization of NS5 protein mainly in the nucleus; this may be related to the modulation of the cytokine gene expression (El Sahili and Lescar, 2017[23]). The NS3 protein initially cleaves the NS5 in HCV, producing the NS5A and NS5B mature proteins. NS5B is an RNA-dependent RNA polymerase involved in the viral replication, with a terminal nucleotide transferase activity using uridine triphosphate as a substrate. Because the NS5 is a highly conserved protein among Flavivirus, and its important role during the viral replication and evasion of the host immune system, has led to the proposal of the NS5 as a target for the development of antivirals against Flavivirus infections (El Sahili and Lescar, 2017[23]; Shi and Gao, 2017[72]).

The Alphavirus genus, consisting of about 29 virus species, is characterized by an icosahedral symmetry and a genome that consists of a positive sense single-stranded RNA of approximately 11.8 Kb length. This ssRNA is composed of two open reading frames (ORFs), the one on the 5' end produces non-structural proteins (nsP1-4), and the second one at the 3' end that produces five structural proteins called: capsid protein (C), envelope glycoproteins (E1 and E2), and two small cleavage products (E3 and 6K). The 5' end of the CHIKV genome also shows a 7-methylguanosine cap (Figure 1(Fig. 1)) (Jose et al., 2009[31]; Shi and Gao, 2017[72]).

The structural proteins are required for entry, assembly, and budding steps. These proteins are processed as a polyprotein, which are cleaved by capsid and host proteases into C, E3, E2, 6K, and E1 proteins. The proteins E3 and E2 have pE2 as intermediary compound, cleaved by host furin protease. The N-terminal domain of E3 is an uncleaved leader peptide for E2, which may help shield fusion peptide in E1 during egress. The E1 glycoprotein mediates the virus-membrane fusion. The E2 mediates the binding to the receptors and attachment factor on the cell membrane and is the major target for neutralizing antibodies. At low pH E1 and E2 dissociated in the endosome, which trigger the fusion of the viral and endosomal membranes, thus releasing the viral genome into the cytoplasm (Li et al., 2010[40]). It has been reported that E2 residue 82, in CHIKV, is determinant of glycosaminoglycan utilization and therefore, in the viral entry, this may have had an impact on the attenuation of the vaccine strain 181/25 (Silva et al., 2014[76]).

Besides, the protein 6K acts as a leader peptide for E1 and facilitates particle morphogenesis, and are involved in the formation of viroporins that play a role in the release of virus progeny, the intracellular trafficking of glycoprotein and the permeability of the membrane. The transframe protein, generated by ribosomal frame shifting, shares N-terminus with 6K, which is also considered as a putative ion channel that may enhance particle release. E1 in CHIKV is a 436 aa type II fusion protein, through fusion of viral envelope and endosomal cellular membrane (Jose et al., 2009[31]; Bautista-Reyes et al., 2017[9])_.

In turn, capsid protein has two domains: The N-terminal RNA binding domain and the C-terminal protease domain. The N-terminal region is indispensable for binding to RNA genome, C protein multimerization and its subsequent nuclear/cytoplasmic translocation. About the nuclear translocation is known that induces host cell transcriptional blockade in encephalitic alphaviruses, however it is not clear in arthritogenic alphaviruses yet (Taylor et al., 2017[83]), however mutation of a nucleolar localization sequence in the CHIKV N-terminal region of capsid protein, is involved in viral attenuation. Thus, this motive capsid protein is a target for the development of vaccines (Taylor et al., 2017[83]; Yap et al., 2017[100]; Sharma et al., 2018[71]).

On the other hand, the alphaviruses present four non-structural proteins (nsP), which are well characterized in CHIKV (Pietilä et al., 2017[59]; Silva and Dermody, 2017[75]):

nsP1 have both guanine-7-methyltransferase and guanylyltransferase activities, thus is responsible for the addition of the 5´Cap to viral genomic and subgenomic RNAs, a step necessary to avoid the viral RNA degradation and for its subsequent translation. This protein is also involved in anchoring the replications complexes to the cellular membranes maybe through interaction with anionic phospholipids from the membrane (Abu Bakar and Ng, 2018[2]; Mutso et al., 2018[53]).

Once the flaviviruses are in the endosomes; the acidic pH triggers the assembly of E trimers that leads to the release of the nucleocapside into the cell cytoplasm. Subsequently the capsid-RNA complex leaves the endosome, presumably through the pores (Figure 2(Fig. 2)). The RNA produces a polyprotein, which is directed to the endoplasmic reticulum, where it is processed by cellular and virus-derived proteases into structural and non-structural proteins (Rodenhuis-Zybert et al., 2010[68]; Garcia-Blanco et al., 2016[26]).

The E and prM proteins are glycosylated and then the NS proteins trigger viral genome replication. Several modifications of viral RNA are observed during its replication, such as methylation, which protects the RNA from quick degradation and manages the transcription efficiency. This modification is necessary for the formation of the virions and it is detected in different flaviviruses (Yap et al., 2017[101]).

