eLife episode 18: TB, and a Handshake

TB, HIV, the smell of a handshake and what makes ants social...
09 March 2015
Presented by Chris Smith




In this episode of the eLife podcast we hear about TB, HIV, social behaviour in ants, genetics in baboons and a surprising twist to the handshake.

In this episode

00:38 - Taking a close look at TB

Whole genome sequencing is providing new insights into the spread of different lineages of TB

Taking a close look at TB
with Judith Glynn, London School of Hygiene and Tropical Medicine

TB kills up to 2 million people per year but despite being relatively well studied, how the bacterium spreads in populations and how differences in the genetic makeup of TB Glynnthe bugs affect that transmission isn't well understood. Judith Glynn from the London School of Hygiene and Tropical Medicine is now using a molecular genetic approach to look for answers.

Judith - What we've done in our study in Northern Malawi is study TB over a very long period and we try and identify everybody with tuberculosis in the population. And then by looking at the actual strain of TB they're infected with, in detail we can begin to work out who might have transmitted to whom and therefore, understand transmission in more detail.

Chris - This is by comparing the genetic sequences of the different isolates from the different patients and so I presume you're finding unique genetic sequences that go with certain  individual cases of TB and you can then track where they go through the population.

Judith - Yes. So, we're looking at the whole of the genetic code of the TB in these different people. Previous studies had used much cruder techniques and they told us quite a lot about where TB was being transmitted, but this new technique known as, whole genome sequencing, we get almost the whole genome. We compare it between people and look at the number of changes in the TB carried by different patients. And by comparing the number of differences we can say who might have transmitted to whom.

Chris - When you look at the trends that come out of the study between who gets infected and who gives what to whom. Are there any sort of generalizations and general points that emerge? And also are there any surprising points? Are there things that emerge from this study that actually we haven't seen before or didn't expect to see?

Judith - To start with, what we expected to see. We find more transmission from people who've got smear positive disease. That's to say they've got so many bacilli in their sputum that you can actually see them under a normal microscope. So, those people transmitted more than people who haven't got that level of disease. We know that, but it shows that the method is working, that we can actually see that in the data. Now to move on to what's sort of newer is the difference between the lineages of TB, these different strains of TB which divided into lineages. So, some previous studies had suggested differences between lineages. What's important in this study is that the population has all of the four different lineages but they're not linked to particular population groups. In studies in the West, you often find a particular lineage is much more common in particular immigrant groups and you might get different transmission dynamics within those groups and different probabilities of transmission. In this population, there was no association with immigration status into this area, which is actually a rural area in Northern Malawi. And so, the differences that we found by lineage are probably more likely to tell us about real differences between lineages in terms of transmissibility.

Chris - And what are the implications of what you found in this study?

Judith - I think it will help us understand more about why some people get ill and some don't. The different lineages really do seem to be behaving differently. For example, the Lineage-1 was the least transmissible and that's been suspected before, Lineage-1 strains seem to come from Southern India. And it's been thought that the TB there was a little bit less virulent. Similarly, Lineage-2, that one seemed to transmit more and again, fits with what people have suspected. Less known was that Lineage-3 also seemed to transmit more and that hasn't been shown previously.

Chris - What keeps these things circulating then if they're at this sort of disadvantage relative to other strains? Why haven't they slowly, by evolution, just been selected out and outcompeted by the more transmissible forms?

Judith - One possibility is the effect of HIV. And interestingly Lineage-1 which is the one which appears to transmit less was more common in people with HIV. So, that would fit with it being less likely to give rise to active disease in people who are not immuno-suppressed.

Chris - Do you think your findings are generalizable if we were to go out of this population in Malawi and drop into another African country, an Asian country, or even into the middle of London. Would we see the same phenomenon?

Judith - I think the differences between the lineages are probably biological in that there don't seem to be differences within the population that explain them here. Whether you would see the same in other populations where there may be differences in whom the different lineages are in if they're in different immigrant groups with different mixing patterns, you might see different effects. But to the extent they're biological differences, and then you should actually see them in different populations. And I think in terms of why it matters is it can give us clues as to what the virulence factors might be. So, the next step would be to actually look in much more detail at what other differences between the lineages at a genome level that might explain some of these differences in virulence and transmissibility. Does that give us clues to, for example, drug design or even an understanding why some people get sicker than others?

06:21 - More to a handshake than meets the eye

Sniffing our hand after a handshake may allow us to detect chemical signals produced by others

More to a handshake than meets the eye
with Idan Frumin, Weizmann Institute

The last time you greeted someone, did you shake their hand? Now a harder question - if you did shake hands, what did you do with your hand afterwards? You probably didn't realize it at the time but the likelihood is that you subconsciously brought it up to your face and sniffed your fingers. Using hidden cameras, Israel's Weizmann Institute researcher, Idan Frumin, has found that we humans effectively do the more socially acceptable equivalent of what two dogs do when they first meet.

