Adaptive metabolic strategies of Mycobacterium tuberculosis to combat stress from antibiotics and ROS

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Request accessible transcript

Transcript (if available)

Content

Adaptive Metabolic Strategies of Mycobacterium tuberculosis
to combat stress from Antibiotics and ROS

By

Philip Jānis Sell

A Thesis Presented to the
FACULTY OF THE USC KECK SCHOOL OF MEDICINE
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
MASTER OF SCIENCE
Molecular Microbiology and Immunology

August 2021

© Copyright 2021 Philip Jānis Sell
ii
TABLE OF CONTENTS

List of Tables............................................................................................................ iii
List of Figures.......................................................................................................... iv
Abstract.................................................................................................................... v
Introduction.............................................................................................................. 1
Results and Discussion............................................................................................ 6
Confirmation of itreS and iotsA strains………………………………………… 6
Suppression of TreS causes defect in biofilm formation…………………….. 6
Increased itreS sensitivity to antibiotics and ROS stress……………………. 7
Fluctuation strain isolation and genetics………………………………………. 9
Flux strain defect in glycerol metabolism……………………………………… 10
Flux strains able to maintain lower ROS levels under stress……………….. 11
Flux glycerol metabolomics……………………………………………………... 12
Confirmation of desaturase A CRISPRi strains………………………………. 14
Suppression of DesA causes defect in biofilm formation……………………. 15
Increased idesA sensitivity to antibiotics and ROS stress…………………… 16
Metabolomics of increased idesA susceptibility to starvation……………….. 17
Materials and Methods............................................................................................. 20
Tables...................................................................................................................... 25
Figures..................................................................................................................... 27
References............................................................................................................... 39
iii
List of Tables

Table 1: Primer List………………………………….…………………………………… 25
Table 2: Sequencing analysis of RRDR region in rpoB gene in RIF
R
mutants.…… 26

iv
List of Figures

Figure 1: Schematic diagram of TMM/TDM in Mycobacterial membrane and
Trehalose metabolism…………………………………………………………………… 27
Figure 2: CRISPRi system knock down of TreS and OtsA enzymes…………........ 28
Figure 3: Suppression of TreS enzyme causes defect of biofilm formation of
PLB………………………………………………………………………………………… 29
Figure 4: Increased itreS sensitivity to antibiotics and ROS………………………… 30
Figure 5: Flux strains are more tolerant to RIF than naïve WT……………………... 31
Figure 6: Flux strains glycerol growth defect is rescued by dextrose
supplementation………………………………………………………………………….. 32
Figure 7: Flux strains maintain lower ROS levels under RIF treatment……………. 33
Figure 8: Flux strain metabolomics for dextrose rescue of glycerol defect………… 34
Figure 9: Suppression of Desaturase A enzymes cause defect of growth in
palmitate…………………………………………………………………………………... 35
Figure 10: Suppression of Desaturase A enzymes cause defect of biofilm
formation of PLB………………………………………………………………………..... 36
Figure 11: Increased idesA sensitivity to antibiotics and ROS……………………… 37
Figure 12: Metabolomics of Desaturase A enzymes requirement for starvation
viability and biofilm..……………………………………………………………………… 38

v
Abstract

Tuberculosis (TB) is the deadliest disease due to a single infectious pathogen,
Mycobacterium tuberculosis (Mtb). Mtb can remain viable after extended periods of antibiotic
treatment, these transiently tolerant tubercle bacilli are called persisters. Not only can the persister
population survive conventional antibiotic treatments but it can also serve as a source for
accumulation of drug resistance mutations. It was previously found that Mtb can form persister-
like bacilli (PLB) by shifting trehalose metabolism for utilization of internal carbon and production
of antioxidants. We confirmed these findings in a common Mtb model organism, Mycobacterium
smegmatis (Msm), and additionally found a link between trehalose metabolism and acquisition
rate of drug-resistance mutations. Paralleled results in this study are reported for three
Desaturase A enzymes involved in fatty acid metabolism which show promising characteristics
as potential metabolic targets for new TB therapeutics. Similar mechanisms to trehalose
metabolism are involved with the desaturases in persister formation, suggested by dysregulation
of glycolysis and the pentose phosphate pathway as defined by metabolomic analysis. Finally,
we identified Msm strains with an irreversible drug tolerance phenotype after single exposure to
antibiotics. Irreversible drug tolerance was related to a defect in growth rate due to altered glycerol
metabolism. Control of oxidative damage was correlated to bacterial survival across all
experiments testing both, transient and irreversible, drug tolerant phenotypes. Mtb uses a range
of adaptive metabolic survival strategies for transient persistence and irreversible tolerance to
antibiotic stress. Inhibiting essential enzymes in these adaptive metabolic pathways could greatly
increase efficacy and decrease length of current TB treatments.

1
Introduction

Tuberculosis (TB) is a top 10 leading cause of death, infecting 10 million and killing about
1.4 million people worldwide in 2019
1
. TB has been the deadliest disease due to a single infectious
pathogen since surpassing HIV/AIDS in 2007. The causative agent of TB, Mycobacterium
tuberculosis (Mtb), is a constant global health threat.
The current treatment regimen for TB is lengthy and drug resistance can develop if not
followed properly
2
. The incidence of rifampicin-resistant TB totaled about 500,000 cases of which
78% were multidrug-resistant
1
. Multidrug-resistant TB is defined as Mtb that is resistant to two or
more TB drugs, including one of the two most potent drugs, rifampicin (RIF) or isoniazid (INH).
Additionally, latent TB infections
3
and formation of drug-tolerant persister populations
4

present unique challenges for developing effective treatments. The major goal of Tuberculosis
research is to shorten the length of treatment by discovering more powerful ways to cure and
prevent disease.
Tuberculosis is spread when airborne droplets containing Mtb are inhaled from the cough
or sneeze of an infected individual. TB of the lung or throat is infectious. The bacteria can also
spread and cause disease in other parts of the body such as the kidneys, brain, and spinal cord.
In productive infections, Mtb prevents fusion of the lysosome to the phagosome after
phagocytosis by alveolar macrophages and alters its environment to support survival
5
. Indeed,
Mtb releases factors necessary for survival
6
by rewiring host metabolism
7,8
, which is vital to host
defense
9
. However, in a more significant and vital role to subsistence, Mtb adjusts its own
metabolism
10
. This survival mechanism occurs within granulomas after the actively replicating
population has been killed by antibiotic treatment and the remaining bacilli are contained by our
immune system. The environment inside the granuloma is not ideal for microbial survival,
featuring conditions such as hypoxia, nutrient scarcity, and presence of reactive oxygen species
2
(ROS)
11–13
. However, Mtb can survive for long periods of time in these harsh conditions
predominantly due to its strategic metabolic shift and with help from its slow replication time.
These tubercle bacilli that remain viable after extended periods in antibiotic stress
conditions are called persisters and are defined by transient drug tolerance. The persister state
is distinct from latent or dormant states
14
. In one published study, Mtb was able to be cultured
from closed lesions of TB patients after antibiotic treatment and found that 7 of the 9 isolates were
still sensitive to antibiotics
15
.
The persister populations are able to survive antibiotic treatments without gaining
antibiotic resistance conferring mutations, and therefore are able to tolerate drug treatment by an
alternative mechanism
16
. It was proposed that a dormant state allowed for persistence because
earlier antibiotics used to treat TB targeted active cell processes like growth and division. Sorting
E. coli cells using a GFP marker expressed from rrnBP1, a ribosomal promoter in which
expression is dependent on growth, it was shown that dim cells were more tolerant to antibiotics
and had different transcriptional profiles to active cells
17
. This finding suggests a strong
connection between metabolism and drug tolerance in persister populations. A similar study using
mCherry under control of isopropyl-β-D-thiogalactopyranoside-inducible promoter and
RedoxSensor green stain to detect if cells were actively replicating or metabolically active,
respectively, showed a similar correlation but found dormancy was not required for persistence
18
.
Later studies using metabolic controllers confirmed the critical role of low metabolic activity in
persister survival
19,20
. Persisters can survive antibiotic treatment because they undergo a reduced
growth rate phenotypic switch allowing for higher tolerance than other bacterial cells in a
genetically homogenous population
21,22
. There is mounting evidence that Mtb persisters use a
metabolic mechanism to survive stress conditions such as antibiotic treatment and host immune
response. Metabolomics is a relatively new field that allows for more comprehensive study of this
metabolic strategy. Using this method, studies have shown that Mtb persister formation relies on
3
an adaptive metabolic shift to mitigate toxicity
23
and ROS
24
, accompanied by cell wall
remodeling
25
.
The cell wall of Mtb is notoriously known for its characteristics of antibiotic resistance
26
,
virulence
27
, and having low permeability to drugs with many efflux pumps
28
. Furthermore, in the
Actinobacteria family, the distinctive mark of Mycobacteria is the presence of mycolic acid (MA)
in the outermost portion of the cell wall Mycomembrane
29
(Figure 1A). Mycolic acids not only
contribute as a virulence factor, but also can serve as a ROS sink.
Two important mycolic acid-containing glycolipids are Trehalose Monomycolate (TMM)
and Trehalose Dimycolate (TDM). They are composed of a trehalose disaccharide and one or
two mycolic acids, respectively. When TMM glycolipids are transported into the cell membrane,
two can be condensed to form TDM and release of one trehalose during mycolate transfer
30
.
Trehalose can also be biosynthesized from uridine diphosphate glucose by the OtsA/B pathway
31
.
Trehalose Synthase (TreS) can convert trehalose into maltose which then can enter central
carbon metabolism (CCM) through glycolysis or the pentose phosphate pathway (PPP)
31,32

