GENETIC DIVERSITY AND MOLECULAR BASIS OF

ANTIMICROBIAL RESISTANCE OF MYCOBACTERIUM

TUBERCULOSIS STRAINS FROM PAKISTAN

By

ASHO ALI

A thesis submitted in partial fulfilment of the requirements for the degree of PhD in Health Sciences

in the discipline of Microbiology

Department of Pathology and Microbiology Faculty of Health Sciences

Aga Khan University Karachi, Pakistan

2009

© Copyright

ii

Genetic Diversity and Molecular basis of Antimicrobial

resistance of Mycobacterium tuberculosis strains from Pakistan

A thesis submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy (Health Sciences) in the discipline of Microbiology

Members of the Thesis Defense Panel appointed to examine the thesis of

Asho Ali

find it satisfactory and recommended that it be accepted

_____________________________ Dr. Rumina Hassan (Thesis Supervisor)

_____________________________ Dr El-Nasir Lalani (Internal Examiner)

_____________________________ Dr. Rana Muzaffar

(External Examiner)

_____________________________ Dr. Anwar-ul-Hassan Gilani

(Convener, Thesis Defense Panel)

Department of Pathology and Microbiology

Aga Khan University Karachi, Pakistan

March 11, 2009

iii

This thesis is dedicated to my parents!

iv

Abstract

Pakistan ranks 8th amongst the 22 high burden tuberculosis (TB) disease countries

with an estimated incidence rate of 181/100,000 population. The high incidence of

tuberculosis in Pakistan is further compounded by the increasing emergence of drug

resistant strains including multi-drug resistant (MDR). Prevalence of MDR-TB in

Pakistan has been shown to be between 2-4% in the untreated patients, while the global

prevalence of MDR is estimated at 3%.

Molecular epidemiological studies, based on the assumption that patients infected

with clustered strains are epidemiologically linked, have helped understand the

transmission dynamics of disease. It has also helped to investigate the basis of variation

in Mycobacterium tuberculosis (MTB) strains, differences in transmission, severity of

disease or drug resistance mechanisms in a defined geographical location. This is in turn

helpful in developing strategies for the treatment and prevention of the disease including

MDR. Molecular epidemiological data using spoligotyping available from Pakistan

shows a predominance of the Central Asian Strain1 (CAS1) (39%), and Beijing strains

(6%). Beijing strains have been shown an association with MDR. Although CAS1 strains

have not shown an association with MDR, about forty percent of total CAS1 strains were

comprised of MDR strains. The data about the types and frequency of drug resistance

gene mutations within these strains is limited.

The overall aims of this study were to explore genetic diversity amongst the

predominant genogroup of Mycobacterium tuberculosis from the country as well as to

investigate genetic basis of drug resistance amongst these predominant MTB strains. In

v

this study variable number of tandem repeat mycobacterial interspersed repetitive unit

(VNTR-MIRU) and IS6110 based restriction fragment length polymorphism (IS6110-

RFLP) was used to study the relationship within CAS1 strains. Knowledge of type and

frequency of mutations amongst prevalent genogroup has been shown to be essential for

the development of appropriate tools for early diagnosis and control of MDR-TB strains.

There is limited information on mutations leading to drug resistance within MTB strains

from the country. Therefore in this study prevalent mutation in the rpoB, katG and inhA

genes for rifampicin (RMP) and isoniazid (INH) resistance were also investigated in

MDR strains of predominant genogroups.

Twelve loci based VNTR-MIRU typing showed highly diverse profile of 367

MTB strains. Of the 178 CAS1 strains studied only 34 (19 %) clustered into groups

based on MIRU profiles, while all 189 ‘unique’ spoligotypes studied had non-matching

MIRU profiles and therefore remained un-clustered. The MIRU-VNTR data shows a

close relationship (70%) within the prevalent CAS1 strains. Therefore it is proposed that

in a region where CAS1 family strains are prevalent most discriminatory MIRU loci 26,

31, 16, 10, 27, 39 and 40, could be used for differentiation and estimation of the

phylogenetic relatedness of Mycobacterium tuberculosis. IS6110-RFLP typing of 78

strains (43 CAS1 and 35 ‘unique’ spoligotypes strains) resulted in 73 different RFLP

types. One cluster of two unique strains, with single copy of IS6110 was identified; the

remaining seventy-two strains revealed unique RFLP patterns.

The most common mutations determined in MDR strains were in codons 531

(60%), 526 (23%) and 516 (5%) of rpoB gene by sequencing, while probe based assay

vi

detected 44% and 21% mutations at codon 531 and 526 respectively. Occurrence of

mutation at codon 526 as well as concurrent mutation in rpoB gene was significantly

higher in CAS1 than Beijing and other un-clustered strains tested. Sixty three percent of

62 MDR isolates showed mutation at codon 315 by sequencing while by probe based

assay sixty percent of resistance was detected. CAS1 family strains exhibited higher rate

of mutation at codon 315 as compared with Beijing.

Multi-drug resistant CAS1 strains are more prone to develop resistance against

RMP and INH through mutations at codon 526 of rpoB gene and codon 315 of katG gene

respectively than non-CAS1 MDR strains. 67% of RMP and INH resistance within MDR

CAS1 strains could be determined by detecting mutations at only three codons 526 and

531 of rpoB and 315 of katG gene.

vii

List of abbreviations

AFB Acid Fast Bacilli ATT Anti-tuberculosis therapy BCG Bacillus Calmette-Guerin bp base pairs CAS1 Central Asian Strain1 DMSO Di methyl sulf oxide DNA Deoxyribonucleic Acid DOTS Direct Observed Treatment Strategy DR Direct Repeat E Ethambutol ETR Exact Tandem Repeats FRET Fluorescence Resonance Energy Transfer HGDI Hunter Gaston Discriminatory Index HBC High Burden Country IDU Injection Drug Users HIV Human Immunodeficiency Virus INH Isoniazid IS Insertion Sequence IUATLD International Union against Tuberculosis and Lung Diseases LiPA Line Probe Assay LJ Lowenstein Jensen LSP Large Sequence Polymorphism MDR Multi Drug Resistance MDR-TB Multi Drug Resistance Tuberculosis MTB Mycobacterium tuberculosis MIRU Mycobacterial Interspersed Repetitive Unit MTC Mycobacterium tuberculosis complex NTP National Tuberculosis control Program PCR Polymerase Chain Reaction PYZ Pyrazinamide RFLP Restriction Fragment Length Polymorphism RMP Rifampicin SNP Single Nucleotide Polymorphism SPSS Special Program for Social Sciences Software TB Tuberculosis UPGMA Un-weighted pair group using arithmetic averages VNTR Variable Number of Tandem Repeat WHO World Health Organization

viii

ACKNOWLEDGMENT

I would like to express my genuine gratitude and sincere thanks to all those who

have helped and encouraged me during this thesis work especially;

Dr Rumina Hasan, my supervisor, who from the first day of her involvement in this

project invested all her time and effort until the day this thesis work became a reality. Her

excellent scientific guidance, constant support and encouragement throughout the thesis

work brought it to completion.

Dr Zahra Hasan, my co-supervisor for making exploration of genomic world of

Mycobacterium tuberculosis possible for me. Her endless support, enduring

encouragement and kindness led me through this thesis.

Dr Rabia Hussain, my thesis committee member, for her expert advisement and

scientific criticism during initial phase of the study.

Dr Rehana Siddiqui, my thesis committee member, for helping me out in statistical

calculations of samples and guidance time to time during this thesis period.

Dr Tariq Moatter, for all his guidance in making detection of drug resistance possible

through real time PCR.

Dr Anwar Ali Siddiqui, Associate Dean Research for his sincere support through out this

thesis period.

Dr Ruth McNerney, from The London School of Hygiene and Tropical Medicine, UK

for guidance in data analysis of MIRU-VNTR typing.

ix

Dr Kristin Kremer from Mycobacteria Reference Unit, Diagnostic Laboratory, The

Netherlands for providing reference strains of Mycobacterium tuberculosis for RFLP.

Ramona Petersson and Solomon Ghebremichael from Swedish Institute for Infectious

Disease Control, Sweden for their help with RFLP typing.

Mr Iqbal Azam from Community Health Sciences, AKU, for statistical support

Mahnaz Tanveer, my colleague and friend, for helping me with the isolation and

culturing of Mycobacterium tuberculosis and supporting me through out without which I

could not have completed my work.

All my colleagues at Clinical Microbiology Lab for providing me the samples and

support in getting things started.

All my PhD and Juma Research Lab colleagues with whom I had great scientific

exchanges and shared my ups and downs of thesis progress.

Jack Fernandes and other staff members of Research department, for their support

through out my stay in the PhD program.

Dr Yasmin Amarsi and all my colleagues at School of Nursing, especially Science

faculty for their great support and forbearance during this thesis period.

University Research Council, AKU for funding this research.

Last but not the least, my dear family; my mother, late father, sisters and brothers for

their love, support and patience from the very beginning until now!

x

DECLARATION

I declare this thesis does not incorporate without acknowledgement any material previously submitted for a degree or diploma in any university and that to the best of my knowledge it does not contain any material previously published or written by another person except where due reference have been made in the text.

______________________________ (Signature of candidate)

December, 2008 Date

xi

List of Tables

No. Title Page #

1.1: Median prevalence of MDR-TB amongst new and previously treated 27 cases by region (%) 1.2: MTB genes associated with resistance to antituberculous agents 28 2.1: Geographical Distribution of selected MTB strains for RFLP 40 2.2: Overview of number of IS6110 element present in 78 MTB Isolates 50 3.1: Geographical Distribution of selected MTB strains for MIRU typing 57 3.2: MIRU-VNTR primers for twelve loci 61 3.3: Allelic diversity of 367 MTB isolates 64 3.4: Twelve MIRU loci analysis of CAS1 and ‘unique’ spoligotypes 65 3.5: MIRU loci analysis of MDR M tuberculosis 66 4.1: PCR primers for drug resistance gene amplification 82 4.2: Detection of Rifampicin Resistance mutations by rpoB gene sequencing 83 of MDR-MTB strains 4.3: Detection of Isoniazid Resistance mutations by katG and inhA gene 84 Sequencing of MDR-MTB strains 5.1: Primers and FRET probes used for PCR amplification and detection 97 of RMP and INH resistance in MTB strains 5.2: Detection of prevalent Rifampicin and Isoniazid resistance by FRET 102 probes using real time PCR 5.3: Summary of mutations identified by Sequencing and Real time PCR in 104 Rifampicin and Isoniazid resistant MTB isolates as compared to phenotypic drug susceptibility testing (DST) 5.4: Comparison of results obtained by LiPA and DNA Sequencing on 106 33 MDR-TB strains

xii

List of Figures

No. Title Page # 1.1: TB incidence rate in four provinces of Pakistan 8 1.2: Tuberculosis infection 13 1.3: MTB genome showing polymorphic repeat sequences 16 2.1: Schematic diagram of steps of IS6110-RFLP method 43

2.2: Autoradiograph of Internal Marker and IS6110 probe hybridization 46

2.3: Dendrogram of IS6110-RFLP typing of Mycobacterium tuberculosis 49

3.1: Schematic diagram of steps of MIRU-VNTR method 62

3.2: Dendrogram of MIRU-VNTR typing of Mycobacterium tuberculosis 66

3.3: Composite dendrogram of RFLP and MIRU-VNTR of 78 MTB strains 67 5.1: Mechanism of action of FRET probes 96 5.2: Melting curves detecting rpoB mutation using FRET probes 101 5.3: Melting curves detecting katG mutation using FRET probes 103 5.4: Melting curves detecting inhA mutation using FRET probes 103 5.5: rpoB mutations identified by InnoLiPA assay 105

xiii

Table of Contents

Page Abstract…………………………………………………………………………… iv List of abbreviations ……………………………………………………………… vii Acknowledgement ………………………………………………………………… viii Declaration ………………………………………………………………………… x List of tables ………………………………………………………………………. xi List of figures……………………………………………………………………….. xii Chapter One: Mycobacterium tuberculosis: An overview

1.1: General Background 1 1.2: History of tuberculosis 2 1.3: Global impact of tuberculosis 4 1.4: Pakistan 7 1.5: Tuberculosis in Pakistan 9 1.6: Natural history of M. tuberculosis 11 1.7: Genome and genomic diversity of M. tuberculosis 12 1.8: Molecular Epidemiology of Tuberculosis & its significance 14 1.9: Molecular epidemiological data from Pakistan 20 1.10: Antituberculous therapy 22 1.11: Drug-resistant tuberculosis 24 1.12: Multiple Drug resistant tuberculosis 25 1.13: MDR-TB in Pakistan 29 1.14: Molecular mechanisms of Drug resistance 29 1.15: Drug susceptibility testing 30 1.16: Goal of present study 32

Chapter Two: Genotyping of Mycobacterium tuberculosis using IS6110-Restriction Fragment Length Polymorphism

2.1: Background 35 2.2: Objectives 38 2.3: Methods

2.3.1: Mycobacterial strains 39 2.3.2: Culture and antibiotic susceptibility 39 2.3.3: DNA extraction 41 2.3.4: IS6110- RFLP typing 42 2.3.5: Computer-assisted phylogenetic analysis 47 2.3.6: Statistical analysis 47

2.4: Results 2.4.1: Diversity of RFLP 48

2.4.2: IS6110-copy number 48 2.5: Discussion 51

xiv

Chapter Three: Genotyping of Mycobacterium tuberculosis using Mycobacterial Interspersed Repetitive Unit typing method 3.1: Background 53 3.2: Objectives 55 3.3: Methods

3.3.1: Mycobacterial strains 56 3.3.2: Culture and antibiotic susceptibility 58 3.3.3: DNA extraction 58 3.3.4: MIRU typing method 58 3.3.5: Phylogenetic analysis 60 3.3.6: Statistical analysis 60

3.4: Results 3.4.1: MIRU typing for CAS1 and ‘unique’ strains 68 3.4.2: Allelic diversity 68 3.4.3: Discriminatory power of MIRU typing for CAS1 70

3.4.4: Comparison of MIRU and RFLP typing profiles 70 3.4.5: Analysis of MIRU typing of MDR isolates 71

3.5: Discussion 72 Chapter Four: Detection of drug resistance gene mutations in MDR CAS1 and Beijing strains by Sequencing 4.1: Background 76

4.2: Objectives 78 4.3: Methods

4.3.1: Mycobacterial strains 79 4.3.2: Culture and antibiotic susceptibility 79 4.3.3: DNA extraction 79 4.3.4: Sequencing of rpoB gene for RMP resistance 80 4.3.5: Sequencing of katG and inhA genes for INH resistance 80 4.3.6: Statistical analysis 81

4.4: Results 4.3.1: rpoB gene mutations for RMP resistance 85 4.3.2: katG and inhA genes mutations for INH resistance 86

4.5: Discussion 87

xv

Chapter Five: Molecular methods for rapid detection of Rifamcin and Isoniazid resistance amongst the MDR strains of predominant genogroups of Mycobacterium tuberculosis 5.1: Background 90 5.2: Objectives 93 5.3: Methods

5.3.1: Mycobacterial strains, Culture & DNA extraction 94 5.3.2: RMP and INH resistance detection using FRET probe 94

based Real Time PCR 5.3.3: RMP resistance detection using InnoLiPA assay 98 5.3.4: Statistical analysis 99

5.4: Results 5.4.1: rpoB gene mutations for RMP resistance using 100

FRET probes 5.4.2: katG and inhA genes mutations for INH resistance 105

using FRET probes 5.4.3: rpoB gene mutations for RMP resistance using InnoLiPA 105

5.5: Discussion 108

Chapter Six: General discussion and Conclusions 110

References 117 Appendices 130 A: Publications 1: “Characterization of Mycobacterium tuberculosis Central Asian Strain1

using Mycobacterial Interspersed Repetitive Unit genotyping” paper published in BMC Microbiology

2: “M. tuberculosis Central Asian Strain 1 MDR isolates have more mutations in rpoB and katG genes as compared with other genotypes” paper published in Scand. Journ. of Inf. Diseases 3: “Molecular Epidemiology of Mycobacterium tuberculosis: Review article” Published in Infectious Diseases Journal, Pakistan 4: “Genotyping and Drug resistance patterns of M. tuberculosis strains in Pakistan” paper published in BMC Microbiology

B: Curriculum vitae

1

Chapter One

Introduction

Mycobacterium tuberculosis: An overview

“If the importance of a disease for mankind is measured by the number of fatalities it causes, then tuberculosis must be considered much more important than those most feared infectious diseases, plague, cholera and the like. One in seven of all human beings dies from tuberculosis. If one only considers the productive middle-age groups, tuberculosis carries away one-third, and often more.”

Robert Koch, March 24, 1882

1.1: General background

More than 125 years after the discovery of its causative agent, tuberculosis (TB) is still a

major killer disease worldwide, with > 8 million new cases and >2 million deaths each

year. It is estimated that today one third of the world population is infected with TB

Chapter preview Page # 1.1: General Background 1 1.2: History of tuberculosis 2 1.3: Global impact of tuberculosis 4 1.4: Pakistan 7 1.5: Tuberculosis in Pakistan 9 1.6: Natural history of M. tuberculosis 11 1.7: Genome and genomic diversity of M. tuberculosis 12 1.8: Molecular Epidemiology of Tuberculosis & its significance 14 1.9: Molecular epidemiological data from Pakistan 20 1.10 Antituberculous therapy 22 1.11: Drug-resistant tuberculosis 24 1.12: Multi-Drug resistant tuberculosis 25 1.13: MDR-TB in Pakistan 29 1.14: Molecular basis of Drug resistance 29 1.15: Drug susceptibility testing 30 1.16: Goals of present study 32

2

(WHO 2008). With the introduction of antituberculosis therapy (ATT) in1950s and the

use of Bacille Calmette Guerin (BCG) vaccine, many experts calculated that TB would

be eradicated in few years. However, TB generally and multidrug resistant (MDR;

defined as resistant to at least rifampicin (RMP) and isoniazid (INH) TB especially is

increasing worldwide since 1990. It is estimated that MDR strains of Mycobacterium

tuberculosis (MTB) constitutes 1-3% of global TB isolates (WHO 2008). World Health

Organization (WHO) estimates that up to 50 million people worldwide may be infected

with drug resistant strains of TB. The fatality rate of MDR-TB is 20-80%. In addition

human immunodeficiency virus (HIV) epidemic has also played a key role in increasing

the number of new TB cases (Aaron, Saadoun et al. 2004; WHO 2008).

Thus global increase in TB, particularly of drug resistant strains emphasize the

need for rapid detection and drug susceptibility testing of MTB in clinical samples. Such

rapid detection is necessary for adequate antituberculous therapy and containment of

resistant strains. In addition, new drugs as well as a better vaccine are also needed for

improved therapy and preventive measures.

1.2: History of Tuberculosis

TB is an ancient disease that has been called King’s Evil, lupus vulgaris,

consumption and white plague during the last several centuries. Archeological findings

from Europe, Egypt, Greece and Rome show evidence of TB (Mathema, Kurepina et al.

2006). For several centuries, scientists and physicians have been trying to understand the

nature of tuberculosis for better diagnoses, prevention, and cure. Hippocrates thought the

3

disease was inherited, while Aristotle (4th century B.C.) pointed out its contagious nature.

This view of TB reemerged in the second half of 17th century, when Italian physicians

supported the idea that TB was contagious. In contrast, doctors in Northern countries,

such as Denmark and Sweden, favored hereditary causes of this disease. They believed

that the theory of TB being contagious was not proven scientifically and did not explain

why some people in urban settings did not get TB even where there was a high incidence

of the disease (Haas 1996, Little, Brown and Co. Boston, Mass). To solve this

philosophic difference, in 1865, Jean-Antoine Villemin, a French military physician,

reported transmitting TB to laboratory rabbits by inoculating tuberculous tissue from a

cadaver. This report was strongly opposed by other French medical practitioners, who

argued that there had to be more modern and more social solutions to the problem of TB,

which arose in the poorer classes due to malnutrition, poor sanitation, and overwork. The

report by Robert Koch 17 years later in 1882, which conclusively showed that TB was

indeed caused by a bacterium discredited this argument (Koch 1982). However, belief in

the societal causes of TB continued in early 20th century and they used TB as an example

of a disease that was caused by overwork and malnutrition (Barnes 2000). Finally this

dichotomy in explaining the etiology of tuberculosis was resolved in early 20th century

by Edward Trudeau's work. He showed that TB could be induced in rabbits with a

purified MTB culture but that the environmental conditions in which the animals were

maintained greatly influenced the course of the disease. This simple experiment gave

4

scientific validity to the treatment of TB in sanitarium for fresh air and healthy food,

started by European physicians in the mid-1800s (Trudeau 1887).

Thus, it was concluded by Rene Dubos that TB is caused by a bacterium, but

environmental factors play a major role, and purely medical solutions alone would not

work to cure and prevent TB (Dubos 1952). This has been clearly seen by the rising

number of TB patients in the last half of the 20th century.

1.3: Global impact of Tuberculosis:

WHO estimates that one third of the world’s population is infected with MTB. In

1993, WHO declared TB a global emergency. TB accounts for 2.5% of the global burden

of disease. Although more cases of TB are reported among men, it is still the commonest

cause of death in young women (Dye, Scheele et al. 1999; Borgdorff, Nagelkerke et al.

2000). The incidence of TB ranges from less than 10 per 100,000 in North America to

100 to 300 per 100,000 in Asia and Western Russia to over 300 per 100,000 in Southern

and Central Africa (WHO 2008).

The TB incidence declined steadily during 1900s in the developed countries in

Europe and United States mainly due to improved socioeconomic conditions, BCG

vaccination and effective antituberculous therapy. This downward trend however, started

increasing in the mid-1980s. Only by extensive efforts mainly through supervised

5

antibiotic therapy has the situation been reversed in Europe and the United States

(Frieden, Fujiwara et al. 1995).

The developing world is however, still suffering from a rising incidence of TB.

Majority of TB cases in the developing world are a result of lack of health care facilities

(Waaler 2002; WHO 2008). According to WHO statistics, the African region has the

highest estimated incidence rate (356/100,000) of TB but the majority of TB patients live

in Asia where a rapid increase in population and HIV cases are the main factors

contributing to the rising number of TB cases in these areas. India, China, Bangladesh,

Indonesia and Pakistan together account for 50% of the new TB cases that arise every

year (Dye 2006; WHO 2008). The other major implication of TB is the occurrence of

disease in 75% of people within the economically productive age group i.e.15-54 years.

Since 95% of all TB cases occur in developing countries, it has a large impact on

socioeconomic development of these countries (Dye 2006; WHO 2008).

WHO has identified 22 highly endemic countries, which contribute about 80% to

the global TB burden. These countries are labeled as being high burden countries (HBCs)

of TB. These HBCs and WHO are making collaborative efforts to achieve the global

target of TB case detection (70%) and cure (85%) through Directly Observed Treatment

Short-course (DOTS) (WHO 2003). To date, however only the Philippines and Vietnam

have succeeded in meeting their targets of case detection and treatment success by the

end of 2004 (Dye 2006; WHO 2008).