Idan - People tend to sniff their own hands following a handshake. We first noticed that anecdotally. We just saw people do that after meeting new people and we set to find out if it's really something that we can describe as an effect. So, we devised a very simple experimental design that would allow us to film people without their knowing it and see if it is the case.

Chris - How did you actually do it then? Were you literally asking people to shake hands and then see what they did afterwards?

Idan - The design was very simple. We put people in a room without them knowing they are being filmed. After about 2 minutes, an experimenter went into the room and either shook or didn't shake their hands. The control, not shaking hands and the actual experiment was with shaking hands. We filmed them for an additional 2 minutes. We measured the time their hands spent near their noses, we focused on the right hand because this was the hand that was shaking but we also looked at the left hand which was not shaking. The results were a very substantial increase in the time that the right hand spent near the nose when there was a handshake.

Chris - Your interpretation of that would be that having shaken someone's hand, there's been a transfer of chemicals from the skin of one to the other. And therefore, if you give your hand a sniff, you're effectively sampling the chemical makeup, the chemical fingerprint, the odour profile of your opposite number.

Idan - Right. This is what we think and we also checked that. We actually measured if we can transfer chemicals from one person to another. We put gloves on an experimenter and shook their hands to see if something is left over from a handshake on this glove. And we saw that there are a host of chemicals that are transferred.

Chris - Is there evidence that when a person brings their hand to their nose they're actually sniffing their fingers? Could it not be that they're just rubbing their face or something?

Idan - This was one of our main concerns. To show that this is indeed something to do with the sense of smell, we measured that using a nasal cannula. It's basically a tube placed under the people's nose. We can measure the airflow using this instrument and we unequivocally saw that when their hand approached the nose, there is a great increase in airflow, meaning there was a sniff involved.

Chris - I don't know about you, Idan, but when I read your paper the first instinct I had was to immediately look down to see what I was doing with my hands and where they were. And I became conscious of every time after that I began to bring them close to my face. It's one of those awful things a bit like when you see someone yawn and you want to yawn, and someone says they're itching and you want to scratch. Are you now sort of obsessed with what you do with your hands?

Idan - Pretty much so, and I also observe that all the time. And it's very funny to see when we present that at conferences how people start to observe each other and themselves, and to see what they do with their hands after they shake hands. It's pretty amusing, yes.

10:11 - HIV Pre-exposure Prophylaxis

What are the side effects when Tenofovir gel is used as a prophylactic against HIV?

HIV Pre-exposure Prophylaxis
with Florian Hladik, Fred Hutchinson Cancer Research Center, University of Washington

Something that you can't pick up from a handshake is HIV, but thousands of people are infected with the virus everyday through sexual contact. One strategy to prevent this is referred to as pre-exposure prophylaxis. The idea is that an at-risk individual can apply a gel laced with an anti-virus agent to potential exposure sites, and this stops the virus gaining a toe-hold, but to make this work, you need to use very high concentrations of the drug and this could have unexpected effects on the tissue as Florian Hladik has discovered...

Florian - If we use a drug topically, it has been shown that the concentrations locally in the mucosa are much higher than if you take the same drug orally. And so there's some concerns that if you achieve much higher concentrations that you might also encounter side-effects. We were interested to look at the effect of microbicides across the whole human genome. Essentially check whether we can find certain patterns of expression that indicate problems.

Chris - And how did you approach this? What were you actually doing in your study?

Florian - The study had four trial arms, and in each arm there were 20 patients. One arm had no drug used, one arm had a control gel, one arm the patients received a gel containing Nonoxynol-9 a detergent that is an irritant as positive control, and then the fourth arm the patient received Tenofovir 1% gel. When the patients were enrolled, before starting the study drug, biopsies were taken from the rectum in all four study arms. Then after the first dosage, so that would be about 1 hour after receiving the first dosage, another biopsy was taken. After that, the patients received the study drug daily over 7 days and at the end of that period we took biopsies again.

Chris - And how did you then process those biopsies? What did you do to see what the Tenofovir was or wasn't doing to the tissue?

Florian - Biopsies were put in a preservative for RNA isolation and then after that we ran Illumina microarray chips where with certain probes, you look at expression changes of genes across the whole human genome.

Chris - So, you're able to then say what genes are being turned on or off in each instance to see what effects there were or weren't on the genes in those tissues?

