Discovering Biology in a Digital World

Zika virus, drug discovery, and student projects

It’s well understood in science education that students are more engaged when they work on problems that matter.  Right now, Zika virus matters.  Zika is a very scary problem that matters a great deal to anyone who might want to start a family and greatly concerns my students.

I teach a bioinformatics course where students use computational tools to research biology.  Since my students are learning how to use tools that can be applied to this problem, I decided to have them apply their new bioinformatics skills to identify drugs that work against Zika virus.

We don’t have the lab facilities to test drug candidates, but it’s nice for students to realize they’re learning skills that could be put to use.

Here’s what we’re doing:

  1. Looking at background information about Zika virus.
  2. Using blastp to identify related proteins that are also bound to drugs.
  3. Using molecular modeling to see if those drugs might also bind to Zika virus proteins.


Getting up-to-speed on Zika virus

We found a great compilation of Zika resources at the NCBI.  CIDRAP has a great set of Zika resources as well.

My students go to the NCBI Zika resource, select the link to publications, and scan the titles to see what’s new.  This list is a bit overwhelming, so I ask them to focus on the first and last sentences in the abstract from P. Brasil et. al., Zika Virus Infection in Pregnant Women in Rio de Janeiro, and on this publication from Tang, et. al.  They need to identify birth defects associated with Zika virus infection and summarize two kinds of data that support the association between infection and birth defects.

Next, they use the Health Map link to see where infections are occurring.  It gets more personal when you see cases happening in your state.

Health Map shows Zika virus cases in real time.










We also look at the ViralZone page from Expasy to learn about the Zika life cycle and see how the Zika polyprotein gets chopped into smaller parts.  This has a link to an interesting Wikipedia page for a Zika virus receptor (DC-SIGN or CD209) that appears to be expressed in the uterus and on brain cells–at least that’s my interpretation of the RNA expression data.

But, it’s easy to get lost clicking too many links, so we go on to protein blast.


Identifying potential drug targets with BLAST

I think the easiest way to find a drug against a virus is to start by looking at compounds we already know about.  We know that many successful antiviral drugs target viral proteases and polymerases, so my students go to the Zika virus reference genome (thanks NCBI!) and get the protein sequences for the Zika virus protease NS3 and the Zika virus RNA dependent RNA polymerase.

Then they use protein blast to search the NCBI structure database and see if there are 3D structures from related viruses that are bound to drugs.

Once they’ve found a structure to work with, they reverse the search and use blastp to compare their new sequence to the sequence of the Zika protein.


Using molecular models to see if drugs might bind to Zika virus

Once our students have found structures that contain a drug, they look at amino acids that are near the drug to see if those residues are similar to those in Zika virus.


Would Sovaldi® (Sofosbuvir) work against Zika virus?

Whenever possible, I like to give examples to show an investigation might work.  When I noticed that some of my blast results included proteins from Hepatitis C virus, I decided to use this as an example.  There’s a drug that works by inhibiting the RNA polymerase in Hepatitis C  (Sovaldi® from Gilead), so I decided to find out if it might work against Zika as well.

Hepatitis C virus RNA polymerase bound to Sovadi® (Sofosbuvir) from 4WTG colored by charge.











First, I did a blastp search and compared the protein sequence from the structure 4WTG against Zika virus RNA polymerase.

blastp results from comparing Zika virus RNA polymerase to the Hepatitis C virus polymerase in 4WTG.







Only 25% of the amino acids are identical, but the E value is 0.007, so that’s encouraging.   I decided to take a closer look.

I used 4WTG as a query sequence in blastp to align it to the Zika virus polymerase sequence.  Then, I downloaded the 4WTG structure and opened it in Molecule World. I selected the drug and used the Select Nearby feature to identify amino acids that might be bound to the drug. Returning to the aligned sequences, I highlighted those amino acids in the alignment.




Interestingly, the drug binds to amino acids that are present in the same positions in both Zika virus RNA polymerase and in the Hepatitis C virus RNA polymerase.  Cool!

I took a closer look.  In the top image, two manganese atoms bound to the drug are also bound to aspartic acid residues.  These are present in both proteins.

Amino acids that interact with Sovaldi® are colored by residue in Molecule World and drawn as tubes.











In the bottom image, I can see an arginine that’s present in both proteins.  Here, it appears to participate in an ionic interaction with the drug.

Amino acids that interact with Sovaldi® are drawn with in a space filling mode and colored by element in Molecule World.











Now, these models don’t prove that Sovaldi would inhibit Zika virus replication.  But it might be worth taking a look.  If I were culturing brain stem cells like Tang, et. al (3), I might take out a loan to buy some Sovaldi® and add it to the growth medium.   Just to see what happens.

For now, I’m looking forward to seeing what my students find.

Note:  All the molecular modeling work described here was carried out with the Molecule World iPad app from Digital World Biology.



  1.  The Zika Virus Resource at the National Center for Biotechnology Information
  2. Brasil P,  Zika Virus Infection in Pregnant Women in Rio de Janeiro N Engl J Med. 2016 Mar 4. [Epub ahead of print]

  3. Tang et al., Zika Virus Infects Human Cortical Neural Progenitors and Attenuates Their Growth, Cell Stem Cell (2016),
  4. Stephen F. Altschul, Thomas L. Madden, Alejandro A. Schäffer, Jinghui Zhang, Zheng Zhang, Webb Miller, and David J. Lipman (1997), “Gapped BLAST and PSI-BLAST: a new generation of protein database search programs”, Nucleic Acids Res. 25:3389-3402.
  5. Appleby TC, Perry JK, Murakami E, Barauskas O, Feng J, Cho A, Fox D 3rd,
    Wetmore DR, McGrath ME, Ray AS, Sofia MJ, Swaminathan S, Edwards TE. Viral
    replication. Structural basis for RNA replication by the hepatitis C virus
    polymerase. Science. 2015 Feb 13;347(6223):771-5. doi: 10.1126/science.1259210.