6

Pakistan and its three neighboring countries India, China and Afghanistan

contribute significantly to the global TB burden. India ranks first on the list of 22 HBCs

in terms of total number of TB patients present. It is estimated that approximately 40% of

people living in Indian subcontinent, China and the Pacific Rim are infected with TB

(Dye, Scheele et al. 1999).

India with its population of over 1000 million is estimated to account for nearly

30% of the global tuberculosis burden. In 1997, the estimated incidence of TB in India

was 187 per 100,000 populations, while the current estimate is 505 per 100,000 (Dye

2006). The situation has been adversely affected by the increasing number of HIV

infected people. The HIV rate in India has been found to vary between 0.4% and 28.8%

(Chadha 2005). Overall about 50-70% of HIV positive patients in India reportedly

develop TB during their lifetime thus leading towards increasing number of TB-HIV co-

infection. It has also been estimated that TB-HIV co-infection in India has increased 2-10

times as compared to the estimates made in 1997 (Swaminathan, Ramachandran et al.

2000; WHO 2008).

China ranks 2nd on the list of 22 HBCs. It has the second largest population of

tuberculosis patients in the world, with more than 1.3 million new cases of tuberculosis

every year. It accounts for 17% of the total global burden of TB (WHO 2008).

Afghanistan is the last country on the WHO list of 22 HBCs. TB is a major public health

burden with 95,000 new TB cases and 26,000 deaths every year in the country (WHO

2008).

7

1.4: Pakistan

Pakistan covers an area of 310,000 square miles in South Asia. It has four major

provinces (Punjab, Sind, North-West Frontier Province and Baluchistan) (Figure 1.1).

The most populous province, Punjab, has only 26 percent of the land area but is home to

about one-half (56% in the 1998 census) of the 143 million people estimated to live in

Pakistan. This population makes Pakistan the seventh most populous country in the world.

Life expectancy is estimated at 66 years in women and 64 years in men (Pakistani

Federal Bureau of Statistics www.statpak.gov.pk/depts/fbs/statistics/pds).

8

Figure 1.1: TB incidence rate in four provinces of Pakistan

(Adapted from pakistan.spe.org/.../clear_pakistan_map2.gif and 2007 data from NTP)

43/100,000

148/100,000

6/100,000

123/100,000

9

1.5: Tuberculosis in Pakistan

Pakistan ranks eight amongst the 22 high burden TB disease countries with an

estimated incidence rate of 181/100,000 while sputum positive TB cases are estimated at

82/100,000 population (WHO 2008). In addition there are estimated 268,000 new cases

and 64,000 deaths from TB each year in Pakistan which contributes about 44% of TB

cases in WHO Eastern Mediterranean region (WHO 2008).The quoted TB burden in

Pakistan is likely to be an underestimated figure as many cases in the country go

unreported due to lack of access to health care facility, over crowding, poverty and other

social constraints (WHO 2003). According to National Tuberculosis Control Program

(NTP) of the country, TB incidence in each of four provinces ranges between 96/100,000

in Baluchistan to 148/100,000 in Punjab (Figure 1.1).

Similar to other developing countries, majority of TB cases occur predominantly

in the economically most productive, 15 to 54-year age group, which further hinders

socioeconomic development. TB is, therefore, not only an important problem in the field

of public health in the country; it is also a socio-economic issue that not only harms

people’s health, but also imposes a heavy economic burden on the family of TB patients

(Dye 2006; Habib and Baig 2006).

In order to control the high incidence rate of TB the government of Pakistan

endorsed DOTS in 1994 in collaboration with WHO. DOTS strategy is being

implemented in Pakistan under NTP. Although treatment success by DOTS program in

Pakistan is reported at more than 70%, DOTS detection rate is below the population

10

coverage of 24%, far below the 70 percent target, suggesting that many patients do not

have access to DOTS (Netto, Dye et al. 1999; WHO 2003). In 2001, the government

declared TB a national emergency, which led to a TB budget increase from US$1.65

million in 2001 to US$26 million in 2006. This was mainly for the development in DOTS

expansion and increases in the case detection and cure rate.

Poor TB patient compliance to treatment has further exacerbated the situation.

Partially treated TB patients are the main source for the spread of drug resistant TB

particularly MDR-TB. Non-availability of drugs, lack of accessibility to the health care

centers and lack of awareness regarding consequences of incomplete therapy are known

causes of poor compliance with TB treatment (Israr 2003; Khan and Malik 2003;

Hussain, Mirza et al. 2005; Habib and Baig 2006).

High treatment cost of drug-resistant TB especially MDR-TB is an additional

economic burden both to the individual patient and society. WHO estimates that 3.4% of

all new cases have MDR-TB while in previously treated patients it is estimated at 36%

(WHO 2008).

Although HIV prevalence is low in Pakistan, HIV/TB co-infection has been

reported as 19% in high risk group (Memon, Memon et al. 2007). The high risk group

mostly include injection drug users (IDU) (Memon, Memon et al. 2007; Vermund and

Yamamoto 2007). Among female sex workers 1-2% HIV prevalence has been reported

(Ministry of Health Pakistan, 2005; National AIDS Control Program Pakistan,

2005;(WHO 2003).

11

1.6: Natural history of Mycobacterium tuberculosis

Mycobacterium tuberculosis (MTB), which causes TB, is one of the most

successful pathogens of mankind and may have killed more people than any other

pathogen (Daniel 2006). These organisms are called “the Koch’s bacillus” after Robert

Koch who first identified tubercle bacilli (Sakula 1983).

MTB is one of the members of Mycobacterium tuberculosis complex (MTC).

Other species in the complex are Mycobacterium africanum, Mycobacterium bovis,

Mycobacterium bovis BCG, and Mycobacterium microti. These are slow growing,

fastidious, aerobic and acid fast rods. On solid culture medium; Lowenstein Jensen (LJ)

and 7H10/7H11 Middlebrook agar, MTB requires 16-18 hours to undergo one cycle of

replication and within 2-3 weeks a single bacterium may produce a visible colony

(Wayne 1977).

Transmission of MTB occurs mainly through air by droplets containing bacteria,

which are formed when infected people cough or sneeze. Inhalation of these small

droplets may cause pulmonary TB amongst healthy people. MTB stay in its human host

in the form of a dormant infection. MTB organisms multiply inside macrophages and

infection may persist in the latent phase for years. Reactivation of latent disease may

occur after several years of initial infection in small proportion of infected people. The

major risk factors for reactivation include old age, malnutrition, HIV co-infection or other

chronic diseases such as diabetes. However, the majority of infected people never

develop clinical disease (Figure 1.2). Prominent symptoms presented by patients are

12

chronic and productive cough, low grade fever, night sweats, fatigue and weight loss.

Extra-pulmonary TB may involve lymph nodes, kidney, bone, joints or meninges etc.

1.7: Genome and genomic diversity of Mycobacterium tuberculosis:

At the end of the 20th century complete genome sequencing of Mycobacterium

tuberculosis H37Rv revealed that it consists of 4.4 mega base pairs with densely packed

coding regions. It is estimated to include 4000 protein coding regions and has a very high

(65%) guanine and cytosine content (Cole, Brosch et al. 1998).

Comparative genome sequencing of several MTB strains such as Mycobacterium

bovis, Mycobacterium tuberculosis H37Rv and Mycobacterium tuberculosis H37Ra, have

revealed that Mycobacterium tuberculosis (MTB), the main causative agent of TB, while

displaying diverse phenotypic characteristics and host ranges, is genetically fairly

consistent with an overall genomic similarity of 99.9% throughout the world

(Boddinghaus, Rogall et al. 1990; Sreevatsan, Pan et al. 1997). Studies have shown that

MTB with highly conserved genome also has polymorphic genomic regions called

repetitive sequences. MTB genome is characterized by the presence of numerous

repeated sequences whereas no plasmid is detected in this species. Genetic diversity

among MTB strains in a particular geographical location has been linked with these

polymorphic repeat sequences (Sreevatsan, Pan et al. 1997). These repeat sequences,

which include insertion sequence (IS), direct repeat (DR) and tandem repeats, are widely

used for MTB strain typing in molecular epidemiology of tuberculosis.

13

Figure 1.2: Tuberculosis infection

(Adapted from www.nature.com/.../v5/n8/full/nri1666.html)

14

Based on these polymorphic repeat sequences various molecular techniques have

been introduced. Molecular epidemiological studies using these molecular techniques

have helped identifying predominant MTB genotype in a defined geographical location.

1.8: Molecular epidemiology of tuberculosis and its significance

Molecular epidemiology is an integration of molecular biology with epidemiology.

Recent developments in molecular biology have resulted in techniques that allow rapid

identification and tracking of specific strains of M. tuberculosis as they spread through

the population (Barnes and Cave 2003; Filliol, Driscoll et al. 2003; Garcia de Viedma

2003; Filliol, Motiwala et al. 2006). While, previous methods, such as colony

morphology, comparative growth rates, susceptibility to antibiotics, and phage typing

were useful but did not provide sufficient information regarding TB epidemiology. In the

early 1990s, different molecular methods were described to discriminate between MTB

isolates (van Embden, Cave et al. 1993; van Soolingen, de Haas et al. 1993). Most of

these techniques use DNA polymorphism based on repetitive DNA elements of M.

tuberculosis as genetic markers. Each of these methods results in strain specific genetic

profiles (fingerprints). Identical strain fingerprints are called clusters, which are usually

associated with recent transmission. While strains with unique fingerprints represent

remote transmission or infection acquired in past.

Several molecular epidemiological studies of tuberculosis have been carried out

using various polymorphic repeat sequences i.e. IS, DR and tandem repeats (Figure 1.3).

IS elements are small DNA segment that can be inserted at multiple sites. These elements

15

show high level of genetic polymorphism and widely used for studying the genetic

variability in MTB species. IS6110-RFLP typing method, based on IS6110 copy numbers

and positions, has been used worldwide as the gold standard TB epidemiology because of

reproducibility of the results (van Embden, Cave et al. 1993). However, there are certain

limitations of this technique. Firstly, around 20% of MTB isolates contains few or no

copies of IS6110 element and the method is unreliable for typing such strains. Secondly,

it needs around 4.5µg of DNA which takes several weeks to culture enough viable

organisms. In addition, the method is labor intensive, technically demanding and

expensive (van Soolingen et al, 1995).

Various PCR based techniques, which target polymorphic loci other than IS6110,

have also been used in epidemiological studies of TB. Spacer oligonucleotide typing

(spoligotyping) based on the polymorphism in DR locus is a widely used molecular

typing method for epidemiological studies worldwide (Kamerbeek, Schouls et al. 1997).

The 36 bp size DRs are interspersed by unique spacer DNA sequences of 35-41 bp.

Spoligotyping identifies the presence or absence of 43 spacer DNA sequences between

the variable direct repeats using PCR in a particular MTB strain. Spoligotyping is simple,

rapid and highly reproducible. However, Spoligotyping has less discriminatory power as

it targets less than 0.1% MTB genomic area as compared to IS6110 based typing which

examines the entire genome (Figure 1.3). Therefore, it cannot totally replace IS6110-

RFLP typing because of its lower discriminatory power, except for in strains with low

copy numbers of IS6110 (van Soolingen et al, 2001).

16

Figure 1.3: MTB genome showing polymorphic repeat sequences

Hypothetical genome of MTB strain X and polymorphic repetitive sequences such as

IS6110, DR and MIRUs for genotyping

(Reference: Adapted from Barnes, P et al, 2003)

Mycobacterial Interspersed Repetitive Units (MIRU)

Direct Repeats (DR)

Insertion Sequence 6110 (IS6110)

17

VNTR-MIRU is another PCR based genotyping method which has higher

discriminatory power than spoligotyping. MIRU-VNTR is based on detection of

independent mini satellite like loci scattered through out the MTB genome and has been

shown to be a reliable and reproducible typing method for studying the MTB population

structure in different geographical locations (Dale, Nor et al. 1999; Supply, Lesjean et al.

2001; Cowan, Mosher et al. 2002; Sola, Filliol et al. 2003). The typed strains are

expressed by a 12-digit numerical code, corresponding to the number of repeats at each

locus (Supply, Mazars et al. 2000; Mazars, Lesjean et al. 2001). This numerical code is

easy to compare and exchange at inter-, and intra-laboratory level. The discriminatory

power of MIRU-VNTR analysis is proportional to the number of loci evaluated. Twelve

loci based MIRU-VNTR analysis has been used in a number of molecular epidemiologic

studies and to elucidate the phylogenetic relationship of clinical isolates (Sola, Filliol et al.

2003; Supply, Warren et al. 2003; Sun, Lee et al. 2004; Warren, Victor et al. 2004;

Kremer, Au et al. 2005). The discriminatory power of standard twelve loci based typing

only slightly lower than that of the IS6110-RFLP, which is currently the gold standard for

MTB genotyping (Supply, Lesjean et al. 2001). However MIRU-VNTR has several

advantages over gold standard IS6110-RFLP method; it requires little culture growth for

DNA and provides comparable, numerical data by using standard gel electrophoresis or

by automated analysis using fluorescence tagged PCR primers and sequencer. MIRU

typing is also a method of choice for MTB strains with 0-5 copies of IS6110 element, as

have been reported from south Asian countries (Mazars, Lesjean et al. 2001).

18

Using these molecular tools molecular epidemiological studies have identified

several MTB strains so far including Beijing family (van Soolingen, Qian et al. 1995),

Haarlem family (Kremer, van Soolingen et al. 1999), Delhi (Bhanu, van Soolingen et al.

2002), the Cameroon family (Niobe-Eyangoh, Kuaban et al. 2004), the Latin American

Mediterranean(LAM) family, the East African Indian (EAI) clade (Sola, Filliol et al.

2001; Filliol, Driscoll et al. 2002; Filliol, Driscoll et al. 2003; Brudey, Driscoll et al.

2006) etc.

Molecular epidemiological studies have also greatly facilitated identification of

TB transmission locally as well as globally. They have helped investigate whether

particular clinical strains of TB differ in infectivity, severity of disease or drug

susceptibility and to differentiate between recent and or previous infection.

Molecular epidemiological investigations to study transmission dynamics by

exploring genetic diversity of MTB strains are of prime importance. These help to

evaluate the patterns and dynamics of TB transmission in a defined geographical location.

When this technique is applied similar MTB strains can be identified, which can lead to

important clues about the pattern and dynamics of transmission. Two of the earliest such

studies conducted in the United States, one in San Francisco and one in New York City,

investigated MTB isolates with matching DNA fingerprints with the assumption that they

were epidemiologically related and represented recent transmission of TB among the

patients (Small, Hopewell et al. 1994). This finding led the authors to conclude that as

many as 40% of TB cases in these two cities were the result of recent transmission and

that TB control practices in San Francisco and New York were not effectively decreasing

19

MTB transmission. Thus these data were used to implement effective interventions to

stop such outbreaks.

Molecular epidemiological studies have also assisted in unraveling the differences

in virulence and drug susceptibility related to bacterial genetic background (Silver, Li et

al. 1998; Valway 1998; Caminero 2001; F.Barnes 2003). Such as Beijing strain (W

strain) have shown higher growth rate in human macrophages as compared to other MTB

isolates (Zhang M 1999). In addition Beijing strains have also found to have a greater

association with drug resistance and have also been found to be less protective against

BCG vaccination in infection as compared to other MTB isolates (H.McShane 2003). A

clinical strain CDC1551 has been shown to induce a stronger immune response in

experimental animals as compared the other isolates (Menca C 1999; Zhang M 1999).

Molecular epidemiological studies help determining genotype specific drug

resistance mutations and prevalence of these mutations in a population (Mathema,

Kurepina et al. 2006). Studies have been conducted to describe clonal expansion of drug

resistant MTB strains between patients in a defined geographical location (Bifani 2001,

Bifani1996, Post FA 2004). Such studies help understanding the relationship between

genetic background, drug resistance and transmission of TB in a population and might

assist in taking measures to prevent the spread of resistant TB strains.

20

Among its applications, molecular epidemiology has also served to elucidate the

poorly understood role of relapses and exogenous reinfection of persons with recurrent

TB after cure (Small, Shafer et al. 1993). In addition the technology has also been applied

as a complementary tool to conventional methods in TB laboratory to investigate and

control cross-contamination.

Thus, to summarize, molecular epidemiology has greatly helped in the analysis of

disease transmission and its prevention. This information plays a pivotal role in

investigating the relationship of a specific genotype with phenotypic characters such as

disease causation and development of antimicrobial resistance.

1.9: Molecular epidemiological data from Pakistan

There is increasing evidence that the genetic difference of MTB is strongly

associated with specific geographical locations (Yang, Barnes et al. 1998; Sola, Devallois

et al. 1999; Soini, Pan et al. 2000; Mazars, Lesjean et al. 2001; Filliol, Driscoll et al.

2003; Hirsh, Tsolaki et al. 2004; Mokrousov, Narvskaya et al. 2004; Gutierrez, Ahmed et

al. 2006). Thus molecular epidemiological studies in Pakistan, a high TB incidence

country may provide unique insights into dissemination dynamics and virulence of the

pathogen. However limited molecular typing data is available from the country. Although

limited, this molecular data shows the predominance of some specific strain types in

Pakistan. A study based on samples collected from Rawalpindi revealed 37% of 113

MTB isolates investigated showed common profile based on five loci based VNTR (A to

21

E exact tandem repeat) analysis. On the other hand a study from Peshawar using IS6110

based typing reported that 89% (n=8/9 ) of isolates were of a common strain type (Sechi,

Zanetti et al. 1996; Gascoyne-Binzi, Barlow et al. 2002). Limitation of these studies is

small sample size and use of only one genotyping technique. In addition the strain types

identified in these studies were not matched with any global database.

Recently spoligotyping was used to investigate strain types amongst 314 MTB

isolates from the country. After comparing the results with international database

(SpolDB3) CAS1 strains lacking spacers 4-7 and 23-34 were found to be the most

prevalent (39%) in Pakistan (Hasan, Tanveer et al. 2006). The CAS1 strain has also been

reported as the second most predominant group in South Asia; India (16-22%) and

Bangladesh (17%) (Bhanu, van Soolingen et al. 2002; Banu, Gordon et al. 2004; Deepak

2005; Hasan, Tanveer et al. 2006). Although Beijing strains, lacking spacers 1-34, are the

most widely reported genotype world wide (Agerton T. B 1999; Bifani 1999; Glynn,

Whiteley et al. 2002; Li, Whalen et al. 2002; Mokrousov, Narvskaya et al. 2002; Reed,

Domenech et al. 2004), their prevalence amongst MTB isolates from Pakistan was found

to be only 6% (Hasan, Tanveer et al. 2006).

Despite the predominance of CAS1 in South Asia, there is limited data related to

its transmission dynamics, disease process and drug resistance. MIRU-VNTR typing

method, as discussed earlier, with higher discriminatory power has been used in several

studies to elucidate intra-strain genetic differences and phylogenetic relationship of

clinical isolates (Sola, Filliol et al. 2003; Supply, Warren et al. 2003; Sun, Lee et al.

2004; Warren, Victor et al. 2004; Kremer, Au et al. 2005). Twelve loci based MIRU-

22

VNTR analysis has been used in a number of molecular epidemiologic studies. It has also

been used to study Beijing strains from East and South Asia (Dale, Nor et al. 1999; Anh,

Borgdorff et al. 2000; Chan, Borgdorff et al. 2001; Prodinger, Bunyaratvej et al. 2001;

Mokrousov, Narvskaya et al. 2004; Chin, Chiu et al. 2007). Available data for MIRU-

VNTR typing for MTB in Pakistan is limited to one report wherein five exact tandem

repeat (ETR) were used to type 113 MTB isolates from Rawalpindi, Pakistan. This

showed clustering of one third of the isolates (Gascoyne-Binzi, Barlow et al. 2002).

However pertaining to CAS1, the predominant strain type in Pakistan no MIRU-VNTR

data is available.

1.10: Antituberculous therapy

Until middle of the 20th century there was no definitive treatment available for TB.

Streptomycin (S) was made available for the TB treatment. Then other drugs including

isoniazid (INH), rifampicin (RMP), pyrazinamide (PYZ) and ethambutol (E) were also

used for antituberculous therapy (ATT). Standard ATT for pulmonary tuberculosis

comprises a six month regimen. For the first two months patients receive a combination

of three to four of these drugs i.e. isoniazid, rifampicin, pyrazinamide and in some cases

ethambutol. Final four months patients continue with isoniazid and rifampicin. This six

months therapy is called “short-course” anti-tuberculous therapy (Blomberg, Spinaci et al.

2001; Moulding, Le et al. 2004). But longer courses of up to two years may be needed

for persons having TB with resistant strains or have TB-HIV co- infection and makes it

100 times as expensive as the first line regimen.

23

In early 1990s World Health Organization introduced directly observed treatment

short course (DOTS) strategy as a cost effective way to control TB (Pfyffer, Welscher et

al. 1997; Sawert, Kongsin et al. 1997; Dye, Garnett et al. 1998). DOTS strategy includes

case detection, standard short-course therapy (with three drugs out of five), regular drug

supply and monitoring and evaluating the program. DOTS has been regarded as the

highly successful and most cost effective intervention for the treatment of drug

susceptible TB control (Burman, Dalton et al. 1997; Chaulk, Friedman et al. 2000;

Chaulk and Grady 2000). However despite of high success rate, DOTS has been found

insufficient for the management of MDR-TB (Becerra, Freeman et al. 2000; Espinal, Kim

et al. 2000). Thus DOTS-Plus has been introduced which takes into account specific

issues such as drug susceptibility and careful use of second-line antituberculous drugs in

order to combat global threat of MDR-TB (Farmer and Kim 1998). In addition

implementation of DOTS-Plus has been shown extremely important to contain super

resistant extensively drug resistant (XDR) MTB strains (Kam and Yip 2004; Laserson,

Thorpe et al. 2005; Caminero 2006). XDR-TB is the disease caused by bacteria that are

resistant to any fluoroquinolones and resistant to at least one second line injectable drug

(amikacin, capreomycin or kanamycin) (Dukes Hamilton, Sterling et al. 2007). The

global emergence of XDR-TB strains poses an additional threat to the success of

tuberculosis therapy. XDR-TB has been reported from throughout the world with highest

known rates in Eastern Europe and Asia (Dukes Hamilton, Sterling et al. 2007). Since

24

XDR-TB are resistant to all the drugs considered essential in the treatment of MDR-TB

cases, the XDR-TB might be almost incurable (Gandhi, Moll et al. 2006).

1.11: Drug-resistant tuberculosis

The resistance of MTB strains to anti-tuberculosis drugs was noted when

streptomycin (S) was first used as monotherapy for TB in the 1940s. In the subsequent

years with the addition of RMP, PYZ and E, multiple antituberculous therapy was

implemented to combat emergence of single drug resistance amongst MTB strains

(Mitchison 1985; Iseman and Madsen 1989).

