Hello, and thank you for listening to the MicroBinFeed podcast. Here, we will be
discussing topics in microbial bioinformatics. We hope that we can give you some
insights, tips, and tricks along the way. There is so much information we all
know from working in the field, but nobody writes it down. There is no manual,
and it's assumed you'll pick it up. We hope to fill in a few of these gaps. My
co-hosts are Dr. Nabil Ali Khan and Dr. Andrew Page. I am Dr. Lee Katz. Both
Andrew and Nabil work in the Quadram Institute in Norwich, UK, where they work
on microbes in food and the impact on human health. I work at Centers for
Disease Control and Prevention and am an adjunct member at the University of
Georgia in the U.S. Hello and welcome to the MicroBinFeed podcast. Today we're
joined by two special guests, Professor Ed Pfeil and Dr. Natasha Kuto. Dr. Ed
Pfeil is a Professor of Bacterial Evolution at the University of Bath. His
interests include genomic evolution of pathogenic bacteria of both men and
animals. Early work is mostly on Staphylococcus aureus, where more recent work
includes gram-negatives such as Klebsiella pneumoniae, particularly in an AMR
and One Health perspective. But he's also worked on Borrelia, Burkholderia,
Wolbachia, Melissococcus, Rhinobacterium, Librio, Septococcus, Neisseria, E.
coli, Mycobacterium bartonella, amongst many, many others. And he also has an
interest in both bees and aquaculture. Dr. Natasha Kuto is a data scientist at
the Center of Genomic Pathogen Surveillance at the University of Oxford. Natasha
is a veterinary doctor and got her PhD in 2016. Her research focuses on the
molecular epidemiology, population genomics, and ecology of a broad range of
bacterial and viral pathogens of both animals and humans. She uses next
generation sequencing and bioinformatics to understand transmission of bacterial
and viral pathogens, and the emergence and spread of AMR between humans and
animals. She's worked on a range of organisms as well, including MRSA, Staph, E.
coli, Klebsiella, all of the enterococci and enteromycobacterium, including
abscesses and TB, and flu, and also pigs. So welcome to you both. It's great to
have you on the show today. Hi. Thank you. Yeah. Hello. And we decided with the
two of you to talk about a couple of different subjects. One of them might be
about multi-locus sequence typing. But today let's have a chat about One Health.
So I'll start off with an easy one. What exactly is One Health, since it was in
both of your background introductions? So really One Health framework is itself
sort of evolved from an earlier kind of idea called One Medicine. And it's
really a synthesis of the, it's an attempt to, let me explain, it's an
acknowledgement that in order to manage AMR effectively, we need to not only
think about what's happening in healthcare settings, in hospitals, but also take
into account our use of antibiotics and the consequences of antibiotics present
in agricultural settings, and also in environmental settings, the rivers and the
soils and such. There's a sense that if we only focus clinical settings, then
we're, in a sense, ignoring the far bigger picture that's happening out there,
where there is plenty of resistance evolving and emerging, which at any point
could perhaps enter communities, enter healthcare settings. So we need to be
much more joined up in our thinking about how to manage this big problem of AMR.
Natasha, anything you want to add to that? I think that was a very good
explanation. Yeah, it was an excellent explanation. I think as a veterinary
doctor, what I would add is just, you know, this One Health approach is not
linked only to AMR, it's linked to disease in any kind, or trying to make, you
know, all of these different sectors healthy. And that by making all of them
healthy, you will have, you know, a better planet, almost. Let me ask you a
question, Natasha. So Ed put it in perspective of AMR, would you say that One
Health is more than just AMR? Are there other components? Yeah, so there's
definitely other components, as there's much more than AMR. I mean, we could
think about diseases other than bacteria that carry AMR, like viruses and fungi
and parasites. So it's definitely more than the name AMR, and you can translate
it into many other diseases. And it's trying also to, you know, learn from these
different perspectives and try to apply it to the other sectors to help improve
health as one thing, let's say, One Health. Yeah, I 100% agree with that. Yeah,
I didn't make that clear. I mean, One Health covers all manner of aspects. And
as Natasha said, it's a general sort of framework for managing the whole kind of
biosphere. AMR has been described as a quintessential One Health problem,
because of this interconnectedness between all the different sort of settings
that we have. So I think we'll get right into something meaty now. I want to
talk about the Confusogram, and just to ruffle some feathers. So what exactly is
the Confusogram, One Health Confusogram? Yeah, so this picture that I had here
was actually taken from a slide from Ed. And I mean, you see this picture where
you have, you know, the hospitals and the humans and the animals in the
environment, and these arrows between these different sectors, you see it very
often, you know, by different authors. And I mean, we agree with it, with the
principle that, you know, these sectors are connected, and that there might be
transmission between these different sectors when it comes to AMR bacteria or
AMR determinants, or plasmids that can carry those determinants, for example.