(Figure 1B). Trehalose replenishes nicotinamide adenine dinucleotide phosphate (NADPH) pools
which function as a critical reducing agent for depleting ROS
24
which is also known to be an
adjunct bactericidal effect of antibiotics
33,34
. Therefore, TreS enzyme has an essential function to
counteract some of the environmental stresses that Mtb encounters within the granuloma and
after antibiotic treatment.
Additionally, TDM hydrolysis is important for formation of biofilms
35
where drug tolerance
persister populations are known to form
36
. An Mtb treS knock-out strain (ΔtreS) was shown to
have increased sensitivity to antibiotics and other stresses when compared with wildtype (WT)
Mtb
24
. Metabolomic analysis showed that the defect of trehalose conversion in ΔtreS Mtb
prevented trehalose entry into glycolysis and the PPP under stress conditions
24
. Glycolysis and
downstream pathways are the main source of ATP production and including the oxidative branch
of the PPP are major sources of NADPH biosynthesis. Also in ΔtreS a decrease in NAPDH
4
concentration was accompanied by depletion of Glutamyl-L-cysteine when compared to WT or
ΔtreS complemented with functional treS or maltose, supporting the metabolic strategy to stress
conditions
24
. Glutamyl-L-cysteine is an important precursor of the antioxidant glutathione
37,38
. The
increased susceptibility of ΔtreS suggests the critical role of TreS activity in effective stress
response in Mtb. Concurrent TMM/TDM accumulation observed in the ΔtreS strain provides
evidence that these glycolipids are the source of trehalose for TreS enzyme in stress conditions
24
.
The utilization of trehalose from TMM and TDM plays a vital role in the adaptive metabolic shift
for Mtb survival and drug tolerance.
Mycolic acids are important for biofilm formation in Mtb and both the FAS-I and FAS-II
systems are required for mycolic acid synthesis
39
. There is considerable potential for additional
metabolic targets involved in this pathway. Mycobacterial virulence and lipogenesis have been
connected to fatty acid (FA) metabolism through elucidation of the role of the FdmR repressor
control of genes involved in FA degradation
40
.
One of the indicated genes was a FA desaturase, DesA3. A requisite step in biosynthesis
of mycolic acids is the desaturation step whereby two cis-double bonds are introduced through
oxidation of saturated fatty acids allowing for addition of other functional groups to mycolic acids
41
.
The genome of Mtb contains three annotated aerobic acyl-ACP desaturases, DesA1 (Rv0824c),
DesA2 (Rv1094), and DesA3 (Rv3229c) that likely play an essential role in this pathway
42
. Both
DesA1 and DesA2 were proposed to desaturate longer acyl chains in mycolic acid synthesis
42,43
.
DesA1 was shown to function in oxidation by Calcium-binding while DesA2 functions by iron-
binding
44,45
. Their functional specificity suggests that they catalyze unique desaturase reactions
within the same MA biosynthesis pathway. DesA1 was shown to be an essential enzyme in
Mycobacterium smegmatis (Msm)
43
. Unlike the other acyl-acyl carrier protein desaturases
involved in MA synthesis, DesA3 was demonstrated to function as a membrane-bound Δ9-
desaturase in Mtb that is involved in the desaturation of palmitoyl and stearoyl for biosynthesis of
oleic acid
46,47
.
5
These desaturases have clear homologies in Msm: MSMEG_5773 (DesA1),
MSMEG_5248 (DesA2), and MSMEG_1886 (DesA3)
43,46,48
. Msm strains were used in this study
as a model organism for preliminary verification of therapeutic targets in Mtb. The presumed
connection of desaturase A enzymes in the MA biosynthesis pathway suggests that they play a
vital role in Mtb utilization of fatty acids and may function in a parallel role as TreS does to
trehalose for the other MA portion of TDM/TMM.
Indeed, similar characteristics such as production of NADPH during desaturation steps for
ROS depletion and downstream catabolism of short chain carbons for ATP generation implicate
similar mechanisms
49
. A Path-seq study showed MadR transcription control for basal levels of
DesA1 and DesA2 to maintain mycolic acid production for new cell wall in normal growth
conditions but under stress conditions an increase of expression leads to cell wall remodeling as
an adaptive response to withstand stress during infection, and finally strong suppression to stop
MA synthesis so bacteria can enter a dormant state
50
. This taken together frames a strong
promise for further elucidation of these desaturases as another set of prime metabolic targets for
future drug development.
We hypothesize that suppression of these fatty acid desaturases may exhibit similar
phenotypes as suppression of TreS in persister-like bacilli (PLB) biofilms and drug tolerance,
providing evidence for their essential role in Mtb persister metabolic adaption strategy.
Lastly, Msm is used in this study as a proven surrogate model for Mtb
51
. The major
advantage of using the non-pathogenic Msm is its fast replication time and decreased biosafety
level compared to Mtb.

6
Results and Discussion

Confirmation of itreS and iotsA strains
The inducible CRISPR-based knock down system
52
was employed to generate
experimental Msm strains in order to interrogate role of potential metabolic targets. Plasmids
containing a catalytically dead cas9 gene and target gene sgRNA are under control of a
tetracycline induced promoter. The dCas9 protein is guided to the target sequence near a
protospacer adjacent motif (PAM) sequence within the DNA to sterically hinder RNA polymerase
from accessing the target gene resulting in transcriptional repression (Figure 2A).
The sgRNA oligonucleotides were designed to be as close to 20 nucleotides in length,
ending with a Cysteine or Glutamine residue, and containing a high fold repression PAM
sequence as previously described
52
. After kanamycin selection of recombinant CRISPRi strains,
cultures were regrown and transcription levels of targeted knock down gene was measured after
treatment with 200 ng/mL anhydrotetracycline (ATc). The mRNA levels relative to WT Msm of
Trehalose Synthase and Trehalose-6-Phosphate Synthase showed statistically significant lower
levels for itreS and iotsA strains, respectively (Figure 2B, t test P value = 0.0003 and 0.0002,
respectively). Addition of the iotsA strain in this study as a negative control was to confirm
phenotypes were not a result of alternate route of trehalose than TMM/TDM.

Suppression of TreS causes defect in biofilm formation
With sufficient suppression in Msm strains, we wanted to characterize and confirm
expected phenotypes to parallel mutant strains of Mtb from previous studies
24
. First, we wanted
to confirm the essentiality of TreS in formation of persister-like bacilli (PLB) using the biofilm
7
assay. A significant defect of biofilm formation was observed on day 6 in the itreS strain treated
with ATc and this is not observed in iotsA or itreS in the absence of ATc as measured by crystal
violet (CV) staining (Figure 3A, 3B, t test P value = 0.0029). Strains were compared to WT biofilm
with itreS not treated with ATc and iotsA treated with ATc as controls. The possible growth rate
differences of strains under Sauton media that is used for the biofilm assay can be ruled out as
all strains grew similarly under these media conditions (Figure 3C). There only showed to be
defect in PLB formation in the absence of TreS, suggesting an important role in the TreS mediated
metabolic shift for PLB formation.

Increased itreS sensitivity to antibiotics and ROS stress
Next, we wanted to test drug sensitivity and identify a potential mechanism for drug
tolerance regarding TreS metabolism. We chose to test the two important first-line TB drugs, INH
and RIF, and recently approved second-line TB drug designed to treat multi-drug resistant TB,
bedaquiline (BDQ)
53
. Small volumes of serial diluted cultures were inoculated on 7H10 solid
medium supplemented with 200 ng/mL ATc and sub-MIC concentrations of isoniazid (4 µg/mL),
bedaquiline (0.0035 µg/mL), or rifampicin (4 µg/mL). 7H10 solid medium supplemented with 200
ng/mL ATc without antibiotics was used as a negative control to ensure colony phenotypes on
agar were identical to be able to compare differences only due to antibiotic treatment. The Msm
itreS displayed an increased sensitivity to all three antibiotic conditions by inability to grow
colonies at spots with equal cell numbers when compared to WT and iotsA (Figure 4A). Again,
only under TreS suppression conditions was an increase in sensitivity to antibiotic treatment
observed. This was a similar result to the drug sensitivity previously found using ΔtreS Mtb and
strongly suggests an important role of TreS in drug tolerance.
As seen in drug sensitivity spot assay, we found that itreS phenocopied ΔtreS Mtb, which
showed drastic depletion of NADPH and accompanied defect in the biosynthesis of
8
glutamylcysteine, a precursor of glutathione, antioxidant. Thus, we hypothesized that ROS
depletion through NADPH synthesis is a likely mechanism for metabolic stress adaption due to
TreS activity. To correlate this, we measured ROS levels by flow cytometry analysis using
dihydroethidium (DHE) stain. WT and itreS were exposed to 10 mM H 2O 2 for 1 hour and levels of
ROS were determined. The ROS levels in itreS were induced over 100-fold while levels in WT
were induced about 7-fold when compared to untreated condition (Figure 4B, t test P value <
0.0001).
Bacterial viability under ROS stress was tested by plating culture dilutions after treatment
with H 2O 2 for 24 hours and counting colony forming units (CFU). The fold change for each strain
was determined from the number of CFU in H 2O 2 treated cultures compared to the starting culture
cell number before treatment and then normalized to the fold change of WT control. In both the
100 µM and 500 µM concentrations of H 2O 2 did suppression of TreS clearly show a significant
decrease in cell viability (Figure 4C, t test P values < 0.0001).
The activity of TreS enzyme proves to play an important role in PLB formation and Msm
survival under stress conditions by depleting intracellular ROS levels through NADPH recycling.
Preventing PLB formation gives clinical applicability for TreS activity as persisters are well known
to be precursor forms to drug resistant bacilli
54–56
. By preventing an earlier stage of formation of
persister populations, we could potentially prevent emergence of drug resistant mutants. Finally,
we wanted to see the effect that suppression of TreS would have on rate of resistance mutations
to first-line antibiotic, rifampicin (RIF). A fluctuation assay was performed to calculate the mutation
rate of WT and itreS strains. Msm cultures were diluted to OD 0.00005 and distributed into parallel
cultures on a 96-well plate. By diluting to such a small OD, we can select cultures from a single
cell and grow homogenous populations. Cultures were incubated at 37 °C for about 4-5 days until
saturation, allowing for spontaneous mutations to occur. Ten to twenty colonies were expanded
and plated. Then after counting number of surviving mutants under treatment of 100 µg/mL
rifampicin, mutation rates were calculated from starting cell numbers. Consistent with our drug
9
sensitivity data, itreS has a statistically significant lower rate of acquiring resistance mutations to
RIF (Figure 4D).
Not only is trehalose metabolism involved in a stress response to deal with ROS depletion,
but it also plays an important role in PLB formation and development of antibiotic resistance
mutations. For these reasons, TreS enzyme is an attractive, clinically relevant target for future
improvement of TB treatments.