Drug resistance is divided into two types: primary and secondary (or acquired)

resistance. Primary resistance is defined as resistance in persons who have not received

anti-tuberculosis drugs for more than 1 month. These patients are presumed to be initially

infected with drug-resistant strains. Acquired resistance is defined as resistance to anti-

tuberculosis drugs, which arises during treatment due to poor compliance or improper

management. Adult patients can be infected with primary drug-resistant strain or acquire

resistance to anti-tuberculosis drugs during the treatment. Usually, children have primary

resistance, as they get infected from adult source with drug-resistant TB (Guidelines

1998; Pablos-Mendez, Raviglione et al. 1998).

WHO and International Union against Tuberculosis and Lung Diseases

(IUATLD) recommends using the terms drug resistance among new cases and drug

25

resistance among previously treated cases. Drug resistance among new cases (formerly:

primary drug resistance) is the presence of drug-resistant strain of M. tuberculosis in a

newly diagnosed patient who has never received anti-tuberculosis drugs or has received

them for less than 1 month. Drug resistance among previously treated cases is that found

in a patient who has previously received at least 1 month therapy with anti-tuberculosis

drugs(WHO 2003).

1.12: Multi-Drug resistant tuberculosis (MDR-TB)

Multiple-drug resistant tuberculosis (MDR-TB) is defined as simultaneous in-

vitro resistance of MTB strains to at least RMP and INH (with or without resistance to

other drugs). As RMP and INH are the most potent first line antituberculous drugs, the

emergence of MDR-TB can give rise to potentially untreatable form of the disease.

MDR-TB treatment require use of second-line drugs (SLDs) that are less effective, more

toxic and more costly than the first line based treatment (Cole 1994; Cole and Telenti

1995). In addition mortality is significantly higher among persons infected with MDR

strain than of those infected with sensitive strain. Patients with multi-drug resistant TB

remain infectious for longer time increasing the risk of disease transmission

(Drobniewski and Wilson 1998; Rattan, Kalia et al. 1998).

Despite of implementation of multiple antituberculous therapies, a steady increase

in the frequency of TB with single and multiple drug resistant MTB strains has been

reported through out the world (Bloch, Cauthen et al. 1994). In early 1990s outbreaks of

26

MDR-TB received global attention (Edlin, Tokars et al. 1992). Nosocomial outbreaks of

MDR-TB have been reported in the USA, France and other countries (CDC 1991; CDC

1991).

A recent WHO survey for the period of 2002-2006 report median prevalence of

MDR as high as 22.3% amongst new tuberculosis cases in Kazakhstan, in contrast to

14.2% in an earlier report (Ang, Ong et al. 2008). The survey also found alarmingly high

rates in other countries such as Ukraine (16% percent), Russia (15 %), and Uzbekistan

(14.8%). Other high MDR-TB areas include Estonia (12%), Lithuania and Latvia (9%),

Russian Federation and China (10%). In Iran and India MDR-TB has been reported at 5%

and 3.4% respectively (WHO 2008). India and China the two most populous countries

are major concerns for MDR-TB as these two countries account for 40% of all TB cases

worldwide. Global occurrence of MDR-TB is about 3% (Espinal, Laszlo et al. 2001;

Sharma and Mohan 2004; Jou, Chen et al. 2005), ranging from 0% in various states of

America to 22.3% in Kazakhstan (Espinal, Laszlo et al. 2001; WHO 2008). Prevalence

of MDR-TB among new and previously treated TB cases is 1.1% and 7% respectively

(Dukes Hamilton, Sterling et al. 2007). 75% of these MDR-TB cases occur in Asia

(Espinal, Laszlo et al. 2001). The global prevalence of MDR-TB amongst new and

previously treated cases has been summarized in Table 1.1.

27

Table 1.1: Median prevalence of MDR-TB amongst new and previously treated cases by region (%)

Region

MDR-TB amongst

new cases

MDR-TB amongst

previously treated cases

Africa 1.4 5.9

America 1.1 7.0

Eastern Mediterranean 0.4 48.3

Europe 0.9 4.7

South East Asia 1.3 20.4

Western pacific 0.9 15.5

Overall median 1.1 7.0

(Adapted from WHO report 2008)

28

Table: 1.2: MTB genes associated with resistance to antituberculous agents

Antituberculous

agents

Gene

Size (bp)

Product

Mutation frequency

among resistant MTB isolates (%)

Rifampicin rpoB 3,534 subunit of RNA polymerase

> 95

Isoniazid katG oxyR-ahpC

inhA

kasA

2,205 585 810

1,251

Catalase-peroxidase Alkylhydroreductase

Enoyl-ACP reductase

-Ketoacyl- ACP reductase

60-70 20 <10

<10

Streptomycin rpsL rrs

372 1,464

Ribosomal proteinS12 16s rRNA

60 <10

Ethambutol embCAB 1,164 Arabinosyltransferases

70

Pyrazinamide pncA 560 Amidase 70-100 Ethionamide inhA

ethA

810

1467

Enoyl-ACP reductase

Flavoprotein monooxygenase

<10

NA*

Kanamycin rrs 1,464 16s rRNA 65 Fluoroquinolone gyrA

gyrB

2,517

2,142

DNA gyrase subunit

DNA gyrase subunit

>90

NA

*data not available

(Adapted from (Musser 1995; Mathema, Kurepina et al. 2006)

29

1.13: MDR-TB in Pakistan

The high incidence of tuberculosis in Pakistan is also compounded by the

increasing emergence of drug resistant strains including MDR. Community based data

from country reports 1.8% prevalence of MDR-TB amongst untreated patients while

laboratory based data reports 4% of MDR-TB in untreated patients (Javaid, Hasan et al.

2008; Rao, Irfan et al. 2008). However WHO estimates 3.4% and 36% MDR-TB in new

TB patients and in previously treated patients respectively, as mentioned earlier. The

reasons for high rate of drug resistance include improper prescription, non-compliance

and over the counter sale of anti-TB drugs (WHO 2008).

Laboratory based studies from urban Rawalpindi showed an increasing frequency

of MDR from 14% in 1999 to 28% in 2004 (Butt, Ahmad et al. 2004) while study from a

tertiary care center in Karachi documented 47% MDR-TB prevalence (Irfan, Hassan et al.

2006).

1.14: Molecular basis of drug resistance

The mechanisms of drug resistance are chromosomal, caused by accumulation of

one or more mutations in independent genes. Accumulation of a number of drug

resistance mutations result in multiple drug resistance (Sreevatsan, Pan et al. 1997). Such

as resistance to RMP is well characterized and more than 95% of the RMP resistant have

mutations in an 81bp hot spot region (codon 507-533), or Rif Resistance Determining

30

Region (RRDR) of the 3534bp rpoB gene (Telenti, Imboden et al. 1993; Musser 1995).

This is the strongest correlation between phenotypic and genotypic resistance in MTB

discovered so far. This rpoB gene encodes the β subunit of RNA polymerase. RMP

interferes with the transcription and elongation of the RNA by binding to the DNA

dependent RNA polymerase (Telenti, Imboden et al. 1993; Telenti 1997).

In contrast to RMP, INH resistance is controlled by a more complex genetic

system, involving several genes such as katG, inhA, kasA, oxyR and ahpC (Zhang, Heym

et al. 1992; Banerjee, Dubnau et al. 1994; Kelley, Rouse et al. 1997). Although the

frequency of mutation at these loci varies between different population, studies show that

70-80% of INH resistance is mostly associated with mutation in codon 315 of katG and

inhA genes (Zhang, Heym et al. 1992; Banerjee, Dubnau et al. 1994). Similarly S, PYZ, E

and Fluoroquinolones resistance have been linked with mutations in rrs, pncA, embB and

gyrA genes respectively (Table 1.2).

1.15: Drug susceptibility testing

Detection of drug resistance is performed by culturing MTB in the presence of

antibiotics. These methods have been performed on egg-based or agar-based solid media

directly or indirectly. For the direct method, media are inoculated with decontaminated

and concentrated clinical specimen, while for indirect method; media are inoculated with

a bacterial suspension of the isolated strains. Based on solid media, there are three

31

conventional phenotypic methods for drug susceptibility testing: the proportion method,

the resistance ratio method and the absolute concentration method (Canetti, Fox et al.

1969). More recent methods are based on liquid media including the BACTEC

radiometric and the Mycobacterial Growth Indicator Tube methods (MGIT) (Pfyffer,

Welscher et al. 1997). However, due to the long time period necessary to obtain results

and laboriousness of these methods molecular approaches have been proposed

Molecular methods are based on the genetic determinants of drug resistance rather

than phenotypic resistance. These including DNA sequencing, real time PCR,

microarrays and kit based line probe assays has enabled detection of drug resistance in

MTB from several weeks to a few days (Torres 2002; Hillemann, Rusch-Gerdes et al.

2007). These methods include amplification of targeted gene segment correlating with

drug resistance by polymerase chain reaction (PCR), sequencing, solid-phase

hybridization, real-time PCR assay or microarray technique (Garcia de Viedma 2003).

Molecular methods have a pivotal role in identifying prevalent drug resistance

mutations amongst MTB population in a particular geographical location. Molecular

tools and genetic determinants of drug resistance has made timely diagnosis of MTB drug

resistance and adequate antituberculous therapy possible (Torres 2002; Hillemann,

Rusch-Gerdes et al. 2007). Although studies have shown both Central Asian strain1

(CAS1) and Beijing family strains as the predominant genogroups (39% and 6%

respectively) in Pakistan and in other South Asian countries there is limited data available

32

pertaining to type and frequency of drug resistance gene mutations amongst these

predominant genogroups (Bhanu, van Soolingen et al. 2002; Glynn, Whiteley et al. 2002;

Banu, Gordon et al. 2004; Almeida, Rodrigues et al. 2005; Hasan, Tanveer et al. 2006). It

is important to understand and study the drug resistance mutations in CAS1 and Beijing,

strains to develop tests for rapid detection of resistance among TB strains from Pakistan.

1.16: Goals of present study

Molecular epidemiological studies, based on the assumption that patients infected

with clustered strains are epidemiologically linked, have helped understanding the

transmission dynamics of disease. Appropriate genotyping tools are a prerequisite for

performing molecular epidemiological studies of MTB strains. Although, these tools are

well established in developed countries, establishment and use of only spoligotyping

technique has been shown in Pakistan (Hasan, Tanveer et al. 2006). However, given the

fact that spoligotyping is limited in its ability to discriminate, additional molecular

methods are required to be established for further analysis of genetic diversity of

predominant MTB genogroups in the country.

Hasan et al have shown significant association of Beijing genotype MTB strains

with MDR (Hasan, Tanveer et al. 2006). Although the most prevalent CAS1 strains have

not shown significant association with drug resistance, given the burden of CAS1 they

constitute almost fifty percent of the total MDR strains in the country.

33

While mutation in MDR Beijing strain is characterized in studies from different

geographical locations (Suresh, Singh et al. 2006; Lipin, Stepanshina et al. 2007), there is

no data on drug resistance genes mutations amongst CAS1 strains. Knowledge of type

and frequency of mutations amongst MDR strains of prevalent genogroup within a

defined geographical location is essential for the development of tools for early diagnosis

and control of MDR-TB strains (Ahmad, Araj et al. 2000; Lipin, Stepanshina et al. 2007),

no such data is available from the country. Thus there is a need to investigate mutations

among MDR CAS1 and Beijing strains from the country.

Therefore the hypotheses for this study were:

1. The prevalent MTB genogroup in the Country has intra strain

genetic diversity.

2. The prevalent drug resistant MTB strains have some common

mutations in the drug resistance genes

Thus the overall aim of this study was to explore genetic diversity amongst the

predominant genogroup of Mycobacterium tuberculosis from the country as well as to

investigate genetic basis of drug resistance amongst these predominant MTB strains.

Specific goals of this study were:

1. To contribute to the development and establishment of additional

molecular typing tools for MTB

2. To study genetic differences amongst identified prevalent genogroup,

CAS1.

34

3. To investigate the genetic basis of drug resistant (MDR) MTB strains

prevalent in our community (CAS1 and Beijing).

Information from this study would help contribute towards an understanding of

molecular epidemiology of MTB in our community. It would enable an analysis and

understanding of strain types prevalent locally. In particular it would greatly

contribute to analyzing factors leading to drug resistance at a molecular level. This

knowledge would assist to  take  appropriate  measures  for the prevention and

adequate treatment of the tuberculosis disease including MDR. 

35

Chapter Two

Genotyping of Mycobacterium tuberculosis using IS6110- Restriction Fragment Length Polymorphism

2.1: Background

Insertion sequences (IS) are small mobile genetic elements which are widely

distributed throughout the MTB genome. Over 14 different kinds of IS have been

identified in the MTB genome which are usually less than 2.5 kb in size (van Soolingen,

de Haas et al. 2000). IS generate genetic polymorphism and therefore are used to

discriminate between different MTB strains. The most widely utilized IS in

epidemiological studies is IS6110. In 1990 Thierry identified IS6110 element in MTB

Chapter preview Page # 2.1: Background 35 2.2: Objectives 38 2.3: Methods

2.3.1: Mycobacterial strains 39 2.3.2: Culture and antibiotic susceptibility 39 2.3.3: DNA extraction 41 2.3.4: IS6110- RFLP typing 42 2.3.5: Computer-assisted phylogenetic analysis 47 2.3.6: Statistical analysis 47

2.4: Results 2.2.1: Diversity of RFLP 48

2.2.2: IS6110-copy number 48 2.5: Discussions 51

36

strains (Thierry, Cave et al. 1990). IS6110 is 1,355 bp in size and has 28 bp inverted

repeats at its ends. It is randomly distributed throughout the MTB genome with copy

numbers ranging between 0-26 (Kurepina, Sreevatsan et al. 1998; McHugh and Gillespie

1998). In 1993, van Embden and colleagues proposed a standardized Southern blot

hybridization method based on the frequency of IS6110 in the MTB genome which

allows strain differentiation.

Restriction fragment length polymorphism (RFLP) using insertion sequence

IS6110 was recommended by van Embden and colleagues for MTB genotyping (van

Embden, Cave et al. 1993). The method is based on the detection of differences in the

numbers and locations of the insertion element IS6110 within the strains of MTB

chromosomes.

Despite certain limitations, such as the need for large amounts of DNA, labor-

intensivity and its low discriminatory ability in cases of strains with few copy of IS6110

element (please see chapter one, section 1.8), IS6110-RFLP typing has played a

significant role in understanding the transmission patterns, source, and spread of MTB

strains (Gopaul, Brown et al. 2006; Mathema, Kurepina et al. 2006). Limited IS6110-

RFLP typing data for MTB strains from Pakistan is available (Sechi, Zanetti et al. 1996;

Dale, Al-Ghusein et al. 2003). The study reports IS6110-RFLP typing of nine MTB

isolates from Peshawar. The results showed that 89% of isolates tested had a common

strain type.

37

Following a preliminary spoligotyping report that described CAS1 strains to be

the most prevalent (39%) genogroup in Pakistan (Hasan, Tanveer et al. 2006), the present

investigation was focused on characterizing CAS1 strains using IS6110-RFLP typing

method. Although spoligotyping is a simple and rapid genotyping technique for MTB, it

is less informative for determining genetic diversity of predominant strains in the high

endemic TB areas like Pakistan (Mathema, Kurepina et al. 2006). IS6110-RFLP is a gold

standard genotyping technique for MTB, with high discriminatory power, and ability to

reveal genetic diversity within predominant strains, such as CAS1, identified on the basis

of spoligotyping.

38

2.2: Objectives

The specific objectives of this study were:

2.2.1: To establish IS6110-RFLP genotyping technique 2.2.2: To identify IS6110-RFLP profile to further characterize and analyze selected CAS1 strains in comparison with ‘orphan’ types strains

2.2.3: To identify utility of IS6110 typing in MTB isolates from Pakistan

2.2.4: To use Bio-numeric software program for the fingerprint analysis

.

39

2.3: Methods

2.3.1: Mycobacterial strains

926 previously spoligotyped MTB strains which were collected between

2003-2005 from across the country were stored within our strain bank (Hasan,

Tanveer et al. 2006; Tanveer, Hasan et al. 2008). 78 isolates (43 CAS1 and 35

orphan types) were randomly selected from 314 spoligotyped strains (Table 2.1)

for characterization with IS6110-RFLP (Hasan, Tanveer et al. 2006). During the

optimization phase of establishment of RFLP, results from MTB DNA grown on

Lowenstein Johnson (LJ) medium was compared with results from MTB DNA

grown on Middlebrook medium. In our experience MTB DNA from LJ gave

better RFLP results. In this study, therefore we used DNA from strains grown on

LJ medium.

M. tuberculosis reference strain 14323 (obtain on request from van

Embden, National Institute of Public Health and the Environment, the

Netherlands) was used as external control and SDL-PvuII marker as internal

control.

2.3.2: Culture and antibiotic susceptibility

All 78 selected strains were cultured on LJ. Drug susceptibility data for all

78 strains was collected from the Clinical Laboratory of Aga Khan University

Hospital, where phenotypic drug resistance testing was performed using the

standard agar proportion method on enriched Middlebrook 7H10 medium (Wayne

and Krasnow 1966; Standards. 1995; Isenberg 2004).

40

Table 2.1: Geographical Distribution of selected MTB strains for RFLP

Location

CAS1 (n: 43)

Unique (n: 35)

Total (n: 78)

Karachi 11 10 21

Sind (other than Karachi)

12 10 22

Punjab 12 11 23

NWFP 7 3 10

Baluchistan 1 1 2

(Reference: (Hasan, Tanveer et al. 2006))

41

The following final drug concentrations were used: rifampicin, 1µg/ml

and 5µg/ml; isoniazid, 0.2µg/ml and 1µg/ml; streptomycin, 2µg/ml and 10µg/ml;

ethambutol 5µg/ml and 10µg/ml. Pyrazinamide was tested using BACTEC (7H12

medium, pH 6.0, at 100µg/ml, Becton Dickinson) as per manufacturer’s

instructions. Strains with a high level of resistance for rifampicin (5µg/ml) and

isoniazid (1µg/ml) were further selected for MDR analysis.

2.3.3: DNA extraction

Genomic DNA was extracted from the MTB strains cultured on LJ and 7H10

agar medium, via cetyl-trimethylamonium bromide (CTAB) method after heat

inactivating the samples at 85 oC for 45 minutes (Honore-Bouakline, Vincensini

et al. 2003). 50 μl lysozyme (10 mg/ml) was added and incubated at 37 oC for

over night. After that 60 μl of 10% SDS and 15 μl Proteinase K (10 mg/ml) was

added and vortexed for few seconds and incubated at 65 oC in a water bath for an

hour. Then 100 μl of CTAB of and100 μl 5M NaCl was added and vortexed. till

all the samples became milky. Then it was incubated at 65 oC for an hour. 750 μl

of Chloroform and Isoamyl (24:1) was added, vortexed for 30 second and then

centrifuged at 14000 rpm at room temperature for 15 min for phase separation.

The aqueous phase was separated in another autoclaved eppendorf and 450 μl

isopropanol was added to precipitate the nucleic acid. The samples were placed at

-20o C for over night. Next day samples were centrifuged at 14000 rpm for 15

minutes. Then supernatant was decanted without loosing the DNA pellet. 1 ml of

42

70% ice cold Ethanol was added and centrifuged at 12000 g for 5 minutes.

Supernatant was discarded and pellet was dried and then reconstituted in 50 -

100μl of 1X TAE. Optical density of DNA was taken and maintained in the

record.

2.3.4: IS6110- RFLP typing method (Standard van Embden method)

The established protocol, used by National Institute of Public

Health and the Environment, the Netherlands, (van Embden, Cave et al.

1993) was followed to type seventy-eight clinical MTB strains DNA,

extracted from MTB culture grown on LJ, was digested with PvuII and

Southern blots were hybridized with IS6110 probe. Blots were normalized

with the standard M. tuberculosis reference strain 14323 (figure 2.1). Brief

outline of the method has been given below.

IS6110 probe synthesis

IS6110-probe (based on 245 bp region of digested IS6110 element)

was prepared by PCR technique, using DNA template of BCG strain (P3)

of Mycobacterium tuberculosis (which was received on request from

National Institute of Public Health and the Environment the Netherlands)

with 50-ng/l of the following primers:

INS-1 (5’ CGT GAG GGC ATC GAG GTG GC)

INS-2 ( 5’GCG TAG GCG TCG GTG ACA AA)

43

Figure 2.1: Schematic diagram of steps of IS6110-RFLP method

(Adapted from Daley, 2005)

44

For PCR one cycle of 3 minutes at 95C, 40 cycles consisting;

1minute at 95C and 1 minute at 62C, 1 cycle of 2 minutes at 72C and

finally 10 minutes extension at 72C. PCR product was then run on 1%

agarose gel to check the size and purity of amplified product.

PCR product was purified using spin column Qiagen PCR Product

Purification kit as per manufacturer’s instructions. The purified IS6110

probe was then labelledwith Horseradish Peroxidase using ECL

(Enhanced Chemiluminesence, Amersham) labeling kit as per the

instructions provided by the manufacture.

PvuII restriction digestion of the sample DNA

Restriction digestion was carried out for 2 - 4.5 g of DNA sample

of each of our strain as well as control strains P3 and MTB 14323 with

PvuII enzyme. The digested DNA was run on a small 0.8% agarose gel in

order to confirm the digestion.

Gel electrophoresis

The digested DNA of each sample as well as control MTB 14323

strains were run on 20 x20 cm size, 0.8% agarose gel on 20V over night

with SDL-PvuII Internal Marker (which was prepared using Supercoiled

DNA ladder, digested with PvuII enzyme, and PhiX174 as per standard

protocol).

45

The gel was then visualized using UV transilluminator to check the

migration of the band of the external marker Mt14323. then the gel was

placed on an UV transilluminator (Biometrica Whatman Co., Model

1003464) until the fluorescence of the ethidium bromide fades completely.

Southern blotting

The blots from the gel were transferred on nitrocellulose

membrane using vacuum blotter.

Probe hybridization

The blots on nitrocellulose membrane were hybridized with

Enhaced Chemiluminescent (ECL) labeled IS6110 probe as well as with

Internal Marker probe over night in the rolling bottle placed in

hybridization chamber at 42C.

Blot development

Blots on the membrane for IS6110 and Internal Marker were

developed on the X-ray film using Amersham ECL detection system

(Figure 2.2).

46

2.2-A

2.2-B

Figure 2.2: Autoradiograph after the hybridization of MTB strains with Internal Marker probe (A) and IS6110 probe (B)

47

2.3.5: Computer-assisted phylogenetic analysis of fingerprints

The autoradiographs of IS6110-RFLP were scanned using Scanjet 3400-

hp scanner. Bionumerics Software (Applied Maths) was used to analyze the

molecular patterns generated by IS6110-RFLP. This experiment was repeated at

least in duplicates for all 78 strains. Scanned images of IS6110 blots were

uploaded in Bionumerics program. Then the dendrogram was generated by

unweighted pair group method analysis (UPGMA). Strains were classified in a

cluster when they shared hundred percent identical IS6110-RFLP patterns.

2.3.6: Statistical analysis

Association of number of IS6110 element and strain type was determined

by chi-square using version 16 of SPSS (Special Program for Social Sciences

Software, USA). A P value of < 0.05 was considered significant.

