genomic epidemiology
molecular epidemiology using whole genome sequencing
source tracking
transmission chains
adaptation and genotype/phenotype links using GWAS
the emergence and spread of AMR
within hospitals
within the environment
pattern of transmission and phylogeography on global, national and local scales: how quickly is the strain spreading and to what degree are local variants genetically distinct from each other
identifying the date and location of clonal origin
when and where did the strain emerge?
how fast is it evolving (mutation rate)?
can we identify triggers leading to the emergence of a new virulent strain?
changes in resistance and virulence properties
reconstructing the genetic events leading to the phenotypic changes underlying increased risk to public health
the virulence/transmission trade-off is complex
not necessarily the case that pathogens evolve to become less virulent
what is molecular epidemiology useful for?
the advantages of whole genome sequencing for molecular epidemiology
patterns of transmission and phylogeography on global, national and local scales
how quickly is the strain spreading and to what degree are local variants genetically distinct from each other?
how much mixture of stains do we see in one location eg. hospital
identifying the date and location of clonal origin
when and where did a strain emerge?
how fast is it evolving (mutation rate)?
can we identify triggers leading to the emergence of a new virulent strain?
is there anything about the host population that might be naive or resistant to a particular strain?
changes in resistance and virulence properties
reconstructing the genetic events leading to the phenotypic changes underlying increased risk to public health
use all available evidence - whole genome data only shows so much of the story
conflict between what a strain IS and what its CAPABLE OF DOING
lots of characteristics that can create clusters, not evolutionary consistent but a method of taxonomy nonetheless
core genome
used to build the phylogeny, tell us about evolutionary relatedness
transmission dynamics
outbreak source
non core genome
accessory genome
what does it do
how does it transmit
is it resistant
phenotypic predictions based on the presence of virulence and resistance determinants
all other epidemiological and clinical metadata. abundance data.
examine an outbreak of tuberculosis in BC canada
used WGS, other molecular methods, contact tracing, social network analysis
found a link to a few users of crack cocaine - super spreaders
super spreader = single source responsible for a large number of cases
reproduction number (R0)
tells us something about how quickly a pathogen is liable to spread through a population
given we know the period of infectiousness
an average number so does not reveal anything about the pattern of spread
dispersal index is an important component of R0 (average number of infections one infected individual will cause
R0 = 3 may mean all infected individuals infect three other people
more likely most infected people infect 0 or 1 other people but a few super spreaders pass it on to many more
particular settings where there were super spreader events
asymptomatic carriage
Mary mallon - Irish born American cook who was possibly born carrying salmonella typhi
persisted in working as a cook, exposing others to the disease - was forcibly quanrantined
died after 30y in isolation
estimated to have infected at least 122 people including 5 dead
identification of high risk clones or variants of concern
the identification and surveillance of strains of increased danger to public health
may cause more serious disease (have heightened virulence)
may be more difficult to treat or prevent (eg. resistance to antibiotics or vaccine escape)
may be more transmissible (eg. through increased rate of replication, or through better colonisation of the host, or survival outside of the host, such as on surfaces)
zoonoses: animal -> human
anthroponoses: human -> animal
antigenic turnover
often seen for seasonal coronaviruses and flu
circulating viruses give rise to new variants that escape prior immunity against infection, immunity builds against these new variants, then these in turn spawn new variants
when infections evolve to escape immunity, we typically end up with an evolutionary tree that looks like a lopsided ladder as new variants sequentially replace their parent variant lineages
tend to emerge in australian winter and migrate to northern hemisphere for our winter - we get a bit of a heads up
the emergence of eMRSA-15 (ST22)
for bacteria, we can identify, for most key pathogenic species, particularly dangerous strains that have a high level of antibiotic resistance, or are highly virulent (high risk clones)
when one hiric replaces another it is called 'clonal' replacement
