Biology Reference
In-Depth Information
The enormous size of our most recent dataset (an
order of magnitude larger than those previously collected)
allowed a more powerful analysis to be performed.
Specifically, to better understand the cellular processes
involved in the general response to chemical stress, we
calculated the frequency of occurrence of significantly
enriched GO biological processes and protein complexes
across all small molecules tested. The result of this
analysis revealed a coherent network of highly inter-
connected cellular processes that define a complex system
of resistance mechanisms activated under conditions of
chemical stress. The existence of such a drug resistance
network is significant, as it implies that the cellular
response to drug is in fact limited, and is defined and
controlled by the regulation of processes identified in this
network (
Figure 8.4
A).
Several of the enriched modules in the MDR network
have been previously observed
[21,26]
; these include the
ergosterol biosynthesis, vesicle-mediated trafficking and
vacuolar biogenesis & regulation of pH modules. Many of
the genes in the ergosterol biosynthetic pathway (highly
homologous to the human cholesterol pathway), when
deleted, are well known to increase the permeability of the
cell to small molecules, likely due to effects on membrane
composition. Several of these genes, particularly ERG2 and
ERG6, are often deleted specifically for the purpose of
increasing the sensitivity of yeast to small molecules in
drug studies. The requirement for the intracellular vesicle-
mediated trafficking network for resistance to drug is not
clear; however, perturbations in the cell wall and/or plasma
membrane are the first cellular components that meet an
incoming small molecule, and the response may require
endocytosis for drug transport and sequestration. It may
also include the activation of downstream signal trans-
duction pathways and upregulation of proteins involved in
remodeling the cell wall and/or plasma membrane,
requiring the trafficking of these parts from the ER to the
cell surface. Such processes require constant turnover in
intracellular vesicular trafficking. This requirement for
alterations in the activities of intracellular trafficking in
response to drug is consistent with similar observations in
mammalian cells
[73
chemical
genetic interaction might be the result of
decreased vacuolar pH and enhanced trapping of CADs in
the vacuole. Consistent with this proposed mechanism,
raising vacuolar pH by treatment with bafilomycin A1,
a specific inhibitor of vacuolar-type H
รพ
-ATPase, conferred
resistance to these drugs. In human patients, CADs are
associated with drug-induced phospholipidosis (DIPL),
a phospholipid storage disorder linked to clinical toxicities
[78,79]
. To explore the potential connection between the
VB module and DIPL, we built a structural model based on
small molecules associated with the VB module and
demonstrated that the model performed well in predicting
small molecules known to cause DIPL (based on the
literature). Thus, yeast chemogenomic profiling is an
effective predictor for small molecules that may cause
DIPL. Moreover, our results identified potential genetic
factors that might influence tolerance of CADs in patient
populations.
Particularly striking in modules our MDR network
involve ribosomal biogenesis and RNA processing,
including core members of stress granules and P-bodies
[80,81]
. For example, the biological role of stress granules
and P-bodies is an area of intensive research that, while well
studied, has thus far escaped characterization, partly
because the members that make up these stress granules and
P-bodies are stress specific, and often include members that
are seemingly involved in fundamental processes with
widely different functional roles (e.g.,
[82,83]
). Five
modules that have not previously been associated with
multidrug resistance include the nuclear pore complex
(NPC), the cytoplasmic exosome, mRNA processing body
and stress granule (SG), translation (TL), RNA polymerases
I, II and III related functions (RNAP) and ribosome
biogenesis (RB). Our data suggests that these modules form
a coherent process that generally captures the cellular
response to chemical stress. Upon close examination of
these functionally linked processes, combined with hierar-
chical clustering of all genes based on co-fitness, a sub-
cluster of ~400 co-fit genes emerged, revealing the processes
required for each step on the road to translation at an
exquisite level of detail (
Figure 8.3
). This road, defined
experimentally, includes all of the steps required for ribo-
some biogenesis; a tightly regulated process that requires the
coordination of several, sometimes simultaneously occur-
ring processes, including the coordinated action of all three
RNA polymerases. Ribosomal biogenesis starts with the
synthesis of ribosomal proteins in the cytoplasm and import
into the nucleus, followed by assembly of the subunits and
synthesis of ribosomal RNA (rRNA) in the nucleolus. As the
ribosome further matures, the subunits are transported
through the nuclear pore complex and into the cytoplasm,
where the final steps of ribosomal maturation occur.
Co-fitness-based gene clusters are enriched for genes
involved in discrete steps of ribosome biogenesis, and the
e
75]
.
The vacuolar biogenesis and regulation of pH VB
module has also been previously recognized as required for
MDR
[21,26,47]
in yeast. Our recent data provided an
additional insight: many of the profile that were enriched in
vacuolar biogenesis were profiles of cationic amphiphilic
drugs
e
(CADs), which often exhibited strong chem-
ical
genetic interactions with NEO1. NEO1 is a member
of the evolutionarily conserved P-type flippases
[76]
and
a major regulator of vacuolar pH. A temperature-sensitive
NEO1 strain exhibits hyper-acidified vacuoles
[77]
, and
because CADs are believed to accumulate in acidic vesicles
due to their basic nitrogen
[78]
, it follows that the NEO1
e