48

2.4: Results

2.4.1: Diversity of RFLP

IS6110-RFLP typing of 78 strains (43 CAS1 and 35 ‘unique’ spoligotyped

strains) resulted in 73 different RFLP types (Figure 2.2). One cluster of two

unique strains, with single copy of IS6110 was identified; the remaining seventy-

two strains revealed unique RFLP patterns.

2.4.2: IS6110-copy number

The copy number of IS6110 in each of the isolates was estimated as

determination of exact copy number was difficult in some of the strains due to

band intensity. Of 78 strains, 4 (5%) strains had ‘zero’ copy, 6 (7.7%) had low

(1-5) copy, 9(11.5%) had medium (6-10) copy and 59(75.6%) had high (11-17)

copy of IS6110 element. Two out of four zero copy strains belonged to CAS1

family while all the low copy isolates belonged to unique strain type (Table 2.2).

Occurrence of low copy IS6110 element (0-5) was significantly associated (P <

0.05) with unique strains.

Of 68 high copy IS6110 strains (>5 copies), 60% belonged to CAS1 strain

type. There was an average of 12.8 IS6110 copies in CAS1 strain, and 9.2 in

unique strains. While CAS1 strains were not clustered a 60% homology was

observed amongst their IS6110 profile. In addition there was no similarity

between IS6110 profiles of MDR strains.

49

Figure 2.3: Dendrogram of IS6110-RFLP typing of Mycobacterium tuberculosis

The figure illustrates analysis of IS6110-RFLP of 78 MTB strains using Bionumerics

software (Applied Maths). The strains included 43 CAS1 and 35 ‘unique’ exhibiting

heterogeneous IS6110-RFLP profiles with one cluster of two strains with one copy

IS6110 while four strains are IS6110 deficient. * Denotes CAS1 strains exhibiting 60%

homology despite of heterogeneous IS6110-RFLP profiles.

* *

*

*

*

4 Zero copy strains

50

Table 2.2: Overview of number of IS6110 element present in 78 MTB Isolates

# of

IS6110 element

# of Strains TOTAL

PercentageCAS1 (43) Unique (35)

MDR S* MDR S

0

2

-

-

2

4

5 %

1

-

-

2

3

5

6.4 %

2-5

-

-

1

-

1

1.2 %

6-10

2

-

2

5

9

12 %

11-15

7

24

2

18

51

65.3 %

16-17

3

5

-

-

8

10 %

14

29

7

28

78

* S = Susceptible MTB strain 0-5 band = 10/78 (13%) 11-17 band = 59/78 (76%)

51

2.5: Discussions

IS6110-RFLP analysis revealed diverse profiles of MTB strains isolated in

Pakistan. RFLP analysis of our strains showed only one cluster of two MTB strains

belonging to unique strain type while none of the CAS1 strains were clustered. This

finding from this study is in agreement with previous studies, which have shown an

association between MTB strain diversity and high TB incidence. (Sahadevan, Narayanan

et al. 1995; Vukovic, Rusch-Gerdes et al. 2003). Strain diversity has also been reported

from neighboring countries; India, Iran and Bangladesh (Farnia, Mohammadi et al. 2004;

Storla, Rahim et al. 2006; Chauhan, Sharma et al. 2007).

Overall 87% of MTB strains showed occurrence of more than six copies and 13%

of MTB strains had 0-5 copies of IS6110 element tested in this study. 41/43 CAS1strains

contained more than six IS6110 copies. This data is comparable with data reported in

other studies from subcontinent (Das, Narayanan et al. 2005; Chauhan, Sharma et al.

2007; Mathuria, Sharma et al. 2008). Two zero copy IS6110 element CAS1strains have

been reported first time in this study (which were ensured by repeating the RFLP

experiment at least thrice for these strains). The high genetic polymorphism in the region

may be due to enhanced genetic variation and transposition of IS6110 element, which

results in variable profiles (Sreevatsan 1997; Brosch 2002).

Overall this study shows that IS6110-RFLP could effectively be used for the

molecular epidemiological studies to characterize majority of MTB isolates from the

country. However despite its use IS6110-RFLP has some practical limitations; in addition

52

to the need of 4.5µg of DNA, the technique is expensive. It is moreover difficult to

characterize large number of MTB isolates using RFLP due to the fact that RFLP testing

is labour intensive. Finally comparison of RFLP base strain information with global data

bases requires expensive and sophisticated computer analysis which is not widely

available (Mathema, Kurepina et al. 2006). In view of these shortcomings an additional

simple method with comparable discriminatory power with IS6110-RFLP is required

which can be used to characterize MTB isolates in general from the country. The next

objective of the study therefore was the characterization of MTB isolates using a second

typing method i.e MIRU-VNTR.

53

Chapter Three

Genotyping of Mycobacterium tuberculosis using Mycobacterial Interspersed Repetitive Unit typing method

3.1: Background

VNTR-MIRU typing method has widely been used for genotyping of clinical

MTB strains. The typed strains are expressed by a 12-digit numerical code,

corresponding to the number of repeats at each locus (Supply, Mazars et al. 2000; Mazars,

Lesjean et al. 2001). This numerical code is easy to compare and exchange at inter-, and

Chapter preview Page # 3.1: Background 53 3.2: Objectives 55 3.3: Methods

3.3.1: Mycobacterial strains 56 3.3.2: Culture and antibiotic susceptibility 58 3.3.3: DNA extraction 58 3.3.4: MIRU typing method 58 3.3.5: Phylogenetic analysis 60 3.3.6: Statistical analysis 60

3.4: Results 3.4.1: MIRU typing for CAS1 and ‘unique’ strains 68 3.4.2: Allelic diversity 68 3.4.3: Discriminatory power of MIRU typing for CAS1 70

3.4.4: Comparison of MIRU and RFLP typing profiles 70 3.4.5: Analysis of MIRU typing of MDR isolates 71

3.5: Discussions 72

54

intra-laboratory level. Therefore a number of studies have used standard twelve loci

based MIRU-VNTR typing method to elucidate the phylogenetic relationship of clinical

isolates in molecular epidemiologic studies (Sola, Filliol et al. 2003; Supply, Warren et

al. 2003; Sun, Lee et al. 2004; Warren, Victor et al. 2004; Kremer, Au et al. 2005).

In this study we have used standard twelve loci based MIRU-VNTR typing

method to characterize CAS1 and ‘orphan’ spoligotyped MTB strains selected from

different geographical locations in Pakistan as described by Supply (Supply, Lesjean et al.

2001).

55

3.2: Objectives

The specific objectives of this part of the study were:

3.2.1: To establish VNTR-MIRU typing method in lab for MTB strains

3.2.2: To identify VNTR-MIRU profile of CAS1 and ‘orphan’ spoligotyped

strains selected from different geographical locations in Pakistan

3.2.3: To identify the most discriminatory MIRU loci for CAS1 as compared

with ‘orphan’ spoligotyped strains

3.2.4: Determine the association of MIRU loci with MDR of clinical isolates

3.2.5: Compare the MIRU-VNTR analysis of selected MTB strains with their

IS6110-RFLP profile.

56

3.3: Methods

3.3.1: Mycobacterial strains

A total of 926 strains of Mycobacterium tuberculosis were spoligotyped in

a parallel study at the Aga Khan University (Hasan, Tanveer et al. 2006; Tanveer,

Hasan et al. 2008). These MTB isolates were collected during the period of 2003–

2005 from the four provinces of Pakistan by a stratified random sampling method

(Hasan, Tanveer et al. 2006). These strains were stored at the MTB strain bank of

the Aga Khan University Hospital. From these stored strains a total of 178 CAS1

strains and 189 ‘unique’ isolates were selected (Table 3.1) for this study using

following statistical formula:

Formula: n = Npq (N-1)D + pq

Where,

N = total no. of isolates available

P = probability (as 40% of CAS1so 0.40 & 60% of Unique so 0.60)

q = probability of failure (1-p)

D = (bound on error) .05 = B2 = (.05)2 = .0006 4 4

57

Table 3.1: Geographical Distribution of selected MTB strains for MIRU typing

Location

CAS1 (n: 178)

Unique (n: 189)

Total (n: 367)

Karachi 49 80 129

Sind (other than Karachi)

33 34 67

Punjab 78 46 124

NWFP 16 25 41

Baluchistan 2 4 6

(Reference: Hasan 2006 & Tanveer 2008)

58

3.3.2: Culture and antibiotic susceptibility

Selected CAS1 and unique MTB strains were taken from the MTB strain

Bank of the Aga Khan University Hospital. Revival and culturing of the strains

were carried out in the Biosafety Level III, Juma Research laboratory of the Aga

Khan University. All mycobacterial strains were cultured on Middlebrook 7H10

agar. Susceptibility testing was performed by the standard agar proportion method

in the Clinical Laboratory of the Aga Khan University Hospital as discussed in

detail in Chapter Two. Strains with a high level of resistance for rifampicin

(5µg/ml) and isoniazid (1µg/ml) were further selected for MDR analysis.

3.3.3: DNA extraction

Genomic DNA was extracted from the MTB strains cultured on 7H10 agar via

cetyl-trimethylamonium bromide (CTAB) method similarly as described in

Chapter Two, Section 2.3.3.

3.3.4: MIRU-VNTR typing

MIRU-VNTR PCR (As described by Supply, 2001)

After extraction of DNA, PCR was performed for twelve MIRU loci (2, 4,

10, 16, 20, 23, 24, 26, 27, 31, 39 and 40) individually for all 367 isolates using

specific primers (Table 2.1) as described previously (Supply, Lesjean et al. 2001).

59

Each of the PCR master mixes contained 0.4µM concentration of specific primers,

0.5mM concentration of dNTPs mix, 1mM concentration of MgCl2, 1x PCR

buffer, 4 % of DMSO and 1U of Super Tth Taq DNA polymerase for a 25 µl

reaction. Master mixes were distributed to 96-well plates. Approximately 40-60ng

of template DNA was added for each sample. M tuberculosis H37Rv DNA used

as a positive control while negative controls lacked DNA. PCR plates were sealed

and placed in PerkinElmer 9700 thermo cycler starting with a denaturing step of

15 min at 95oC, followed by 35 cycles of 1 min at 94oC, 1 min at 59oC, and 1 min

30 s at 72oC, followed by an extension of 72oC. After the thermo cycling step, all

367 MTB isolates were analyzed using a simple gel electrophoresis method. The

MIRU-VNTR method has been outlined in Figure 2.1.

Allele scoring of MIRU loci

The PCR products were electrophoresed on a 2.5% agarose gel and sized

with a 100-bp ladder (Promega). Band sizes were measured using Geldoc

Quantity-one (Bio-RAD) soft ware and allelic numbers were determined using the

MIRU-VNTR allele scoring table from international database link

www.ibl.fr/mirus.html.

60

3.3.5: Phylogenetic analysis

The twelve digits MIRU-VNTR allele score obtained for each MTB strain

was then entered into Bionumerics soft ware (Applied Maths, St. Martens

Latem, Belgium) as a character set and used to generate a dendrogram by un-

weighted pair group using arithmetic averages (UPGMA). To compare isolates

combining both methods, a multi experiment composite data set with MIRU and

Spoligotyping was created by using the available tools in Bionumerics.

3.3.6: Statistical analysis

The Hunter Gaston Discriminatory Index (HGDI) was calculated for

comparison of discriminatory power of MIRU-VNTR typing for different loci

(Hunter and Gaston 1988).Non parametric analysis was carried out using the

Mann-Whitney test to determine the utility of MIRU typing to distinguish

between CAS1 and ‘unique’ as well as CAS1-MDR and ‘unique’ MDR. A P

value of < 0.05 was considered significant. This analysis was carried out using

version 14 of SPSS (Special Program for Social Sciences Software, USA).

61

Table 3.2: MIRU-VNTR primers for twelve loci

Loci # PRIMER sequences (5' - 3')

MIRU 2 2F - TGGACTTGCAGCAATGGACCAACT

2R - TACTCGGACGCCGGCTCAAAAT

MIRU 4

4F - GCGCGAGAGCCCGAACTGC

4R - GCGCAGCAGAAACGTCAGC

MIRU 10

10F - GTTCTTGACCAACTGCAGTCGTCC

10R - GCCACCTTGGTGATCAGCTACCT

MIRU 16

16F - TCGGTGATCGGGTCCAGTCCAAGTA

16R - CCCGTCGTGCAGCCCTGGTAC

MIRU 20

20F - TCGGAGAGATGCCCTTCGAGTTAG

20R - GGAGACCGCGACCAGGTACTTGTA

MIRU 23

23F - CTGTCGATGGCCGCAACAAAACG

23R - AGCTCAACGGGTTCGCCCTTTTGTC

MIRU 24

24R - CGACCAAGATGTGCAGGAATACAT

24F - GGGCGAGTTGAGCTCACAGAA

MIRU 26

26F - TAGGTCTACCGTCGAAATCTGTGAC

26R - CATAGGCGACCAGGCGAATAG

MIRU 27

27F - TCGAAAGCCTCTGCGTGCCAGTAA

27R - GCGATGTGAGCGTGCCACTCAA

MIRU 31

31F - ACTGATTGGCTTCATACGGCTTTA

31R - GTGCCGACGTGGTCTTGAT

MIRU 39

39F - CGCATCGACAAACTGGAGCCAAAC

39R - CGGAAACGTCTACGCCCCACACAT

MIRU 40

40F - GGGTTGCTGGATGACAACGTGT

40R - GGGTGATCTCGGCGAAATCAGATA

(Reference: Supply, 2001)

62

Figure 3.1: Schematic diagram of steps of MIRU-VNTR method (*Adapted from Supply 2000 )

Amplicon size is measured using agarose gel

Based on the size of the amplicons, 12 digit code is assigned to each strain

*MIRU typing is based on 12 out of 41 Polymorphic MIRUs

PCR for each of the 12 MIRU loci Is carried out for each strain

The numerical code is then entered in Bio-numeric Software program

63

Figure 3.2: Dendrogram of MIRU-VNTR typing of Mycobacterium tuberculosis

Three hundred and sixty seven strains were typed and a cluster analysis was carried out using Bionumerics software using the unweighted pair group method. The 178 CAS1 strains studied showed an overall homology of >70%. No MIRU clusters were observed between any of the 187 ‘unique’ strains studied.

64

Table 3.3: Allelic diversity of 367 MTB isolates from Pakistan

Discriminatory Index: #≥ 0.6= Highly Discriminant and *0.3-0.59= Moderately Discriminant.

MIRU loci

Allele Number Allelic Diversity

Rank

Conclusion

0 1 2 3 4 5 6 7 8 92 18 85 262 2 0.4355 11 *Moderately

discriminant

4 21 13 290 9 11 22 1 0.3670 12 Moderately discriminant

10 18 4 29 48 73 133 50 8 4 0.7862 3 #Highly discriminant

16 43 19 46 126 83 40 9 1 0.7885 2 Highly discriminant

20 22 60 248 29 3 4 1 0.5080 10 Moderately discriminant

23 11 4 4 10 49 234 42 8 4 1 0.5616 8 Moderately discriminant

24 77 238 34 2 3 12 1 0.5271 9 Moderately discriminant

26 9 24 20 12 31 58 96 84 30 3 0.8337 1 Highly discriminant

27 20 23 107 167 42 8 0.6893 7 Highly discriminant

31 18 4 30 101 116 73 25 0.7731 4 Highly discriminant

39 22 60 153 116 15 1 0.6962 6 Highly discriminant

40 11 31 116 150 51 6 2 0.7073 5 Highly discriminant

Average 0.6394

65

Table 3.4: Twelve MIRU loci analysis of CAS1 and ‘unique’ spoligotypes.

MIRU Loci

HGDI values for P- value

CAS1 strains (n=178)

Conclusion ‘unique’ strains (n= 189)

Conclusion

2 0.4953 Moderately discriminant

0.3859 Moderately discriminant 0.105

4 0.2676 Poorly discriminant

0.4540 Moderately discriminant 0.000*

10 0.7449 Highly discriminant 0.8190 Highly

discriminant 0.004*

16 0.7503 Highly discriminant

0.7760 Highly discriminant 0.325

20 0.5993 Moderately discriminant 0.4242 Moderately

discriminant 0.124

23 0.5211 Moderately discriminant 0.6099 Highly

discriminant 0.126

24 0.5709 Moderately discriminant 0.4966 Moderately

discriminant 0.436

26 0.8117 Highly discriminant 0.8511 Highly

discriminant 0.000*

27 0.7588 Highly discriminant 0.6151 Highly

discriminant 0.909

31 0.7772 Highly discriminant 0.7756 Highly

discriminant 0.340

39 0.7090 Highly discriminant

0.6775 Highly discriminant 0.732

40 0.6970 Highly discriminant 0.7228 Highly

discriminant 0.661

Average: 0.6419 Average: 0.6339

Significantly different loci are indicated by ‘*’ (P<0.05)

66

Table 3.5: MIRU loci analysis of MDR M tuberculosis

MIRU Loci

HGDI values for P- value

MDR CAS1 (n=62)

Conclusion

MDR ‘unique’ (n= 54)

Conclusion

2 0.5944 Moderately discriminant

0.4983 Moderately discriminant 0.463

4 0.2871 Poorly discriminant

0.4689 Moderately discriminant 0.007*

10 0.7536 Highly discriminant 0.8092 Highly

discriminant 0.142

16 0.7784 Highly discriminant 0.7582 Highly

discriminant 0.188

20 0.6076 Highly discriminant

0.3662 Moderately discriminant 0.098

23 0.5463 Moderately discriminant

0.5311 Moderately discriminant 0.862

24 0.5219 Moderately discriminant 0.4780 Moderately

discriminant 0.266

26 0.8080 Highly discriminant 0.8609 Highly

discriminant 0.100

27 0.7504 Highly discriminant

0.5542 Moderately discriminant 0.955

31 0.7583 Highly discriminant

0.7659 Highly discriminant 0.934

39 0.7155 Highly discriminant 0.6003 Highly

discriminant 0.465

40 0.7150 Highly discriminant 0.7358 Highly

discriminant 0.332

Average: 0.6530

Average: 0.6189

*Locus 4 between the two is significantly different (P value < 0.05)

67

Figure 3.3: Composite dendrogram of IS6110-RFLP and MIRU-VNTR of 78 MTB strains The figure illustrates a composite analysis of IS6110-RFLP and MIRU-VNTR of 78 MTB strains using Bionumerics software (Applied Maths).

68

3.4: Results

3.4.1: MIRU typing for the predominant CAS1 genogroup and ‘unique’

strains from Pakistan

The twelve loci MIRU-VNTR analysis detected a total of 349 MIRU

patterns in our sample size of 367 strains (Figure 3.2). The 178 strains of the

CAS1 genogroup were found to be more than 70 % homologous, but were further

divided into 160 distinct patterns comprising of; 15 clusters of two strains each, 1

cluster of four strains and with 144 non-matching patterns. The 189 strains

previously identified by spoligotyping as ‘unique’ remained un-clustered after

MIRU analysis. The distribution of the MIRU alleles is summarized in Table 3.3.

3.4.2: Allelic diversity

Allelic diversity of clinical isolates was determined by twelve MIRU

loci analysis using the Hunter Gaston Discriminatory Index (HGDI). Overall,

MIRU-VNTR typing of 367 MTB strains indicated a discriminatory power of

0.999. Diversity of CAS1 (n: 178) and ‘unique’ (n: 189) strains was further

calculated separately (Table 3.3 and 3.4). Allelic analysis of 178 CAS1 strains

showed a HGDI of 0.998.

69

Allelic diversity for each locus was calculated in order to determine the

discriminatory power of these loci in a combined group for the MTB population

studied. Overall, the average allelic diversity of loci studied in these strains was

found to be 0.6394 (Table 3.3). Based on their discriminatory index (DI), seven

MIRU loci 10, 16, 26, 27, 31, 39 and 40 were designated as “highly discriminant”

(DI 0.6). While, MIRU loci 2, 4, 20, 23 and 24 were designated as “moderately

discriminant” (0.3 DI 0.6) (Sola, Filliol et al. 2003). In our MTB population

locus 26 was found to be the most discriminatory allele in order to distinguish

between CAS1 strains and ‘unique’ spoligotypes. Locus 26 provided a 10 allelic

discrimination with a HGDI of 0.833. This was followed by loci 16, 10, 31, 40,

39 and 27 respectively in order of decreasing discrimination. Locus 4 was found

to be the least discriminatory with 7 alleles and a HGDI of 0.367.

As shown in Table 3.4, the average allelic diversity of CAS1 strains was found to

be 0.6419. Of these, seven MIRU loci, numbers 26, 31, 27, 16, 10, 39, and 40,

were “highly discriminant” (DI: ≥ 0.6); four MIRU loci, 20, 24, 23, and 4 were

“moderately discriminant” (DI: 0.3-0.59); while locus number 4 was “poorly

discriminant” (DI< 0.3) for CAS1 isolates.

The average allelic diversity of ‘unique’ strains was found to be 0.6339 (Table

3.4). The diversity patterns observed for ‘unique’ strains was similar that found

70

for CAS1 strains , i.e. eight MIRU loci, number 26, 10, 16,31, 40, 39, 27 and 23

were “highly discriminant” (DI: ≥ 0.6) and four loci numbers 24, 4, 20 and 2 were

“moderately discriminant” (DI:0.3-0.59). However, no loci for ‘unique’ strains

were identified to be “poorly discriminant”.

3.4.3: Discriminatory power of MIRU typing for CAS1

Further statistical analysis was carried out to investigate the utility of each

of the twelve loci of MIRU typing to distinguish between CAS1 and ‘unique’

strains. Data was analyzed using the non-parametric Mann-Whitney test. Results

revealed that differences in loci 4, 10 and 26 were statistically significant (P-value

< .01).

3.4.4: Comparison of MIRU and RFLP typing profiles

To further investigate the heterogeneous pattern shown by MIRU-VNTR

and IS6110-RFLP typing, MIRU and RFLP profiles of 78 MTB strains were

compared which comprised a subset of strains; 43 CAS1 and 35 ‘unique’

spoligotypes. IS6110-RFLP typing of these 78 strains resulted in 73 different

RFLP types (Figure 3.3). One cluster of two strains, with single copy of IS6110

was further discriminated into individual patterns by MIRU-VNTR typing. The

remaining seventy-two strains revealed unique RFLP patterns.

71

3.4.5: Analysis of MIRU typing of MDR isolates

We analyzed MIRU patterns for all the MDR strains in order to investigate

an association between resistance and MIRUs. Of the CAS1 strains studied, 62

were MDR (35%) while 54 ‘unique’ strains were MDR (29%). HGDI values of

MIRU loci in MDR strains are shown in Table 3.5. Locus 4 was found to be

statistically significant in discriminating between CAS1 and ‘unique’ MDR

strains. Overall, no significant difference could be established between MIRU

patterns of MDR isolates belonging to either CAS1 or ‘orphan’ strains.

72

3.5: Discussion

Using 12 loci based MIRU-VNTR typing 367 MTB strains were studied and were

found to be highly diverse. Of the 178 CAS1 strains studied only 34 (19%) clustered into

groups based on MIRU profiles, while all 189 ‘unique’ spoligotypes studied had non-

matching MIRU profiles and therefore remained unclustered.

