In silico prediction of the human-malaria parasite interactome

Posted by Kasra

In silico prediction of protein-protein interactions within a species is an advancing field. Especially now that relatively large amounts of empirical data are available for training and validation, more and more in silico methods are being presented. However, as a host-parasite interactions enthusiast, I always had the question if the interactions between the host and pathogen proteins can be predicted. Although already done a few times for viruses such as HSV, HCV, HIV and Influenza, creation of an empirical inter-species interactome is not an easy and always affordable task. Still having an interactome database provides very valuable data in host-pathogen research. It not only reveals systemic overviews about the nature of the interaction occuring, but it can also open doors for more accessible and feasible research by suggesting a shorter list of proteins, pathways or interactions to focus on.

With this in mind, I had great pleasure in reading this single author paper by Stefan Wuchty published recently in PloS ONE that provides a computational interactome of Plasmodium falciparum and Homo sapiens. This paper actually led me to a whole body of research that has been done on experimental and computational determination of host-pathogen interactomes, albeit mostly on viruses.

In order to map P. falciparumH. sapiens interactome, Wuchty used experimentally determined host-parasite interactions plus orthologous protein groups between the two species as starting point. The false-positives were removed with various filtering methods that are beyond the reach of my knowledge of mathematics and informatics. Biological criteria such as co-expression of the interacting proteins in the host cell and specific parasite phase were also required.  Interstingly, the pattern of interactions he found, showed similarity with what was already observed with viral pathogens. In order to take control of the cell, intracellular pathogens appear to attack the host both at the protein as well as the pathway level. Hub proteins – proteins that interact with many other proteins and are envolved in multiple metabolic/siganling pathways – have been found to be an attack target not only by P. falciparum but also other pathogens. In addition, Wutchy saw that a relatively small number of human proteins interact with a big number of parasite proteins, suggesting that the parasite utilizes all it has got to take over the key proteins of the host.

This study and other works of the same style certainly provide precious knowledge about host-parasite interactions both in terms of systems biology and also hints for hands on research. I look forward to seeing these interactomes created and expanded for other pathogens and also their experimental validation and usage. Furthermore, an online database of these host-parasite interactomes would definitely make them more accessible.


Wuchty S (2011). Computational Prediction of Host-Parasite Protein Interactions between P. falciparum and H. sapiens. PloS one, 6 (11) PMID: 22114664

Blood invasion of Plasmodium falciparum is dependent on a single receptor on the surface of red blood cells

Posted by Kasra

Plasmodium falciparum parasites invade different groups of cells during their life cycle. Upon injection into humans, sporozoites pass through the skin and travel in the blood to be picked up by hepatocytes. After completion of the liver phase, merozoites come back to the blood and invade red blood cells. Finally, there is another sort of invasion happening inside the midgut of the mosquito, different from the vertebrate host. The invasion of red blood cells by merozoites is the most accessible of the three for studying. Of great scientific and theraputic interest are the proteins that allow for this binding and internalization of the parasites to occur. Blocking the blood stage of malaria would essentially abrogate the majority of its pathological complications.

Many surface proteins of Plasmodium and red blood cells had been previously proposed to be involved in this host-parasite interaction and binding. However, in almost all cases, great redundancy was shown in the protein-protein interactions; meaning that knocking out one surface protein or blocking one interaction would only replace it by another one. In some cases this would result in a change in the tendency of the Plasmodium parasites to bind mature vs. Immature red blood cells, but invasion would still happen. But now, recent work has found a definitive receptor for invasion of red blood cells by P. falciparum merozoites. The beauty of this work is not only in finally finding a ligand and receptor for this stage of P. falciparum life-cycle, but also in my opinion in making great use scientific knowledge already available for making this discovery. So here is the brief story for those of you who do not feel like reading this short yet elegant letter to Nature:

In search for a definitive receptor, Crosnier and colleagues decided to study a surface protein of P. falciparum that was found by another group to be essential for parasite growth: PfRh5. In order to find its binding protein, they went through the already published proteome of the red blood cell and picked up the secreted and surface expressed proteins. Using an ‘Avidity-based extracellular interaction screen’ (AVEXIS) they screened for binding of PfRh5 to recombinantly produced secreted or surface proteins of the red blood cell and luckily they got a single hit: Basigin or BSG. They then validated this direct interaction using Surface Plasmon Resonance and showed that the interaction occurs independently from glycosylations. Next, they showed that adding soluble BSG, blocking it by a specific antibody or shRNA knockdown inhibits invasion of red blood cells by all clinical and lab strains of P. falciparum. As I previously mentioned, inhibition of invasion was never seen before for any of the receptor-ligand pairs suggested.

Finding such a well-fit receptor for invasion of P. falciparum brings up an evident question: Are people with mutations in the bsg gene resistant to malaria? The authors found very few nonsynomymous single nucleotide polymorphisms (meaning SNPs that lead to a different protein sequence) in some populations in the databases. Blood donations from some of those SNP carriers showed actually resistance to P. falciparum invasion. Unfortunately, population genetics data is seriously lacking in areas afflicted with malaria, so whether this gene has been through positive selection or not cannot be determined at the moment. The authors are hopefull to be able to answer this question after some genome projects currently in progress in Africa are complete.