Fluctuation strain isolation and genetics
To further analyze the role of an adaptive metabolic shift as a mechanism to develop drug
resistance in Mtb, ten colonies were randomly selected out of surviving WT mutant colony plates
from the fluctuation assay. Individual colonies were picked from 7H10 agar and transferred to
liquid cultures, allowed to regrow to OD 1.0, and used to make cell stocks in a 15% final glycerol
concentration. These colonies will be termed “Flux” and numbered 1-10 and abbreviated as F1-
10 henceforth.
First, we wanted to identify if their drug tolerance had a transitory or permanent nature. A
spot assay with the ten Flux colonies and naïve WT, all matched to OD 0.5, was inoculated on
7H10 solid medium containing varying concentrations of RIF (0 – 128 µg/mL). The WT colony
could not be formed starting in the plate containing 32 µg/mL RIF, as consistent with the previously
reported RIF MIC for Msm
57,58
, while all Flux colonies were able to grow. The Flux colonies could
grow in the highest tested concentration of 128 µg/mL RIF with slightly different success (Figure
5). Identical drug sensitivity results from separate stock inoculations and after maintaining Flux
cultures for many generations in non-selective complete media suggest that their drug tolerance
is irreversible and not transitory, which is different from the characteristic of persisters.
We performed sequence analysis of a RIF target gene, rpoB, in the 10 Flux strains to
identify resistance mutations to rifampicin. Surprisingly, only the F1 and F2 strains had mutations
10
in the 81 bp Rifampicin Resistant Determination Region (RRDR) of rpoB (Table 2)
59
. Over 95%
of rifampicin resistant isolates have a mutation in the RRDR region, however, there are fewer
common mutations that could cause rifampicin resistance outside of this region. Sequencing of
the full rpoB gene found that there were no other mutations in the other Flux strains that would
confer resistance to rifampicin (data not shown). Further experiments were carried out to identify
phenotypes that differed from WT to understand how these Flux strains were maintaining
rifampicin tolerance after initial exposure without RIF
R
mutations.

Flux strain defect in glycerol metabolism
Replicates of growth curve experiments continued to show differences in growth rate
between naïve WT and Flux strains when glycerol and not dextrose was provided as a single
carbon source. This difference was exaggerated when using Sauton minimal media instead of
7H9 liquid media. The optimal concentration of glycerol was found to be 1.5% where the most
significant defect in growth rate relative to WT was observed for all Flux strains (Figure 6A, t test
P value < 0.0001).
The glycerol growth defect in Flux strains could be rescued with including a supplement
of 0.2% dextrose; there were no consistently significant growth differences between WT and Flux
strains in dextrose alone (Figure 6C). Since a previous study showed that the efficacy of some
antitubercular compounds was dependent on glycerol metabolism
60
and slower growth in glycerol
metabolism has been linked to drug tolerance
61
, we wanted to explore the potential role that the
glycerol growth defect might have in the drug tolerance of the Flux mutants.
For the next experiments, two experimental groups were formed: the RRDR mutation
strains comprising of F1 and F2 and the non-mutation Flux strains which had the most consistently
slowed growth rate compared to WT from all repeated replicates, comprising of F7 and F8 (Figure
11
6B). Interestingly, the F1 and F2 strains with RRDR mutations typically had the fastest growth
rates out of all the Flux strains.

Flux strains able to maintain lower ROS levels under stress
To find out if there was a phenotypic relation between drug tolerance and glycerol
metabolism, we examined the nature of these strains under antibiotic treatment and response to
ROS. An initial experiment was performed to monitor how ROS levels between the Flux strains
and naïve WT might differ when treated with various concentrations of RIF. The ROS levels of
WT, F1, and F5 strains were measured by DHE staining with flow cytometry after treatment of
RIF for 4 hours in 7H9 complete media and fold change ROS was based on the untreated
condition for each strain. The F1 and F5 strains showed lower intrinsic ROS levels compared to
wildtype in the untreated condition. Similar ROS levels were observed at 16 µg/mL RIF which is
0.5X MIC for Msm. When RIF concentration reached at least the MIC of 32 µg/mL RIF, clearly
Flux strains were able to maintain lower levels of ROS as concentrations increased with respect
to WT (Figure 7A). This could be due to a more vigorous adaptive response to ROS in Flux cells
when RIF effect is high and not as necessary at sub-MIC concentrations.
Similar results were observed in a separate experiment with the F1, F2, F7, and F8 strains
against WT with longer RIF exposure to test conditions more consistent with the CFU assay and
higher concentrations comparable to the earlier fluctuation assay. ROS in WT control increased
about 3-fold while ROS in Flux strains increased about 2 to 2.5-fold when exposed to 48 µg/mL
RIF for 24 hours and the ROS in WT reached almost a 7-fold increase while the Flux strains
remained at an average increase around 5-fold when exposed to 96 µg/mL RIF and 128 µg/mL
RIF (Figure 7C). There did not seem to be a clear or consistent pattern in ROS levels, glycerol
metabolism, or RRDR mutations within the Flux strains.
12
The CFU enumeration assay of these strains without RIF treatment showed similar colony
number increase in 7H9 complete media and noticeably large differences with treatment of 48
µg/mL RIF plated after 6 hours and 24 hours of incubation (Figure 7B). Flux strains can maintain
cell viability in RIF stress conditions by maintaining a lower level of ROS at bactericidal levels.
The fold change ATP level of Flux strains was significantly lower than WT in response to
rifampicin treatment. The mean fold change of Flux strains was 0.5-fold while WT increased
almost 2-fold after treatment of 16 µg/mL RIF for 4 hours in 7H9 complete media when compared
to untreated condition (Figure 7D, t test P value < 0.0001). ATP level can be an indicator of cell
viability. Interestingly, ATP pools of Flux cells maintained lower as compared to that of WT in
response to treatment with sub-MIC concentrations of RIF. This could possibly be due to an
expedited adaptive response to RIF stress through an anticipatory mechanism obtained from
previous exposure to RIF, while the naïve WT adaptive response is slower. Furthermore, previous
studies have shown that persister formation is enhanced by depletion of ATP levels
62,63
. This
provides evidence that drug tolerant persister cells and the Flux biochemically similar, suggesting
a possible metabolic mechanism for transition from transient drug tolerance to irreversible drug
tolerance.

Flux glycerol metabolomics
A possible explanation for Flux growth defect was buildup of toxic metabolites in glycerol
metabolism, specifically, the methylglyoxal detoxification pathway which has been previously
implicated
60
. To test this, we measured the expression of glpK mRNA in Flux strains and WT after
culturing in 1.5% glycerol in Sauton minimal media. GlpK is the initial enzyme catalyzing glycerol
as a substrate of CCM intermediate and precursor of methylglyoxal in the methylglyoxal
detoxification pathway of glycerol metabolism that uses ATP to convert glycerol into glycerol 3-
phosphate. The glycerol growth defect and depleted levels of ATP observed in the Flux strains
13
could be explained by increased glpK expression. Interestingly, there were no significant
differences in glpK expression (data not shown). This was consistent in transcript levels extracted
from both the liquid culture and filter culture methods. Although glpK expression was similar, there
were increased levels of methylglyoxal metabolite abundance in Flux strains compared to WT
from filter culture metabolomics (Figure 8E, t test P value = 0.0457, 0.0041, 0.0024, and 0.3197
for F1, F2, F7, and F8, respectively).
To investigate metabolic differences between the Flux strains and WT Msm, we grew filter
cultures on Sauton agar with 1.5% glycerol as the sole carbon source. To identify metabolic
rescue of growth, we also grew filter cultures in parallel on Sauton agar plates containing 0.2%
dextrose only and plates containing both 1.5% glycerol and 0.2% dextrose. Testing the targeted
metabolites uniquely changed in Flux cells under glycerol growth could provide insight on the
mechanistic basis of their irreversible drug tolerance. Metabolites from Msm were extracted
64
,
identified and analyzed using liquid chromatography/mass spectrometry (LC-MS)
65
, and then
statistical analysis was performed using bioinformatic tools available in MetaboAnalyst
66
.
Global metabolic changes showed clear clustering between F1 and F2 strains, F7 and F8
strains, and WT when grown on filter cultures in 1.5% glycerol in Sauton agar for 2 days (Figure
8A). In similar filter culture conditions but 1.5% glycerol in Sauton agar supplemented with
additional 0.2% dextrose, the clear clustering between these groups is diminished (Figure 8B).
These clustering patterns were also observed in the PCA coordinate analysis of these
samples with glycerol only condition (Figure 8D, left) versus glycerol and dextrose conditions
(Figure 8D right). These data support underlying metabolic differences for the glycerol growth
defect in Flux strains when compared to WT.
Overall, ROS levels were induced in Flux strains when cultured in Sauton liquid media
with 1.5% glycerol as the only carbon source (Figure 8B). Increased ROS levels and increased
accumulation of the toxic metabolite, methylglyoxal, under glycerol conditions could explain the
delayed growth of Flux strains. Higher ROS levels would be counteractive towards the Flux drug
14
tolerance, however, can be explained by the differences in ROS associated antioxidant
metabolites: Alanine, Proline, and Glutamate
67–69
. In the glycerol only condition, significant
depletion of Alanine and Proline is seen in Flux cells which would prevent the antioxidant effect
of these metabolites to scavenge and decrease ROS levels (Figure 8E). These metabolites are
generally restored in the glycerol supplemented with 0.2% dextrose. Alanine is partially or fully
restored dependent on the Flux strain, Proline is largely restored, and Glutamate is restored to
similar abundance levels to that of WT (Figure 8E).
These data provide some insight on the metabolic mechanism of glycerol metabolism in
Flux cells and irreversible drug tolerance. Future studies are needed to further explore the
biochemical consequences of Flux cells under glycerol only and dextrose rescue conditions to
define how this metabolism defect can manifest phenotypically.