Twelve loci based MIRU-VNTR typing has been extensively used to study

Beijing strains. Results of these studies indicates Beijing strains to display variable

clustering, between 53-100% (Banu, Gordon et al. 2004; Mokrousov, Narvskaya et al.

2004; Kovalev, Kamaev et al. 2005; Kremer, Au et al. 2005; Nikolayevskyy 2006).

Amongst the Beijing isolates, locus 10 has been found to be the most discriminatory

followed by locus 26 and 31, while other loci being almost monomorphic (Mokrousov,

Narvskaya et al. 2004; Nikolayevskyy 2006). In contrast, 12 loci based MIRU-VNTR

analysis of CAS1 strains exhibited diverse MIRU-VNTR profiles, which did not reveal

any monomorphic loci within the CAS1 genogroup. Our data showed MIRU loci 26, 31,

16, 10, 27, 39 and 40, in decreasing order, to be the most discriminatory for the CAS1

genogroup of Mycobacterium tuberculosis. Despite exhibiting genetic variability CAS1

strains studied also revealed more than 70% homology in their MIRU profile. Altogether

these results are in accordance with previous findings which have suggested that the

definition of ongoing transmission in high TB incidence areas should include closely

73

related MIRU-VNTR genotypes (Yeh, Ponce de Leon et al. 1998; Hanekom, van der

Spuy et al. 2008).

Overall allelic diversity and discriminatory power of the VNTR loci in the MTB

isolates of CAS1 and ‘ unique’ spoligotypes studied in this study were higher than that

reported earlier for strains from Singapore, Russia and South Africa (Mokrousov,

Narvskaya et al. 2004; Sun, Bellamy et al. 2004). The greater diversity observed can be

attributed to continual import of new strains due to traffic of people between Pakistan and

neighboring countries endemic for tuberculosis such as, migration of populations from

Afghanistan, and also travel between neighboring countries including China, Iran, the

Middle East, India and Bangladesh.

To further understand the genetic character of MTB strains studied, we compared

a subgroup of CAS1 and ‘unique’ strains to IS6110-RFLP typing (discussed in Chapter

2). One cluster of two strains detected by RFLP typing containing one copy of IS6110

was further differentiated by MIRU-VNTR typing, further supporting the higher

discriminatory ability of MIRU-VNTR typing especially for low copy IS6110 strains

(Blackwood, Wolfe et al. 2004) .

We also compared our MIRU profiles of the CAS1 family isolates with studies

from Russia, Singapore and Bangladesh (Banu, Gordon et al. 2004; Mokrousov,

Narvskaya et al. 2004; Sun, Bellamy et al. 2004; Gutierrez, Ahmed et al. 2006), and also

74

with CAS strains from India (Gutierrez, Ahmed et al. 2006). However, none of the CAS1

MIRU types identified in this study were shared by those reported previously.

We have used the standard 12 loci based method of MIRU-VNTR typing.

However, recent studies have identified increasing numbers of related MIRU loci which

may help in further discrimination between strains. Supply et al. used 29 loci based

typing and subsequently recommended 24 loci based typing for phylogentic analysis and

15 loci typing for improved epidemiological studies (Supply, Allix et al. 2006). They

identified MIRUs 10, 26, 40, 31, 4 and 16 as being highly discriminatory (in decreasing

order) for routine epidemiological studies (Supply, Allix et al. 2006). On the other hand

Gutierrez et al used 21 loci based VNTR typing to study 91 MTB isolates from India

(Gutierrez, Ahmed et al. 2006).

Overall analysis of MIRU loci for MTB strains revealed loci 26, 16, 10, 31, 40, 39,

27, 23, 24, 20, 2 and 4 to be in descending order of discrimination for allelic diversity.

Loci 4, 10 and 26 had a significantly lower discriminatory index with a P-value <0.05 in

CAS1 strains than in ‘unique’, suggesting these loci to be the most conserved in CAS1

strains. In a region where CAS1 family of strains are the most prevalent spoligotype we

found MIRU loci 26, 31, 16, 10, 27, 39 and 40, in decreasing order, to be the most

discriminatory for differentiation of Mycobacterium tuberculosis.

75

In addition, locus 4 of CAS1 MDR strains showed significantly lower

discriminatory index with a P-value <0.05 when compared with MDR ‘unique’

spoligotype strains.

Prevalence of MDR-TB amongst untreated patients in Pakistan is reportedly 1.8%

(Javaid, Hasan et al. 2008). However, laboratory based studies have reported MDR-TB

prevalence of up to 47% in their samples (Irfan 2006; Butt 2004). CAS1 is the most

prevalent MTB genotype in the country followed by the Beijing genogroup (Hasan 2006).

The Beijing family of strains has been shown to be associated with drug resistance in

China, Russia, Vietnam, New York, and Estonia (Glynn 2002), no such association with

drug resistance has been demonstrated for CAS1 strain.

In view of increasing MDR burden detailed analysis of MDR strains is required.

Therefore we further studied the common mutations in drug resistance genes in the MDR

strains of predominant genogroups CAS1 and Beijing as well as in unique MTB strains.

76

Chapter Four

Detection of drug resistance gene mutations in MDR CAS1 and Beijing strains by Sequencing

4.1: Background

Drug resistance in MTB results from accumulation of resistance mutations within

chromosomes (Sreevatsan, Pan et al. 1997). More than 95% of the RMP resistance have

mutations in the rpoB gene. Whereas mutations in katG and inhA genes are associated

with 70-80% of INH resistance (Zhang, Heym et al. 1992; Banerjee, Dubnau et al. 1994;

Kelley, Rouse et al. 1997). Knowledge of type and frequency of mutations amongst

Chapter preview Page # 4.1: Background 76 4.2: Objectives 78 4.3: Methods

4.3.1: Mycobacterial strains 79 4.3.2: Culture and antibiotic susceptibility 79 4.3.3: DNA extraction 79 4.3.4: Sequencing of rpoB gene for RMP resistance 80 4.3.5: Sequencing of katG and inhA genes for INH resistance 80 4.3.6: Statistical analysis 81

4.4: Results 4.3.1: rpoB gene mutations for RMP resistance 86 4.3.2: katG and inhA genes mutations for INH resistance 87 4.5: Discussions 88

77

prevalent genogroup has been shown to be essential for the development of appropriate

tools for early diagnosis and control of MDR-TB strains (Lipin, Stepanshina et al. 2007).

There is limited information on mutations leading to drug resistance within MTB strains

from the country. One study reports transition at codon 531 of the rpoB gene in 7 out of

9 rifampicin resistant MTB isolates from Pakistan (Rossau, Traore et al. 1997). Studies

from India and Russia have shown an association of Beijing genogroup strains with

mutation at codon 531 of rpoB gene (Suresh, Singh et al. 2006; Lipin, Stepanshina et al.

2007). However no information regarding drug resistance gene mutations in CAS1 strains

is available so far.

Given the fact that CAS1 is the most predominant genotype in Pakistan their

association with drug resistance needs to be investigated. Therefore in this part of study

prevalent mutations in the drug resistance genes for RMP and INH resistance were

investigated in MDR strains of predominant genogroups.

78

4.2: Objectives

The specific objectives of this study were to:

4.2.1: Identify mutations in rpoB gene associated with Rifampicin resistance amongst CAS1, Beijing and ‘orphan’ Multi-drug resistant strains

4.2.2: Identify mutations in katG and inhA genes associated with Isoniazid

resistance amongst CAS1, Beijing and ‘orphan’ MDR strains

4.2.3: Analyze possible associations of specific mutation conferring Rifampicin

and Isoniazid resistance with prevalent genogroups as well as with ‘orphan’ MTB isolates.

79

4.2: Methods

4.2.1: Mycobacterial strains

All the available CAS1, Beijing and unique MDR strains, which were

genotyped previously, were included in this study. Thus a total of 62 MDR strains

comprised of 30 CAS1, 12 Beijing and 20 unique strains were selected for this

study. 10 susceptible MTB Susceptible strains including eight CAS1 and two

Beijing strains were also included. Laboratory strain Mycobacterium tuberculosis

H37Rv was used as control (wild type).

4.2.2: Culture and antibiotic susceptibility

All mycobacterial strains were cultured on Middlebrook 7H10 agar.

Susceptibility testing was performed by the standard agar proportion method as

discussed in chapter two. Strains with a high level of resistance for rifampicin

(5µg/ml) and isoniazid (1µg/ml) were further selected for MDR analysis.

4.2.3: DNA extraction

Genomic DNA was extracted from the MTB strains cultured on 7H10

agar via cetyl-trimethylamonium bromide (CTAB) method similarly as described

in Chapter Two, Section 2.3.3.

80

4.2.4: Sequencing of rpoB gene for Rifampicin resistance

rpoB gene was amplified using specific primers and cycles as reported

previously (Ma, Wang et al. 2006), using Perkin Elmer thermo cycler. 495 bp

rpoB region covering 81bp hyper variable region was amplified using specific

primers (Table 4.1). The larger rpoB region was targeted in order to explore

mutations outside hyper-variable region as well. For PCR one cycle of 15 minutes

at 95C, 35 cycles consisting; 30 seconds at 95C, 1 minute at 62C, 30 seconds

at 72C and finally 07 minutes extension at 72C was performed. Amplicons were

purified using QIA quick Qiagen PCR purification kit. Purified amplicons were

then sent for the direct sequencing of both the strands to Macrogen Company,

Korea. DNA sequences were then compared using BLAST from NCBI web link

(www.ncbi.nlm.nih.gov/BLAST).

4.2.5: Sequencing of katG and inhA genes for Isoniazid resistance

For detection of INH resistance, katG, and promoter region of inhA genes

were amplified using specific primers and cycles as reported reviously (Gonzalez,

Torres et al. 1999; Ahmad, Fares et al. 2002; Ma, Wang et al. 2006) using Perkin

Elmer thermo cycler.

428 bp katG region covering codon 315 and inhA region covering

regulatory region of inhA gene was amplified using specific primers (Table 4.1).

81

For PCR one cycle of 15 minutes at 95C, 35 cycles consisting; 30

seconds at 95C, 1 minute at 67C, 30 seconds at 72C and finally 07 minutes

extension at 72C was performed. Similarly amplicons were purified using QIA

quick Qiagen PCR purification kit and then sent for sequencing to Macrogen

Company, Korea. DNA sequences were then compared using BLAST from NCBI

web link (www.ncbi.nlm.nih.gov/BLAST).

4.2.6: Statistical analysis

Association of a particular gene mutation with a genogroup was

determined by chi-square using version 16 of SPSS (Special Program for Social

Sciences Software, USA). A P- value of < 0.05 was considered significant.

82

Table 4.1: PCR primers for drug resistance gene amplification

Antimicrobial agent

Target gene

Primers

Amplicon size (bp)

Reference

Rifampicin rpoB F: GACGACATCGACCACTTC R: GGTCAGGTACACGATCTC

495 Ma X, 2006

Isoniazid katG F: CACTGGCCGCGGCGGTCGACATT R: GTCAGTGGCCAGCATCGTCGGGGA

423 Ahmad, 2002

inhA F: CCTCGCTGCCCAGAAAGGGA R: ATCCCCCGGTTTCCTCCGGT

248 Gonzalez, 1999

83

Table 4.2: Detection of Rifampicin Resistance mutations by rpoB gene sequencing of MDR-MTB strains

# tcc(2),ccg(6),ctc(2),tac(1), cgc(1), aac(1),gac(1) *denotes significant difference (P < 0.05) between the occurrence of a specific site mutation as compared with other mutations as analyzed for each subgroup (CAS1, Beijing and Unique strains) ^ Mutations observed in susceptible strains also

Geno-type (N)

Number of Codon mutation

Hyper variable region (codon 507-533)

Outside Hyper variable region (codon 534-604) 511 gct

ccg

513 caa cta

515 atg ata

516 gac gtc

522 tcg ttg

526 cac#

527 aag cag

531 tcg ttg

539 tca gtc

541 gac tac

544 ctg ccg

548 cgc ctc

549 gac tac

550 gtg ttg

561 Atc gtc

573 gaa gca

^592 ggg gag

^595 tac acc

None

CAS1 (30)

1 - - 1 2 13* 1 17 2 - - 1 1 1 - - 4 3 1

Beijing (12)

- - - 2 - - - 8 - 1 - - - 1 1 9 - 1

Unique (20)

1 1 1 - - 1 - 12 1 - - - - - - 1 - 4

Total (62)

2 1 1 3 2 14 1 37 2 1 1 1 1 1 1 1 14 3 6

85

Table 4.3: Detection of Isoniazid Resistance mutations by katG and inhA gene sequencing of MDR-MTB strains

Genotype (N)

katG gene mutation

inhA gene mutation

Number of codon mutation Number of nucleotide mutation

315 agc acc

318 gag cag

352caa aaa

361gcc gtc

371cca tca

379gcc gtc

285ggc gtc

283ctg atg

274aag aac

251 acg aag

244gcg gag

242gcg

ggg

216 (C T)

CAS1 (30)

20 - - -

- - 1** 1 - 1 1 1 1

Beijing (12)

4* 1

1 1 1 - - - 2 - - - -

Unique (WHO)

13 - - -

- 1 - - - - - - -

Total (62) 37 1 1 1 1 1 1 1 2 1 1 1 1 *Denotes marginal significance of low occurrence of codon 315 mutation in Beijing strains than CAS1 and Unique (P= 0.052) ** Novel mutation

86

4.3: Results

4.3.1: Detection of rpoB gene mutations for Rifampicin resistance

Mutations in 495 bp region including 81bp hyper-variable region (RRDR)

of rpoB gene leading to rifampicin resistance were investigated. Using DNA

sequencing, mutations were detected in 56 (90%) of sixty two MDR strains

including 30 CAS1, 12 Beijing and 20 un-clustered spoligotyped strains.

Seventeen different combinations of mutations were identified in the rpoB gene of

MDR strains (Table 4.2). In addition mutations at codon 592 and 595 of rpoB

gene were noted. These however were seen in both resistant and susceptible

strains and therefore are not included in Table 4.2.

The mutations most commonly seen in MDR strains were in codons 531

(60%), 526 (23%) and 516 (5%) of rpoB gene. Most variability was seen in codon

526 with nine different combinations of mutations. A significantly higher

frequency of codon 526 mutation (93% vs. 7%, P= 0.008) was noted in CAS1 as

compared to Beijing and other un-clustered strains. Occurrence of more than one

mutation in rpoB gene was also significantly higher (40% vs. 12.5%, P= 0.014) in

CAS1 than Beijing and other un-clustered strains tested. Eighty-six percent of

total mutations were identified in RRDR region of the rpoB gene. 10% (n=6) of

the MDR strains tested did not show any mutations in the rpoB region.

87

4.3.2: Detection of mutations in katG and inhA genes for INH resistance

The presence of mutations in katG and inhA genes was explored further.

428 bp katG gene sequence covering codon 315 revealed twelve different sites of

mutations (Table 4.3). Sixty three percent (39/ 62) MDR isolates had a mutation

at codon 315. Our analysis (Table 4.3) revealed that CAS1 family strains

exhibited higher rate of mutation at codon 315 as compared with Beijing with

marginal significance (42% vs. 8%, P= 0.052). None of the mutations were

observed in the ten susceptible strains tested. Characterization of promoter region

of InhA gene revealed only one T-A mutation at nucleotide 216 of a CAS1 strain

as shown in Table 4.3.

88

4.5: Discussion

Our data reports commonly found mutations in the rpoB and katG genes of

predominant CAS1, Beijing and unique MDR strains. Overall the most affected codons

identified in rpoB gene were 531 (60%), 526 (23%) and 516 (5%). These findings are

comparable with previously reported data (Musser 1995). In agreement with previously

reported data from India (Siddiqi, Shamim et al. 2002) and China (Jiao, Mokrousov et al.

2007), all the mutations at codon 531 of rpoB gene were S531L. Occurrence of highest

frequency (67%) of mutation at codon 531 amongst the Beijing genogroups is also

comparable with data reported earlier (Banerjee, Dubnau et al. 1994; Sharma, Sethi et al.

2003).

Mutation at codon 526 of rpoB gene exclusively amongst the CAS1 strains

suggests an association between prevalence of this mutation and the CAS1 genogroup.

Codon 526 was furthermore found to be the most variable codon. Variability at this

codon has also been reported in a study from Russia (Afanas'ev, Ikryannikova et al.

2007). However our MTB isolates exhibited higher variablity at this codon as compared

to the data reported from other Asian countries (Hirano, Abe et al. 1999; Jiao, Mokrousov

et al. 2007). Our finding imply that 526 mutation results in no significant loss of fitness

and survival in CAS1 strains. Similar mutation and genogroup association was also

observed previously in a study where 75% of mutation at codon 516 of rpoB gene was

found amongst Latino-American and Mediterranean (LAM) genogroup (Lipin,

Stepanshina et al. 2007). Although studies from East Asian countries report 12%

89

mutation at codon 526 in Beijing family strains (Qian, Abe et al. 2002), we were unable

to confirm this finding.

In contrast, mutation at codon 516 was observed less frequently (5%) amongst our

isolates as compared to the data reported from other countries including those from Asia

(Hirano, Abe et al. 1999; Yuen, Leslie et al. 1999; Jiao, Mokrousov et al. 2007). Absence

of any mutation in the rpoB gene amongst ten percent of our MTB isolates is in

agreement with earlier reports (Telenti, Imboden et al. 1993; Hirano, Abe et al. 1999).

This finding may point towards presence of mutation in another region of rpoB or

presence of some other gene contributing towards rifampicin resistance.

Another important finding was the significant association of frequency of double

mutation (40%) with CAS1 genogroup. Although similar association has been reported

with Beijing genogroup strains from Latvia (Tracevska, Jansone et al. 2003), Beijing

strains in this study had not shown the same.

Overall, occurrence of mutation in codon 315 of katG gene correlates with the

range (60-90%) reported globally (Ramaswamy and Musser 1998). Although this,

frequency is not as high as reported from certain regions such as Russia (95%) and

Australia (91%) (Lavender, Globan et al. 2005; Lipin, Stepanshina et al. 2007). A

relatively high frequency of 315 mutation amongst our CAS1 isolates suggests that these

strains have an enhanced potential to retain peroxidase activity and decreased ability to

activate isoniazid by this mutation (Wengenack, Uhl et al. 1997). We further identified

one strain with single mutation at codon 285 (GGC-GTC) of katG that might be

responsible for isoniazid resistance. This mutation has not been described previously. In

contrast to katG gene, only one isolate showed mutation in the promoter region of inhA

90

gene. While most of the studies show around 15% of mutation in inhA gene (Baker,

Brown et al. 2005), relatively much lower frequency of this mutation amongst our

isolates may reflect local strain phenomena.

This study reports preliminary information regarding prevalent drug resistance

gene mutations amongst the MDR strains of predominant genogroups; CAS1 and Beijing,

from Pakistan. Data revealed that MDR CAS1 strains, which constitute half of the total

MDR strains in the country, were more prone to developing resistance against rifampicin

and isoniazid through mutations at codon 526 of rpoB gene and codon 315 of katG gene

respectively than Beijing and unique MDR strains. Within the MDR CAS1 strains 67%

of rifampicin and isoniazid resistance could be determined by detecting mutations at only

three codons 526 and 531 of rpoB gene and 315 of katG gene.

Knowledge of type and frequency of mutations amongst prevalent genotype

within defined geographical locations is essential for the development of appropriate

tools for early diagnosis and control of MDR-TB strains (Lipin 2007).A rapid diagnosis

of antituberculous drug resistance, especially RMP and INH, is particularly essential for

appropriate anti-tuberculous therapy and containment of the resistant strains

Therefore, using information of prevalent mutations for RMP and INH resistance

rapid molecular methods for MDR-TB detection amongst predominant genogroups;

CAS1 and Beijing were evaluated in the next part of this study.

91

Chapter Five

Molecular methods for rapid detection of Rifampicin and Isoniazid resistance amongst the MDR strains of predominant

genogroups of Mycobacterium tuberculosis 5.1: Background

It has been reported that the type and frequency of mutations varies for resistant

MTB strains within defined geographical locations (Lipin, Stepanshina et al. 2007).

Therefore distribution of rifampicin and isoniazid resistance mutations was determined in

the last part of this study using standard sequencing method (Chapter Four). Findings of

this study revealed rifampicin resistance mutation at codon 531 (60%), 526 (23%) and

Chapter preview Page #

5.1: Background 91 5.2: Objectives 94 5.3: Methods

5.3.1: Mycobacterial strains, Culture & DNA extraction 95 5.3.2: RMP and INH resistance detection using FRET 95 probe based Real Time PCR 5.3.3: RMP resistance detection using InnoLiPA 99 line probe assay 5.3.4: Statistical analysis 100

5.4: Results 5.4.1: Detection of mutation in rpoB gene for RMP 101

Resistance using FRET probes 5.4.2: Detection of mutation in katG & inhA genes 106 for INH resistance using FRET probes 5.4.3: Detection of rpoB RMP resistance by 106

INNO-LiPA assay 5.5: Discussions 109

92

516 (5%) of rpoB gene amongst the MDR-TB isolates. While for INH resistance, 63%

MDR isolates showed mutation at codon 315 of katG gene. These are important

preliminary findings for the development of rapid methods for detection of rifampicin

and isoniazid resistance in majority of MDR isolates from the country.

Different molecular assays have been proposed for the detection of mutations

associated with resistance to anti-TB drugs beside DNA sequencing which is considered

gold standard for molecular detection of drug resistance. These include real time PCR,

microarrays and kit based line probe assays (Torres 2002; Hillemann, Rusch-Gerdes et al.

2007). Use of rapid, simple and cost effective hybridization probe based real time PCR

method has been reported to detect between 60 -100 % mutations associated with

resistance to rifampicin and isoniazid (Torres 2002; Sajduda, Brzostek et al. 2004).

Limited data is available regarding use of this technique from within the Asian region.

Although a study based on isolates from India and Mexico reports detection rates of

100% (n=16) and 86% (n=64) respectively with the probe based method (Varma-Basil,

El-Hajj et al. 2004).

Commercially available kit based line probe assay (INNO-LiPA, Innogenetics

Zwijndrecht, Belgium) detects mutations in the rpoB gene for rifampicin resistance

(Rossau, Traore et al. 1997). The LiPA kit contains 10 oligonucleotide probes (one

specific for the M. tuberculosis complex, five overlapping wild-type S probes, and four R

probes for detecting specific mutations of resistant genotypes) immobilized on

nitrocellulose paper strips (Rossau, Traore et al. 1997). Since rifampicin resistance

93

usually develops in conjunction with isoniazid resistance reportedly in more than 90% of

rifampicin resistant isolates, this rapid and simple method could predict MDR (Traore,

Fissette et al. 2000). A number of studies have evaluated the diagnostic accuracy of LiPA

for the detection of rifampicin resistance in diverse geographic settings (Hirano, Abe et

al. 1999; Bartfai, Somoskovi et al. 2001; Johansen, Lundgren et al. 2003).