This ligand-receptor interaction appears to be specific to P. falciparum and other Plasmodium species have not been mentioned in the paper (I am pretty sure they have been checked). This makes in vivo drug testing difficult as P. falciparum is not used in the murine malaria infection model. Nontheless, P. falciparum is the most prevalent and lethal malaria-causing species. The discovery of BSG as a receptor for invasion opens many doors towards a generation of theraputics and prophylaxis, bringing us hopefully one step closer to its elimination.

Crosnier C, Bustamante LY, Bartholdson SJ, Bei AK, Theron M, Uchikawa M, Mboup S, Ndir O, Kwiatkowski DP, Duraisingh MT, Rayner JC, & Wright GJ (2011). Basigin is a receptor essential for erythrocyte invasion by Plasmodium falciparum. Nature, 480 (7378), 534-7 PMID: 22080952

What kills YOU may make THEM stronger – cross-resistance between an antimalarial and an antibiotic

Posted by Marie-Luise Winz

Ross J. Davidson et al. recently published their article on the occurrence of resistance against ciprofloxacin, a fluoroquinolone antibiotic among E.coli strains isolated from rectal swabs of patients from six remote villages in Guyana between 2002 and 2005. Ciprofloxacin is a fluoroquinolone broad-spectrum antibiotic that inhibits the bacterial gyrase and topoisomerase IV. Both type II topoisomerases are responsible for relaxing positive supercoils prior to DNA replication. Their inhibition ultimately inhibits cell replication.

Surprisingly, more than 5% of the samples showed evidence of ciprofloxacin resistant E.coli, although all the patients denied previous use of this antibiotic. This is a dramatically high percentage, even superior to the 4% of resistant strains found in an intensive care unit survey in the United States, where ciprofloxacin is actually being used. In order to determine which resistance mechanisms were used by the bacteria, the quinolone-resistance determining regions (QRDR) of the genes gyrA, gyrB (subunits of gyrase), parC, and parE (subunits of topoisomerase IV) were amplified by PCR and analysed, and the presence of quinolone-resistance (qnrA/qnrB) genes that protect gyrase and topoisomerase IV from inhibition by quionolones (Jacoby et al., 2006), was examined. Furthermore, the presence of energy-consuming export systems was analysed.
Ciprofloxacin resistance was found to be caused mostly by mutations of gyrA and parC, whereas mutations of gyrB, parE, or presence of qnrA, qnrB, and export systems could not be demonstrated.
Exposure to ciprofloxacin or other fluoroquinolones could be ruled out, due to the remoteness of the area. In contrast, a correlation between a significantly higher proportion of resistant E.coli strains and a P.vivax epidemic in 5 of the 6 villages included in this study, in late 2002 was observed – 10.2% resistant strains in February 2003 versus 3.8% in February 2002 (before the epidemic) and 3.5% in February 2005 (after the epidemic).
Chloroquine is one of the most common and most affordable antimalarial drugs used in many tropical countries, despite a high proportion of resistant P.falciparum strains, especially in Africa. The drug is still commonly used, and effective against infection with P.vivax. Chloroquine has been demonstrated to have weak antibiotic effects on some bacteria, including E.coli (Jain et al., 2003). Fluoroquinolones were actually derived from the chemically related quinolines, the family of compounds, to which chloroquine belongs.

ciprofloxacin (a) and chloroquine (b)

ciprofloxacin (a) and chloroquine (b)

By an in vitro assay, the authors were able to demonstrate that serial treatment with chloroquine was able to confer resistance to ciprofloxacin, leading to mutations of gyrA and parC, the genes that were also mutated in the resistant strains from Guyana.
However, the occurrence of bacterial resistance to ciprofloxacin and the intake of chloroquine as antimalarial treatment in the individual patient did not correlate. Nevertheless, it is possible, that either chloroquine was taken up unintentionally e.g., through contaminated drinking water or that resistant bacteria were passed on to other patients who hadn’t undergone antimalarial treatment, through faecal contamination.
The levels of chloroquine that were observed in the drinking water (not during, but after the epidemic) do not seem to be likely to cause bacterial resistance, and only one single resistant E.coli strain was found in a water sample in 2004. Yet, it is very improbable, that the use of chloroquine and the high incidence of E.coli resistance to ciprofloxacin are unrelated events.
The authors point out that the use of chloroquine as an antimalarial drug should not be discontinued, despite their findings, owing to the good efficiency of this drug against P.vivax and to the fact that it is one of the most cost-effective antimalarial drugs at hand. However, they are planning to carry out further experiments in the future, in which they want to test other quinoline antimalarials for their ability to cause bacterial fluoroquinolone resistance, and may be able to suggest safer antimalarials for future use.
They also emphasize that the use of fluoroquionolone antibiotics in regions with high use of quinoline antimalarials might have to be reconsidered.
I found this article very interesting, having already heard about cross-resistance to different antibiotics among bacteria, but not between different drugs that are used against pathogens from two different domains – eukaryotes and (eu-)bacteria. This demonstrates, once more, how tightly seemingly different things are related in our world. For us as (future) scientists, it is thus important to keep our eyes and minds open to findings that don’t seem related to our field of interest at first sight, but might be, after all, of high relevance.