Structure based design towards the identification of novel binding sites and inhibitors for the chikungunya virus envelope proteins.

TLDR: The identification of novel sites for potential binding on the chikungunya glycoprotein complexes and the identification of possible antagonists for these sites are described through virtual screening using two successive docking scores; FRED docking for fast precise screening, with the top hits then subjected to a ranking scoring using the AUTODOCK algorithm.

Abstract: Chikungunya virus is an emerging arbovirus that is widespread in tropical regions and is spreading quickly to temperate climates with recent epidemics in Africa, Asia, Europe and the Americas. It is having an increasingly major impact on humans with potentially life-threatening and debilitating arthritis. Thus far, neither vaccines nor medications are available to treat or control the virus and therefore, the development of medicinal chemistry is a vital and immediate issue that needs to be addressed. The viral envelope proteins play a major role during infection through mediation of binding and fusion with the infected cell surfaces. The possible binding target sites of the chikungunya virus envelope proteins have not previously been investigated; we describe here for the first time the identification of novel sites for potential binding on the chikungunya glycoprotein complexes and the identification of possible antagonists for these sites through virtual screening using two successive docking scores; FRED docking for fast precise screening, with the top hits then subjected to a ranking scoring using the AUTODOCK algorithm. Both the immature and the mature forms of the chikungunya envelope proteins were included in the study to increase the probability of finding positive and reliable hits. Some small molecules have been identified as good in silico chikungunya virus envelope proteins inhibitors and these could be good templates for drug design targeting this virus.

Figures

Table 3 Top 5 hits identified for site 2 using the mature glycoprotein receptor (3N42), showing the molecular weights, calculated logP (clogP), predicted binding energies, inhibitory constants (Ki) and the interaction residues.

Table 1 The identified receptor cavities in the immature (3N40) and mature (3N42) crystal structures, grid coordinates x, y and z, cavity volumes, points count and location for each cavity site.

Fig. 2 Crystal structure of (a) the immature glycoproteins (generated from file pdb: 3N40) and (b) the mature glycoproteins (generated from the file pdb: 3N42) showing the identified binding cavities as solid filled surface.

Table 2 Top 5 hits identified for site 2 using the immature glycoprotein receptor (3N40), showing the molecular weights, calculated logP (clogP), predicted binding energies, inhibitory constants (Ki) and the interaction residues.

Fig. 3 Work-flow diagram of the virtual screening procedures used for CHIKV envelope proteins.

Fig. 6 3D representation of the predicted docking pose of compound 2 (Table 5) within site 4 binding pocket (transparent surface) of the mature chikungunya envelope protein showing the H-bonding (green dashes) with the fusion loop amino acid Phe87 (shown in orange).

Full text

University of Wollongong
Research Online
(*<3;@6-$*0,5*, ,+0*05,(5+,(3;/"(7,9: (*<3;@6-$*0,5*, ,+0*05,(5+,(3;/

Structure based design towards the identication of
novel binding sites and inhibitors for the
chikungunya virus envelope proteins
Adel A. Rashad
University of Wollongong(9<6>4(03,+<(<
Paul A. Keller
University of Wollongong2,33,9<6>,+<(<
#,:,(9*/!5305,0:;/,67,5(**,::05:;0;<;065(39,76:0;69@-69;/,&50=,9:0;@6-'63365.65.69-<9;/,905-694(;065*65;(*;;/,&!'0)9(9@
9,:,(9*/7<):<6>,+<(<
"<)30*(;065,;(03:
#(:/(+,33,9"$;9<*;<9,)(:,++,:0.5;6>(9+:;/,0+,5;0C*(;0656-56=,3)05+05.:0;,:(5+05/0)0;69:-69;/,
*/02<5.<5@(=09<:,5=,367,796;,05:6<95(36- 63,*<3(99(7/0*:(5+ 6+,3305.