It's just that what we need is to actually quantify this risk, quantify the
transmission. And I think that's what's been lacking, and we are getting more
information with these new, you know, studies that have included these big data
sets. And so what we are trying to see here is indeed, to try to quantify the
risk and say, and, you know, put a kind of a, how do you say, well, basically
quantify the risk of transmission between these different sectors. And if
there's this, this flow is impeded by the ecological barriers. And so if a
certain strain goes from one of these sectors to the other, what's going to
happen? Is it going to adapt to this new sector? Is it going to spread? Or is it
going to only infect one person, but then there's no onward person or animal,
but there's no onward transmission. And so the risk is low because it's just one
spillover event, but then there is no onward transmission. So that's what we're
trying to do. And what we've been trying to do for the past three years, well,
ADD has been working on this for a longer time. But at least for me, when I
started working with ADD was, we're trying to answer this question. We're trying
to quantify the risk of transmission. Yes. So people will try and add the
picture to the show notes, but if you just look up something like One Health
Internet, any search engine, you'll see this, you'll see lots and lots of these
delightful pictures that show little circles of animal and clinical and human.
And they'll usually put a nice picture of a globe on it as well. And they'll
have arrows of this interconnect these with all with arrows. And I think,
Natasha, what you're saying is we've established that there is the circulation
between all these different niches and sectors, but it's not enough to say that
it's circulating. We need to actually understand what's the mechanism that this
is occurring and understand to what extent each of these are jumping through the
different sectors. I mean, this is a hard question, but in a given situation,
what is the comparative rate of zoonoses occurring and then the subsequent
actual establishment of, say, an outbreak from that? So do you see lots and lots
of intermittent cases, like multiple jumps over and over again? And then what
rate of that actually kicks off when we see it, we see it in the clinic? I think
I could take a step back almost, and this might sound slightly tangential, but
the reason I've been thinking about the confusogram specifically feeds into
something that's been sort of increasingly bothering me for a few years,
actually. And it's just the power of figures. We're not very good at scientists
in acknowledging and appreciating just how much of a sucker we are for a nice
figure. And actually, there's how much sort of confirmation bias can go on when
we see a figure, when we see a figure like the confusogram, where we've just got
lots of settings, lots of arrows going everywhere.  but there's nothing
explicitly stated there, but there is nevertheless this impression that
everything can flow unimpeded from everywhere to basically everywhere else.
There's a hypothesis intrinsic to that figure, which we don't explicitly state
and we don't explicitly examine. And that's the kind of thing that I'd quite
like to sort of expose and bring out to the light a little bit more, that we
could actually look at that figure and ask the question, well, that's kind of
stating a null in a sense, that's stating a null hypothesis where flow is in a
sense unimpeded, but is that really true? Let's start from that basic
assumption. Can, and if it's not true, where were the barriers? What were the
mechanisms of the possible barriers out there? Were there ecological or
biological or whatever? And then let's start from there. Let's examine this
figure as a hypothesis. And I think that's really what we've been trying to do.
And this idea of figures, I mean, as I say, it's sort of, it's been of
increasing worry over recent years because especially in the genomics
literature, there's so many beautiful figures and so many papers now. They
almost seem to be taking over from the words somehow. And they look lovely
because the software is so good and the analysis software is so good. But I
quite like to see some papers with just really, very minimal figures
occasionally, where it's actually the hypothesis is actually written down, not
somehow embedded in the lovely figures. So that's really, I mean, I'm basically
saying the same thing that you've said already that we wanted to test this idea
that, okay, there's clearly potential for everything to move everywhere else.