Florian - Yeah, the first readout really is the number of genes that are changing. So, essentially how many genes are upregulated and how many genes are downregulated by the treatment. And that was a real surprise to us because we were expecting all the changes in the Nonoxynol-9 arm which we had used as a positive control arm. There was one arm that showed by far the most changes and we were assuming it must be the N-9 arm but it actually turned out the Tenofovir arm.

Chris - What is the Tenofovir doing then? The fact that it's actually producing these gene changes argues that you're getting what are called off-target effects because Tenofovir should be affecting only virus function, shouldn't it? So the fact that you're seeing gene changes argues that it is doing something to those cells. Which genes were being impacted?

Florian - We saw a lot of impact on transcription factors. Tenofovir, overall had relatively strong inhibitory effect on many transcription factors. One of them was a transcription factor called, CREB1 and also another one, CREB-binding protein. And these are transcription factors that are important for transcription of Interleukin-10. The Interleukin-10 is regulating, toning down an immune response that has been set in motion by a pathogen and by inflammatory cytokines that has this kind of action. That would indicate that Tenofovir inhibits the anti-inflammatory arm of immunity.

Chris - So, if Tenofovir in this setting can de-repress inflammation, in other words it can facilitate an inflammatory process. That means long-term exposure to it might have consequences that we hadn't anticipated then.

Florian - That could be. And the hypothesis from our data is that this inflammation prolonging effect of Tenofovir would come into play only if you already have a background level of inflammation due to some other reasons. In particular in people who are in need of using these kinds of strategies to prevent HIV transmission, inflammation is often present.

Chris - Certainly quite a worry.

14:49 - The division of labour in ants

Neuroscientists are studying the link between genetics and social behaviour in ants

The division of labour in ants
with Timothy Linksvayer, University of Pennsylvania

Ants, bees, and wasps - these are social insects and they live in colonies containing thousands of individuals and everyone in those colonies knows their Ant behaviourrole. Younger individuals tend to adopt nest-based nursing and cleaning tasks while more senior insects take on riskier foraging roles outside. To find out what role genetics plays in controlling these behaviours. Tim Linksvayer has been looking at which genes are turned on or off in younger and older ants, as well as in other social insect species.

Tim - We knew that a worker individual's age influenced it's behaviour. So, we looked at the patterns of gene expression. So, which genes were turned off or on in worker individuals as they aged. So, across age categories and we also looked across behaviours - specifically nursing and foraging. And then by comparing those different categories of individuals, those different samples of individuals, we could see which genes are turned off or turned on in say nurses compared to foragers or in individuals that were young versus old.

Chris - And how many genes did change when you compared the animals that are behaving inside the nest as nurses, compared to the ones then on foraging duty?

Tim - So, overall we found about 2400 genes that changed. So, 1200 genes that were upregulated in foragers, and 1200 genes that were upregulated in nurses.

Chris - Given that you've got this long list of genes now, and you can look across the spectrum of social insects. Why is this important? What can we do with this gene family that you've identified that we couldn't do based on previous studies?

Tim - I actually think the most exciting part of this research is that we asked not only what proportion of high conserved genes versus novel genes are involved in social behaviour. We asked what features of genes affect their evolutionary rate and affect how they influence social behaviour. And specifically, we looked at the regulatory context. So, it's known in other model systems that genes that are highly connected - you can imagine a gene exists within a cascade so one gene is turned on and then it affects whether another gene is turned on or another gene. And it's been found that genes that influence many other genes then have a strong effect on the traits expressed by the organism. And these genes are more conserved, whereas genes that have a lesser effect and that are loosely connected, that are not very connected, these genes are not conserved, which makes sense. Genes that are doing a lot, that are really important, are conserved. Genes that aren't doing as much are less conserved. And so, our study, we showed how this regulatory context affects the evolution of genes associated with social behaviour. As expected, we found that genes that were highly connected and that were strongly affecting the trait, we looked at the social behaviour, we looked at those genes were more conserved than genes that were less highly connected. Those genes were more rapidly evolving.

Chris - And can you go a step further and say whether or not there are more conserved genes in the nursing behaviour than the foraging behaviour, or are they equally represented between both behaviours?

Tim - Yeah. That was one interesting, and I would say, unanticipated result. We found that the genes that were upregulated in foragers, we found that those genes were more highly conserved. And keep in mind that the foragers are old, so it's like kind of amusing result where the older individuals, their behaviour is controlled by more conserved genes where the younger individuals, the nurses, their behaviour was controlled with more rapidly evolving genes.

Chris - Did that surprise you? Because it's hard to envisage how you can have one behaviour without the other and they're both equally important. So, why should one particular behaviour be more conserved at a genetic level than the other?

