MOLECULAR STRATEGIES TO CONTROL TUBERCULOSIS AND DEVELOP NOVEL 
VACCINES AGAINST THIRD WORLD DISEASES 
William R. Jacobs, Jr., Ph.D., Assistant Investigator 
Tuberculosis, caused by Mycobacterium tuber- 
culosis, has reemerged in the United States as a tre- 
mendous pubhc heahh problem, with increasing 
numbers of new cases for the last six years. In addi- 
tion, M. tuberculosis strains have emerged that are 
resistant to all known chemotherapeutic agents. 
Worldwide, over 8 million new cases of tuberculo- 
sis are reported each year, since the disease has re- 
mained a problem throughout the developing 
world. The 3 million deaths it causes annually make 
it the leading infectious killer in the world today. 
Dr. Jacobs' laboratory uses molecular genetic ap- 
proaches to investigate the biology of M. tubercu- 
losis. 
Epidemiological Analysis 
of Tuberculosis Infections 
Restriction fragment length polymorphism (RFLP) 
analysis can play a key role in determining the mode 
of transmission of tuberculosis, as individual iso- 
lates can be tracked from one infected individual to 
the next. The recent increases in the incidence of 
tuberculosis in the United States seem to be asso- 
ciated with the AIDS epidemic. However, it is un- 
clear whether the tuberculosis of AIDS patients 
results from reactivated disease, reflecting an expo- 
sure to the tubercle bacillus prior to infection with 
AIDS, or from a first-time infection. In collabora- 
tions with others, including Drs. Peter Small and 
Gary Schoolnik (HHMI, Stanford University), analy- 
ses were undertaken to distinguish between these 
two possibilities. 
Difl'erent isolates of M. tuberculosis have differ- 
ent RFLP patterns when probed with a particular 
DNA element found in M. tuberculosis strains. RFLP 
analysis of the strains isolated from a recent tubercu- 
losis outbreak among AIDS patients in a group home 
revealed that 1 1 individuals had all been infected 
with the identical strain of M. tuberculosis. This 
demonstrates that AIDS patients are highly suscepti- 
ble to infection with M. tuberculosis and confirms 
that tuberculosis is highly contagious. The clear re- 
sults of these studies should be translatable into bet- 
ter public health care policies. Similar analyses are 
under way to track M. tuberculosis isolates that are 
multiply drug resistant. 
Genetic Analysis of Mycobacterial 
Virulence Determinants 
By understanding how mycobacteria infect and 
cause disease in mammalian hosts, it should be pos- 
sible to develop effective measures to treat and 
prevent mycobacterial disease. One approach to 
understanding the virulence determinants of M. tu- 
berculosis is to identify and characterize the genes 
responsible for the virulence phenotype. Gene 
function, such as virulence, can be defined by gen- 
erating well-defined mutants and comparing the 
mutant strain with the wild-type bacterium. Dr. Ja- 
cobs' laboratory has focused on developing methods 
to generate well-defined mutations in mycobacteria. 
Methods of mutagenesis using chemical or physi- 
cal agents to damage DNA, coupled with the screen- 
ing of colonies arising from single clones of cells, 
are particularly unsuitable for the mycobacteria. 
The organisms grow in clumps, not as single-cell 
suspensions, as a result of their complex cell wall 
structure. Insertional mutagenesis would be a more 
effective way of mutagenizing mycobacteria, as this 
method most often employs a transposon that con- 
tains a selectable marker gene. Colonies of mutated 
cells are selected for by plating cells in the presence 
of a selecting agent. The agent prevents the growth 
of the nonmutagenized cells clumped to cells that 
contain a transposon, and thus the colonies that 
arise are a pure population of a mutagenized clone. 
When Dr. Jacobs' laboratory initiated this work, 
no useful transposons that functioned in mycobac- 
teria had been identified. In the absence of a trans- 
poson, a shuttle mutagenesis approach was success- 
fully employed. This involved the cloning of DNA 
fragments from Mycobacterium smegmatis, BCG 
(bacille Calmette-Guerin), and M. tuberculosis 
into vectors that could not replicate in mycobac- 
teria. Transposons were introduced into the cloned 
mycobacterial DNA fragments in Escherichia coli. 
Then the transposon-mutated DNA fragment was 
reintroduced into the mycobacterial chromosome 
via homologous recombination of the DNA frag- 
ments on the adjacent sides of the transposon. 
This system was useful for generating mutations 
in both M. smegmatis and M. tuberculosis. Surpris- 
ingly, however, analyses revealed that M. tubercu- 
losis displayed a high degree of illegitimate recom- 
bination as compared with M. smegmatis and that 
the illegitimate recombination process was not terri- 
bly efficient. 
Serendipitously, a novel insertion element, 
IS 1096, was discovered in Dr. Jacobs' laboratory. 
IS 1096 has been demonstrated to possess many 
properties that make it attractive as a potential inser- 
tional mutagen of M. tuberculosis. It is small, it 
IMMUNOLOGY 335 