The aims of this study were to explore utility of two rapid molecular methods;

probe based real time PCR to identify common mutations present in rpoB, katG and inhA

genes and reverse hybridization Line probe assay (LiPA) to identify mutations in rpoB

gene amongst the prevalent genogroups; CAS1 and Beijing MDR-TB strains (identified

by sequencing, chapter four). Unique strains were also tested in order to compare the

results.

94

5.2: Objectives

The overall aim of this study was to evaluate a rapid test to detect rifampicin and

isoniazid resistance in prevalent MDR strains. Specific objectives were to

evaluate:

5.2.1: Real Time PCR method using Fluorescence resonance energy

transfer (FRET) probes for the detection of rifampicin and

isoniazid resistance in prevalent CAS1 and Beijing MDR-TB

strains from the country

5.2.2: InnoLiPA assay for the detection of rifampicin resistance in

MDR strains from prevalent genogroups

95

5.3: Methods 5.3.1: Mycobacterial strains, culture & DNA extraction

All 62 MDR-MTB and 10 susceptible strains which were used for

sequencing (Chapter Four) were selected for this study as well. These CAS1,

Beijing and un-clustered spoligotyped strains were cultured and DNA was

extracted by similar methods as described earlier in this study.

5.3.2: Rifampicin and Isoniazid resistance detection using FRET probe based Real Time PCR: (As described by Torres 2000 and Torres 2002)

Detection of mutation in rpoB katG and inhA genes was performed using

pairs of fluorescence resonance energy transfer (FRET) probes using Light Cycler

instrument (Torres, Criado et al. 2000; Torres, Criado et al. 2002). Briefly, FRET

probes consist of two probes anchor and sensor that are designed to hybridize

adjacent to each other on the complementary DNA sites. In this study 3’ end of

anchor probe was labeled with fluorescein, and the adjacent 5’end of the sensor

probe was labeled with Cal-635, another fluorescent dye (Figure 5.1). The Light

Cycler instrument activated fluorescein, which caused activation of the adjacent

dye, Cal-635. This resulted in emission of fluorescence at a different wavelength

(630-705). The detection of mutations within the DNA regions covered by the

FRET probes is based on the differential patterns of denaturation of the probes

which are bound either to homologous sequences or to sequences with a mutation.

The Tms in each of the cases will be different. Therefore, differences in the Tms for

96

the probes with respect to those obtained when assaying the probes with wt

sequences indicate the presence of a mutation in the DNA region covered by the

probes. Probes were labeled with fluorescein at the 3’ end and CAL-635 at the 5’

end. Probes and primers are listed in Table 5.1 for respective gene.

PCR reaction:

PCR reactants in a final volume of 20 μl included 2 μl of a commercial

ready-to-use reaction mix for PCR (Light-Cycler-DNA master hybridization

probes, Roche Diagnostics). MgCl2 was added to a final concentration of 4 mM.

The primers and probes were added to final concentrations of 1.4 and 0.3 μM,

respectively. 200ng of template DNA was used. The 20 μl (final volume) reaction

mix was placed in glass capillary cuvettes, which were filled by centrifugation in

a microcentrifuge.

PCR cycling condition:

Conditions for cycling were 95°C for 15 s, followed by 35 cycles of 94°C

for 1 s, 57°C for 45 s and 72°C for 30s. A melting program of 50 to 85°C at 0.1°

C/s with continuous monitoring of the fluorescence followed it. The Tm of each of

the FRET probes was calculated as the average value of the Tms obtained in at

least two independent experiments. The probe was homologous to the wild type

(wt) sequence of rpoB gene and led to different melting temperatures (Tms) when

hybridized with target DNA region with mutation. When deviations in Tm were

more than two standard deviations, a mutation was suspected.

97

Figure 5.1: Mechanism of action of FRET probes. A) Annealing step: the two FRET probes (anchor and sensor) bind to the template DNA. The head-to-tail positioning of the probes allows the two dyes to be close to each other. By a process of energy transfer, excitation of the anchor dye stimulates the sensor dye, which in turn emits fluorescence. B) The behavior of the sensor probe in the post-PCR melting step is shown for a wild-type and mutant sequence. The sensor probe is melted at lower temperature when a mutated sequence is found owing to a thermodynamically impaired binding of this probe. C) The presence of a mutation is detected by a deviation in the melting temperature of the probe. (Reference: Adapted from Garcia de Viedma 2003)

A

B

C

98

Table 5.1: Primers and FRET probes used for PCR amplification and detection of RMP and INH resistance in MTB strains

Target Gene

Primer or FRET Probe

Sequence Product size

RpoB rpoB F-primer TCGCCGCGATCAAGGAGT 158

rpoB R-primer GTGCACGTCGCGGACCTCCA rpoB probe sensor Cal635-ACCCACAAGCGCCGACTGCTGG-P rpoB probe anchor TTCATGGACCAGAACAACCCGCTGTCGGT-F

KatG katG F-primer GAAACAGCGGCGCTGATCGT 209

katG R-primer GTTGTCCCATTTCGTCGGGG katG probe sensor Cal635TCACCAGCGGCATCGAGGTCGT-P katG probe anchor CGTATGGCACCGGAACCGGTAAGGACGC-F

InhA inhA F-primer CCTCGCTGCCCAGAAAGGGA 248

inhA R-primer ATCCCCCGGTTTCCTCCGGT inhA probe sensor Cal635-CCCGACAACCTATCATCTCGCC-P inhA probe anchor CCCCTTCAGTGGCTGTGGCAGTC-F

Adapted from: (Torres, Criado et al. 2000; Sajduda, Brzostek et al. 2004)

99

5.3.3: Rifampicin resistance detection using InnoLiPA line probe assay

A subset of MDR CAS1 (n=26) and Beijing (n=7) were subjected to

INNO-LiPA analysis. Mycobacterium tuberculosis H37Rv strain was used as a

control. The hybridization assay was performed according to manufacturer’s

instructions. Brief account of the method is as follows:

PCR reaction:

The target DNA was amplified in a nested PCR using following

biotinylated primers and cycling conditions:

LIPAOP1 (outer primer), 59-GAGAATTCGGTCGGCGAGCTGATCC-39,

LIPA OP2 (outer primer), 59-CGAAGCTTGACCCGCGCGTACACC-39;

LIPA IP1 (inner primer), 59-GGTCGGCATGTCGCGGATGG-39; and

LIPA IP2 (inner primer), 59-GCACGTCGCGGACCTCCAGC-39.

The first-round PCR consisted of 30 cycles of 95°C for 60 s, 58°C for 30s, and

72°C for 90 s. The second-round PCR consisted of 30 cycles of 95°C for 20s,

65°C for 30 s, and 72°C for 30 s.

100

Probe hybridization:

Each PCR product was denatured and hybridized to membrane- bound

capture probes i.e. S-type probes (SI, S2, S3, S4 and S5) and four R- type probes,

which are specifically designed to hybridize to the sequences of the four most

frequently observed mutations; R2 (Asp-516-Val), R4a (His-526-Tyr), R4b (His-

526-Asp), and R5 (Ser-531-Leu).

Interpretation of the banding pattern:

The hybridization of the product of amplification with probes immobilized

on a nitrocellulose support was revealed by reaction of the biotine with the

streptavidine coupled with alkaline phosphate, which allowed detection of

respective mutation. (Rossau, Traore et al. 1997; Matsiota-Bernard, Vrioni et al.

1998; Watterson, Wilson et al. 1998; Hirano, Abe et al. 1999; Bartfai, Somoskovi

et al. 2001)

5.3.4: Statistical analysis

Association of a particular gene mutation with a genogroup was

determined by Fisher exact test using version 16 of SPSS (Special Program for

Social Sciences Software, USA). A p-value of < 0.05 was considered significant.

101

5.4: Results

5.4.1: Detection of mutations in rpoB gene for Rifampicin resistance using

FRET probes

Mutations in the rpoB, katG and inhA genes detected using FRET probes

are shown in Figure 5.2. The temperatures at which the probes melted (Tm) from

PCR products during the melting program are also shown. Tm was calculated

using the Light Cycler software. The average Tm for rpoB gene in ten susceptible

strains was 65.4 °C. Twenty seven (44%) MDR strains showed mutations in

codon 531 with an average >2 C increase while 6 (10%) strains showed

mutations in codon 526 with an average >2C decrease in the Tm of probe as

compared to the susceptible strains (Figure 5.1). Concurrent mutation of 531 and

526 in seven MDR strains resulted in 2 C increase in the Tm of melting probes.

The number of mutation detected in rpoB, gene of 62 MDR strains by probe

based real time PCR assay was 52. Sensitivity and Specificity of probe based

assay for detection of mutation in rpoB was 84% and 100% in comparison to

culture susceptibility testing and 93% and 100% respectively compared to

sequencing.

102

Figure 5.2: Melting curves detecting rpoB mutation using FRET probes with three different types of samples. Rifampicin resistant MTB strains, with 531 codon mutation shows higher Tm while MTB strains with 526 codon mutation exhibit lower Tm compared to wild type MTB strain.

103

Table 5.2: Detection of prevalent Rifampicin and Isoniazid resistance by FRET probes using real time PCR Antimicrobial

agent Target gene

Mutation at codon(s)*/nucleotide

Number of strains

Mean Tm with standard deviation (°C )

Rifampicin rpoB Susceptible strains 10 65.4 0.29

531 27 68.81 1.21

531 + 526 07 67.46 0.64

526 06 63 1 516 03 65.82 0.79 531+522 02 68.9 0.14 513 01 67.2 531 + 527 01 67.2 526+ 511 01 62.2 515+511 01 65.1 No mutation 06 65.2 0.68

Isoniazid katG Susceptible strains 10 73.61 0.59 315 33 69.78 1.06 285 01 73.9 315 +242 01 74 315+379 01 69.18 315+352+274 01 70.7 315+383+351+244 01 69.4 318+361+371+274 01 71.3 No mutation 23 73.61 0.54 inhA Susceptible strains 10 64.76 0.74 216 nucleotide 01 59.7 No mutation 61 64.6 0.77

* Mutations arranged by decreasing occurrence

104

Figure 5.3: Melting curves detecting katG mutation using FRET probes with two different types of samples; wild type (WT) strains have perfect match to the sensor probe and a Tm of 73.6C; R type (INH resistant) MTB strains with the mutation at codon 315 have lower Tm 71.0C.

Figure 5.4: Melting curves detecting inhA mutation using FRET probes. Strains with wild type gene have a Tm of 64.6 C; similar to susceptible strains while only strain with a point mutation in the gene has almost 5 0C lower Tm.

105

Table 5.3: Summary of mutations identified by Sequencing and Real time PCR in Rifampicin and Isoniazid resistant MTB isolates as compared to phenotypic drug susceptibility testing (DST)

Geno-group

(N)

Rif resistance N (%)

INH resistance N (%)

Resistance

by DST rpoB gene mutation by

Resistance by DST

katG gene mutation

by

inhA gene mutation by

Sequenc-

ing

Real Time

Sequenc-

ing

Real Time

Sequenc-

ing

Real Time

CAS1(30)

30 29/30

(97)27/30 (87)

30

20/30 (66)

20/30

(66) 1/30 (3.3)

1/30 (3.3)

Beijing (12) 12 11/12

(92)9/12 (75)

12 5/12 (42)

5/12 (33) 0/12 0/12

Unique (20)

20 16/20 (85)

16/20 (80)

20 12/20 (60)

12/20

(55) 0/20 0/20 Concordan

ce With DST

56/62 (90)

52/62 (82)

37/62 (60)

37/62

(60) 1/62 (2)

1/62 (2)

106

Number of Strains

Mutation at rpoB codons

Figure 5.5: rpoB mutations identified by InnoLiPA assay

0

2

4

6

8

10

12

14

∆S1 ∆S2 ∆S1& S2

516 526 531 526&531

ND

CAS1

Beijing

107

Table 5.5: Comparison of results obtained by LiPA and DNA Sequencing of 33 MDR-TB strains

LiPA pattern

DNA sequencing result

# of isolates

CAS1 (n=26)

Beijing (n=7)

R5 ( Ser531Leu) TCGTTG ( Ser531Leu) 14 5 R2 ( Asp516Val) GACGTC (Asp516Val) 1 1 R4a ( His526Tyr) CACTCC ( His526 Ser) 3 0

R2 &R5 ( Asp516Val + Ser531Leu)

GACGTC (Asp516Val) TCGTTG ( Ser531Leu)

1 0

S1 CACCCC ( His526Pro) 1 0 S2 GACGTC (Asp516Val)

CACCCC ( His526Pro) 1 1

0 0

S1 & S2 CTGCCG (Leu511Pro) CACCCC ( His526Pro)

1 0

Wild type No mutation detected 3 1

108

5.4.2: Detection of mutations in katG and inhA genes for isoniazid resistance using FRET probes

Average Tm for katG gene of ten susceptible strains was 73.6 °C. Thirty -

seven (60%) MDR strains showed mutations in codon 315 detected by a decrease

of > 3C of temperature in probe’s Tm as compared to the susceptible strains

(Figure 5.2). One mutation in the promoter region of inhA gene was also detected

using inhA hybridization probes, with decrease of 5 C in the Tm as compared to

the 64.7 C Tm of susceptible strains (Figure 5.3). The number of mutations

detected in katG and inhA genes of 62 MDR strains by probe based real time PCR

assay was 37 and 1 respectively. Sensitivity and Specificity for detection of

mutation in katG was 60% and 100% in comparison to culture susceptibility

testing and 95% and 100% respectively compared with sequencing.

5.4.3: Detection of Rifampicin resistance by INNO-LiPA assay

Rifampicin resistance mutations were detected by LiPA in 29/33

rifampicin-resistant MTB strains. Nineteen MTB isolates showed S531L, three

isolate had H526Y, and two isolated showed D516V mutation while five strains

showed mutations other than these codons (Figure 5.4).

LiPA results were also compared to culture susceptibility and sequencing

as well. The sensitivity and specificity of the LiPA assay for detection of rpoB

mutation in comparison with culture susceptibility testing and sequencing was

88% and 100% respectively. Overall 29/33 resistant MTB strains were found

resistant by LiPA assay. The resistant strains, which were not detected by LiPA

were also found to be wild type by sequencing (Table 5.3).

109

3.6: Discussion

Using the information of common mutations for rifampicin and isoniazid

resistance (discussed in Chapter Four) previously described FRET probe based real time

PCR method was evaluated for the detection of mutations in rpoB, katG and inhA genes

(Torres, Criado et al. 2000; Torres 2002).

Using rpoB probes, mutations were detected in 84% of the MDR isolates. Overall

the frequency of mutations identified were in agreement with data reported earlier (Torres,

Criado et al. 2000; Garcia de Viedma 2003; Sajduda, Brzostek et al. 2004). These

findings supports that probe based assay may be useful as a rapid diagnostic tool for

detection of rifampicin resistance.

While using katG probe, mutation in 60% of the isoniazid resistant isolates was

detected at codon 315 similar to that of sequencing. Detection of mutation at codon 315 is

clinically important, as it has been linked with high level of isoniazid resistance (van

Soolingen, de Haas et al. 2000; Marin, Garcia de Viedma et al. 2004). Additionally katG

315 mutations have also been found to be associated with successful transmission of

MDR-TB within the population (van Soolingen, de Haas et al. 2000). Thus, the utility of

katG probes for detection of this mutation in this geographical location may be vuseful

for rapid detection of MDR-TB strains.

110

With inhA probes only one mutation was detected which was in accordance with

sequencing results. This finding is in contrast to the study from Equatorial Guinea in

which none of the mutations were observed in katG gene while 80.5% mutations were

detected in inhA gene (Tudo, Gonzalez et al. 2004). Thus, low prevalence of mutation in

inhA gene of MDR strains from this country might be a regional phenomenon, which

needs further investigation.

LiPA assay has been used in several studies for the rapid detection of resistance to

rifampicin in MTB isolates. These studies have shown greater than 95% of sensitivity and

specificity (Morgan, Kalantri et al. 2005) and allowed prediction of MDR in more than

95% of the cases (Traore, Fissette et al. 2000). In this study LiPA assay showed 88% and

100% of sensitivity and specificity respectively. High (100%) specificity and rather lower

sensitivity has also been reported in other studies (Morgan, Kalantri et al. 2005; Traore,

van Deun et al. 2006). The high specificity suggests that LiPA test is a good predictor for

MDR and may be used in this setting for detection of rifampicin resistance from majority

of resistant isolates. The cost of the test on the hand is the major limitation for the

application of this test at large (Piersimoni and Scarparo 2003).

In conclusion, FRET probe based methods are less expensive and rapid, and have

the potential of detecting wide variety of resistance mutations in majority of clinical

MTB isolates in our geographical area. Use of this method in conjunction of routine

susceptibility testing for MDR-TB detection might have major impact on the

management of multidrug-resistant tuberculosis.

111

Chapter six

General discussion and Conclusions

World Health Organization declared tuberculosis as a global emergency in

1993. Increase in population, migration and high incidence of immuno deficiency

virus (HIV) infection are major contributory factors for the resurgence and spread of

tuberculosis in the late 20th and early 21st century. The understanding of tuberculosis

transmission dynamics has been greatly enhanced by the availability of genotyping

tools. Using appropriate genotyping methods molecular epidemiological studies of

Mycobacterium tuberculosis have provided novel insights into the biogeography of

tuberculosis, which has been shown to play a significant role in proposing new and

innovative strategies for control of tuberculosis.

Pakistan accounts for approximately half of the total tuberculosis cases in the

Eastern Mediterranean region with estimated number of 423000 cases. This figure

implies that a large number of people in the country are serving as disease reservoir.

Additionally, approximately 75% of the infected cases are in the productive years

(15-59) of their life, the illness exerting an additional economical burden on their

families and on the country at large (WHO 2008). Moreover, non compliance and

premature cessation of treatment by one out of every four-five tuberculosis patients

results in development of drug resistant tuberculosis (Liefooghe, Michiels et al. 1995).

According to recent estimate prevalence of MDR-TB amongst untreated cases in

Pakistan is 1.8% (Javaid, Ghafoor et al. 2008).

112

This study provides first insight into the genetic differences of

Mycobacterium tuberculosis, including Central Asian Strain1 (CAS1), identified as a

predominant genogroup from Pakistan (Hasan, Tanveer et al. 2006). CAS1 family

strains investigated in this study are a closely related strain type of Mycobacterium

tuberculosis and are wide spread in South Asian countries including India and

Bangladesh (Banu, Gordon et al. 2004; Bhanu, Banavalikar et al. 2004). In addition

these strains have also been reported from Sudan and were also found amongst South

Asian immigrants in England (Gascoyne-Binzi, Barlow et al. 2002; Brudey, Driscoll

et al. 2006). Thus it is important to carry out in depth molecular studies using more

discriminatory methods than spoligotyping to study these strains further.

This study used MIRU-VNTR and IS6110-RFLP molecular typing methods

to unravel intra-strain differences amongst CAS1 strains from the country. IS6110-

RFLP and MIRU-VNTR typing revealed highly discriminative profiles with

relatively lower clustering of MTB strains from this geographical setting. This

relatively high genetic diversity in MTB population including CAS1 strains was

unexpected, since studies have demonstrated lower levels of genetic diversity in high-

incidence communities (Pineda-Garcia, Ferrera et al. 1997; Easterbrook, Gibson et al.

2004; Verver, Warren et al. 2004; Nikolayevskyy, Gopaul et al. 2006). The diverse

genotypic profiles might support the hypothesis that majority of cases in the country

arise from endogenous reactivation of diverse latent tuberculosis infection as opposed

to cross transmission. Such high genetic diversity has also been reported from other

113

countries including India and Bangladesh and some African regions (Tazi, Reintjes et

al. 2007; Hanekom, van der Spuy et al. 2008; Sharma, Kalyani et al. 2008).

There may be three major reasons for the high genetic polymorphisms in the

MTB strains from the country; diverse host population with different ethnicity,

ancient TB endemicity and genetic variability. Studies have suggested co-evolution of

the MTB strains with diverse population over a long period of time in any

geographical region (Brosch, Gordon et al. 2002; Gagneux, DeRiemer et al. 2006).

Pakistan; part of Indian Subcontinent, is an old endemic TB region. Millions of

people have migrated to Pakistan from India after the partition of Subcontinent in

1947. Many of them must have brought strains from different parts of the

subcontinent. Besides, Pakistan also faced large number of influxes of immigrants

after the liberation of Bangladesh in 1971 and after Soviet Union invasion of

Afghanistan in 1979 (Meulemans 2000). It has been proposed that TB endemicity

and host-pathogen coexistence in a population requires a social network of

approximately 200-400 persons (McGrath 1988). In our geographical location many

communities may have experienced the exposure of fundamental genotypes of

Mycobacterium tuberculosis in past. Thus these fundamental MTB genotypes may

have had ample time to create a large number of population adapted genetic variants

(Sola, Filliol et al. 2001; Gagneux, DeRiemer et al. 2006). High genetic diversity may

also be due to genetic variation resulting from insertion/deletion events (Brosch,

Gordon et al. 2002).

114

Studies have grouped CAS1 strains under ‘modern’ MTB lineage due to absence

of Mycobacterium tuberculosis specific ‘TbD1 region’ (Sreevatsan, Pan et al. 1997;

Brosch, Gordon et al. 2002; Gutierrez, Ahmed et al. 2006). It has also been suggested

that the clones which take longer to evolve result in more extensive genetic diversity

(Sola, Filliol et al. 2001). In addition, Region of difference (RD) 750 has also been

reported deleted from CAS family strains. This deletion has been suggested to be

involved in the lower production of protective cytokines (Gagneux, DeRiemer et al.

2006; Newton, Smith et al. 2006; Gagneux and Small 2007), which might had a role

in persistence and spread of CAS1 strains in the region.

Beside genetic diversity MIRU-VNTR typing also revealed seven most

discriminatory MIRU loci i.e. 26, 31, 16, 10, 27, 39 and 40 for the molecular

epidemiological studies in the geographical area where CAS1 strains are common.

Five of these seven loci identifies (26, 31, 10, 39 and 40) were also the most

discriminatory in other studies (Mazars, Lesjean et al. 2001; Supply, Warren et al.

2003), suggesting that the relative degree of information carried by the different loci

are globally conserved in MTB population. These limited numbers of polymorphic

loci demonstrate high discriminatory value for the prevalent MTB population from

this region and thus could be used as a cost effective molecular typing method.

Application of limited number of loci may be a useful and rapid genotyping tool for

studying transmission dynamics in the population where CAS1 strains are prevalent.

Study from our lab also revealed that almost half of the predominant CAS1 strains

comprise of MDR (Hasan, Tanveer et al. 2006). In addition, drug resistance in MTB

115

has been shown to result from acquisition of resistance mutations within the

chromosomes (Sreevatsan, Pan et al. 1997). Therefore mutations in the drug

resistance genes for MDR amongst the predominant genogroups were also

investigated. Results from this study suggest that more than 60% of rifampicin and

isoniazid resistance within MDR CAS1 strains could be determined by detecting

mutations at only three codons 526 and 531 of rpoB and 315 of katG gene. Study

further revealed that MDR CAS1 strains were more prone to develop resistance

against rifampicin and isoniazid through mutations at codon 526 of rpoB gene and

codon 315 of katG gene respectively as compared to non-CAS1 MDR strains. This

data is in contrast to the study suggesting an association between codon 526 mutation

and a decrease in fitness (Gagneux, Long et al. 2006). Predominance of CAS1 MDR

strains in our population on the other hand suggests that 526 mutation does not result

in any significant loss of fitness and survival in CAS1 strains. This implication is also

supported by an earlier study where six out of seven CAS1 MDR strains had

mutations at codon 526 of rpoB gene (Githui, Jordaan et al. 2004).

