Download - Evolution of MDR-TB - APHL TB Conference... · Evolution of MDR-TB ... 106–110 In clinico Fitness of rpoA/C Mutations ... Ruedi Aebersold Uwe Sauer Tanja Stadler Jörg Stelling

Evolution of MDR-TB

Sébastien Gagneux, PhD10th National TB Conference

Washington DC, 19th April 2017

Dept. of Medical Parasitology & Infection BiologyTuberculosis Research Unit

Global TB Estimates (2016)

Number of cases

Number of deaths

1.8 million10.4 million

190,000480,000

All forms of TB

MDR-TB

XDR-TB ~ 48,000 ~ 24,000

>50% Case-fatality:→ Entering Post-Antibiotic Era !“Only” ~ 5% MDR / XDR

“Drug‐Resistant Bacteria Are Less Fit”

DOGMA:

DS

DR

Fitness

Time

Proportion MDR among new TB (2015)

WHO REPORT 2016

0.01

0.1

1

10

100

29 76 54 67 4 32 78 4 29 79 34 33 77 30 31

Source reference

Rel

ativ

e fit

ness

of r

esis

tant

stra

ins

(log)

The Relative Fitness of DR Mtb is Heterogeneous

Borrell & Gagneux 2009 IJTLD 13: 1456-66

Exploring the Role of Epistasis

Epistasis

Drug-resistancemutation(s)

Compensatorymutations

Strain geneticbackground

“Success”of MDR Mtb

Borrell & Gagneux 2011 Clin Microbiol Infect 17: 815–820

DR

DS

Fitn

ess DR

Compensation

TimeBorrell & Gagneux 2009 IJTLD 13: 1456-66

Evolution of Drug Resistance

Compensatory Mutations in rpoA/C

Comas et al. 2012 Nature Genetics 44: 106–110

Comas et al. 2012 Nature Genetics 44: 106–110

In clinico Fitness of rpoA/C Mutations

High-confi. CMs

All CMs

*

*

* P < 0.05

Song et al. 2014 Mol Microbiol 91: 1106–19

Genetics in M. smegmatis

In vitro growth Transcription efficiency

Gygli et al. 2017 FEMS Microbiol Rev, in press.

• RIFR strains: N=1,488• + RpoA: N=73• + RpoC: N=729

Fenner et al. AAC 2012 56: 3047-53

Mtb Lineage Impacts INH Resistance LevelsP

erce

ntag

e

0

20

40

60

80

100

MIC <3.0mg/L

MIC >3.0mg/L

MIC <3.0mg/L

MIC >3.0mg/L

0.0110.004

5

5

2

2

3

1764

24

katG 315mutations

inhA promotor-15mutations

2

2

2

3

InhA pro ‐15mutations

KatG 315mutations

MIC MIC • 134 clinical INHR isolates

0.005

Lineage 4

Lineage 3

Lineage 2

Lineage 6

Animalstrains

Lineage 5

Lineage 1

Lineage 7

100

100

100

100100

100

100

99

100100

Positive Sign Epistasis in DoubleR

Borrell et al. 2013 Evol Med Publ Health eot003: 65-74

rpoB / gyrA mutations

Epistasis

Drug-resistancemutation(s)

(e.g. rpoB/gyrA)

Compensatorymutations

(e.g. rpoA/C)

Strain geneticbackground

(e.g. L1 vs L2)

• Success of MDR Mtb• MICs• Patient outcome (?)

Borrell & Gagneux 2011 Clin Microbiol Infect 17: 815–820

Conclusion (1): Epistasis matters!

A Web of Epistasis Mediates MDR in Mtb

Trauner et al. 2014 Drugs 74: 1063-72

What about within-host evolution?

Simple population

Antibiotic treatment

Time

Simple population + 1

Antibiotic treatmentTime

Case Study From Switzerland

• Tibetan refugee (HIV-neg.)