Structure based design towards the identication of novel binding sites and
inhibitors for the chikungunya virus envelope proteins
Abstract
/02<5.<5@(=09<:0:(5,4,9.05.(9)6=09<:;/(;0:>0+,:79,(+05;9670*(39,.065:(5+0::79,(+05.8<0*23@;6
;,47,9(;,*304(;,:>0;/9,*,5;,70+,40*:05-90*(:0(<967,(5+;/,4,90*(:;0:/(=05.(505*9,(:05.3@
4(169047(*;65/<4(5:>0;/76;,5;0(33@30-,;/9,(;,505.(5++,)030;(;05.(9;/90;0:B<:-(95,0;/,9=(**05,:
5694,+0*(;065:(9,(=(03()3,;6;9,(;69*65;963;/,=09<:(5+;/,9,-69,;/,+,=,3674,5;6-4,+0*05(3
*/,40:;9@0:(=0;(3(5+044,+0(;,0::<,;/(;5,,+:;6),(++9,::,+B,=09(3,5=,367,796;,05:73(@(4(169
963,+<905.05-,*;065;/96<./4,+0(;0656-)05+05.(5+-<:065>0;/;/,05-,*;,+*,33:<9-(*,:B,76::0)3,
)05+05.;(9.,;:0;,:6-;/,*/02<5.<5@(=09<:,5=,367,796;,05:/(=,56;79,=06<:3@),,505=,:;0.(;,+>,
+,:*90),/,9,-69;/,C9:;;04,;/,0+,5;0C*(;0656-56=,3:0;,:-6976;,5;0(3)05+05.65;/,*/02<5.<5@(
.3@*6796;,05*6473,?,:(5+;/,0+,5;0C*(;0656-76::0)3,(5;(.650:;:-69;/,:,:0;,:;/96<./=09;<(3:*9,,505.
<:05.;>6:<**,::0=,+6*205.:*69,:#+6*205.-69-(:;79,*0:,:*9,,505.>0;/;/,;67/0;:;/,5:<)1,*;,+
;6(9(5205.:*6905.<:05.;/,&%!!(3.690;/46;/;/,044(;<9,(5+;/,4(;<9,-694:6-;/,
*/02<5.<5@(,5=,367,796;,05:>,9,05*3<+,+05;/,:;<+@;605*9,(:,;/,796)()030;@6-C5+05.76:0;0=,(5+
9,30()3,/0;:$64,:4(33463,*<3,:/(=,),,50+,5;0C,+(:.66+05:030*6*/02<5.<5@(=09<:,5=,367,796;,05:
05/0)0;69:(5+;/,:,*6<3+),.66+;,473(;,:-69+9<.+,:0.5;(9.,;05.;/0:=09<:
Keywords
*/02<5.<5@(=09<:,5=,367,796;,05:;6>(9+:0+,5;0C*(;06556=,3)05+05.:0;,::;9<*;<9,+,:0.5
05/0)0;69: 
Disciplines
,+0*05,(5+,(3;/$*0,5*,:A$6*0(3(5+,/(=069(3$*0,5*,:
Publication Details
#(:/(+,33,9"$;9<*;<9,)(:,++,:0.5;6>(9+:;/,0+,5;0C*(;0656-56=,3)05+05.:0;,:
(5+05/0)0;69:-69;/,*/02<5.<5@(=09<:,5=,367,796;,05:6<95(36- 63,*<3(99(7/0*:(5+ 6+,3305.

B0:16<95(3(9;0*3,0:(=(03()3,(;#,:,(9*/!5305, /D796<6>,+<(<:4/7(7,9:

‐1‐

Structure Based Design towards the Identification of Novel Binding Sites and Inhibitors for
the Chikungunya Virus Envelop Proteins
Adel A. Rashad and Paul A. Keller
*
Centre for Medicinal Chemistry, School of Chemistry, University of Wollongong, Wollongong, Australia, 2522
* Correspondence to:
Paul A. Keller:
keller@uow.edu.au
Tele: +61 2 4221 4692
Fax: +61 2 4221 4287
ABSTRACT
Chikungunya virus is an emerging arbovirus that is widespread in tropical regions and is spreading quickly to temperate
climates with recent epidemics in Africa, Asia, Europe and the Americas. It is having an increasingly major impact on
humans with potentially life-threatening and debilitating arthritis. Thus far, neither vaccines nor medications are
available to treat or control the virus and therefore, the development of medicinal chemistry is a vital and immediate
issue that needs to be addressed. The viral envelope proteins play a major role during infection through mediation of
binding and fusion with the infected cell surfaces. The possible binding target sites of the chikungunya virus envelope
proteins have not previously been investigated; we describe here for the first time the identification of novel sites for
potential binding on the chikungunya glycoprotein complexes and the identification of possible antagonists for these
sites through virtual screening using two successive docking scores; FRED docking for fast precise screening, with the
top hits then subjected to a ranking scoring using the AUTODOCK algorithm. Both the immature and the mature forms
of the chikungunya envelope proteins were included in the study to increase the probability of finding positive and
reliable hits. Some small molecules have been identified as good in silico chikungunya virus envelope proteins
inhibitors and these could be good templates for drug design targeting this virus.
Keywords Alphaviruses; Chikungunya virus; Envelope proteins; Virtual screening
1. Introduction
Chikungunya virus (CHIKV) is an emerging mosquito-borne arthrogenic member of the alphavirus genus (family
Togaviridae) that has caused widespread outbreaks of debilitating human disease in the past five years [1].
Chikungunya fever (CHIKF) caused by the virus was first described in 1952 [2], and currently has been identified in

‐2‐

nearly 40 countries. In 2008 it was listed as a US National Institute of Allergy and Infectious Diseases (NIAID)
category C priority pathogen because of the high morbidity and mortality rates and major health impact [3, 4].
The symptoms of chikungunya fever infection generally start 4–7 days after the mosquito bite. Infection
usually presents in two phases; the first is acute, while the second stage is persistent (chronic), causing disabling
polyarthritis [5]. Acute infection lasts 1–10 days and is characterized by a painful polyarthralgia, high fever, asthenia
(weakness), headache, vomiting, rash, and myalgia [6]. The persistent chronic stage of CHIKF is characterized by
polyarthralgia that can last from weeks to years beyond the acute stage [7]. Neurological disorders including
encephalitis, myelopathy, peripheral neuropathy, myeloneuropathy and myopathy have also been reported [8].
The CHIKV genome is approximately 11.8 Kb in size and consists of a single stranded, positive sense RNA
genome with two open reading frames (ORFs) [9], one in the 5` end which encodes two polyproteins, the precursors of
the non-structural proteins. The second ORF at the 3` end encodes the structural proteins, the capsid (C), envelope
glycoproteins E1 and E2 and two small cleavage products (E3, 6K). Similar to other members of the alphaviruses, the
CHIKV starts the life cycle by entering the target host cells by pH dependent endocytosis via a receptor mediated
interaction [10]. A recent study identified prohibitin1 (PHB1) as a microglial cell expressed CHIKV binding protein
[11].
After entering the cell, the endosome acidic environment triggers conformational changes in the viral envelope
complex made of E1 and E2 proteins, resulting in dissociation of the E2-E1 heterodimers, and the formation of E1
homotrimers. The E1 trimer inserts into the target cell membrane via its hydrophobic fusion peptide (fusion loop) and
refolds to form a hairpin-like structure. Exposure of the E1 fusion peptide leads to releasing of the nucleocapsid into the
host cell cytoplasm [12],[13]. During the replication cycle inside the host cell, the capsid protein is released, and the
pE2 and E1 glycoproteins are translated in the Golgi and are moved to the plasma membrane, where pE2 is cleaved by
furin-like protease activity into E2 and E3 [14].
Glycoprotein E2 is responsible for receptor binding whereas E1 is responsible for membrane fusion [4]. E3
contains the 64-amino-terminal residues of p62 and mediates the correct folding of pE2 and its subsequent association
with E1 [15]. E3 also protects the E2-E1 heterodimer from premature fusion with cellular membranes [16]. Furin
maturation of p62 into E3 and E2 during transport to the cell surface primes the spikes for subsequent fusogenic
activation for cell entry. Mature virions bud at the plasma membrane via interactions between E2 and genome-
containing viral nucleocapsids present in the cytoplasm [17], ready for infecting new cells. The crystal structures of
both the immature and the mature glycoprotein complexes have recently been solved [17], (Fig. 1).