Tim - Yes. It is somewhat surprising. We explained it by arguing that the nurses are in fact doing more different things. So nurses, nurse individuals. So, they provide food. But they also do other things in the nest, so they clean the nest, they groom each other, they groom brood, as opposed to foragers - you could say have more programmed tasks. They're leaving the nest going to collect food, coming back, and they're also older. And actually, realistically, the age is probably very important as well. As the ants age, their metabolism is changing dramatically. And you might imagine that older individuals, there's essentially less physiological processes turned on. That could explain why it's essentially a more simple set of genes, it's more highly connected, and more conserved set of genes that's affecting behaviour in the older foragers.

19:56 - What baboons can tell us

Researchers are exploring the genetic architecture of gene expression in baboons and other primates

What baboons can tell us
with Jenny Tung, Duke University

Individuals can vary because the genetic sequences they carry will be different. But variation can also occur, not just by changing the sequence of a gene but by Baboonsaltering its level of expression. As Jenny Tung, who's been looking at this question in wild baboons puts it, it's a bit like making two different cake recipes. Both are going to contain eggs, and milk, and flour. But if you alter the relative proportions of each, you get quite a different cake.

Jenny - It turns out that the protein coating sequence, so the ingredients in the cake as it were, are often very, very similar between different primate species or among different individuals within the same species. That suggests the amount of gene product that's being produced might be really important. And until recently, it was very hard to measure those values, at least on a genome wide scale. We knew that, theoretically, this should be important but we haven't been able to back that up with empirical data from real primates living in real populations until now.

Chris - What's changed that means that you can now do that?

Jenny - Well. There have been a lot of technical advances in really the last five to ten years that have meant you can measure whether genes are turned up or turned down in pretty much any species that you can get your hands on. Now we can actually look directly at the sequence products of the genes. And there's more of it if the gene is expressed a lot and there's less of that sequence product in the cells if the genes aren't expressed very much. You can also get a sense of how genetically different individuals are, and it's connecting those two things that we did in this paper to ask about whether genetic differences between individuals play a role in how many eggs go in that cake.

Chris - So, which species did you study?

Jenny - So, here we are studying Yellow Baboons. And the reason we did so is because we were able to take advantage of samples we collected from a really special population that I'm very fortunate to work with. It's a wild population of baboons in Kenya that's been studied since 1971. So, we actually know a lot about these animals, although we didn't know very much about their gene expression profiles until now.

Chris - And what did you take away from them to measure?

Jenny - So, these were blood samples. We just purify RNA from the cells and we're able to actually look at the DNA sequence that produced that RNA.

Chris - And you're using the RNA as an index of gene expression. The more RNA there is, the more gene products there is, and therefore, the more active that particular gene that produced that RNA molecule originally must be?

Jenny - That's exactly right. Yes. And because we're using actual sequence data, so we're actually looking at the sequence of the genes as a measure of how much the gene was expressed. We also kind of get information about genetic variation in that sequence for free.

Chris - You study a big group of animals, you do this, you get the RNA profiles from them. What did it tell you?

Jenny - One thing it told us is that genetic differences between individual baboons play a really important role in determining, whether, gene expression levels tend to be high or low. This is pretty consistent with what we know in humans. But it turned out that it was actually much easier to identify these kinds of genetic effects on gene expression levels in the baboons than it has been in humans so far, suggesting that baboons might in fact be more genetically variable than your average human population. Those genes in baboons were also likely to be related to genetic differences among individuals in humans. So, genetic effects on gene expression levels in baboons are, to some degree, similar to the sets of genes that show the same pattern in humans. And that was interesting and surprising because it suggests that some genes are kind of more free to vary than others. Natural selection makes some genes very constant in the way that they are expressed, whereas other genes are allowed to vary in populations more freely.

Chris - What sorts of things do these genes that are changing very rapidly in the baboons specify in the body? What jobs do they do and what jobs are they controlling?

Jenny - Some of those genes are involved in immune defence, again, suggesting that when you're job is to cope with exposure to pathogens, being able to have a lot of variation and being able to evolve very rapidly might be important. Other kinds of things that we see are variation in genes that would respond to other kinds of environmental stimuli. Things that help your body cope with the environmental changes that it faces day to day.

Chris - Have you got a handle on the role of epigenetics here? Because obviously you're looking at gene expression levels but have you any handle on the role that the epigenetic influence might be playing in what you're seeing or is that mixed in there and you can't tell?

Jenny - Based on the data set that we reported on here, we can't tell because we didn't do any direct measures of the epigenome. But it's very tempting to speculate that some of the, particularly, the long-term influences of social environment on gene expression might be mediated by aspects of the epigenome because some of those things are known to be stable over quite some time. It's actually a great question. It's something that we're looking at in my lab now.


Add a comment