Confirmation of desaturase A CRISPRi strains
The same CRISPRi knock down system employed for itreS and iotsA mutants was used
for generation of experimental Msm Desaturase A mutants. After identical transformation protocol
and following treatment with 200 ng/mL anhydrotetracycline, total mRNA was extracted and
quantified. The relative mRNA levels for each Desaturate A1, A2, and A3, showed statistically
significant lower levels when compared to their respective expression levels in WT (Figure 9A, t
test P value = 0.0001, < 0.0001, and < 0.0001 for A1, A2, and A3, respectively). The relatively
low level of knock down in idesA1 mutant suggests that DesA1 is an essential enzyme that
requires a higher level of expression for survival of Msm, consistent with previous literature
43
.
Despite this, clear phenotypes were still observed in all following experiments. Growth
curve experiments showed that all three idesA mutants had similar growth rates to WT in 6%
glycerol, 0.1% propionate, and 0.2% acetate, however a severe growth defect when 5 µM
palmitate was the single carbon source in Sauton minimal media (Figure 9B). The fact that these
15
idesA mutants can utilize short-chain carbons like propionate and acetate but not longer-chain
carbons like palmitate suggest that they are involved in the desaturase pathway downstream of
palmitate and upstream of propionate and acetate.

Suppression of DesA causes defect in biofilm formation
Matching the results of TreS knock down, the suppression of Desaturase A enzymes also
caused a defect in biofilm formation. Biofilm cultures were incubated at 37 °C without agitation
and the A1, A2, and A3 strains were treated with ATc. Metabolites were extracted from parallel
cultures and pictures taken of CV-stained biofilms at 5, 6, and 7 days of incubation. Although
pictures of CV-stained biofilms are not as visually clear (Figure 10A), quantification by
measurement of CV stain absorbance (data not shown) shows similar trend to protein
concentration of cell biomass for strains at each time point.
The protein concentrations were determined by BCA protein assay and used for
metabolite abundance normalization. The protein concentration is also a measure of PLB density
in biofilms like CV staining. All desaturase A mutants showed significant defect in biofilm formation
at days 6 and 7 by lower protein concentrations compared to WT, with idesA1 having the most
severe defect (Figure 10B, t test P value = 0.0005, 0.0008, and 0.0039 [Day 6]; < 0.0001, 0.0019,
and 0.0008 [Day 7] for A1, A2, and A3, respectively).
Growth rates of desaturase A mutants was similar to that of WT in Sauton minimal media
liquid culture (Figure 10C). Biofilm formation defect is the likely result of desaturase A suppression
as these mutant strains had identical growth rates to WT in Sauton minimal media. Since
desaturase A enzymes cause deficiency of PLB formation, they may play a role in mycolic acid
desaturation from TDM or TMM for energy production when entering CCM as well as ROS
mediation through production of NADPH scavenger.

16
Increased idesA sensitivity to antibiotics and ROS stress
Since similar PLB phenotypes to itreS were seen in desaturase A mutants, we suspect
they also had increased sensitivity to antibiotics and ROS damage. To characterize the effect of
antibiotic treatment on idesA mutants and their ability to respond to ROS, we performed CFU
enumeration assays, DHE staining FACS analysis for ROS level, and a fluctuation assay like the
itreS experiments.
For the CFU enumeration assay, WT, A1, A2, and A3 strains were treated with 0.007
µg/mL BDQ or 0.035 µg/mL BDQ, 1X and 5X MIC for Msm, respectively, and no antibiotic
treatment control in 7H9 complete media were allowed to grow for 24 hours. At 0 hour before
treatment, 6 hours post exposure, and 24 hours post exposure the cultures were diluted and
inoculated to determine cell numbers. The fold change CFU of idesA mutants after the 6-hour
treatment of 1X BDQ showed almost 0.5-fold difference relative to WT and was even more
significantly decreased after 24 hours of 5X BDQ (Figure 11A, t test P value = 0.0153, 0.0132,
and 0.0325 [1X BDQ]; 0.0009, 0.0085, and 0.0021 [5X BDQ] for A1, A2, and A3, respectively).
All cultures showed similar growth in untreated 7H9 complete media as confirmed by CFU counts.
Similarly, a CFU enumeration assay was performed with treatment of 500 µM H 2O 2 in 7H9
complete media for 24 hours. The mean WT fold change of about 0.5-fold was much greater than
all idesA mutants. The idesA2 strain average of 0.15-fold CFU change was 3.4 times less than
WT, the idesA3 strain average of about 0.1-fold CFU change was a little more than 5 times less
than WT, and the idesA1 strain average of 0.026-fold CFU change was 20 times less than WT
(Figure 11B, left, t test P value = 0.0068, 0.0314, and 0.0161 for A1, A2, and A3, respectively).
Although the error of WT CFU fold change was large, the CFU fold change was still statistically
significantly decreased in idesA strains compared to WT. Interestingly, the ROS level after
treatment of 10 mM H2O2 in 7H9 complete media for 1 hour showed no difference between WT
and idesA1, but almost twice as much fold induction for the idesA2 and idesS3 strains when
17
compared to WT (Figure 11B, right, t test P value = 0.9050, < 0.0001, and 0.0004 for A1, A2, and
A3, respectively). Based upon the results, we hypothesize that DesA enzymes are involved in
dealing with antibiotic induced ROS to increase the survival rate as seen in TreS mediated
metabolism remodeling.
To validate the functional relevance of DesA mediated adaptive strategy in developing
drug resistant mutations, we decided to seek the drug resistant mutation rates of idesA and WT.
The rate of acquisition of RIF
R
as determined by a fluctuation assay in 100 µg/mL of RIF was
significantly decreased in idesA mutants and consistent with significant ROS level increase in
idesA mutants after treatment of 48 µg/mL of RIF in 7H9 complete media for 4 hours. The idesA
mutants had at least 2 times increased in fold induction of ROS level compared to WT after
treatment of RIF with the most increase in idesA2 of a 7-fold induction of ROS after treatment
(Figure 11C, right, t test P value < 0.0001 for A1, A2, and A3).
The idesA1 strain has similar ROS level induction as WT after H 2O 2 treatment but
significantly less cell viability than WT from CFU enumeration. This could be due to the delayed
expression of desA mRNA in response to stress under transcriptional control of the mycolic acid
desaturase regulator (MadR)
50
. Since desA2 is also under control of the MadR transcription factor,
it is possible that the essentiality of desA1 causes the more significant decrease in cell viability.
All idesA strains are much more susceptible to RIF treatment and is consistent with ROS level
increase after treatment which could suggest major mode of action.

Metabolomics of increased idesA susceptibility to starvation
Desaturase A enzymes show an important role in drug tolerance and surviving ROS
stress. We wanted to explore the metabolic implications of the defect in PLB formation in biofilm
cultures and in starvation conditions using PBS
10
. A CFU enumeration assay of WT, idesA1,
idesA2, and idesA3 strains under PBS starvation conditions showed that suppression of each
18
desaturase A enzyme significantly reduced viability while WT was able to maintain its viable cell
number over 3 days (Figure 12A).
Metabolomic analysis of metabolite abundances at each time point showed significant or
full depletion of several TCA cycle intermediates such as malate, fumarate, and αketoglutarate in
the idesA mutants while WT can sustain sufficient levels of these metabolites at 1 and 2 days of
starvation (data not shown).
In trehalose metabolism to CCM, trehalose is converted to maltose which can be formed
into maltose 6-phosphate or glucose 6-phosphate, which then enter the PPP or glycolysis,
respectively (Figure 12C). One of the final metabolites of the PPP is sedoheptulose 7-phosphate.
There is significant depletion of maltose, glucose 6-phosphate, and sedoheptulose 7-phosphate
in idesA strains, which indicates that there is little or no carbon flux into glycolysis or PPP when
compared to WT. These two pathways are vital for survival in stress conditions like starvation
because glycolysis and downstream TCA cycle are major metabolic sources of ATP and PPP is
an important source of NADPH for scavenging ROS. Accumulation of ROS and decreased ability
to produce ATP to survive are the likely causes of the significant CFU decrease seen in idesA
strains while WT can maintain cell number viability due to functional glycolysis and PPP pathways
for survival.
Similar patterns are seen in the Maltose 6-phosphate, Glucose 6-phosphate, and
Sedoheptulose 7-phosphate metabolites extracted from biofilm cultures at 6 and 7 days of
incubation (Figure 12D). Finding similar results under stress conditions point to a dysfunctional
PPP pathway in the idesA mutants.
19
More studies need to be done to understand how suppression of the Desaturase A
enzymes can affect the PPP pathway and consequent deficiency in controlling ROS levels under
stress conditions. It is promising that the idesA strains present similar phenotypes to itreS. If more
biochemical studies can verify the essentiality of fatty acid Desaturase A enzymes, they may
serve as metabolic targets for development of novel TB therapeutics.
20
Materials and Methods

