Reference: Muller et al. “Passenger deletions generate
therapeutic vulnerabilities in cancer.” Nature (2012) 488, pgs 337 – 341
Reference: Lehner and Park. “Exploiting collateral damage.”
Nature (2012) 488, pgs 284 – 285
Reference: Johnston, Iain. “The chaos within: Exploring
noise in cellular biology.” Significance (2012) August, pgs 17 – 21
Innovative
ways to specifically kill cancer cells within the context of a living human
body are necessary. The scientific
community has plenty of ideas: Trojan horse proteins bearing chemotherapeutics,
exploiting the nature of cancer cell surface receptors, intense high throughput
studies to identify cancer-sensitive compounds and targeted therapies towards
known oncoproteins. Unfortunately, just
as each human is unique, cancer is an all-encompassing term for hundreds of
different diseases that each has their own set of complications to be
overcome. Forward thinking and ingenuity
are keys to successful progression. To
this end, Muller et al. report in this week’s Nature magazine about an
experimental design that identifies key vulnerabilities in cancer cells by
highlighting what proteins are not present.
Enolase
is an essential enzyme necessary for the second to last step of glycolysis. Three homologues of this enzyme exist with three
different gene expression profiles: ENO1
is ubiquitous, ENO2 is restricted to
neurons, and ENO3 is only in
muscle. It has been shown that
invertebrates and mice carry several homologous genes that encode for proteins
capable of doing each other’s jobs. The
beauty of this redundancy occurs when one of the genes is knocked out: the
other proteins are able to pick up the slack and death isn’t an inevitable
result. In the case of enolase, both ENO1 and ENO2 are expressed in neural cells and both are capable of
performing the same function. But,
imagine that one gene becomes mutated.
The cell would then have an unhealthy reliance on the other ENO gene. Since the enzyme is essential, a blow to the
other enolase homologue should result in death to that cell.
Cancer cells
have an unhealthy habit of collecting mutations. The Cancer Genome Atlas Research Network has
sought to study the genomes of cancer to establish what mutations have turned a
once healthy cell into a feast of illness.
Gliobastomas are a type of brain tumor that affects glial cells, which
expresses both ENO1 and ENO2, but is far more reliant on ENO1.
Interestingly, the lp36 locus, home of the ENO1 gene (among others) is often deleted in glioblastoma. In theory, this should create an Achilles
Heel out of ENO2.
Muller et al. began with two different cells
lines: one expressed both ENO1 and ENO2 (ENO1 wild type) while the other only expressed ENO2 (ENO1-null). Upon two independent shRNA-mediated
knockdowns, only the cells only expressing ENO2
displayed marked inhibition of proliferation.
Wanting to further prove their concept, the authors then treated the
cells with the enolase inhibitor phosphonoacetohydroamate (PHAH). The compound displayed potent toxicity towards
the ENO1-null cells and little impact
on the ENO1 wild type cells. Finally, PHAH was titrated into cells with
varying degrees of both ENO1 and ENO2 expression. Intriguingly, the data showed a direct
relationship between the sensitivity of the cells to PHAH and their enolase
activity profiles.
The
lp36 locus contains other essential housekeeping proteins. In addition to these cells being reliant on ENO2, they might also be unnaturally
resting on other individual proteins whose homologues have been knocked
out. Determining what they are and
inhibiting them as well could lead to even greater effectiveness at cancer
cell-specific death. The paper ends up
with this thought: “By one estimate, 11% of all protein-coding genes in the
human genome are deleted in human cancers.
Thus, given the large number of homozygous deletions across many
different cancer types spanning many hundreds of genes, the model described
here for [glioblastoma] should be applicable to the development of personalized
treatments for many other cancer types.”
While
reading this paper, I kept thinking back to another article I read in the
journal Significance concerning the random events and complete chaos that is
the inside of a cell. Take two cells
with exactly the same genome. Variability
exists in the expression profiles of all genes between those two cells due to
random chance. In the Muller et al. example, one single cancer cell
in a gliobastoma may express ENO2 at
a much higher level than its neighboring cell, which means that Cell A will
need more drug to kill it than Cell B even though both have the exact same
genome. But scientists don’t consider
Cell A and Cell B, we consider whole cell populations and assume they are all
acting identically. One IC50
value represents the cells in that particular plate at that particular time and
is averaged with other cell populations at different times. Eventually, scientists are looking at
averages of averages. We make broad assumptions on broad pieces of data that are based on what is happening right then. But cancer isn’t
static; tumors within a human body are an ever evolving entity that picks up
more mutations and creates more roadblocks as time goes on. Cancer is a many-headed hydra.