There's no reason why there should, on the face of it, be any barriers between
what people in the communities and the pigs or whatever, bacteria, bacteria,
they can move around in all sorts of ways. But let's think about what might
actually limit that transmission. And that comes down to ecology, actually, I
think, as much as anything. And that's something we know very little about. We
know very little. It comes down to host adaptation. It comes down to the role of
the pan genome. It comes down to the movement of individual elements. So this is
a transmission. We've been talking about transmission of the bugs, but then
overlaid that we've got the transmission of the genes as well. So that's a whole
different layer of complexity. But if there's a lot of local adaptation, if
there's a lot of, if the selective landscape is very, very rugged, then you
would intuitively imagine that actually there might be quite substantial
barriers of movement between one setting and another, yeah. So I was wondering,
how much of a feedback loop is there? Are we seeing things that are, say, highly
AMR resistant in hospitals, bleeding into wild birds and into farming? Does it
get down that far? We do see it. There are studies where they've looked into
wildlife, birds, for example, and also into wastewater, wastewater coming out of
the hospitals. And if you do look into that, of course, you will find it. I
mean, we did a study in Bath where we looked at the river, right? It was at the
river, and we looked for Klebsiella, and we did find, interestingly, Klebsiella
ornithinolitica, which is not a common pathogen. So it usually lives in the
environment with E8, for example. And it was very, it was unexpected also
because it was not, at least to our knowledge, it's not causing any sort of
major outbreaks or infections within the hospital. But if we look for it, we do
find it in the environment around the hospitals, for example. I mean, how much
is it, how much of it is really coming from humans, from the hospitals, or it's
just a matter of survival of these species within the environment because of
local adaptation, as Ad said, and competition with other microorganisms. I tend
to say that people tend to forget that antimicrobials are actually derived from,
most of them, at least, are actually derived from natural compounds that have
existed for thousands of years in our planet, so on Earth. So it's just normal
that we do find this AMR determinants, of course, in a lower level, in the
environment and in animals and even in humans. But of course, the pressure that
we've created due to the use of these antibiotics increased the frequency of
these genes. But to answer what you just asked, yes, if you do look for it, you
will find it. You don't have anything to add to that. Yeah, so as Natasha says,
if you look for it, then you will find it. But when people, so many of the sort
of environmental surveys are carried out where you actually select for
antibiotic resistance, so you put antibiotics in the media. And this, of course,
gives you the best chance of finding something and therefore having something to
write about. But it doesn't necessarily give you a very good picture of how
common that resistant gene or resistant strain is in that particular setting
where you sampled. So when we, I guess we'd come on to talk about the SPARC
study later, which has just been accepted in ancient microbiology, by the way,
we had proofs yesterday, so that's very exciting. So in this study, this was a
big One Health study in and around the city in Northern Italy, Pavia. And that
region was chosen because it's a hotspot for these carbapenemase-producing
Klebsiella that have the KPC gene. There's very high prevalence of these
carbapenem not susceptible clones in the hospitals of this region. And we took a
big sample from the hospitals, from the community, from the various farms, from
the environment, all taken, all the samples were contemporaneous, all within
like a sort of 15 month period, all within a defined region, which is very
important, of course, because you want to give yourself the best chance of
finding transmission and finding this evidence for sort of leakage from the
hospital into the environment. We didn't select for antibiotic resistance in
this study. And we didn't find a single example. We ended up sequencing 3,500
genomes, something like that. And we didn't find a single example of carbapenem
non-susceptible Klebsiella pneumonia outside of the hospital. Not one, either
genotypically or phenotypically in this big, big sample. So it's not to say it's
not there, but I think even in that setting, even in Northern Italy, it seems
that there is a, it's likely that these resistant clones can be, hospital-
adapted clones can be out-competed quite quickly in the environment. And that's
something which has been, you know, proposed many times over the years with
different bugs with MRSA, hospital-acquired MRSA as well. So it's a slightly
confused picture because of the differences in methodologies. And also because
of course, what's true in Northern Italy might not be true for Chad or for
Cambodia, or places where there's a lot more, where the antibiotic stewardship
is different and maybe there's a lot more contact between humans and animals and
that sort of thing. So there doesn't seem to be penetration. There's certainly
not unimpeded flow of hospital-type bugs in the environment. But as Natasha
says, there's plenty of resistances out there, which is what you'd expect
anyway, because, you know, antibiotics have been around a lot longer than we
have. And the resistance genes have been around as long as antibiotics have. So
it's a complicated picture. And I wouldn't say that there's no resistance out
there, but I'm saying that in terms of the direct feed through from the hospital
into the environment, it may be less than perhaps some people imagine it to be.