Bacterial strains
Wildtype Mycobacterium smegmatis (Msm) mc
2
155 laboratory strain (WT), Msm
Trehalose Synthase knock-down strain (itreS), Msm Trehalose-6-Phosphate Synthase knock-
down strain (iotsA), Msm Desaturase A1 knock-down strain (idesA1), Msm Desaturase A2 knock-
down strain (idesA2), and Msm Desaturase A3 knock-down strain (idesA3) were used in this
study; Naïve Msm strains refer to WT bacteria that have never been exposed to antibiotics and
Flux Msm strains were surviving WT colonies isolated after exposure to concentrations of
rifampicin (RIF) exceeding MIC.

Culture conditions
All bacterial strains used in experiments were collected at mid-logarithmic phase growth,
cultured at 37 °C in Middlebrook 7H9 liquid medium (Difco) supplemented with 0.5% bovine serum
albumin, 0.085% NaCl, 0.2% glycerol, 0.2% dextrose, and 0.04% tyloxapol. Bacteria were plated
on Middlebrook 7H10 with agar medium (Difco) supplemented with 0.5% bovine serum albumin,
0.085% NaCl, 0.2% glycerol, and 0.2% dextrose. Luria-Bertani with agar medium (Sigma-Aldrich)
was used only for plating bacterial culture dilutions in CFU experiments. Sauton minimal medium
was used for biofilm assays and in experiments where indicated.

mRNA quantification
Msm cells were harvested at a mid-logarithmic phase of growth by centrifugation
(Centrifuge 5424 R, Eppendorf). Cells were resuspended in TRI-reagent (Sigma-Aldrich) and
mechanically lysed in tissue homogenizer (Precellys Evolution, Bertin Technologies) with 0.1 mm
zirconia beads once for 10 minutes at 6000 rpm. Total RNA was isolated from lysates using RNA
miniprep kit (Zymo Research) and then converted into cDNA with iScript cDNA synthesis kit
21
(BioRad) in a thermal cycler (Vapo.Protect Mastercycler Pro, Eppendorf). Levels of cDNA were
quantified by quantitative real-time polymerase chain reaction (qRT–PCR) in a light cycler
(LightCycler 96, Roche) using iQ SYBR Green Supermix (BioRad). Signals were normalized to
sigA housekeeping gene transcript levels and calculated by ΔΔCT values. Error bars are the
standard deviation from technical triplicates.

Plasmid construction
Guide RNA (sgRNA) was designed to target the 3’-end of the non-template strand open
reading frame for target genes, comprising of a ~20 nucleotide sequence that is 5’ from an
effective protospacer adjacent motif sequence within this region. The PLJR962 Msm CRISPRi
backbone plasmid was amplified in E. coli, selected with kanamycin, digested with BsmBI
restriction enzymes (Thermo Fischer Scientific), and then gel purified. The designed oligo primers
were annealed and ligated into the BsmBI-digested vector backbone.

Cell transformation
Competent Msm cells were prepared by multiple washes of mid-log phase cultures in 10%
glycerol solution. The recombinant plasmid of interest was added to competent cells and
introduced through electroporation (Pulse Controller II; Capacitance Extender Plus; and Gene
Pulser II, BioRad). Transformed liquid cultures were grown to mid-log phase growth and plated
on 7H10 media containing kanamycin (50 µg/mL) to select for recombinant colonies. Colonies
were regrown in 7H9 broth and incubated with anhydrotetracycline (200 ng/mL) for at least one
doubling time (4 hours) to induce targeted gene repression. Endogenous knock-down of targeted
gene of interest was confirmed by mRNA quantification.

Metabolomics
22
Msm cells were harvested by centrifugation at specific time points as indicated. Cells were
resuspended in pre-cooled acetonitrile:methanol:H 2O solution (40:40:20, v:v:v) and mechanically
lysed in tissue homogenizer (Precellys Evolution, Bertin Technologies) with 0.1 mm zirconia
beads twice for 10 minutes at 6000 rpm. Metabolites were collected from lysates by filtration
through a 0.22 µm spin-X column. Samples were collected and combined with equal volume of
4% formic acid in acetonitrile, centrifuged, and then supernatant was collected in metabolomics
tube. Compounds in samples were identified using the 1290 Infinity II liquid chromatography
system (Agilent) coupled with 6230 time-of-flight mass spectrometer (Agilent). Metabolite
abundances were normalized to cell biomass by sample protein concentration using BCA protein
assay kit (Thermo Fischer Scientific). Error bars are the standard deviation from biological
triplicates.

CV staining
Crystal Violet (CV) staining was performed on biofilm cultures after washing with 10% PBS
solution. Crystal Violet stain was added to wells and incubated for 10 minutes at room
temperature. Stain was then removed, wells allowed to dry for an additional 10 minutes, and
pictures of wells were taken. To quantify CV intensity, stained biofilms were resuspended in
99.9% ethanol, diluted 1:100 in 10% PBS solution, and then absorbance was measured at 600
nm wavelength. Error bars are the standard deviation from biological or technical replicates
dependent on number of culture wells available.

Biofilm assay
All cultures were OD-matched and CRISPRi strains were treated with anhydrotetracycline
at least 4 hours before harvesting. Msm cells were harvested at a mid-logarithmic phase of growth
by centrifugation and then resuspended in Sauton minimal media to OD 0.01. Cultures were
distributed in 24-well plates and wrapped in parafilm to prevent gas exchange. Cultures were
23
incubated at 37 °C without agitation until formation of biofilms occurred. Sample collection for
metabolomics and CV staining were performed as previously described.

Spot assay
All cultures were OD-matched and CRISPRi strains were treated with anhydrotetracycline
at least 4 hours before plating. Bacterial cultures were serial diluted and 1-2 µL of each culture
dilution was plated on 7H10 solid medium with antibiotics added as indicated. Negative control
plates to observe the untreated colony phenotypes contained no antibiotics. Plate images were
captured with ChemiDoc MP Imaging System (BioRad).

FACS analysis
All cultures were OD-matched and CRISPRi strains were treated with anhydrotetracycline
at least 4 hours before experimental conditions were applied. Untreated control cultures were
matched to low OD (0.05 to 0.3) and treated experimental cultures were matched to higher OD
(0.5 to 1) depending on time of exposure and treatment concentration. Upon completion of
experimental exposure, cultures were stained with 10 µM dihydroethidium (Abcam) and incubated
at 37 °C for 30 minutes. Cultures were then sorted in a flow cytometer (Attune NxT, ThermoFisher
Scientific) and reported values represent the R4 %-gated cell fraction. Error bars are the standard
deviation from biological replicates.

Fluctuation analysis
Msm cultures were diluted to OD 0.00005 and distributed into parallel cultures on a 96-
well plate. Cultures were incubated at 37 °C for about 4-5 days until saturation, allowing for
spontaneous mutations to occur. Ten to twenty colonies were expanded and then plated on 7H10
media containing rifampicin (100 µg/mL); the number of mutants was counted. Dilutions of
expanded colonies were also plated on 7H10 without antibiotics to determine the culture cell
24
number. Mutation rates were calculated with respective mutant numbers and cell numbers using
the mutation rate calculator website: bz-rates - mutation rates calculator for fluctuation assays
(upmc.fr). Values reported are the mutation rate per cell per division corrected by the plating
efficiency with 95% confidence limit.

25
Tables

26

27
Figures

Figure 1: Schematic diagram of TMM/TDM in Mycobacterial membrane and Trehalose metabolism.
(A) The location of TMM and TDM is in outermost layer of the Mycomembrane. The molecular structure of TDM includes
a trehalose disaccharide connecting two mycolic acids. Adapted from Marrakchi et al
29
. (B) TMM/TDM and UDP-
Glucose are major sources of Trehalose, the latter is converted into Trehalose by OtsA/B. TreS catalyzes the
conversion of Trehalose into Maltose which then can enter Glycolysis or PPP. AG, Arabinogalactan; PG, Peptidoglycan;
TMM, Trehalose Monomycolate; TDM, Trehalose Dimycolate; UPD-Glucose, Uridine Diphosphate Glucose; TreS,
Trehalose Synthase; OtsA/B Trehalose-6-Phosphate Synthase; PPP, Pentose Phosphate Pathway.

