Document Text (Pages 71-80) Back to Document

In Silico Drug Design of Biofilm Inhibitors of Staphylococcus epidermidis

by Al-mulla, Aymen Faraoun, MS

Page 71

4. Results and Discussion
4.1 Drug Design
4.1.1 Target Preparation:
I. Target Determination:
Some of the proteins and genes of S.epidermidis were studied to identify
the most effective target for biofilm maturation.
The following targets were nominated:

1. sarA proteins and genes.
2. IcaR proteins and genes.
3. IcaC genes.
4. PIA polysaccharides.
5. AtlE.
6. RNA-Polymerase.
7. Sigma factors of RNA-Polymerase.
8. SesC proteins.
9. Quorum sensing autoinducers.
10. Histidine kinases.
Proteins rather than genes are favored as drug targets because any
mutation in the gene will produce strains resistant to the drug. The targeting
of gene products directly associated with virulence will likely to be more
reasonable, since mutations in the targeted genes will not stimulate
resistance and the drug can still attenuate the pathogen (Garbom et al.,
2004). A sarA protein was chosen from these targets as it acts as an

Page 72

icaADBC operon stimulator (Tormo et al., 2005a). SarA is a 124-residue
DNA binding protein encoded by the sarA locus, which consists of three
overlapping transcripts, driven by three distinct promoters, P1, P3 and P2
(Bayer et al., 1996).
DNA binding and profiling studies suggest that the sarA protein may
regulate target genes by directly binding to target gene promoters or
indirectly via downstream effects on regulons (e.g. binding to
the agr promoter) or by stabilizing mRNA during the log phase (Cheung
et al., 2004; Roberts et al., 2006).
SarA, like its homolog sarR, is a dimeric winged helix structure with each
monomer consisting of 5 α-helices, 3 β-strands and several loops (α1α2β1α3α4-β2β3-α5).
The sarA dimer possesses a central helical core and two
winged helix motifs. Within each winged helix motif there is a helix-turnhelix
motif (α3α4) and a β-hairpin turn wing (β2β3), both of which are
putative DNA binding domains (Liu et al., 2006).
The importance of this protein comes from its multifunctional regulatory
activity. First of all, it acts at the initiation step of biofilm production by
direct binding to icaA promoter enhancing transcription of icaADBC
operon. Furthermore, sarA influences the regulation of biofilm formation
via an agr-dependent pathway. It has also been found that sarA enhances
the proteolytic enzymes activity, which has an important rule in the
regulation of biofilm development (Tormo et al., 2005a). So, blocking this
protein will hit the biofilm development process in many stages (Figure

Page 73

Figure 4.1 Schematic overview from String database of the effect of sarA on many
proteins and transcriptional factors involved in biofilm development process

The sequences of the protein were downloaded from the National Center
for Biotechnology Information (NCBI); Uniprot database blast was then
used to determine its existence in most bacterial strains (Figure 4.2).

Page 74

Figure 4.2 Uniprot blast engine searching all bacterial strains containing sarA protein

SarA protein of 25 strains of S. epidermidis were aligned with Mega5.1
software to determine the most popular one (Figure 4.3).

Page 75

Figure 4.3 Alignment of sarA protiens from 25 different S.epidermidis strains by
Mega 5.1 software

Page 76

The WebLogo website was used to observe differences between the 25
sequences of the protein (Figure 4.4).

Figure 4.4 WebLogo results of comparison between the 25 sarA sequences

The results of Mega software show that the differences between sarA
sequences aligned were very low. The WebLogo website gave a more
accurate view as it displays small differences at amino acids No. 49 and
116. Therefore, sarA protein of the VCU144 strain of S. epidermidis was
chosen as the target protein sequence (Figure 4.5).

Page 77

Figure 4.5 Primary structure of sarA protein in S. epidermidis VCU144

Further investigations about the protein were done to determine its
location in bacteria, its virulence and antigenicity. A CELLO subcellular
localization predictor was used for localizing the protein and it was
determined as cytoplasmic.
A BTXpred server was used for virulence determination; the protein was
not a toxin. VaxiJen server was used for antigenicity prediction; the protein
was classified as non- antigen.

II.Homology Modelling:
Unfortunately the sarA protien of S. epidermidis has not been crystallized
yet, and no 2D NMR studies have been found for it. So, the next choice
was to model it from the most identical protein. RaptorX website (Källberg
et al., 2012) was used for modeling the protein (Figure 4.6).

Page 78

Figure 4.6 SarA protein modelling by Raptorx website

Page 79

The protein was modeled by using 2fnpA and 2frhA proteins (codes of
sarA protiens of different S. aureus strains in Protein Data Bank) as
templates. The identity score was 85% and 90% respectively. Figure 4.7
shows the modelled protein in a PDB format.

Figure 4.7 SarA 3D Homolog of S. epidermidis

RaptorX website estimated two hypothetical structures. These results
were analysed by Qmean website for determining the reliability of the
structure, where the first one gave 70% score and the other gave 66% score
of reliability. The first model, then, was chosen as the PDB format of the
sarA protein (Figure 4.8).

Page 80

Figure 4.8 Z-score for the most reliable sarA model from QMEAN website

4.1.2 Ligands Search:
The first step in ligand determination was the search for antibiofilm groups
from research papers. Antibiofilm groups were:

1. NSAID(Alem and Douglas, 2004)
2. Furanones (Lӧnn-Stensrudet al., 2009)
3. Anthelmintic drugs(Imperiet al., 2012)
4. Polyphenols (Binket al., 2011)
5. Miscellaneous (Rifampicin, vancomycin, Diacetyl , Acetic acid,

Ferric ammonium citrate, Allicin)
From these groups, 20 molecules were chosen, depending on their relation
to biofilm as initial molecules:

© 2009 All Rights Reserved.