eMRSA-15 emerged in the uk, became the most rapidly spreading mrsa clone in europe
lots of strains are given st numbers - refers to legacy technique from before the capacity for wgs
MLST
st stands for sequence type and is derived from multi locus sequence typing (used prior to wgs)
based on the variation of multiple genes (typically 7)
genes chose typically encode core metabolic enzymes (housekeeping genes) as these are likely to be under stabilising selection
not used very much anymore
with advent of wgs you tend to group similar variations together in one strain
major spread in early 2000 almost completely down to ST22 strain
two different dynamics
local spreading within one hospital
longer distance transmission which creates more distinct strains
emerged in Birmingham area from a methicillin susceptible population of st22 by the acquisition of a type IVh SCCmec cassette
also acquired resistance to fluoroquinolones through two point mutations generating amino acid substitutions Ser8oPhe in topoisomerase IV (GrlA) and Ser84Leu in gyrase A (GyrA)
co-incident with the first preclinical trials of fluoroquinolones in the same place
acquired resistance to methicillin and then to fluoroquinolons (antibiotics given pre-surgery) - resistance to fq is probably what gave it such a high selective advantage
comet shaped phylogenetic tree = big explosion with tail of initial strain
high risk clones of mrsa - st239
at its peak the most common mrsa strain - accounts for 90% of mrsa cases in mainland Asia and is also very common in South America and eastern europe
multiply resistant (exhibits a large SCCmec element) and is only found within the hospital environment
relates to concept of fitness cost
see hospitals as an ecological niche, exposure to antibiotics, unique surfaces, routes of transmission through contact
bacteria adapt to the niche, mrsa is a massively adapted bacterium
mr is a massive gene that takes up a lot of the genome - lots of the consequences that come with it
like a suit of amour in a medieval battle - great for the setting but very impractical anywhere else
hence only being found in hospitals (specifically niche)
epidemiological and experimental evidence for increased virulence and transmissibility of ST239 within hospitals
wgs on a collection of 62 st239 isolates
42 from europe, north and South America, Asia, Australia 20 year temporal range
20 strains from a single hospital in northern Thailand collected over a single year
examine global spread and potential for local epidemiology
illustrates 2 concepts
how to measure mutation rate (and hence date of emergence) and
click to edit
of the 4310 sites exhibiting an SNP, 38 were homoplasmic, meaning they occurred independently on different lineages in the tree
conflict with the tree
tree is based on commonality of snps
things close on the tree are close bc they have more snps in common
homoplasy - same things happening independently on different branches
what's going on - the mutation is random, it doesn't care, but selection does
if a particular mutation confers a strong adaptive advantage and occurs independently, we're likely to see it at multiple points in the population
notably many of the homoplasic saps were inn genes involved in drug resistance, with 10 corresponding to mutations known to confer resistance
many homoplasmic mutations observed in the Sars-CoV2 genomes that have a role in immune evasion and/or increased transmissibility
Klebsiella pneumoniae
kp is a major cause of hospital acquired infections and neonatal sepsis
ha infections
can lead to pneumonia, wound, soft tissue and urinary tract infections
tend to be more resistant as exposed to abs as selective pressure
similar to e coli
community acquired infections
tends to be more virulent
can lead to pyogenic liver abscess, pneumonia, meningitis
causes commercially important livestock disease eg. bovine mastitis
commonly isolated from the guts of humans and other mammals, vertebrates and invertebrates, from soil and water, and plants
doesn't replicate great in water but can survive for a long time
ecological generalist
high priority AMR pathogen
AMR in Kp is typically plasmid borne
Kp can carry beta lactamases (broad and extended spectrum with lactamase inhibitors)
CTX-M
SHV
TEM
carbapenems among the last-line drugs used for enterobacteriaceae infections
most important mechanism of carbapenem resistance is the production of carbapenemase enzymes
kpc
ndm
vim
oxa-48
carbapenemase genes associated with a small number of widespread kp clones, mostly responsible for hospital-acquired infections
ST258/512
ST307
usually carried on plasmids
burden of carbapenem resistant k pneumoniae increased by a factor of 6-16 times in terms of number of infections and number of deaths, between 2007 -> 2015
virulence in the context of one health
key virulence genes of klebsiella spp.