28

Figure 2: CRISPRi system knock down of TreS and OtsA enzymes.
(A) Schematic representation of the CRISPRi system knock down mechanism. Plasmids containing dcas9 and target
gene sgRNA are induced by tetracycline. The dCas9 is guided to the target DNA sequence by sgRNA to sterically
hinder RNAP from accessing target gene resulting in transcriptional repression. Adapted from Rock et al. (B) Relative
mRNA level of Trehalose Synthase (TreS) and Trehalose-6-Phosphate Synthase (OtsA) of CRISPRi Msm strain itreS
and iotsA, respectively, for knock down induced with anhydrotetracycline normalized to WT. The average percent
transcript level relative to WT of the CRISPRi Msm strain is displayed above bars. PTet, tetracycline induced promoter;
dcas9, catalytically dead cas9 gene; dCas9, catalytically dead cas9 protein; sgRNA, single guide RNA; RNAP, RNA
Polymerase; PAM, protospacer adjacent motif; NT, non-template strand; T, template strand; WT, Msm wildtype strain;
itreS, Msm Trehalose Synthase knock-down strain; iotsA, Msm Trehalose-6-Phosphate Synthase knock-down strain.
Significance levels: ns, not significant; *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001; calculated by two-sided
Student’s t-test.

29

Figure 3: Suppression of TreS enzyme causes defect of biofilm formation of PLB.
(A) Pictures taken of CV-stained biofilm cultures after 6 days of incubation. (B) Quantification of CV stain intensity from
6-day biofilm cultures. (C) Growth curve of WT, itreS, and iotsA in Sauton minimal media liquid culture. The itreS and
iotsA strains were treated with Atc. WT, Msm wildtype strain; itreS, Msm Trehalose Synthase knock-down strain; iotsA,
Msm Trehalose-6-Phosphate Synthase knock-down strain; ATc, anhydrotetracycline. Significance levels: ns, not
significant; *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001; calculated by two-sided Student’s t-test.

30

Figure 4: Increased itreS sensitivity to antibiotics and ROS.
(A) Spot assay plates of WT, itreS, and iotsA (top to bottom) with decreasing OD (left to right) in no treatment control
7H10 solid medium and 7H10 containing antibiotics. (B) ROS level as determined by DHE staining FACS analysis after
treatment of 10 mM H2O2 in 7H9 complete media for 1 hour, fold change relative to untreated condition. (C) CFU
enumeration assay of culture dilutions plated after treatment of 100 µM or 500 µM H2O2 in 7H9 complete media for 24
hours, fold change relative to 0-hour cell number and normalized to WT. (D) Rate of acquisition of RIF
R
as determined
by fluctuation assay in 100 µg/mL of RIF (N=20). The itreS and iotsA strains were treated with ATc prior to and during
experimental conditions, all spot assay plates and fluctuation assay plates for itreS contained ATc. WT, Msm wildtype
strain; itreS, Msm Trehalose Synthase knock-down strain; iotsA, Msm Trehalose-6-Phosphate Synthase knock-down
strain; ATc, anhydrotetracycline; OD, optical density; INH, isoniazid; BDQ, bedaquiline, RIF, rifampicin; CFU, colony
forming units. Significance levels: ns, not significant; *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001; calculated by
two-sided Student’s t-test.

31

Figure 5: Flux strains are more tolerant to RIF than naïve WT.
Spot assay plates of WT and Flux strains 1-10 on 7H10 solid medium with varying concentrations of RIF. All cultures
were matched to OD 0.5 and 2 µL were inoculated. Concentrations of RIF used are indicated above plate pictures and
the orientation of colonies on the plate is shown in the diagram on the left. The numbers in diagram refer to the Flux
strain number; WT, Msm wildtype strain; RIF, rifampicin.

32

Figure 6: Flux strains glycerol growth defect is rescued by dextrose supplementation.
(A) Growth curve of WT and Flux strains (1-10) grouped in Sauton minimal media with 1.5% glycerol as single carbon
source. (B) Growth curve of WT, Flux strains 1+2, and Flux strains 7+8 grouped in Sauton minimal media with 1.5%
glycerol as single carbon source. (C) Growth curve of WT and Flux strains (1-10) grouped in Sauton minimal media
with 1.5% glycerol as single carbon source (left), 1.5% glycerol and 0.2% dextrose (middle), and 0.2% dextrose as
single carbon source (right). WT, Msm wildtype strain; Flux, Msm Fluctuation strains 1-10 grouped. Significance levels:
ns, not significant; *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001; calculated by two-sided Student’s t-test.

33

Figure 7: Flux strains maintain lower ROS levels under RIF treatment.
(A) ROS levels of WT, F1, and F5 strains as determined by DHE staining FACS analysis after treatment of varying
concentrations of RIF (X-axis) for 4 hours in 7H9 complete media. Fold change was calculated from untreated level for
each strain and normalized to WT. (B) Cell number counts from CFU enumeration assay of WT, F1, F2, F7, and F8
strains without antibiotic treatment and treatment with 48 µg/mL RIF plated after 6 hours and 24 hours of incubation.
(C) ROS levels of WT, F1, F2, F7, and F8 strains after treatment of 48 µg/mL, 96 µg/mL, and 128 µg/mL RIF for 4
hours. Fold change values were calculated in relation to each respective strain without RIF treatment condition. (D)
ATP level of WT and Flux strains after treatment of 16 µg/mL RIF for 4 hours in 7H9 complete media. Fold change
values were calculated relative to untreated group conditions and normalized to culture OD. WT, Msm wildtype strain;
Flux, Msm Fluctuation strains 1-10 grouped; ROS, reactive oxygen species; RIF, rifampicin; ATP, adenosine
triphosphate. Significance levels: ns, not significant; *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001; calculated by
two-sided Student’s t-test.

34

Figure 8: Flux strain metabolomics for dextrose rescue of glycerol defect.
(A) Global metabolite profile of 173 metabolites extracted from WT, F1, F2, F7, and F8 strains grown on filter cultures
in 1.5% glycerol on Sauton agar media for 2 days. (B) Global metabolite profile of 162 metabolites extracted from WT,
F1, F2, F7, and F8 strains grown on filter cultures on 1.5% glycerol and 0.2% dextrose in Sauton agar media for 2 days.
(C) ROS levels of WT, F1, F2, F7, and F8 strains after liquid culture growth in 1.5% glycerol in Sauton media for 1 day.
(D) PCA plot of WT (red), F1 (green), F2 (blue), F7(light blue), and F8 (purple) strains in 1.5% glycerol (left) or 1.5%
glycerol and 0.2% dextrose (right) on Sauton agar media. (E) Relative abundances of Methylglyoxal, Alanine, Proline,
and Glutamate extracted from WT, F1, F2, F7, and F8 strains grown on filter cultures in 1.5% glycerol on Sauton agar
media for 2 days. WT, Msm wildtype strain; F1,2,7,8, Msm Fluctuation strains 1,2,7,8, respectively; ROS, reactive
oxygen species; G, 1.5% glycerol only condition; G+D, 1.5% glycerol and 0.2% dextrose condition. Significance levels:
ns, not significant; *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001; calculated by two-sided Student’s t-test.
35

Figure 9: Suppression of Desaturase A enzymes cause defect of growth in palmitate.
(A) Relative mRNA levels of DesA1, DesA2, and DesA3 of CRISPRi Msm strains idesA1, idesA2, and idesA3,
respectively, for knock down induced with anhydrotetracycline relative to WT. (B) Growth curve of WT, A1, A2, and A3
strains in Sauton minimal media with 5 µM palmitate, 6% glycerol, 0.1% propionate, or 0.2% acetate as single carbon
source. The A1, A2, and A3 strains were treated with ATc. WT, Msm wildtype strain; A1, Msm Desaturase A1 knock-
down strain (idesA1); A2, Msm Desaturase A2 knock-down strain (idesA2); A2, Msm Desaturase A3 knock-down strain
(idesA3); ATc, anhydrotetracycline. Significance levels: ns, not significant; *P < 0.05; **P < 0.01; ***P < 0.001; ****P <
0.0001; calculated by two-sided Student’s t-test.

36

Figure 10: Suppression of Desaturase A enzymes cause defect of biofilm formation of PLB.
(A) Pictures taken of CV-stained biofilm cultures after 6 and 7 days of incubation. (B) Quantification of biofilm cell mass
by BCA assay protein concentration from 6- and 7-day biofilm cultures. (C) Growth curve of WT, A1, A2, and A3 in
Sauton minimal media liquid culture. The A1, A2, and A3 strains were treated with ATc. WT, Msm wildtype strain; A1,
Msm Desaturase A1 knock-down strain (idesA1); A2, Msm Desaturase A2 knock-down strain (idesA2); A2, Msm
Desaturase A3 knock-down strain (idesA3); ATc, anhydrotetracycline. Significance levels: ns, not significant; *P < 0.05;
**P < 0.01; ***P < 0.001; ****P < 0.0001; calculated by two-sided Student’s t-test.

37

Figure 11: Increased idesA sensitivity to antibiotics and ROS.
(A) CFU enumeration assay of WT, A1, A2, and A3 strains with or without under treatment of BDQ. The fold change
CFU of 6-hour treatment with 0.007 µg/mL BDQ and 24-hour treatment with 0.035 µg/mL BDQ are presented in bar
graphs, and CFU counts of no treatment condition over time is shown in the line graph. The CFU fold change was
calculated from 0 hour starting cell number for each respective strain and normalized to WT. (B) CFU enumeration
assay WT, A1, A2, and A3 strains after treatment of 500 µM H2O2 in 7H9 complete media for 24 hours, fold change
relative to 0-hour cell number. ROS level as determined by DHE staining FACS analysis after treatment of 10 mM H2O2
in 7H9 complete media for 1 hour, fold change relative to untreated condition. (C) Rate of acquisition of RIF
R
as
determined by fluctuation assay in 100 µg/mL of RIF (N=10). ROS level as determined by DHE staining FACS analysis
after treatment of 48 µg/mL of RIF in 7H9 complete media for 4 hours, fold change relative to untreated condition. The
A1, A2, and A3 strains were treated with ATc, and fluctuation assay plates contained ATc. WT, Msm wildtype strain;
A1, Msm Desaturase A1 knock-down strain (idesA1); A2, Msm Desaturase A2 knock-down strain (idesA2); A2, Msm
Desaturase A3 knock-down strain (idesA3); ATc, anhydrotetracycline; CFU, colony forming units; BDQ, bedaquiline;
RIF, rifampicin. Significance levels: ns, not significant; *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001; calculated
by two-sided Student’s t-test.