Yeah, I want to touch on that because we're talking about antimicrobial
resistance, which must be a great metabolic load on the organism to keep these
systems available. I mean, some systems probably are not as complicated as
others, but surely this is detrimental to the bug that it has to tolerate and
avoid antimicrobials in the long run. So surely it doesn't really want this.
Yeah, well, that's often the assumption. And in some cases, there's experimental
data to back that up. There's a fitness cost to having harboring resistance
genes or resistance plasmids if you haven't got the antibiotic around. But
there's other experiments. And plasmids where there doesn't seem to be a very
big cost at all. I mean, the plasmid, the study that Natasha mentioned before
with the Ornitholitica having, it was a Blaroxa 48, it's another carbapenemase
gene. There's not so much evidence that that has a really big fitness cost to
the cell. And in fact, it might not even have been doing anything, that plasmid
in those cells. It might have just been sort of stealthily hanging out. They
weren't phenotypically resistant those cells, which is kind of interesting. So
if we had selected for resistance, we wouldn't have.  found those resistance
plasmids, which is kind of, you know, That's not how it works, Ed. Yeah. But
it's, but it's quite, quite interesting, indeed. And I mean, during EMAM, there
was a talk about also in human patients, how this exactly the plasmids that
carry the, this oxo 48, and can, we can kind of switch on and switch off the
production of the protein, let's say, of the of the mechanism of resistance. So
they, they, the plasmid mutates to kind of silence the gene. And so there's not
not much cost, if there's no production of the of the gene and of the protein.
And I remember that it was a recent tweet as well, kind of, you know, talking
about this, you know, this hypothesis that indeed, these plasmids carry, they
are very costly for the for the cell. But in truth, you can, you can actually
see when when you draw some of, you know, follow it do some population structure
analysis, that some of these plasmids are actually expanding with, with their,
with the specific clone. So there isn't an obvious association between one thing
and the other. And so if these plasmids were very costly, you wouldn't see this,
at least for some, you know, plasmid host associations. Yes, I think that's more
like of an assumption. But in reality, that's, it's not as simple as that. No,
it's not sounding very simple at all. We're not having any clear cut black and
white on on any of this, which is which is a real shame. So if I find a gene, it
doesn't mean it's resistant. If it if it, if it's resistant, it might not have
the gene. I mean, where do we, how should we then think about this? Is it more
thinking of an activation energy of series of different factors as just coming
back to what you're saying about risk, Natasha, like we would assess the AMR,
the disease potential, the zoonotic potential, should we think of it that way,
rather than this clear cut, pathogen, bad, commensal good, that sort of attitude
that we've we've seen in the past? Yeah, I think, I mean, having worked on on a
specific disease in dogs, while I was doing my my PhD, I was working on atopic
dermatitis, which is a condition that affects dogs. And most of of these dogs
develop secondary infections due to a cousin of Staph aureus, Staphylococcus
intimidus. And, you know, dogs, healthy dogs live with Staphylococcus intimidus
all their lives without developing any sort of infection. But some people would
consider the Staphylococcus intimidus as pathogens, as really bad guys, like you
said. And I don't see it that way. I mean, I think in, it's a matter of
opportunity. It's a matter of chance, to be honest, I think, you know, if we
coexist with these, with these microorganisms, they must be beneficial to us as
well. And we are beneficial to them. So it's a matter of balance. And if there's
a disruption of this balance, then there is a chance that they will, you know,
that they will invade and cause a more a more serious infection. And of course,
there are certain species that are more prone to do to doing this, like Staph
aureus, E. coli, Klebsiella. But it doesn't necessarily mean that they are the
bad guys. I mean, if we, if if they exist in our guts, or in our skin, it's, it
must be because they're beneficial to us. So I don't really believe they're,
they're bad guys, or good guys. I think, you know, it's a matter of balance. And
not disrupting the balance is the best thing. If you were presented with a, with
a, with a culture of a nice, harmless lactic acid bacteria and a culture of
Shigella or something, you go for the lactic acid, you drink the lactic acid
bacteria first, right? So there are clearly some bugs which are more dangerous
than others. But I think Natasha's absolutely right in pointing out that context
is really important. And actually, there's a very fine line between a bug living
quite blamelessly as a commensal and it being quite a dangerous pathogen. And,
and this context can be on all sorts of levels, it could be subtle genomic
changes in the bug, it could be changes in the microflora that the community
that the bug is living, I mean, C. diff is not a problem, when it's at a very
low level in the gut, it's only when it overgrows, that it becomes a problem. It
could be the context of where it is, where it happens to be in the body. I mean,
it's a bug can be completely harmless in the gut. But if it gets E. coli is a
great model for you know, all the different E. pecs and U. pecs and all these
other things they have, they have particular virulence attributes according to
where they are. So context is everything really with virulence. And it's and the
distinction between a bad guy and a good guy is is can be extremely subtle, like
the difference in one gain or even loss of a single element or a single gene.
