Tuberculosis: it’s not the clone wars

Digitally colored scanning electron microscope image of Mycobacterium tuberculosis
Digitally colored scanning electron microscope image of Mycobacterium tuberculosis

Whole genome sequencing allows researchers to quantify diversification of Mycobacterium tuberculosis within a host, showing that the individual bacteria comprising an infection are not all identical clones. Using a known TB cluster, Emerging Pathogens Institute researchers confirm that hosts can transmit all — or just some — of the genetic diversity unique to their mTB population to a new victim.

Tuberculosis genotyping is commonly used to trace transmission between victims in settings where the disease is uncommon. Conventional whole genome sequencing of this pathogen is rooted in the idea that the disease is caused by specific bacterial clones. But, unlike Star Wars’ fictive Clone Wars where each droid is a precise replica, real life Mycobacteria tuberculosis invaders harbor small but detectable differences.

New research by UF Emerging Pathogens Institute and Danish Statens Serum Institut investigators, published this month in EBioMedicine, reveals that not only does the bacterial agent which causes tuberculosis diversify slightly within a host, but also that some or all of this variation can be passed to subsequent hosts. The finding that Mycobacterium tuberculosis, or mTB, variation is transmissible holds implications for how both sampling and genotyping techniques can be modified to better track the disease’s spread.

Once a victim becomes infected with mTB, its population within a single host begins to change in minute by trackable ways. While some people develop active tuberculosis right away, others may remain carriers while the germ “sleeps” latently within their body. (Compared to other infectious agents, mTB is notably slow growing and slow to mutate.)

It was once thought that latency itself caused bacterial diversification — small repeating changes in DNA base pairs — as well as selective pressures from drug therapies.

But a new study, led by UF’s Nancy Séraphin, Ph.D., and colleague Anders Norman, confirms that while latency does fuel this mini-diversification, other forces are at play too and they can be used to better understand and trace transmission links. UF’s Michael Lauzardo, M.D., and colleague Troels Lillebaek are the paper’s senior authors. Séraphin and Lauzardo are affiliated with both the EPI and UF’s College of Medicine, department of medicine, division of infectious diseases and global medicine.  Norman and Lillebaek are with the International Reference Laboratory of Mycobacteriology, Statens Serum Institut, and the  University of Copenhagen’s department of public health in Denmark.

Working with the Florida Department of Health TB Program and Bureau of Public Health Laboratories, the team used a well-known tuberculosis cluster in Florida where the transmission sequence was already established to test how whole genome sequencing techniques could be used to further refine existing genotyping methods used to determine transmission sequences. They cultured samples from three patients, who coughed up sputum, and then mapped deviations from the parental “clone” genetic profile as small as just a few switched base pairs.

When base pairs located at the same position on a genome are found to be switched or to have a varied pattern, it is called a single nucleotide polymorphism, or SNP for short. This is the most common type of genetic variation between organisms of the same type.

“When we sequence the bacteria, we get these pairs of letters G-C, and A-T,” Séraphin says. “But the sequencer can also capture instances or positions in the genome where it could be one or the other of these two letters. When you compare these to a reference, the sequencer will tell you that 80 percent of the time it is at A and 20 percent of the time it is at C, for example.” The frequency of these swapped base pairs, or minority alleles, comprise what the researchers term “sub-clones.”

“These sub-clones, or variations, may dictate the type of disease someone gets,” Séraphin says. “How long it takes them to be treated, or how responsive they are to the drugs used to treat them, and the fitness or success of that strain of mycobacteria to transmit within a population.”

The different sub-clones detected in the study were too small to qualify as comprising a new strain of tuberculosis, but detectable enough to track their journey from a primary host into a second.

But transmission, too, is variable. Secondary hosts may be exposed to the full genetic scope of differences, or they may acquire only a small part of it. Imagine standing near someone who tosses a handful of marbles in the air. One marble may strike you, or six.

It’s similar when a TB carrier coughs or sings: tiny droplets carrying mTB hitchhikers become airborne, and when someone else in close proximity inhales, they may breathe in bacteria representing the full suite of sub-clones unique to the primary carrier, or perhaps just one.

The research team, with the assistance of the Florida Department of Health TB program, identified that one person from the cluster, the index case, gave the disease to two other secondary hosts.  “We found six sub-clonal populations of mTB in the index case,” Séraphin says. “We then found all six sub-clones in one secondary case, but only one sub-clone in the other secondary case. This shows variation and selective forces in transmission.”

They also found that the variability within a single host’s mTB population changed over time; and that even though variation can be seeded during the transmission process, latency can then select against specific sub-clones, furthering obscuring transmission links.

A seminal study previously proposed using a cut-off of five SNPs to determine transmission links between infected people. “Looking at sub-clonal populations can help us to better define and analyze transmission links,” Séraphin says. “But it also shows that a standard cutoff is not applicable to every situation.” In other words, every outbreak may have a different profile for what sub-clonal variation looks like, so any “cut off” threshold would need to be derived from the specific traits unique to each outbreak.

“We knew from the beginning that variation matters,” Séraphin says. “But here, we show how important the role of selection now becomes in transmission. This can be used to rethink how to use whole genome sequencing for establishing transmission links.”

EPI Explainer

What caused the genetic variation the team found? One possibility is that it was due to the founder effect. This describes what happens when a small sample of organisms, derived from a wider and more varied population, establishes a new colony; and the reduced genetic variation of this colony and its offspring reflects the genomes of the founders. But Séraphin and colleagues ruled this out, instead interpreting the variability as more in line with genetic drift. This is another evolutionary concept which describes changes in allele frequencies within a population as being selected for, or against, due to random chance. In other words, some allele frequencies may become amplified, and others may blink out, even in the absence of adaptive selective forces; genetic variation that is selected for, or against, in this fashion tends to have no effect on an organism’s fitness.

By: DeLene Beeland