38

Figure 12: Metabolomics of Desaturase A enzymes requirement for starvation viability and biofilm.
(A) CFU enumeration assay of WT, A1, A2, and A3 strains under PBS starvation conditions. The CFU fold change was
calculated relative to 0-day count for each respective strain. (B) Relative abundance of Maltose, Glucose 6-phosphate,
and Sedoheptulose 7-phosphate extracted from WT, A1, A2, and A3 strains grown in PBS starvation conditions at 1
and 2 days. (C) Schematic representation of specific metabolites involved in trehalose metabolism and PPP (D)
Relative abundance of Maltose 6-phosphate, Glucose 6-phosphate, and Sedoheptulose 7-phosphate extracted from
WT, A1, A2, and A3 strains grown in PLB biofilm assay collected at 6 and 7 days. The A1, A2, and A3 strains were
treated with ATc. WT, Msm wildtype strain; A1, Msm Desaturase A1 knock-down strain (idesA1); A2, Msm Desaturase
A2 knock-down strain (idesA2); A2, Msm Desaturase A3 knock-down strain (idesA3); ATc, anhydrotetracycline; CFU,
colony forming units; PPP, Pentose Phosphate Pathway.

39
References

1. World Health Organization. Global tuberculosis report 2020. (World Health Organization,
2020).
2. Chan, E. D. Current medical treatment for tuberculosis. BMJ 325, 1282–1286 (2002).
3. Chee, C. B. E., Reves, R., Zhang, Y. & Belknap, R. Latent tuberculosis infection:
Opportunities and challenges: Latent tuberculosis infection review. Respirology 23, 893–900
(2018).
4. Zhang, Y., Yew, W. W. & Barer, M. R. Targeting Persisters for Tuberculosis Control.
Antimicrob. Agents Chemother. 56, 2223–2230 (2012).
5. Podinovskaia, M., Lee, W., Caldwell, S. & Russell, D. G. Infection of macrophages with
Mycobacterium tuberculosis induces global modifications to phagosomal function: Mtb
infection modifies phagosome function. Cell. Microbiol. 15, 843–859 (2013).
6. Ehrt, S. et al. Reprogramming of the Macrophage Transcriptome in Response to Interferon-γ
and Mycobacterium tuberculosis. J. Exp. Med. 194, 1123–1140 (2001).
7. Rahman, Md. A. et al. Hydrogen sulfide dysregulates the immune response by suppressing
central carbon metabolism to promote tuberculosis. Proc. Natl. Acad. Sci. 117, 6663–6674
(2020).
8. Makowski, L., Chaib, M. & Rathmell, J. C. Immunometabolism: From basic mechanisms to
translation. Immunol. Rev. 295, 5–14 (2020).
9. Phelan, J. J. et al. Modulating Iron for Metabolic Support of TB Host Defense. Front. Immunol.
9, 2296 (2018).
10. Rizvi, A. et al. Rewiring of Metabolic Network in Mycobacterium tuberculosis During
Adaptation to Different Stresses. Front. Microbiol. 10, 2417 (2019).
11. Remot, A., Doz, E. & Winter, N. Neutrophils and Close Relatives in the Hypoxic Environment
of the Tuberculous Granuloma: New Avenues for Host-Directed Therapies? Front. Immunol.
10, 417 (2019).
12. Betts, J. C., Lukey, P. T., Robb, L. C., McAdam, R. A. & Duncan, K. Evaluation of a nutrient
starvation model of Mycobacterium tuberculosis persistence by gene and protein expression
profiling: Nutrient starvation of M. tuberculosis. Mol. Microbiol. 43, 717–731 (2002).
13. Guirado, E. & Schlesinger, L. S. Modeling the Mycobacterium tuberculosis Granuloma – the
Critical Battlefield in Host Immunity and Disease. Front. Immunol. 4, (2013).
14. Gomez, J. E. & McKinney, J. D. M. tuberculosis persistence, latency, and drug tolerance.
Tuberculosis 84, 29–44 (2004).
15. Vandiviere, H. M., Loring, W. E., Melvin, I. & Willis, S. THE TREATED PULMONARY LESION
AND ITS TUBERCLE BACILLUS.*: II. THE DEATH AND RESURRECTION. Am. J. Med. Sci.
232, 30–37 (1956).
16. Lewis, K. Persister Cells. Annu. Rev. Microbiol. 64, 357–372 (2010).
17. Shah, D. et al. Persisters: a distinct physiological state of E. coli. BMC Microbiol. 6, 53 (2006).
40
18. Orman, M. A. & Brynildsen, M. P. Dormancy is not necessary or sufficient for bacterial
persistence. Antimicrob. Agents Chemother. 57, 3230–3239 (2013).
19. Li, Y. & Zhang, Y. PhoU is a persistence switch involved in persister formation and tolerance
to multiple antibiotics and stresses in Escherichia coli. Antimicrob. Agents Chemother. 51,
2092–2099 (2007).
20. Zhang, L. et al. The catabolite repression control protein Crc plays a role in the development
of antimicrobial-tolerant subpopulations in Pseudomonas aeruginosa biofilms. Microbiol.
Read. Engl. 158, 3014–3019 (2012).
21. Balaban, N. Q., Merrin, J., Chait, R., Kowalik, L. & Leibler, S. Bacterial persistence as a
phenotypic switch. Science 305, 1622–1625 (2004).
22. Wakamoto, Y. et al. Dynamic persistence of antibiotic-stressed mycobacteria. Science 339,
91–95 (2013).
23. Lee, J. J. et al. Glutamate mediated metabolic neutralization mitigates propionate toxicity in
intracellular Mycobacterium tuberculosis. Sci. Rep. 8, 8506 (2018).
24. Lee, J. J. et al. Transient drug-tolerance and permanent drug-resistance rely on the trehalose-
catalytic shift in Mycobacterium tuberculosis. Nat. Commun. 10, 2928 (2019).
25. Eoh, H. et al. Metabolic anticipation in Mycobacterium tuberculosis. Nat. Microbiol. 2, 17084
(2017).
26. Becker, K. & Sander, P. Mycobacterium tuberculosis lipoproteins in virulence and immunity -
fighting with a double-edged sword. FEBS Lett. 590, 3800–3819 (2016).
27. Forrellad, M. A. et al. Virulence factors of the Mycobacterium tuberculosis complex. Virulence
4, 3–66 (2013).
28. Rodrigues, L., Viveiros, M. & Aínsa, J. A. Measuring Efflux and Permeability in Mycobacteria.
in Mycobacteria Protocols (eds. Parish, T. & Roberts, D. M.) vol. 1285 227–239 (Springer
New York, 2015).
29. Marrakchi, H., Lanéelle, M.-A. & Daffé, M. Mycolic Acids: Structures, Biosynthesis, and
Beyond. Chem. Biol. 21, 67–85 (2014).
30. Goins, C. M., Dajnowicz, S., Smith, M. D., Parks, J. M. & Ronning, D. R. Mycolyltransferase
from Mycobacterium tuberculosis in covalent complex with tetrahydrolipstatin provides
insights into antigen 85 catalysis. J. Biol. Chem. 293, 3651–3662 (2018).
31. Kalscheuer, R. & Koliwer-Brandl, H. Genetics of Mycobacterial Trehalose Metabolism.
Microbiol. Spectr. 2, (2014).
32. Nobre, A., Alarico, S., Maranha, A., Mendes, V. & Empadinhas, N. The molecular biology of
mycobacterial trehalose in the quest for advanced tuberculosis therapies. Microbiology 160,
1547–1570 (2014).
33. Kohanski, M. A., Dwyer, D. J., Hayete, B., Lawrence, C. A. & Collins, J. J. A Common
Mechanism of Cellular Death Induced by Bactericidal Antibiotics. Cell 130, 797–810 (2007).
34. Dwyer, D. J. et al. Antibiotics induce redox-related physiological alterations as part of their
lethality. Proc. Natl. Acad. Sci. 111, E2100–E2109 (2014).
41
35. Ojha, A. K., Trivelli, X., Guerardel, Y., Kremer, L. & Hatfull, G. F. Enzymatic Hydrolysis of
Trehalose Dimycolate Releases Free Mycolic Acids during Mycobacterial Growth in Biofilms.
J. Biol. Chem. 285, 17380–17389 (2010).
36. Ojha, A. K. et al. Growth of Mycobacterium tuberculosis biofilms containing free mycolic acids
and harbouring drug-tolerant bacteria. Mol. Microbiol. 69, 164–174 (2008).
37. Rushworth, G. F. & Megson, I. L. Existing and potential therapeutic uses for N-acetylcysteine:
the need for conversion to intracellular glutathione for antioxidant benefits. Pharmacol. Ther.
141, 150–159 (2014).
38. Anderson, M. E. & Meister, A. Transport and direct utilization of gamma-glutamylcyst(e)ine
for glutathione synthesis. Proc. Natl. Acad. Sci. U. S. A. 80, 707–711 (1983).
39. Takayama, K., Wang, C. & Besra, G. S. Pathway to Synthesis and Processing of Mycolic
Acids in Mycobacterium tuberculosis. Clin. Microbiol. Rev. 18, 81–101 (2005).
40. Dong, W. et al. Mycobacterial fatty acid catabolism is repressed by FdmR to sustain
lipogenesis and virulence. Proc. Natl. Acad. Sci. U. S. A. 118, (2021).
41. Yuan, Y., Crane, D. C., Musser, J. M., Sreevatsan, S. & Barry, C. E. MMAS-1, the Branch
Point Between cis- and trans-Cyclopropane-containing Oxygenated Mycolates in
Mycobacterium tuberculosis. J. Biol. Chem. 272, 10041–10049 (1997).
42. Cole, S. T. et al. Deciphering the biology of Mycobacterium tuberculosis from the complete
genome sequence. Nature 396, 190–190 (1998).
43. Singh, A. et al. Identification of a Desaturase Involved in Mycolic Acid Biosynthesis in
Mycobacterium smegmatis. PLOS ONE 11, e0164253 (2016).
44. Yeruva, V. C. et al. The Mycobacterium tuberculosis desaturase DesA1 (Rv0824c) is a Ca2+
binding protein. Biochem. Biophys. Res. Commun. 480, 29–35 (2016).
45. Dyer, D. H., Lyle, K. S., Rayment, I. & Fox, B. G. X-ray structure of putative acyl-ACP
desaturase DesA2 from Mycobacterium tuberculosis H37Rv. Protein Sci. 14, 1508–1517
(2009).
46. Di Capua, C. B., Doprado, M., Belardinelli, J. M. & Morbidoni, H. R. Complete auxotrophy for
unsaturated fatty acids requires deletion of two sets of genes in Mycobacterium smegmatis:
Unsaturated fatty acid synthesis in M. smegmatis. Mol. Microbiol. 106, 93–108 (2017).
47. Phetsuksiri, B. et al. Unique Mechanism of Action of the Thiourea Drug Isoxyl on
Mycobacterium tuberculosis. J. Biol. Chem. 278, 53123–53130 (2003).
48. Mohan, A., Padiadpu, J., Baloni, P. & Chandra, N. Complete Genome Sequences of a
Mycobacterium smegmatis Laboratory Strain (MC
2
155) and Isoniazid-Resistant (4XR1/R2)
Mutant Strains. Genome Announc. 3, e01520-14, /ga/3/1/e01520-14.atom (2015).
49. Harwood, J. L. et al. Lipids: biology and health. (John Wiley & Sons Inc, 2016).
50. Peterson, E. J. et al. Path ‐seq identifies an essential mycolate remodeling program for
mycobacterial host adaptation. Mol. Syst. Biol. 15, (2019).
51. Lelovic, N. et al. Application of Mycobacterium smegmatis as a surrogate to evaluate drug
leads against Mycobacterium tuberculosis. J. Antibiot. (Tokyo) 73, 780–789 (2020).
42
52. Rock, J. M. et al. Programmable transcriptional repression in mycobacteria using an
orthogonal CRISPR interference platform. Nat. Microbiol. 2, 16274 (2017).
53. Olaru, I. D., Heyckendorf, J., Andres, S., Kalsdorf, B. & Lange, C. Bedaquiline-based
treatment regimen for multidrug-resistant tuberculosis. Eur. Respir. J. 49, 1700742 (2017).
54. Huemer, M., Mairpady Shambat, S., Brugger, S. D. & Zinkernagel, A. S. Antibiotic resistance
and persistence-Implications for human health and treatment perspectives. EMBO Rep. 21,
e51034 (2020).
55. Fisher, R. A., Gollan, B. & Helaine, S. Persistent bacterial infections and persister cells. Nat.
Rev. Microbiol. 15, 453–464 (2017).
56. Bakkeren, E. et al. Salmonella persisters promote the spread of antibiotic resistance plasmids
in the gut. Nature 573, 276–280 (2019).
57. Lechner, D., Gibbons, S. & Bucar, F. Modulation of isoniazid susceptibility by flavonoids in
Mycobacterium. Phytochem. Lett. 1, 71–75 (2008).
58. Chakravorty, S. et al. Rifampin Resistance, Beijing-W Clade-Single Nucleotide Polymorphism
Cluster Group 2 Phylogeny, and the Rv2629 191-C Allele in Mycobacterium tuberculosis
Strains. J. Clin. Microbiol. 46, 2555–2560 (2008).
59. Zaw, M. T., Emran, N. A. & Lin, Z. Mutations inside rifampicin-resistance determining region
of rpoB gene associated with rifampicin-resistance in Mycobacterium tuberculosis. J. Infect.
Public Health 11, 605–610 (2018).
60. Pethe, K. et al. A chemical genetic screen in Mycobacterium tuberculosis identifies carbon-
source-dependent growth inhibitors devoid of in vivo efficacy. Nat. Commun. 1, 57 (2010).
61. Safi, H. et al. Phase variation in Mycobacterium tuberculosis glpK produces transiently
heritable drug tolerance. Proc. Natl. Acad. Sci. 116, 19665–19674 (2019).
62. Shan, Y. et al. ATP-Dependent Persister Formation in Escherichia coli. mBio 8, e02267-16,
/mbio/8/1/e02267-16.atom (2017).
63. Conlon, B. P. et al. Persister formation in Staphylococcus aureus is associated with ATP
depletion. Nat. Microbiol. 1, 16051 (2016).
64. Eoh, H. & Rhee, K. Y. Multifunctional essentiality of succinate metabolism in adaptation to
hypoxia in Mycobacterium tuberculosis. Proc. Natl. Acad. Sci. 110, 6554–6559 (2013).
65. Zhou, B., Xiao, J. F., Tuli, L. & Ressom, H. W. LC-MS-based metabolomics. Mol BioSyst 8,
470–481 (2012).
66. Pang, Z., Chong, J., Li, S. & Xia, J. MetaboAnalystR 3.0: Toward an Optimized Workflow for
Global Metabolomics. Metabolites 10, 186 (2020).
67. Hu, Z. et al. Synthesis of beta-alanine C60 derivative and its protective effect on hydrogen
peroxide-induced apoptosis in rat pheochromocytoma cells. Cell Biol. Int. 31, 798–804 (2007).
68. Meena, M. et al. Regulation of L-proline biosynthesis, signal transduction, transport,
accumulation and its vital role in plants during variable environmental conditions. Heliyon 5,
e02952 (2019).
69. Altman, B. J., Stine, Z. E. & Dang, C. V. Erratum: From Krebs to clinic: glutamine metabolism
to cancer therapy. Nat. Rev. Cancer 16, 773–773 (2016).