The coupling of protein synthesis, RNA synthesis, and the virion assembly on membranous structures assures that the newly synthesized RNA genome can be associated with C protein, thus initiating the assembly process. RNA encapsidation initiates the budding of particles into the ER-derived membrane vesicles, formation of which is believed to be induced by prM/E heterodimers (Welsch et al., 2009[97]; Rodenhuis-Zybert et al., 2010[68]). These immature virions that have budded into the ER, are then processed by carbohydrate addition and modification as they proceed through the Golgi membrane system. It is likely that the transportation through the trans-Golgi network requires the presence of the glycosylated prM protein, which is obtained after prM/E heterodimers dissociation. prM is cleaved, just prior virus release, by the host-encoded furin, into a pr peptide and a membrane-associated M protein, allowing for the cleavage of E proteins and thus helping in the virion maturation. Virions follow the exocytosis pathway and are released into the extracellular space by fusion of vesicles containing virions with the plasma membrane (Keelapang et al., 2004[33]; Rodenhuis-Zybert et al., 2010[68]). Mature viruses are released from the host cell by exocytosis and the new mature virus can infect more cells (Suwanmanee and Luplertlop, 2017[81]).

After internalization of alphaviruses in endosomes, the reduction in pH level, due to the H⁺ pump, leads to the conformational reorganization of viral heterodimer E1-E2, releasing E1 as a result. Thereby, the domain II is exposed and the viral particle fuses with the membrane of the host cell (Schuffenecker et al., 2006[70]). The E3 protein interacts with the E2 protein, stabilizing a region called the “acid-sensitive region”, indirectly facilitating the activation of E1 that is required for the membrane fusion (Sjöberg et al., 2011[78]). The endosome releases the nucleocapsid and the RNA genome; both coupled by interaction of the major RNA subunit 60S with the signal protein of the nucleocapsid. This mechanism is highly conserved among alphaviruses (Sjöberg et al., 2011[78]) (Figure 3(Fig. 3)).

Subsequently the genome is translated to generate the non-structural and structural proteins. The non-structural proteins (nsP1-4) and their precursors are necessary for the replication of the viral RNA. The five structural proteins (C, E3, E2, 6K, E1) and their precursors are necessary for viral encapsidation and budding. Viral replication occurs in the host cytoplasm in close connection with the Golgi apparatus (Thiberville et al., 2013[85]).

The non-structural proteins (nsP1-4) are translated as polyprotein: P123 and P1234 (in smaller proportion) (Solignat et al., 2009[79]). First nsP4 is cleaved off from nsP123 to form the replication complex, which produces the negative strand (Barton et al., 1991[8]); this is detected during the early phases of CHIKV replication. When nsP123 concentration is enough to support an efficient reaction, it is cleaved into mature nsP1, nsP2, nsP3, and nsP4 (Solignat et al., 2009[79]). These proteins, together with host cell proteins, act as a plus-strand RNA replicase, which produces the 26S sub-genomic plus-strand RNA using the negative-strand RNA as a template (Shirako and Strauss, 1994[74]). This 26S sub-genomic RNA encodes the polyprotein precursor for structural proteins, which is cleaved to yield C, pE2, 6K, and E1 by an autoproteolytic serine protease (Solignat et al., 2009[79]). C and 6K proteins are accumulated in cell's cytoplasm for the formation of new nucleocapsids, which are assembled with a copy of the viral genome by means of proteolytic cuts. E3, E2, and E1 proteins, are post-transcriptionally modified and initiate their glycosylated in the endoplasmic reticulum and translocated to the Golgi apparatus and pack in vesicles to be delivered in the cell membrane. When the nucleocapsid interacts with the glycoproteins accumulated in the cell's membrane, the viral particles under the maturation process acquire a membrane envelope and then they are finally released by exocytosis (Bautista-Reyes et al., 2017[9]).

To date there is not a specific treatment to eradicate the infection by arboviruses, the treatment choice is basically to reduce the symptoms, although, the viral replication is not affected. A vaccine for DENV has been recently implemented to be used in people aged 9 to 45 years living in endemic areas (Barrows et al., 2016[7]). However, there are limitations to the application of this vaccine, since it has been reported that people who have never had dengue are more susceptible to developing a more severe disease, if they are vaccinated. In addition, the Philippine government suspended the use of the vaccine after the death of three people due to dengue despite being vaccinated (Dyer, 2017[22]). Therefore, there is no vaccine available for the treatment of DENV (Lim et al., 2013[42]).

In the middle of this bleak panorama, several candidates for arbovirus vaccines have been proposed, however, these are only tested in animal models. For example, Weger-Lucarelli and colleagues (2014[96]) describe the construction and characterization of a modified Ankara virus-based vaccine, and expressing CHIKV E3 and E2 proteins, the authors demonstrated that the vaccine protected different type of mice against CHIKV infection (Weger-Lucarelli et al., 2014[96]). Prow and colleagues, described a single-vector construct-based vaccine, that expresses the prME protein of zika virus and the polyprotein conformed by C-E3-E2-6K-E1 proteins of chikungunya virus, having as a result that a single dose of the vaccine in different type of mice, can elicit neutralizing antibodies, therefore, it provides protection against both viruses (Prow et al., 2018[63]).