And to some extent, the same can be said of the genes themselves. I mean, many
of the things that we talk about as virulence factors, they are genes which will
increase the ability of bacteria to adhere to host tissue, they may have been
horizontally acquired from other species where they're where they're actually,
these genes are carrying out quite different. So there's not even good and bad
genes in that sense. It's all I hate to use the horrible, horrible term emergent
property. You don't hear that very often anymore. Thank God. But it's kind of it
kind of sums up what the is the virulence is, is is a consequence of a whole
load of of other factors other than just the bacteria or just the genes that
happen to be there. In context, like you said, context, it's all about context.
I have a I have a question that bugs me all the time. I was asked this as a
student and I don't have the answer. Is flagella a virulence factor? Well, it's
a case in point, isn't it? So it can be. But it doesn't have to be. I mean, if
you're just a free, free swimming, Vibrio splendidus or something, then, you
know, you're not, you're not bothering anybody, but or even in E. coli. Yeah.
Even in E. coli. Yeah. Right. I was wondering, with all this one health
approach, right, all of our sequencing so far has really been targeted towards
pathogens in humans, which cause disease in humans, and there's a little bit for
veterinary pathogens. Then there's not very much for commensals. So how much are
we actually missing, in general, when we look at the one health approach, you
know, if we're only really, you know, if the bulk of the work is on human
pathogens? Yeah, it's a really good question. And I've, I've had a lot of
discussions about that, with the work that that I've been doing at the
University of Oxford, that's GPS, how would be, you know, how would we frame
what we need to do to have, you know, better, a better overview of what's, of
what's happening? Should we aim to sequence only the resistance, the resistance
strains? Or should we actually look at the commensal ones as well? Should we
focus on the ones that cause disease in hospitals, because there's, you know,
that's where humans are dying of AMR infections? Or should we, should we look
into, you know, should we look at the other non pathogenic strains or species
and try to find AMR determinants before they become before they expand and, and
cause and, and be transferred to more pathogenic species? So I don't have an
answer to that. And I really would like to hear what Ed has to say about that.
And I think, you know, when when we were at EMAM, and Hajo was, I'm sorry, if I
don't know if I can say names, but Hajo was was presenting his his view on on
where we should focus the resources. I, you know, I kind of agree with him that
if the main problem is, is within the hospitals that people are dying of
infections that cannot be treated with antibiotics, then we should focus on
those pathogens that are causing infections. But in a way, if we want to prevent
it preventing from happening, you know, preventing a certain AMR determinant, or
a certain lineage, for example, from expanding, and becoming a global problem,
then we need to start looking before that happens. So we need to start looking
at the susceptible commensal population. But I don't think we have the right
answer. At least I don't have. I agree. I mean, there's there's different
questions here. So there's different, there's kind of different levels of one
health. So the question of where should we prioritise our day to day management
of AMR and infectious disease in general, I think it's clear that we concentrate
on where it's happening, where we can see it spread, which is  in the healthcare
system predominantly, in the community outbreaks, of course, but probably not
worried so much, at least in high resource settings, about there being the risk
of there being some really horrible virulent resistant thing pop out from the
local cows. So that's at that sort of immediate level, which is what, you know,
the direct management of disease, but then there's this existential threat or
risk of, you know, where does the resistance come from? Where does virulence
come from? It all comes from out there originally. So if we want to understand
the drivers of resistance, why these benign bugs can turn nasty, what are the
ecological conditions that drive virulence, then we need to have a much better
understanding of and absolutely take on board what's going on in non-human
systems. We need to think about not just, you know, livestock, but think about
plants, think about insects, think about wild animals, which are really hard to
work on when it comes to infectious disease, because they tend not to present to
hospitals. They just die. Fish in particular just sink. So they're really hard.