Asset Metadata

Creator Sell, Philip Jānis (author)

Core Title Adaptive metabolic strategies of Mycobacterium tuberculosis to combat stress from antibiotics and ROS

Contributor Electronically uploaded by the author (provenance)

School Keck School of Medicine

Degree Master of Science

Degree Program Molecular Microbiology and Immunology

Degree Conferral Date 2021-08

Publication Date 07/29/2021

Defense Date 05/25/2021

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag antibiotics,central carbon metabolism,desaturase,fatty acid,glycerol metabolism,glycolysis,metabolite,metabolomics,methylglyoxal detoxification,Mycobacterium,OAI-PMH Harvest,pentose phosphate pathway,reactive oxygen species,ROS,smegmatis,Stress,TCA cycle,trehalose,trehalose synthase,Tuberculosis

Format application/pdf (imt)

Language English

Advisor Eoh, Hyungjin (committee chair), Luna, Brian (committee member), Yuan, Weiming (committee member)

Creator Email psell@usc.edu,theole.psell@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-oUC15669265

Unique identifier UC15669265

Legacy Identifier etd-SellPhilip-9935

Document Type Thesis

Format application/pdf (imt)

Rights Sell, Philip Jānis

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the author, as the original true and official version of the work, but does not grant the reader permission to use the work if the desired use is covered by copyright. It is the author, as rights holder, who must provide use permission if such use is covered by copyright. The original signature page accompanying the original submission of the work to the USC Libraries is retained by the USC Libraries and a copy of it may be obtained by authorized requesters contacting the repository e-mail address given.

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA

Repository Email uscdl@usc.edu

Abstract (if available)

Abstract Tuberculosis (TB) is the deadliest disease due to a single infectious pathogen, Mycobacterium tuberculosis (Mtb). Mtb can remain viable after extended periods of antibiotic treatment, these transiently tolerant tubercle bacilli are called persisters. Not only can the persister population survive conventional antibiotic treatments but it can also serve as a source for accumulation of drug resistance mutations. It was previously found that Mtb can form persister-like bacilli (PLB) by shifting trehalose metabolism for utilization of internal carbon and production of antioxidants. We confirmed these findings in a common Mtb model organism, Mycobacterium smegmatis (Msm), and additionally found a link between trehalose metabolism and acquisition rate of drug-resistance mutations. Paralleled results in this study are reported for three Desaturase A enzymes involved in fatty acid metabolism which show promising characteristics as potential metabolic targets for new TB therapeutics. Similar mechanisms to trehalose metabolism are involved with the desaturases in persister formation, suggested by dysregulation of glycolysis and the pentose phosphate pathway as defined by metabolomic analysis. Finally, we identified Msm strains with an irreversible drug tolerance phenotype after single exposure to antibiotics. Irreversible drug tolerance was related to a defect in growth rate due to altered glycerol metabolism. Control of oxidative damage was correlated to bacterial survival across all experiments testing both, transient and irreversible, drug tolerant phenotypes. Mtb uses a range of adaptive metabolic survival strategies for transient persistence and irreversible tolerance to antibiotic stress. Inhibiting essential enzymes in these adaptive metabolic pathways could greatly increase efficacy and decrease length of current TB treatments.