The use of natural products, as therapeutic tools, is an ancestral practice in different cultures. According to WHO 80 % of the population of some countries from Africa and Asia depend on the use of traditional medicines to deal with some diseases (Oliveira et al., 2017[56]). The neglected disease, as some of the mentioned in this work, required the urgent description of new molecules with antiviral capacity for their treatment, as a sample; there are multiple articles that describe the use, in vitro, of natural products as inhibitors of some arboviruses infection.

For instance, a compilation of studies, from 1997 to 2012, describes the use of natural products of 31 species (that belong to 24 different plant families). Those studies include the analysis of species as Cladosiphon okamuranus, Leucaena leucocephala, Mimosa scabrella, Tephrosia madrensis, Cryptonemia crenulata, Gymnogongrus griffithsiae, Meristiella gelidium, among others. From which, were isolated approximately ten phytochemicals: as Fucoidan, Galactomannan, Glabranine, 7-O-methyl-glabranina, Galactane, hyperóeside, Kappa carragenane, Zosterice acid, 4-hydroxypanduratine A, panduratine A. These compounds belong to various kinds of chemical families such as: polysaccharides sulfated, flavonoids, quercetine and natural compounds of chalcone, which have anti-proliferative activity against dengue virus (Abd Kadir et al., 2013[1]).

In relation to zika, curcumin, a common food additive, was considered capable of reducing ZIKV infectivity by preventing the virus binding to cells, without showing adverse effects on cell viability (Sharma et al., 2018[71]). Quercetine, a flavonoid present in fruits, vegetables, leaves and grains, inhibits the enzymatic activity of NS2B-NS3 in a dose-dependent manner. Also, it has been reported that commercially available quercetine inhibits ZIKV protease with an IC₅₀ of 26.0 ± 0.1 μM (Mounce et al., 2017[49]). Although all these studies are promising, more researches are still needed, in order to validate them as therapeutic tools.

Despite the fact the use of natural products to inhibit different viruses has proven to be feasible and with good results in vitro. The knowledge of the biology of Flavivirus and Alphavirus have accelerated the discovery of selective targets to attack, with possibilities to inhibit their replication. In the case of ZIKV these proteins are well known as possible targets: NS2B-NS3 protease, NS3 helicase, NS5 polymerase, NS5 methyltransferase, the envelope and the capsid. To DENV, the research has been focused on multifunctional enzymes NS3, NS5 and the capsid protein. In the case of CHIKV one of the most suitable targets is the cysteine protease domain nsP2. All these proteins were selected as target because of their important role in the respective viral cycle replication.

However, Roy et al. (2017[69]) assessed a set of 9 natural products against zika virus, their results show that 6 of these compounds may inhibit the virus replication and have interaction as no competitive inhibitors of the NS2B-NS3 protein. In another study, that involved over 1350 extracts of plants as inhibitors of the DENV polymerase using the domain RdRp of DENV-2 NS5, 49 extracts showed an inhibition of at least 80 % infection, and one of these extracts, obtained from Cryptocarya chartacea exhibited 90 % inhibition at a concentration of 10 µg/ml. Subsequent purification of this extract reported 10 compounds: 1 pinocembrin and 9 chartaceone derivatives of pinocembrin. The pinocembrin did not show antienzymatic activity, however all the chartaceone derivatives showed different values of inhibition (Allard et al., 2011[4]).

Pohjala et al. evaluated 356 (123 natural products and 233 clinical approved drugs) as inhibitors of CHIKV infection, using a stable replication cell line. Their results showed that under experimental conditions the flavonoids apigenin, chrysin, naringenin and silybin, were the most actives against CHIKV (Pohjala et al., 2011[60]). The Harringtonine, a natural product obtained from Cephalotaxus sp, shows a high antiproliferative activity against CHIKV, with EC₅₀ of 0,24 µM and a very low cytotoxicity. These results suggest a possible inhibition at early stages of viral replication; additionally this compound was able to inhibit the replication of Sindbis virus too, suggesting a possible inhibition of other alphaviruses (Kaur et al., 2013[32]). In accordance with the previously indicated and, also, including other authors who have made relevant contributions on this subject, Table 1(Tab. 1) (References in Table 1: Brecher et al., 2017[14]; Carneiro et al., 2016[16]; Kaur et al., 2013[32]; Khan et al., 2010[35]; Khan et al., 2011[34]; Kiat et al., 2006[36]; Oliveira et al., 2017[56]; Rausch et al., 2017[67]; Roy et al., 2017[69]) shows the main natural products that have been evaluated against Flavivirus and Alphavirus.