AIDSrelated outbreak RFLP

1
AIDS-related outbreak - RFLP

• First 2 cases unrelated
• 3 case started outbreak
• 12 cases in 100 days
• Min. incubation period just 24 days
• Recommend changes to TB control in face of AIDS

From Daley et al., NEJM 1992
2
Molepi of outbreaks

• Dozens of papers
• Spread in unexpected settings
• Gas station / Airplane / Lap dancers
• Spread across species
• Humans to circus elephants in US
• Spread across borders
• MDR-TB from NY repeats American in Paris
• Russian strains do Moscow on the Hudson

3
Population based studies

• Outbreak isolates share genotypes in face of
  diverse background
• Therefore If all isolates in community typed,
  those with same genotype likely represent
  outbreaks
• Called clustering
• Clustered cases surrogate for recent transmission
  
• Unique cases surrogate for reactivation of
  organism acquired remotely (time/place)

4
Population studies - Why?

• Incubation period highly variable
• Outbreaks not evident
• 200 cases in a community may be
• 200 people reactivating
• 100 reactivators with 1 secondary each
• Many small clusters, e.g. Foreign-born TB,
  Montreal
• 1 reactivator with 199 secondary cases
• New introduction in naïve population, e.g.
  European settlers in Western Canada / Innuit
• Implications for TB control

5
Molepi of TB in communities

• Clustering as surrogate for transmission
• 30-40 TB in SF, NY due to ongoing epidemic
  spread
• Small et al, NEJM, 1994
• Alland et al, NEJM, 1994
• Only 10 of transmission suspected by routine
  public health invest.
• Risk factors for clustering US born, males,
  middle age, white, black

6
Molepi around the world

• South Africa
• 50-60 TB recent transmission
• Higher in gold mines
• Montreal
• TB rates low
• lt 20 TB recent transmission
• Higher clustering in some foreign-born groups

7
Lessons from TB molepi

• TB is an old disease, epidemiology already worked
  out by Koch in theory
• Molecular approaches have provided refined
  picture of
• Current situation
• Aid clinical studies
• Measure control approaches
• Molecular clusters serve to tag circulating
  clones
• Permit further bacteriologic study

8
What would you want to know about a M.
tuberculosis isolate?

• Are all bacteria created equal?
• If not, are some burly and other wimpy?
• Are there means of studying them to try to
  determine the good/bad/ugly?
• Genetic basis
• Previously piecemeal, now genomic
• Phenotypic study
• Models, e.g. mice
• Natural settings epidemiology

9
Genetics starting point is M. tuberculosis
H37Rv

• 4.4 million base pairs
• 65.6 GC content
• available on internet
• 3924 predicted genes
• what do they do?
• how do they work together to kill?

M. Tb genome
Cole ST et al. Nature 1998 393 537.
10
Example 1 Genomic study of Bacillus of Calmette
Guerin

• Original strain Mycobacterium bovis
• Attentuated 1908-1921 by 230 in vitro passages on
  potato-bile
• Further passaged in BCG labs for half century
  until lyophilization

11
Why study BCG vaccines?

• What is known
• 118 million newborns vaccinated
• Good safety profile
• Not about to disappear

• What is not known (partial list)
• Do BCG vaccines prevent TB?
• How safe where most needed (high HIV)

What are we giving and why are they attenuated?
12
Predicted genetics of BCG

• Since BCG came from M. bovis, compared
  to H37Rv, expect
• Differences between M. tuberculosis M. bovis
• Differences between virulent M. bovis and BCG
• Differences between different BCG strains

M. tb genome