For isoniazid resistance CAS1 isolates showed relatively higher frequency of

codon 315 katG gene mutation compared to Beijing strains. Similar findings have

also been reported previously in a study, which noted codon 315 mutation amongst

six out of seven CAS1 MDR strains (Githui, Jordaan et al. 2004). There is a

possibility that with codon 315 katG mutation CAS1 strains have an enhanced

potential to retain peroxidase activity and decreased ability to activate INH by this

mutation as suggested by earlier studies (Wengenack, Uhl et al. 1997; Baker, Brown

116

et al. 2005). On the other hand, studies have suggested an inverse relationship

between propagation of resistant MTB strains and other than codon 315 katG

mutations in a defined geographical location (Pym, Saint-Joanis et al. 2002; Gagneux,

Burgos et al. 2006). Thus, significantly higher katG mutations in codons other than

315 in Beijing strains might explain comparatively lower prevalence of Beijing

strains than CAS1 in this geographical setting. However, further investigation is

required to validate this implication.

In addition, for the development and implementation of rapid MDR detection

amongst MTB stains circulating within Pakistan probe based real time PCR methods

have shown promising results. Using rpoB, katG and inhA Florescence Resonance

Energy Transfer probes (FRET), common mutations in 84% and 60% of the

rifampicin and isoniazid resistant respectively isolates were detected. These data

suggest that probe based assay may be useful as a rapid diagnostic tool. However

findings from this study further show that reliance on probe based assay alone will be

unable to detect all the resistance cases and that such analysis will require to be

complemented with routine susceptibility testing for MDR detection. In future probe

based real time PCR method could be tested on larger population of MTB strains.

Additionally, method needs to be compared with recently introduced line probe

assays for the detection of resistance in MDR-TB strains in terms of cost and time in

order to make recommendations for its application in detection of rifampicin and

isoniazid resistance (Hillemann, Rusch-Gerdes et al. 2007).

117

In conclusion, molecular typing techniques established in this study, could enhance

understanding of transmission dynamics of disease and could be used to improve

current control programs for tuberculosis in the country. Community based studies

using these established techniques may shed more light on the transmission of

susceptible and drug resistant MTB strains by tracking these strains in the community.

Genotyping information of MTB strains together with epidemiologic investigations

may provide important information about the spread of MTB strains by identifying

factors related to transmission and progression to tuberculosis disease. This in turn

could greatly assist in formulating strategies for control of tuberculosis. The data

obtained in this study further demonstrates that use of rapid drug resistance detection

methods in conjunction of routine susceptibility testing, such as FRET probe based

method used in this study for MDR-TB detection, could further assist in the

management of multidrug-resistant tuberculosis in the country by timely and adequate

use of anti-tuberculosis therapy. As Ian Sutherland noted: “one man’s cure is many

men’s prevention” (Meulemans 2000).

118

References:

Aaron, L., D. Saadoun, et al. (2004). "Tuberculosis in HIV-infected patients: a comprehensive review." Clin Microbiol Infect 10(5): 388-98.

Afanas'ev, M. V., L. N. Ikryannikova, et al. (2007). "Molecular characteristics of rifampicin- and isoniazid-resistant Mycobacterium tuberculosis isolates from the Russian Federation." J Antimicrob Chemother 59(6): 1057-64.

Agerton T. B (1999). "Spread of strain W, a highly drug resistant strain of Mycobacterium tuberculosis, across the United States." Clin Infect Dis 29: 85-95.

Ahmad, S., G. F. Araj, et al. (2000). "Characterization of rpoB mutations in rifampin-resistant Mycobacterium tuberculosis isolates from the Middle East." Diagn Microbiol Infect Dis 38(4): 227-32.

Ahmad, S., E. Fares, et al. (2002). "Prevalence of S315T mutation within the katG gene in isoniazid-resistant clinical Mycobacterium tuberculosis isolates from Dubai and Beirut." Int J Tuberc Lung Dis 6(10): 920-6.

Almeida, D., C. Rodrigues, et al. (2005). "High incidence of the Beijing genotype among multidrug-resistant isolates of Mycobacterium tuberculosis in a tertiary care center in Mumbai, India." Clin Infect Dis 40(6): 881-6.

Ang, C. F., C. S. Ong, et al. (2008). "An overview of the phenotypic and genotypic characteristics of multidrug-resistant Mycobacterium tuberculosis isolates from four Asian countries." J Med Microbiol 57(Pt 8): 1039-40.

Anh, D. D., M. W. Borgdorff, et al. (2000). "Mycobacterium tuberculosis Beijing genotype emerging in Vietnam." Emerg Infect Dis 6(3): 302-5.

Baker, L. V., T. J. Brown, et al. (2005). "Molecular analysis of isoniazid-resistant Mycobacterium tuberculosis isolates from England and Wales reveals the phylogenetic significance of the ahpC -46A polymorphism." Antimicrob Agents Chemother 49(4): 1455-64.

Banerjee, A., E. Dubnau, et al. (1994). "inhA, a gene encoding a target for isoniazid and ethionamide in Mycobacterium tuberculosis." Science 263(5144): 227-30.

Banu, S., S. V. Gordon, et al. (2004). "Genotypic analysis of Mycobacterium tuberculosis in Bangladesh and prevalence of the Beijing strain." J Clin Microbiol 42(2): 674-82.

Barnes, D. S. (2000). "Historical perspectives on the etiology of tuberculosis." Microbes Infect 2(4): 431-40.

Barnes, P. F. and M. D. Cave (2003). "Molecular epidemiology of tuberculosis." N Engl J Med 349(12): 1149-56.

Bartfai, Z., A. Somoskovi, et al. (2001). "Molecular characterization of rifampin-resistant isolates of Mycobacterium tuberculosis from Hungary by DNA sequencing and the line probe assay." J Clin Microbiol 39(10): 3736-9.

Becerra, M. C., J. Freeman, et al. (2000). "Using treatment failure under effective directly observed short-course chemotherapy programs to identify patients with multidrug-resistant tuberculosis." Int J Tuberc Lung Dis 4(2): 108-14.

Bhanu, N. V., J. N. Banavalikar, et al. (2004). "Suspected small-scale interpersonal transmission of Mycobacterium tuberculosis in wards of an urban hospital in Delhi, India." Am J Trop Med Hyg 70(5): 527-31.

119

Bhanu, N. V., D. van Soolingen, et al. (2002). "Predominace of a novel Mycobacterium tuberculosis genotype in the Delhi region of India." Tuberculosis (Edinb) 82(2-3): 105-12.

Bifani, P. J. (1999). "Identification of a W variant outbreak of Mycobacterium tuberculosis via population based molecular epidemiology." Jama 282: 2321-2327.

Blackwood, K. S., J. N. Wolfe, et al. (2004). "Application of mycobacterial interspersed repetitive unit typing to Manitoba tuberculosis cases: can restriction fragment length polymorphism be forgotten?" J Clin Microbiol 42(11): 5001-6.

Bloch, A. B., G. M. Cauthen, et al. (1994). "Nationwide survey of drug-resistant tuberculosis in the United States." Jama 271(9): 665-71.

Blomberg, B., S. Spinaci, et al. (2001). "The rationale for recommending fixed-dose combination tablets for treatment of tuberculosis." Bull World Health Organ 79(1): 61-8.

Boddinghaus, B., T. Rogall, et al. (1990). "Detection and identification of mycobacteria by amplification of rRNA." J Clin Microbiol 28(8): 1751-9.

Borgdorff, M. W., N. J. Nagelkerke, et al. (2000). "Gender and tuberculosis: a comparison of prevalence surveys with notification data to explore sex differences in case detection." Int J Tuberc Lung Dis 4(2): 123-32.

Brosch, R., S. V. Gordon, et al. (2002). "A new evolutionary scenario for the Mycobacterium tuberculosis complex." Proc Natl Acad Sci U S A 99(6): 3684-9.

Brudey, K., J. R. Driscoll, et al. (2006). "Mycobacterium tuberculosis complex genetic diversity: mining the fourth international spoligotyping database (SpolDB4) for classification, population genetics and epidemiology." BMC Microbiol 6: 23.

Burman, W. J., C. B. Dalton, et al. (1997). "A cost-effectiveness analysis of directly observed therapy vs self-administered therapy for treatment of tuberculosis." Chest 112(1): 63-70.

Butt, T., R. N. Ahmad, et al. (2004). "Multi-drug resistant tuberculosis in Northern Pakistan." J Pak Med Assoc 54(9): 469-72.

Caminero, J. A. (2001). "Epidemiological evidence of the spread of a Mycobacterium tuberculosis strain of the Beijing Genotype on Gran Canaria Island." Am J Resp Crit Care Med 164: 1165-1170.

Caminero, J. A. (2006). "Treatment of multidrug-resistant tuberculosis: evidence and controversies." Int J Tuberc Lung Dis 10(8): 829-37.

Canetti, G., W. Fox, et al. (1969). "Advances in techniques of testing mycobacterial drug sensitivity, and the use of sensitivity tests in tuberculosis control programmes." Bull World Health Organ 41(1): 21-43.

CDC (1991). "Nosocomial transmission of multidrug-resistant tuberculosis among HIV-infected persons--Florida and New York, 1988-1991." MMWR Morb Mortal Wkly Rep 40(34): 585-91.

CDC (1991). "Transmission of multidrug-resistant tuberculosis from an HIV-positive client in a residential substance-abuse treatment facility--Michigan." MMWR Morb Mortal Wkly Rep 40(8): 129-31.

Chadha, V. K. (2005). "Tuberculosis epidemiology in India: a review." Int J Tuberc Lung Dis 9(10): 1072-82.

120

Chan, M. Y., M. Borgdorff, et al. (2001). "Seventy percent of the Mycobacterium tuberculosis isolates in Hong Kong represent the Beijing genotype." Epidemiol Infect 127(1): 169-71.

Chauhan, D. S., V. D. Sharma, et al. (2007). "Molecular typing of Mycobacterium tuberculosis isolates from different parts of India based on IS6110 element polymorphism using RFLP analysis." Indian J Med Res 125(4): 577-81.

Chaulk, C. P., M. Friedman, et al. (2000). "Modeling the epidemiology and economics of directly observed therapy in Baltimore." Int J Tuberc Lung Dis 4(3): 201-7.

Chaulk, C. P. and M. Grady (2000). "Evaluating tuberculosis control programs: strategies, tools and models." Int J Tuberc Lung Dis 4(2 Suppl 1): S55-60.

Chin, P. J., C. C. Chiu, et al. (2007). "Identification of Beijing lineage Mycobacterium tuberculosis with combined mycobacterial interspersed repetitive unit loci 26, 31, and ETR-A." J Clin Microbiol 45(3): 1022-3.

Cole, S. T. (1994). "Mycobacterium tuberculosis: drug-resistance mechanisms." Trends Microbiol 2(10): 411-5.

Cole, S. T., R. Brosch, et al. (1998). "Deciphering the biology of Mycobacterium tuberculosis from the complete genome sequence." Nature 393(6685): 537-44.

Cole, S. T. and A. Telenti (1995). "Drug resistance in Mycobacterium tuberculosis." Eur Respir J Suppl 20: 701s-713s.

Cowan, L. S., L. Mosher, et al. (2002). "Variable-number tandem repeat typing of Mycobacterium tuberculosis isolates with low copy numbers of IS6110 by using mycobacterial interspersed repetitive units." J Clin Microbiol 40(5): 1592-602.

Dale, J. W., H. Al-Ghusein, et al. (2003). "Evolutionary relationships among strains of Mycobacterium tuberculosis with few copies of IS6110." J Bacteriol 185(8): 2555-62.

Dale, J. W., R. M. Nor, et al. (1999). "Molecular epidemiology of tuberculosis in Malaysia." J Clin Microbiol 37(5): 1265-8.

Daniel, T. M. (2006). "The history of tuberculosis." Respir Med 100(11): 1862-70. Das, S. D., S. Narayanan, et al. (2005). "Differentiation of highly prevalent IS6110

single-copy strains of Mycobacterium tuberculosis from a rural community in South India with an ongoing DOTS programme." Infect Genet Evol 5(1): 67-77.

Deepak, A. (2005). "High Incidence of The Beijing Genotype." CID 40: 881-6. Drobniewski, F. A. and S. M. Wilson (1998). "The rapid diagnosis of isoniazid and

rifampicin resistance in Mycobacterium tuberculosis--a molecular story." J Med Microbiol 47(3): 189-96.

Dubos, R. (1952). "The White plague." Rutgers, The State University. Dukes Hamilton, C., T. R. Sterling, et al. (2007). "Extensively drug-resistant

tuberculosis: are we learning from history or repeating it?" Clin Infect Dis 45(3): 338-42.

Dye, C. (2006). "Global epidemiology of tuberculosis." Lancet 367(9514): 938-40. Dye, C., G. P. Garnett, et al. (1998). "Prospects for worldwide tuberculosis control under

the WHO DOTS strategy. Directly observed short-course therapy." Lancet 352(9144): 1886-91.

Dye, C., S. Scheele, et al. (1999). "Consensus statement. Global burden of tuberculosis: estimated incidence, prevalence, and mortality by country. WHO Global Surveillance and Monitoring Project." Jama 282(7): 677-86.

121

Easterbrook, P. J., A. Gibson, et al. (2004). "High rates of clustering of strains causing tuberculosis in Harare, Zimbabwe: a molecular epidemiological study." J Clin Microbiol 42(10): 4536-44.

Edlin, B. R., J. I. Tokars, et al. (1992). "An outbreak of multidrug-resistant tuberculosis among hospitalized patients with the acquired immunodeficiency syndrome." N Engl J Med 326(23): 1514-21.

Espinal, M. A., S. J. Kim, et al. (2000). "Standard short-course chemotherapy for drug-resistant tuberculosis: treatment outcomes in 6 countries." Jama 283(19): 2537-45.

Espinal, M. A., A. Laszlo, et al. (2001). "Global trends in resistance to antituberculosis drugs. World Health Organization-International Union against Tuberculosis and Lung Disease Working Group on Anti-Tuberculosis Drug Resistance Surveillance." N Engl J Med 344(17): 1294-303.

F.Barnes, P. (2003). "Molecular epidemiology of Tuberculosis." New England J of Medicine 349(12): 1149-1155.

Farmer, P. and J. Y. Kim (1998). "Community based approaches to the control of multidrug resistant tuberculosis: introducing "DOTS-plus"." Bmj 317(7159): 671-4.

Farnia, P., F. Mohammadi, et al. (2004). "Evaluation of tuberculosis transmission in Tehran: using RFLP and spoligotyping methods." J Infect 49(2): 94-101.

Filliol, I., J. R. Driscoll, et al. (2002). "Global distribution of Mycobacterium tuberculosis spoligotypes." Emerg Infect Dis 8(11): 1347-9.

Filliol, I., J. R. Driscoll, et al. (2003). "Snapshot of moving and expanding clones of Mycobacterium tuberculosis and their global distribution assessed by spoligotyping in an international study." J Clin Microbiol 41(5): 1963-70.

Filliol, I., A. S. Motiwala, et al. (2006). "Global phylogeny of Mycobacterium tuberculosis based on single nucleotide polymorphism (SNP) analysis: insights into tuberculosis evolution, phylogenetic accuracy of other DNA fingerprinting systems, and recommendations for a minimal standard SNP set." J Bacteriol 188(2): 759-72.

Frieden, T. R., P. I. Fujiwara, et al. (1995). "Tuberculosis in New York City--turning the tide." N Engl J Med 333(4): 229-33.

Gagneux, S., M. V. Burgos, et al. (2006). "Impact of bacterial genetics on the transmission of isoniazid-resistant Mycobacterium tuberculosis." PLoS Pathog 2(6): e61.

Gagneux, S., K. DeRiemer, et al. (2006). "Variable host-pathogen compatibility in Mycobacterium tuberculosis." Proc Natl Acad Sci U S A 103(8): 2869-73.

Gagneux, S., C. D. Long, et al. (2006). "The competitive cost of antibiotic resistance in Mycobacterium tuberculosis." Science 312(5782): 1944-6.

Gagneux, S. and P. M. Small (2007). "Global phylogeography of Mycobacterium tuberculosis and implications for tuberculosis product development." Lancet Infect Dis 7(5): 328-37.

Gandhi, N. R., A. Moll, et al. (2006). "Extensively drug-resistant tuberculosis as a cause of death in patients co-infected with tuberculosis and HIV in a rural area of South Africa." Lancet 368(9547): 1575-80.

122

Garcia de Viedma, D. (2003). "Rapid detection of resistance in Mycobacterium tuberculosis: a review discussing molecular approaches." Clin Microbiol Infect 9(5): 349-59.

Gascoyne-Binzi, D. M., R. E. Barlow, et al. (2002). "Predominant VNTR family of strains of Mycobacterium tuberculosis isolated from South Asian patients." Int J Tuberc Lung Dis 6(6): 492-6.

Githui, W. A., A. M. Jordaan, et al. (2004). "Identification of MDR-TB Beijing/W and other Mycobacterium tuberculosis genotypes in Nairobi, Kenya." Int J Tuberc Lung Dis 8(3): 352-60.

Glynn, J. R., J. Whiteley, et al. (2002). "Worldwide occurrence of Beijing/W strains of Mycobacterium tuberculosis: a systematic review." Emerg Infect Dis 8(8): 843-9.

Gonzalez, N., M. J. Torres, et al. (1999). "Molecular analysis of rifampin and isoniazid resistance of Mycobacterium tuberculosis clinical isolates in Seville, Spain." Tuber Lung Dis 79(3): 187-90.

Gopaul, K. K., T. J. Brown, et al. (2006). "Progression toward an improved DNA amplification-based typing technique in the study of Mycobacterium tuberculosis epidemiology." J Clin Microbiol 44(7): 2492-8.

Guidelines (1998). "Guidelines for surveillance of drug resistance in tuberculosis. WHO Geneva/IUATLD Paris. International Union Against Tuberculosis and Lung Disease." Int J Tuberc Lung Dis 2(1): 72-89.

Gutierrez, M. C., N. Ahmed, et al. (2006). "Predominance of ancestral lineages of Mycobacterium tuberculosis in India." Emerg Infect Dis 12(9): 1367-74.

H.McShane (2003). "Susceptibility to tuberculosis- the importance of the pathogen as well as the host." Clin Exp Immunol 133: 20-21.

Haas, F. (1996, Little, Brown and Co. Boston, Mass). "The origions of Mycobacterium tuberculosis and the notion of its contagiousness." Little, Brown and Co. Boston, Mass p.3-19.

Habib, F. and L. Baig (2006). "Cost of DOTS for tuberculous patients." J Pak Med Assoc 56(5): 207-10.

Hanekom, M., G. D. van der Spuy, et al. (2008). "Discordance between MIRU-VNTR and IS6110 RFLP genotyping when analyzing Mycobacterium tuberculosis Beijing strains in a high incidence setting." J Clin Microbiol.

Hasan, Z., M. Tanveer, et al. (2006). "Spoligotyping of Mycobacterium tuberculosis isolates from Pakistan reveals predominance of Central Asian Strain 1 and Beijing isolates." J Clin Microbiol 44(5): 1763-8.

Hillemann, D., S. Rusch-Gerdes, et al. (2007). "Evaluation of the GenoType MTBDRplus assay for rifampin and isoniazid susceptibility testing of Mycobacterium tuberculosis strains and clinical specimens." J Clin Microbiol 45(8): 2635-40.

Hirano, K., C. Abe, et al. (1999). "Mutations in the rpoB gene of rifampin-resistant Mycobacterium tuberculosis strains isolated mostly in Asian countries and their rapid detection by line probe assay." J Clin Microbiol 37(8): 2663-6.

Hirsh, A. E., A. G. Tsolaki, et al. (2004). "Stable association between strains of Mycobacterium tuberculosis and their human host populations." Proc Natl Acad Sci U S A 101(14): 4871-6.

123

Honore-Bouakline, S., J. P. Vincensini, et al. (2003). "Rapid diagnosis of extrapulmonary tuberculosis by PCR: impact of sample preparation and DNA extraction." J Clin Microbiol 41(6): 2323-9.

Hunter, P. R. and M. A. Gaston (1988). "Numerical index of the discriminatory ability of typing systems: an application of Simpson's index of diversity." J Clin Microbiol 26(11): 2465-6.

Hussain, A., Z. Mirza, et al. (2005). "Adherence of private practitioners with the National Tuberculosis Treatment Guidelines in Pakistan: a survey report." J Pak Med Assoc 55(1): 17-9.

Irfan, S., Q. Hassan, et al. (2006). "Assessment of resistance in multi drug resistant tuberculosis patients." J Pak Med Assoc 56(9): 397-400.

Iseman, M. D. and L. A. Madsen (1989). "Drug-resistant tuberculosis." Clin Chest Med 10(3): 341-53.

Isenberg, H. (2004). "Clinical microbiology procedure handbook." 2nd ed, vol.2: 7.8.2.1-7.8.2.3.

Israr, S. M. (2003). "Is Ministry of Health fully prepared to implement an effective DOTS program in Pakistan? An operations research on TB control program in the public health sector in Sindh." J Pak Med Assoc 53(8): 324-7.

Javaid, A., A. Ghafoor, et al. (2008). "Primary drug resistance to antituberculous drugs in NWFP Pakistan." J Pak Med Assoc 58(8): 437-40.

Javaid, A., R. Hasan, et al. (2008). "Prevalence of primary multidrug resistance to anti-tuberculosis drugs in Pakistan." Int J Tuberc Lung Dis 12(3): 326-31.

Jiao, W. W., I. Mokrousov, et al. (2007). "Molecular characteristics of rifampin and isoniazid resistant Mycobacterium tuberculosis strains from Beijing, China." Chin Med J (Engl) 120(9): 814-9.

Johansen, I. S., B. Lundgren, et al. (2003). "Direct detection of multidrug-resistant Mycobacterium tuberculosis in clinical specimens in low- and high-incidence countries by line probe assay." J Clin Microbiol 41(9): 4454-6.

Jou, R., H. Y. Chen, et al. (2005). "Genetic diversity of multidrug-resistant Mycobacterium tuberculosis isolates and identification of 11 novel rpoB alleles in Taiwan." J Clin Microbiol 43(3): 1390-4.

Kam, K. M. and C. W. Yip (2004). "Surveillance of Mycobacterium tuberculosis susceptibility to second-line drugs in Hong Kong, 1995-2002, after the implementation of DOTS-plus." Int J Tuberc Lung Dis 8(6): 760-6.