• Primary resistance to: isoniazid rifampicin (+ compensatory mutation in rpoC) pyrazinamid streptomycin ethionomide fluoroquinolones linezolid (!)

“Clinical Cure” Relaps

Bloemberg et al. 2015 NEJM 373: 1986-8

Resistancemutationsto 7 drugs


bedaquilineR

clofazimineR



capreomycinR

bedaquilineR

clofazimineR



delamanidR

capreomycinR

bedaquilineR

clofazimineR



bedaquilineR

clofazimineR

capreomycinR

delamanidR


What about

other

patients? AndrejTrauner

Mtb in the host

Rare(10)

Minor(2)

Major(1)

Antibiotic treatment

Time

How stable are individual clones?

time (weeks)0 2 4 6 8 16 24

P01P02P03P04P05P06P07P08P09P10P11P12 D

Non-efficacious treatmentEfficacious treatment

Treatment

PhenotypicDST

INH

RIF

EMB

STR

GenotypicDST

INJ

FQ PZA

DS DR

Context of Treatment Efficacy

Trauner et al. 2017 Genome Biol, in press.

1 out of 12 patients.

Emergence of FQ resistance


Total variable SNPs = 492

Site Frequency Spectrum


Multiple isolatedpopulations

Diversity driven by drift

Purifying selectionon minor clones

Predominant clone


D: Detected ND: Not detectedpD : Stable pND : AbsentpDND: Loss pNDD: Gain

Quantifying the Trajectory of Mutations


Non-efficacious treatment: ALL SNPs NSY SNPs SYN SNPsEfficacious treatment: ALL SNPs NSY SNPs SYN SNPs


Purifying Selection in Efficaciously Treated Patients


Efficaciously treated Nonefficaciously treated

Gene set Na Excessive mutationb Excess NSYc Excessive

mutation Excess NSY

DrugResistanced

13 0/100 (0.501) 0/0 (1.000) 5/87 (0.001) 5/5 (0.177)

DrugResistanceAssociatede,f

166 10/100 (0.545) 4/10 (0.987) 6/87 (0.946) 6/6 (0.121)

MycolateSuperpathwayg

54 3/100 (0.881) 1/3 (0.964) 5/87 (0.229) 3/5 (0.876)

MTBC T-CellAntigensh

300 14/100 (0.153) 10/14 (0.426) 6/87 (0.550) 6/6 (0.121)

a Number of genes in the gene set. b Proportion of mutations in gene set, p-value calculated with a one-sided binomial test. c Proportion of NSY mutations ingeneset, p-value calculated with a one-sided binomial test. d Walker et al., e Zhang et al., f Farhat et al., g O’Neill et al., h Coscolla et al.

Non-efficaciously Treated Patients Accumulate Mutations in DR Genes


• Most new mutants are unstable.

• Mutations with no (little) functional effect

are more stable.

• This is particularly true in efficaciously

treated patients.

• Non-efficaciously treated patients

accumulate mutations in DR genes.

Conclusions (2)

Thanks to… CollaboratorsUniversity of Valencia

Iñaki Comas

New York UniversityJoel Ernst

University of Cape TownRob WilkinsonHelen Cox

Stellenbosch UniversityRob Warren

University of BernMatthias Egger

University of ZurichErik Böttger

UCSFMidori Kato-Maeda

Sanger InstituteSimon Harris

University of GhanaDorothy Yeboah-Manu

Institute Trop. Med. AntwerpBouke de Jong

Fudan UniversityQian Gao

ETH ZurichRuedi AebersoldUwe SauerTanja StadlerJörg StellingChristian Beisel

Forschunszentrum BorstelStefan Niemann

Max-Planck Institute JenaJohannes Krause

Thanks to…

• Sonia Borrell• Mireia Coscolla• Daniela Brites• Andrej Trauner• Julia Feldmann• Miriam Reinhard• Levan Jugheli• Sebastian Gygli• Liliana Ruthaiwa• Rhastin Castro• Chloe Loiseau• Monica Ticlla• Peter Major

[email protected]