‐3‐


Fig. 1 Crystal structure of the immature envelope glycoprotein complex of Chikungunya. Generated from the pdb file:
3N40 [17].
E1 is folded into three β-sheet rich domains (I, II and III). E2 is an all β protein belonging to the
immunoglobulin superfamily, with three domains A, B and C. Domain B is at the membrane upper end and domain C
is towards the viral membrane: Domain A is at the centre while domain C binds to the adjacent domain II of E1. The
long β – ribbon of E2 makes most of the connection with E3. Furin loop is E2-E3 junction in the immature complex;
this junction contains a functional proprotein convertase motif which is cleaved by the cellular proteases; furin-like
proprotein convertases during the maturation of the glycoproteins [14]. The amino acid His60 in this junction is the
critical residue that determines the spectrum of furin and furin-like convertases that process E2-E3 glycoprotein
complex [18]. The U shaped fusion loop of E1 is inserted in a groove between E2 domains A and B being stabilized by
hydrogen bonds with E2 histidine side chains [17]. In the neutral pH, E3 maintains E2 domain B in an orientation with
respect to domain A in such way that it creates the groove accommodating the E1 fusion loop, protecting the virus from
premature fusion with other cellular membranes [17],[19]. Some residues in domain B of E2 are believed to be
associated with cell recognition [16]. The fusogenic activity of the E1 fusion peptide is highly dependent on pH change.
The histidine residues of E2 are believed to be involved as the pH sensor for the activation of the fusion protein at
lower pH [17] due to the increased probability of histidines to become positively charged at lower pH values, based on
the fact that the imidazole ring of the histidine residue is the only amino acid side chain whose apparent dissociation
constant from protons (pKa) falls within the physiological range. Within the E1 fusion peptide sequence, the glycine
residue (Gly91) is critical for the fusion process. Also, it was found that one histidine residue at E1 230, which is
located outside of the fusion sequence, is also critical for the fusion [20].
During the chikungunya fever, some limited symptomatic treatments including corticosteroids may be used in
cases of debilitating chronic CHIKV infection [21],[22], and only in the last 24 months have efforts for development of
therapeutics been reported such as arbidol [23], mycophenolic acid [24], daphnane-type diterpenoids [25],
harringtonine [26], purines and β-lactams based inhibitors [27], and the immunostimulant polycytidylic acid [Poly

Frequently asked questions

Q1. What amino acids are able to form hydrophobic interactions?

valine, alanine and proline amino acids within this pocket are also able to participate in the hydrophobic interactions. 

Q2. What have the authors contributed in "Structure based design towards the identification of novel binding sites and inhibitors for the chikungunya virus envelope proteins" ?

The possible binding target sites of the chikungunya virus envelope proteins have not previously been investigated ; the authors describe here for the first time the identification of novel sites for potential binding on the chikungunya glycoprotein complexes and the identification of possible antagonists for these sites through virtual screening using two successive docking scores ; FRED docking for fast precise screening, with the top hits then subjected to a ranking scoring using the AUTODOCK algorithm. Both the immature and the mature forms of the chikungunya envelope proteins were included in the study to increase the probability of finding positive and reliable hits. 

Q3. What is the function of the furin loop?

Furin loop is E2-E3 junction in the immature complex; this junction contains a functional proprotein convertase motif which is cleaved by the cellular proteases; furin-like proprotein convertases during the maturation of the glycoproteins [14]. 

Q4. What is the likely result of the ligands identified for this site?

Being able to bind to residues in both E1 and E2, the ligands identified for this site are most likely to confirm their hypothesis and stabilize the E1-E2 heterodimer and prevent the dissociation.‐ 

Q5. What is the common type of interaction between the ligands in the mature form?