Kamerbeek, J., L. Schouls, et al. (1997). "Simultaneous detection and strain differentiation of Mycobacterium tuberculosis for diagnosis and epidemiology." J Clin Microbiol 35(4): 907-14.

Kelley, C. L., D. A. Rouse, et al. (1997). "Analysis of ahpC gene mutations in isoniazid-resistant clinical isolates of Mycobacterium tuberculosis." Antimicrob Agents Chemother 41(9): 2057-8.

Khan, J. A. and A. Malik (2003). "Tuberculosis in Pakistan: are we losing the battle?" J Pak Med Assoc 53(8): 320-1.

Koch, R. (1982). "Classics in infectious diseases. The etiology of tuberculosis: Robert Koch. Berlin, Germany 1882." Rev Infect Dis 4(6): 1270-4.

124

Kovalev, S. Y., E. Y. Kamaev, et al. (2005). "Genetic analysis of mycobacterium tuberculosis strains isolated in Ural region, Russian Federation, by MIRU-VNTR genotyping." Int J Tuberc Lung Dis 9(7): 746-52.

Kremer, K., B. K. Au, et al. (2005). "Use of variable-number tandem-repeat typing to differentiate Mycobacterium tuberculosis Beijing family isolates from Hong Kong and comparison with IS6110 restriction fragment length polymorphism typing and spoligotyping." J Clin Microbiol 43(1): 314-20.

Kremer, K., D. van Soolingen, et al. (1999). "Comparison of methods based on different molecular epidemiological markers for typing of Mycobacterium tuberculosis complex strains: interlaboratory study of discriminatory power and reproducibility." J Clin Microbiol 37(8): 2607-18.

Kurepina, N. E., S. Sreevatsan, et al. (1998). "Characterization of the phylogenetic distribution and chromosomal insertion sites of five IS6110 elements in Mycobacterium tuberculosis: non-random integration in the dnaA-dnaN region." Tuber Lung Dis 79(1): 31-42.

Laserson, K. F., L. E. Thorpe, et al. (2005). "Speaking the same language: treatment outcome definitions for multidrug-resistant tuberculosis." Int J Tuberc Lung Dis 9(6): 640-5.

Lavender, C., M. Globan, et al. (2005). "Molecular characterization of isoniazid-resistant Mycobacterium tuberculosis isolates collected in Australia." Antimicrob Agents Chemother 49(10): 4068-74.

Li, Q., C. C. Whalen, et al. (2002). "Differences in rate and variability of intracellular growth of a panel of Mycobacterium tuberculosis clinical isolates within a human monocyte model." Infect Immun 70(11): 6489-93.

Liefooghe, R., N. Michiels, et al. (1995). "Perception and social consequences of tuberculosis: a focus group study of tuberculosis patients in Sialkot, Pakistan." Soc Sci Med 41(12): 1685-92.

Lipin, M. Y., V. N. Stepanshina, et al. (2007). "Association of specific mutations in katG, rpoB, rpsL and rrs genes with spoligotypes of multidrug-resistant Mycobacterium tuberculosis isolates in Russia." Clin Microbiol Infect 13(6): 620-6.

Ma, X., H. Wang, et al. (2006). "rpoB Gene mutations and molecular characterization of rifampin-resistant Mycobacterium tuberculosis isolates from Shandong Province, China." J Clin Microbiol 44(9): 3409-12.

Marin, M., D. Garcia de Viedma, et al. (2004). "Rapid direct detection of multiple rifampin and isoniazid resistance mutations in Mycobacterium tuberculosis in respiratory samples by real-time PCR." Antimicrob Agents Chemother 48(11): 4293-300.

Mathema, B., N. E. Kurepina, et al. (2006). "Molecular epidemiology of tuberculosis: current insights." Clin Microbiol Rev 19(4): 658-85.

Mathuria, J. P., P. Sharma, et al. (2008). "Role of spoligotyping and IS6110-RFLP in assessing genetic diversity of Mycobacterium tuberculosis in India." Infect Genet Evol 8(3): 346-51.

Matsiota-Bernard, P., G. Vrioni, et al. (1998). "Characterization of rpoB mutations in rifampin-resistant clinical Mycobacterium tuberculosis isolates from Greece." J Clin Microbiol 36(1): 20-3.

125

Mazars, E., S. Lesjean, et al. (2001). "High-resolution minisatellite-based typing as a portable approach to global analysis of Mycobacterium tuberculosis molecular epidemiology." Proc Natl Acad Sci U S A 98(4): 1901-6.

McGrath, J. W. (1988). "Social networks of disease spread in the lower Illinois valley: a simulation approach." Am J Phys Anthropol 77(4): 483-96.

McHugh, T. D. and S. H. Gillespie (1998). "Nonrandom association of IS6110 and Mycobacterium tuberculosis: implications for molecular epidemiological studies." J Clin Microbiol 36(5): 1410-3.

Memon, A. R., M. A. Memon, et al. (2007). "Frequency of dual tuberculosis/human immunodeficiency virus infection in patients presenting at tertiary care centers at karachi." J Coll Physicians Surg Pak 17(10): 591-3.

Menca C, e. a. (1999). "Mycobacterium tuberculosis CDC 1551 induces more vigorous host response in vivo and in vitro, but is not more virulent than other clinical isolates." J Immunology 162: 6740-6746.

Meulemans, H. (2000). "Tuberculosis in Pakistan: The forgotten Plague." Mitchison, D. A. (1985). "[Mechanisms of the action of drugs in the short-course

chemotherapy]." Bull Int Union Tuberc 60(1-2): 36-40. Mokrousov, I., O. Narvskaya, et al. (2004). "Analysis of the allelic diversity of the

mycobacterial interspersed repetitive units in Mycobacterium tuberculosis strains of the Beijing family: practical implications and evolutionary considerations." J Clin Microbiol 42(6): 2438-44.

Mokrousov, I., O. Narvskaya, et al. (2002). "Phylogenetic reconstruction within Mycobacterium tuberculosis Beijing genotype in northwestern Russia." Res Microbiol 153(10): 629-37.

Morgan, M., S. Kalantri, et al. (2005). "A commercial line probe assay for the rapid detection of rifampicin resistance in Mycobacterium tuberculosis: a systematic review and meta-analysis." BMC Infect Dis 5: 62.

Moulding, T. S., H. Q. Le, et al. (2004). "Preventing drug-resistant tuberculosis with a fixed dose combination of isoniazid and rifampin." Int J Tuberc Lung Dis 8(6): 743-8.

Musser, J. M. (1995). "Antimicrobial agent resistance in mycobacteria: molecular genetic insights." Clin Microbiol Rev 8(4): 496-514.

Netto, E. M., C. Dye, et al. (1999). "Progress in global tuberculosis control 1995-1996, with emphasis on 22 high-incidence countries. Global Monitoring and Surveillance Project." Int J Tuberc Lung Dis 3(4): 310-20.

Newton, S. M., R. J. Smith, et al. (2006). "A deletion defining a common Asian lineage of Mycobacterium tuberculosis associates with immune subversion." Proc Natl Acad Sci U S A 103(42): 15594-8.

Nikolayevskyy, V. (2006). "Differentiation of tuberculosis strains in a population with mainly Beijing- family Strains." EID 12(9).

Nikolayevskyy, V., K. Gopaul, et al. (2006). "Differentiation of tuberculosis strains in a population with mainly Beijing-family strains." Emerg Infect Dis 12(9): 1406-13.

Niobe-Eyangoh, S. N., C. Kuaban, et al. (2004). "Molecular characteristics of strains of the cameroon family, the major group of Mycobacterium tuberculosis in a country with a high prevalence of tuberculosis." J Clin Microbiol 42(11): 5029-35.

126

Pablos-Mendez, A., M. C. Raviglione, et al. (1998). "Global surveillance for antituberculosis-drug resistance, 1994-1997. World Health Organization-International Union against Tuberculosis and Lung Disease Working Group on Anti-Tuberculosis Drug Resistance Surveillance." N Engl J Med 338(23): 1641-9.

Pfyffer, G. E., H. M. Welscher, et al. (1997). "Comparison of the Mycobacteria Growth Indicator Tube (MGIT) with radiometric and solid culture for recovery of acid-fast bacilli." J Clin Microbiol 35(2): 364-8.

Piersimoni, C. and C. Scarparo (2003). "Relevance of commercial amplification methods for direct detection of Mycobacterium tuberculosis complex in clinical samples." J Clin Microbiol 41(12): 5355-65.

Pineda-Garcia, L., A. Ferrera, et al. (1997). "DNA fingerprinting of Mycobacterium tuberculosis strains from patients with pulmonary tuberculosis in Honduras." J Clin Microbiol 35(9): 2393-7.

Prodinger, W. M., P. Bunyaratvej, et al. (2001). "Mycobacterium tuberculosis isolates of Beijing genotype in Thailand." Emerg Infect Dis 7(3): 483-4.

Pym, A. S., B. Saint-Joanis, et al. (2002). "Effect of katG mutations on the virulence of Mycobacterium tuberculosis and the implication for transmission in humans." Infect Immun 70(9): 4955-60.

Qian, L., C. Abe, et al. (2002). "rpoB genotypes of Mycobacterium tuberculosis Beijing family isolates from East Asian countries." J Clin Microbiol 40(3): 1091-4.

Ramaswamy, S. and J. M. Musser (1998). "Molecular genetic basis of antimicrobial agent resistance in Mycobacterium tuberculosis: 1998 update." Tuber Lung Dis 79(1): 3-29.

Rao, N. A., M. Irfan, et al. (2008). "Primary drug resistance against Mycobacterium tuberculosis in Karachi." J Pak Med Assoc 58(3): 122-5.

Rattan, A., A. Kalia, et al. (1998). "Multidrug-resistant Mycobacterium tuberculosis: molecular perspectives." Emerg Infect Dis 4(2): 195-209.

Reed, M. B., P. Domenech, et al. (2004). "A glycolipid of hypervirulent tuberculosis strains that inhibits the innate immune response." Nature 431(7004): 84-7.

Rossau, R., H. Traore, et al. (1997). "Evaluation of the INNO-LiPA Rif. TB assay, a reverse hybridization assay for the simultaneous detection of Mycobacterium tuberculosis complex and its resistance to rifampin." Antimicrob Agents Chemother 41(10): 2093-8.

Sahadevan, R., S. Narayanan, et al. (1995). "Restriction fragment length polymorphism typing of clinical isolates of Mycobacterium tuberculosis from patients with pulmonary tuberculosis in Madras, India, by use of direct-repeat probe." J Clin Microbiol 33(11): 3037-9.

Sajduda, A., A. Brzostek, et al. (2004). "Molecular characterization of rifampin- and isoniazid-resistant Mycobacterium tuberculosis strains isolated in Poland." J Clin Microbiol 42(6): 2425-31.

Sakula, A. (1983). "Robert Koch: Centenary of the Discovery of the Tubercle Bacillus, 1882." Can Vet J 24(4): 127-131.

Sawert, H., S. Kongsin, et al. (1997). "Costs and benefits of improving tuberculosis control: the case of Thailand." Soc Sci Med 44(12): 1805-16.

127

Sechi, L. A., S. Zanetti, et al. (1996). "Molecular epidemiology of Mycobacterium tuberculosis strains isolated from different regions of Italy and Pakistan." J Clin Microbiol 34(7): 1825-8.

Sharma, M., S. Sethi, et al. (2003). "Rapid detection of mutations in rpoB gene of rifampicin resistant Mycobacterium tuberculosis strains by line probe assay." Indian J Med Res 117: 76-80.

Sharma, R., M. Kalyani, et al. (2008). "Genetic diversity in clinical Mycobacterium tuberculosis isolates from Punjab." Int J Tuberc Lung Dis 12(10): 1122-7.

Sharma, S. K. and A. Mohan (2004). "Multidrug-resistant tuberculosis." Indian J Med Res 120(4): 354-76.

Siddiqi, N., M. Shamim, et al. (2002). "Molecular characterization of multidrug-resistant isolates of Mycobacterium tuberculosis from patients in North India." Antimicrob Agents Chemother 46(2): 443-50.

Silver, R. F., Q. Li, et al. (1998). "Expression of virulence of Mycobacterium tuberculosis within human monocytes: virulence correlates with intracellular growth and induction of tumor necrosis factor alpha but not with evasion of lymphocyte-dependent monocyte effector functions." Infect Immun 66(3): 1190-9.

Small, P. M., P. C. Hopewell, et al. (1994). "The epidemiology of tuberculosis in San Francisco. A population-based study using conventional and molecular methods." N Engl J Med 330(24): 1703-9.

Small, P. M., R. W. Shafer, et al. (1993). "Exogenous reinfection with multidrug-resistant Mycobacterium tuberculosis in patients with advanced HIV infection." N Engl J Med 328(16): 1137-44.

Soini, H., X. Pan, et al. (2000). "Characterization of Mycobacterium tuberculosis isolates from patients in Houston, Texas, by spoligotyping." J Clin Microbiol 38(2): 669-76.

Sola, C., A. Devallois, et al. (1999). "Tuberculosis in the Caribbean: using spacer oligonucleotide typing to understand strain origin and transmission." Emerg Infect Dis 5(3): 404-14.

Sola, C., I. Filliol, et al. (2001). "Spoligotype database of Mycobacterium tuberculosis: biogeographic distribution of shared types and epidemiologic and phylogenetic perspectives." Emerg Infect Dis 7(3): 390-6.

Sola, C., I. Filliol, et al. (2003). "Genotyping of the Mycobacterium tuberculosis complex using MIRUs: association with VNTR and spoligotyping for molecular epidemiology and evolutionary genetics." Infect Genet Evol 3(2): 125-33.

Sola, C., I. Filliol, et al. (2001). "Mycobacterium tuberculosis phylogeny reconstruction based on combined numerical analysis with IS1081, IS6110, VNTR, and DR-based spoligotyping suggests the existence of two new phylogeographical clades." J Mol Evol 53(6): 680-9.

Sreevatsan, S., X. Pan, et al. (1997). "Restricted structural gene polymorphism in the Mycobacterium tuberculosis complex indicates evolutionarily recent global dissemination." Proc Natl Acad Sci U S A 94(18): 9869-74.

Standards., N. C. f. C. L. (1995). "Antimycobacterial susceptibility testing for Mycobacterium tuberculosis. 15M24T[15]." National Committee for Clinical Laboratory Standards, Villanova, Pa.

128

Storla, D. G., Z. Rahim, et al. (2006). "Heterogeneity of Mycobacterium tuberculosis isolates in Sunamganj District, Bangladesh." Scand J Infect Dis 38(8): 593-6.

Sun, Y. J., R. Bellamy, et al. (2004). "Use of mycobacterial interspersed repetitive unit-variable-number tandem repeat typing to examine genetic diversity of Mycobacterium tuberculosis in Singapore." J Clin Microbiol 42(5): 1986-93.

Sun, Y. J., A. S. Lee, et al. (2004). "Characterization of ancestral Mycobacterium tuberculosis by multiple genetic markers and proposal of genotyping strategy." J Clin Microbiol 42(11): 5058-64.

Supply, P., C. Allix, et al. (2006). "Proposal for standardization of optimized mycobacterial interspersed repetitive unit-variable-number tandem repeat typing of Mycobacterium tuberculosis." J Clin Microbiol 44(12): 4498-510.

Supply, P., S. Lesjean, et al. (2001). "Automated high-throughput genotyping for study of global epidemiology of Mycobacterium tuberculosis based on mycobacterial interspersed repetitive units." J Clin Microbiol 39(10): 3563-71.

Supply, P., E. Mazars, et al. (2000). "Variable human minisatellite-like regions in the Mycobacterium tuberculosis genome." Mol Microbiol 36(3): 762-71.

Supply, P., R. M. Warren, et al. (2003). "Linkage disequilibrium between minisatellite loci supports clonal evolution of Mycobacterium tuberculosis in a high tuberculosis incidence area." Mol Microbiol 47(2): 529-38.

Suresh, N., U. B. Singh, et al. (2006). "rpoB gene sequencing and spoligotyping of multidrug-resistant Mycobacterium tuberculosis isolates from India." Infect Genet Evol 6(6): 474-83.

Swaminathan, S., R. Ramachandran, et al. (2000). "Risk of development of tuberculosis in HIV-infected patients." Int J Tuberc Lung Dis 4(9): 839-44.

Tanveer, M., Z. Hasan, et al. (2008). "Genotyping and drug resistance patterns of M. tuberculosis strains in Pakistan." BMC Infect Dis 8: 171.

Tazi, L., R. Reintjes, et al. (2007). "Tuberculosis transmission in a high incidence area: a retrospective molecular epidemiological study of Mycobacterium tuberculosis in Casablanca, Morocco." Infect Genet Evol 7(5): 636-44.

Telenti, A. (1997). "Genetics of drug resistance in tuberculosis." Clin Chest Med 18(1): 55-64.

Telenti, A., P. Imboden, et al. (1993). "Detection of rifampicin-resistance mutations in Mycobacterium tuberculosis." Lancet 341(8846): 647-50.

Thierry, D., M. D. Cave, et al. (1990). "IS6110, an IS-like element of Mycobacterium tuberculosis complex." Nucleic Acids Res 18(1): 188.

Torres, M. J. (2002). "Rapid detection of resistance associated mutations in Mycobacterium tuberculosis by Light Cycler PCR." Rapid cycle real time PCR- methods and applications: p. 83-91.

Torres, M. J., A. Criado, et al. (2002). "Rifampin and isoniazid resistance associated mutations in Mycobacterium tuberculosis clinical isolates in Seville, Spain." Int J Tuberc Lung Dis 6(2): 160-3.

Torres, M. J., A. Criado, et al. (2000). "Use of real-time PCR and fluorimetry for rapid detection of rifampin and isoniazid resistance-associated mutations in Mycobacterium tuberculosis." J Clin Microbiol 38(9): 3194-9.

129

Tracevska, T., I. Jansone, et al. (2003). "Prevalence of Beijing genotype in Latvian multidrug-resistant Mycobacterium tuberculosis isolates." Int J Tuberc Lung Dis 7(11): 1097-103.

Traore, H., K. Fissette, et al. (2000). "Detection of rifampicin resistance in Mycobacterium tuberculosis isolates from diverse countries by a commercial line probe assay as an initial indicator of multidrug resistance." Int J Tuberc Lung Dis 4(5): 481-4.

Traore, H., A. van Deun, et al. (2006). "Direct detection of Mycobacterium tuberculosis complex DNA and rifampin resistance in clinical specimens from tuberculosis patients by line probe assay." J Clin Microbiol 44(12): 4384-8.

Trudeau, E. L. (1887). "Environment in its relation to the progress of bacterial invasion in tuberculosis." Am. J. Sci 94: 118-123.

Tudo, G., J. Gonzalez, et al. (2004). "Study of resistance to anti-tuberculosis drugs in five districts of Equatorial Guinea: rates, risk factors, genotyping of gene mutations and molecular epidemiology." Int J Tuberc Lung Dis 8(1): 15-22.

Valway, S. E. (1998). "An outbreak involving extensive transmission of a virulent strain of Mycobacterium tuberculosis." N Engl J Med 338(10): 633-639.

van Embden, J. D., M. D. Cave, et al. (1993). "Strain identification of Mycobacterium tuberculosis by DNA fingerprinting: recommendations for a standardized methodology." J Clin Microbiol 31(2): 406-9.

van Soolingen, D., P. E. de Haas, et al. (1993). "Comparison of various repetitive DNA elements as genetic markers for strain differentiation and epidemiology of Mycobacterium tuberculosis." J Clin Microbiol 31(8): 1987-95.

van Soolingen, D., P. E. de Haas, et al. (2000). "Mutations at amino acid position 315 of the katG gene are associated with high-level resistance to isoniazid, other drug resistance, and successful transmission of Mycobacterium tuberculosis in the Netherlands." J Infect Dis 182(6): 1788-90.

van Soolingen, D., L. Qian, et al. (1995). "Predominance of a single genotype of Mycobacterium tuberculosis in countries of east Asia." J Clin Microbiol 33(12): 3234-8.

Varma-Basil, M., H. El-Hajj, et al. (2004). "Rapid detection of rifampin resistance in Mycobacterium tuberculosis isolates from India and Mexico by a molecular beacon assay." J Clin Microbiol 42(12): 5512-6.

Vermund, S. H. and N. Yamamoto (2007). "Co-infection with human immunodeficiency virus and tuberculosis in Asia." Tuberculosis (Edinb) 87 Suppl 1: S18-25.

Verver, S., R. M. Warren, et al. (2004). "Transmission of tuberculosis in a high incidence urban community in South Africa." Int J Epidemiol 33(2): 351-7.

Vukovic, D., S. Rusch-Gerdes, et al. (2003). "Molecular epidemiology of pulmonary tuberculosis in belgrade, central serbia." J Clin Microbiol 41(9): 4372-7.

Waaler, H. T. (2002). "Tuberculosis and poverty." Int J Tuberc Lung Dis 6(9): 745-6. Warren, R. M., T. C. Victor, et al. (2004). "Clonal expansion of a globally disseminated

lineage of Mycobacterium tuberculosis with low IS6110 copy numbers." J Clin Microbiol 42(12): 5774-82.

Watterson, S. A., S. M. Wilson, et al. (1998). "Comparison of three molecular assays for rapid detection of rifampin resistance in Mycobacterium tuberculosis." J Clin Microbiol 36(7): 1969-73.

130

Wayne, L. G. (1977). "Synchronized replication of Mycobacterium tuberculosis." Infect Immun 17(3): 528-30.

Wayne, L. G. and I. Krasnow (1966). "Preparation of tuberculosis susceptibility testing mediums by means of impregnated disks." Am J Clin Pathol 45(6): 769-71.

Wengenack, N. L., J. R. Uhl, et al. (1997). "Recombinant Mycobacterium tuberculosis KatG(S315T) is a competent catalase-peroxidase with reduced activity toward isoniazid." J Infect Dis 176(3): 722-7.

WHO (2008). "WHO Report 2008, Global tuberculosis control - surveillance, planning, financing."

WHO, R. (2003). "WHO annual report on global TB control--summary." Wkly Epidemiol Rec 78(15): 122-8.

WHO, R. (2008). "World Health Organization report: Global tuberculosis control." Yang, Z., P. F. Barnes, et al. (1998). "Diversity of DNA fingerprints of Mycobacterium

tuberculosis isolates in the United States." J Clin Microbiol 36(4): 1003-7. Yeh, R. W., A. Ponce de Leon, et al. (1998). "Stability of Mycobacterium tuberculosis

DNA genotypes." J Infect Dis 177(4): 1107-11. Yuen, L. K., D. Leslie, et al. (1999). "Bacteriological and molecular analysis of rifampin-

resistant Mycobacterium tuberculosis strains isolated in Australia." J Clin Microbiol 37(12): 3844-50.

Zhang M, e. a. (1999). "Enhanced capacity of a widespread strain of Mycobacterium tuberculosis to grow in human macrophages." J Infect Dis 179: 1213-1217.

Zhang, Y., B. Heym, et al. (1992). "The catalase-peroxidase gene and isoniazid resistance of Mycobacterium tuberculosis." Nature 358(6387): 591-3.

131

Appendices