That's what I was saying earlier about virulence genes. I mean, some of the key,
the toxin genes in cholerae, they look similar to the symbiont of the squid,
fissure, adibio fissure. So there's genes which are similar in the, you know,
they've evolved in a completely different context. And it's only when we get,
you know, fully understand, if we want to understand virulence, which is
actually quite a lot harder than resistance, I think, because it's a much more
multidimensional thing, then we need to consider the whole virulome, if that's a
word, if it's not, it is now. So yeah, I mean, if that's what you're getting at,
Andrew, I think that that's absolutely right. But if our question is, you know,
how do we stop this particular strain of MRSA or whatever spreading, then we
stop it spreading by infection controlled in the hospitals. So I had a follow-up
question to that. Since both of you have looked at both angles, how much overlap
is there between a project that's looking in zoonoses versus looking at, say, a
hospital outbreak? Is the knowledge from one transferable to another? Is, or do
we need to start thinking of these as separate streams with separate approaches?
That's a really tough one. So some of the methods are the same. So it's all A's,
C's and G's and C's. I mean, what we're interested in here, right? So a lot of
the sort of bioinformatics and the analysis tools will be the same. The models
of transmission and vector-borne diseases, you have to take into account the
reservoir hosts, right? So if you're talking about live, you have to have some
idea of what the, not just what the ticks are doing, but what the deer or the
pheasants or the shrews or whatever it is that the ticks are feeding on what
they're doing. So you have another layer, massive layer of complexity there. So
we can't just do the simple, Bill infects Fred on ward A and then the nurse goes
to the next ward. Those sort of outbreak analysis just won't work in those more
complex environmental settings, especially for vector-borne diseases. In terms
of like host switches, I mean, we've seen what can happen with host switches
with COVID, right? But that was very much an exception in that I think probably
these things jump between hosts all the time, but they almost always just fizzle
out. The fact that COVID could hit the ground and go straight into human
transmission was an aberration in many. And that's something that we need to
really try and understand the best we can, whether this is, because there do
seem to be more frequent events of these host spillovers in recent decades. So
whether that's something, how we're changing the habitat, how we're changing
animals coming to contact with each other and pathogens are responding by being
able to jump from one animal to the other much more easily. There may be some
general sort of framework that we can understand about that, but again, that's
very different from the sort of hospital epidemiology that we're used to. And
Tash, anything to add to that? I know it's a very hard one. No, I just, I agree
with that. I think the tools available are the same, but the questions are very
different. I mean, at least when it comes to, trying to find the drivers of AMR
transmission. I mean, I guess if we were talking about, analyzing an outbreak in
a hospital or analyzing an outbreak in a farm, I mean, there are parallels, of
course, like Ed said, the behavior, the human behavior is different. I mean, I
think we are the main drivers of outbreaks transmission anyways. The way we
behave with the nurses or veterinarians or medical doctors. But yeah, so I think
it's different. I mean, if we're also trying to understand like AMR transmission
in human communities, then there's a whole other layer of social science as well
that we need to get on board with. People's behavior, people's risk perception,
people's, you know, how well they comply with the advice they're given, whether
they take antibiotics when they need them or whether they don't take them when
they need them, you know, all these things, we need to get a full grasp of. All
right. And I think on that note, I think we'll draw to a close on that forward-
looking outlook from Ed. So I want to thank both Ed and Natasha for their time
with us today. And thank you to listening to the Micro Binfy podcast. We'll see
you next time. Thank you so much for listening to us at home. If you liked this
podcast, please subscribe and rate us on iTunes, Spotify, SoundCloud, or the
platform of your choice. Follow us on Twitter at Micro Binfy. And if you don't
like this podcast, please don't do anything. This podcast was recorded by the
Microbial Bioinformatics Group. The opinions expressed here are our own and do
not necessarily reflect the views of CDC or the Quadram Institute.