The presence of an electron rich system results in strong noncovalent molecular interactions, e.g. the π-cation interaction between E2 Arg100 (Arg36 in the mature form) and E1 Lys52. 

Q6. What was the use of the autogrid 4.2 algorithm?

AutoGrid 4.2 algorithm was used to evaluate the binding energies between the inhibitors and the enzyme and to generate the energy maps for the docking run. 

Q7. What is the mechanism of inhibition of the furin loop?

The indirect allosteric inhibition mechanism might be through the inhibition of the interaction between the CHIKV envelop protein, and hence the furin susceptible motif (furin loop), and the acting protease (therefore, the Life Chemicals protein-protein inhibitors library was used here), or through trapping the glycoprotein conformation in one inactive form (relative to the furin cleavage step), which does not interact with the acting proteases. 

Q8. What was the result of the cluster analysis?

Cluster analysis was performed on docked results, with a root-mean-square tolerance of 2.0 Å, the docked poses were ranked according to the binding energies and ligand efficiencies, and finally the five lowest energy poses (Tables 2-5) were selected as the resultant complexes with the proteins. 

Q9. What is the role of E3 in the formation of pE2?

E3contains the 64-amino-terminal residues of p62 and mediates the correct folding of pE2 and its subsequent association with E1 [15]. 

Q10. What is the role of histidine in the chikungunya fever?

During the chikungunya fever, some limited symptomatic treatments including corticosteroids may be used incases of debilitating chronic CHIKV infection [21],[22], and only in the last 24 months have efforts for development of therapeutics been reported such as arbidol [23], mycophenolic acid [24], daphnane-type diterpenoids [25], harringtonine [26], purines and β-lactams based inhibitors [27], and the immunostimulant polycytidylic acid [Poly‐ 4 ‐   (I:C)]  [28]. 

Q11. What software was used to extract the top 20 docked poses?

The top 20 docked poses ranked in each of the four binding sites were then extracted as pdb files, and were processed with AutoDock Tools 1.5.6rc3 (ADT) graphical interface [40]. 

Q12. What is the effect of the furin susceptible motif on the chikv envelop?

being a groove in this area looking like the enzyme mouth (Fig. 4), bound small molecules in this site might act as indirect allosteric inhibitors for the furin susceptible peptide motif, and therefore, might impair the cleavage step by the furin proteases. 

Q13. What is the likely site for the fusion loop?

Small molecules binding to this narrow channel will have significant effects; this might not only freeze the relative movement of E2 domains A and B, but might also freeze the fusion loop through stabilizing interactions, and consequently, prevent the exposure of the fusion loop. 

Q14. What are the sites that look critical to the protein functions?

The authors managed to identify two sites that look critical to the protein functions; mainly the fusion process, based on the functionality and the location of the sites. 

Q15. What is the role of histidine in the chikunya virus envelope?

the authors report for the first time the novel binding sites in the CHIKV envelope glycoproteins that can be used as sites for inhibitors that could alter the function of the envelope proteins and consequently, inhibit the virus fusion function. 

Q16. What is the effect of small molecules on the fusion process?

small molecules that bind to this site may stabilize the E1-E2 heterodimer and prevent their dissociation during the fusion process. 

Q17. How many hits for each site in both forms of the CHIKV envelope proteins were identified?

Five hits for each site in both forms of the CHIKV envelope proteins were identified revealing some important features for further developing antagonists for these proteins. 

Q18. What are the advantages of using Fred and Autodock together?

Fred and Autodock are powerful tools for the preliminary identification of hits [41], and have previously been used together successfully for the in silico identification of potential inhibitors [42]. 

Q19. What is the common sequence in the top 5 docked poses?

inspection of the top 5 docked poses in each site reveals that they have the common sequence: heterocycle-S-CH2-CO-N, the amidic nitrogen in this sequence might be NH, and also can be a part of another ring system, Fig. 5 shows the 2D representations of the top docked poses in site 2 for both the immature and the mature forms of the envelope glycoproteins.