Xavier Lab

Program for Computational Biology | Memorial Sloan-Kettering Cancer Center


We investigate how cell-cell interactions govern complex behavior in cell populations using combined experimental and computaitonal approaches. We work on three systems of biomedical relevance:


Cooperation and conflict in pathogenic bacteria

Rather than being solitary organisms pathogenic bacteria have important social traits. Bacterial cells cooperate in many ways by communicating via signaling molecules, exchanging metabolites, building biofilms and moving collectively as swarms. However, collaboration is vulnerable to the evolution of selfish ‘cheaters’ who may exploit the advantages generated by the collective without contributing themselves. How do cooperative behaviors evolve in spite of the evolutionary conflict between individual- and population-level interests?

Biofilm formation requires that individual cells produce and secrete copious amounts of extracellular polymeric substances (EPS). Polymer production can carry a cost to individual cells, which would make polymer secretion an exploitable trait. We investigated this problem using individual-based modeling. We observed that in spite of the cost, polymer-producers can have a competitive advantage by pushing their own lineage towards oxygen rich regions. Thus, we propose that rather than being purely cooperative communities, biofilms are in fact stabilized by the competitive secretion of EPS.
See: Xavier JB, Foster KR (2007) Cooperation and conflict in microbial biofilms. Proc Natl Acad Sci U S A. 104(3):876-81. [pdf]

Text Box:  Swarming in Pseudomonas aeruginosa (an opportunistic human pathogen) is another trait open to exploitation. Swarming requires the production of a copious film of rhamnolipid surfactants. We conducted experiments and found that a mutant rendered incapable of emitting surfactant (green) could swarm along the film from a wild-type strain (red) without adversely affecting the producer, yet it overwhelmed a different strain engineered to produce surfactant continuously with no regulation. By investigating the regulation of the genes responsible for rhamnolipid synthesis we found that wild-type P. aeruginosa has a mechanism to prevent takeover by cheaters. Rhamnolipid production is only synthesized when carbon is abundantly available and when growth is limited by a lack of nitrogen. By being frugal with surfactant production, producers can escape cheaters waiting for a free ride.

Xavier JB, Kim W, Foster KR (2011) A molecular mechanism that stabilizes cooperative secretions in Pseudomonas aeruginosa. Mol Microbiol. 79(1):166-79. [pdf]

van Ditmarsch D, Boyle KE, Sakhtah H, Oyler JE, Nadell CD, Deziel E, Dietrich LEP, Xavier JB. Convergent evolution of hyperswarming leads to impaired biofilm formation in pathogenic bacteria. Cell Reports. [open access]

The study of bacterial cooperation can assist in the rational development of novel therapies that specifically target interaction in pathogenic populations. Drugs that target individual level traits of bacteria, such as antibiotics and other antimicrobials, create a strong selective pressure for resistance. Our growing understanding of social interaction opens the way to novel therapies that do not select for resistance.

Exploiting social evolution in biofilms. Boyle KE, Heilmann S, van Ditmarsch D, Xavier JB. Current Opinion in Microbiology. 16:1–6 [PDF]

Species interaction in the gut microbiota

Text Box:  Our gut hosts bacteria in such high numbers that they outnumber our own cells by 10:1. The gut microbiota is a complex symbiotic microbial community that helps digest nutrients and fight off invasion by pathogens. Yet, in spite of its central role for human health, this complex microbial ecosystem remains largely unknown to us. Many of the bacterial species are unculturable which makes even a simple census difficult.
This is changing in recent years thanks to metagenomics. By sequencing variable regions of the 16S ribosomal RNA we are now able to probe the specific content of complex microbial communities in unparalleled ways. The application of this technology to the gut microbiota is revealing that species composition and metabolic funciton correlate with the host health and suggests a range of medical applications of microbiome biology. We can envision that the analysis of microbiota composition will be used in the future for diagnostics and therapeutics to restitute healthy balances in this complex system.
Before the clinical translation of microbiome biology is possible we must seek to understand the ecological processes governing its composition dynamics and function thoroughly. Our goal is to characterize the human gut microbiota through computational models that quantitatively describe its ecology, i.e. the dynamics of the species, how they interacts with the innate immune system and, ultimately, how species dynamics influences host health and disease. We are applying modeling techniques previsouly developed for complex microbial communities in environmental engineering (e.g. pdf).
This project is supported by the Lucille Castori Center for Microbes, Inflammation and Cancer (established June 2010).

Taur Y, Xavier JB, Lipuma L, Ubeda M, Goldberg J, Gobourne A, Joo Lee Y, Dubin KA, Socci ND, Viale A, Perales MA, Jenq RR, van den Brink M, Pamer EG. Intestinal Domination and the Risk of Bacteremia in Patients Undergoing Allogeneic Hematopoietic Stem Cell Transplantation. Clin Infect Dis. [PDF]

Bucci V, Bradde S, Biroli G, Xavier JB. Social Interaction, Noise and Antibiotic-Mediated Switches in the Intestinal Microbiota. PLoS Computational Biology. 8(4): e1002497 [PDF]

Cell-cell communication in cancer

Text Box:  Tumors establish interactions with their microenvironment that are essential to cancer progression and invasiveness. Uncovering the mechanisms of cancer-stromal interactions can lead to novel therapies that specifically target the tumor microenvironment. The interactions between cancers and normal host cells involve many molecular players but is also dependent on physical properties of the tumor microenvironment.
A hallmark of cell-cell signaling processes is that they depend not only on biochemical factors, such as cell signal production rate and sensitivity to signal, but also on physical factors, most notably the extracellular diffusion of signaling molecules and the spatial organization of the tumor. Cell-cell signaling becomes very complex when many cell types and different signals are involved in a population. Often, the signaling pathways interact in coplex feedbacks that bring even more complexity. Mathematical formalism brings this complexity to a tractable form.
We are adapting spatial models developed for other fields (see biofilms above and pdf) to develop computer simulations that quantitatively describe cell-cell singling in cancer in spatially structured environments. This can lead to novel predictive tools for tumor progression and eventual assist in the rational design of therapies.
This project is part of the Center for Cancer Systems Biology “Systems Biology of Cancer Diversity”.

Nadell CD, Foster KR, Xavier JB (2010) Emergence of spatial structure in cell groups and the evolution of cooperation.PLoS Comput Biol. 6(3):e1000716. [pdf]

Chang WK, Carmona Fontaine C, Xavier JB. Tumour–stromal interactions generate emergent persistence in collective cancer cell migration Interface Focus. [PDF]