gene clusters associated with heightened virulence
seems the more individual genes the strain has the more virulent it is
access to iron is critical for pathogenicity
yersiniabactin
aerobactin
salmochelin
colibactin (toxin)
the regulators of mucoid phenotype
rmpA
rmpA2
both up regulate capsule production
hypervirulent lineages contain all 4 siderophores plus rmpA and rmpA2 eg. CG23 in Kp
associated with community-acquired infection resulting in liver abscesses
the "EuScape" Study
epidemic of carbapenem-resistant klebsiella pneumoniae in europe is driven by nosocomial spread
analysed the genome sequences and epidemiological data of >1,700 K. pneumoniae samples isolated from patients in 244 hospitals in 32 countries over a 6 month period in 2014
477 of 682 (69.9%) carbapenemase-positive isolates are concentrated in four clonal lineages, sequence types 11, 15, 101, 258/512 and their derivatives
more highly resistant K pneumoniae are more likely to spread through the health-care network
carbapenemase-positive isolates have the highest transmissibility
over half of the hospitals that contributed carbapenemase-positive isolates probably experienced within-hospital transmission, and inter hospital spread is far moe frequent within, rather than between countries
oxo-48 plasmid - almost identical in each strain across many bacteria - conserved but doesn't seem to confer resistance
non resistant plasmid found in sampling nowhere near any carbapenem resistant outbreaks
could transfer easily into a bug
cryptic resistance gene - biding its time before it transfers and is able to confer resistance
susceptible isolates show no geographical structure
isolates from a single hospital are phylogenetically divergent
carbpenemase producing isolates show strong geographical structure, and tend to correspond to a single major clone within hospital or local region
one health: the intersection of human health, animal health and environmental health
combines human and veterinary medicine in response to zoonoses
one health and amr
mopping the floor with the tap wide open
fear is that hospital adapted clones will also adapt to life outside of hospitals
have to consider all strains emerging from places like wastewater treatment plants or evolving due to agricultural methods
one health priorithy research agenda for AMR
a joint initiative of the food and agriculture organisation of the un, the un environment programme, the world organisation for animal health, and WHO, collectively known as the quadripartite
a guide to one health AMR to help policymakers, researchers and a multidisciplinary scientific community work together to mitigate AMR within the one health approach
surveilllance
interventions
transmission
economics and policy
behavioural insights
accessible methodologies to identify sources and drivers of AMR between one health sectors
impact of infection prevention and control practices on mitigating AMR
how does AMR transmission across one health sectors in HICs vary from LICs/LMICs? and according to antibiotic/pathogen?
the one health confusogram
there are likely to be plenty of opportunities for bacteria/plasmids to transmit from one setting to another
is this flow impended by ecological barriers? that is, does local adaptation limit to the competitiveness of migrants entering a new setting or niche?
understanding which of these routes is the most important is key
not just opportunity, but capability and likelihood - ecological dimensions
eg. how much are bacteria found in pigs adapted to pigs vs. adapted to survival in humans
WHAT LIMITS TRANSMISSION BETWEEN DIFFERENT SETTINGS?
opportunity
the means by which bacteria can move from one compartment to another is limited. seems unlikely
ecology
different strains are adapted to specific niches, and they are unlikely to thrive or expand in the wrong niche
a key example of ecological adaptation is specialisation to different hosts
spillover refers to when a pathogen jumps from one host to another
the ecological and evolutionary dynamics are very complex, but have direct relevance to one health
host specialism
interactions between species
phylogenetic distance between host species
environmental factors
source host factors
recipient host factors
intermediate host factors
characteristics of the pathogens
clinically important kp clones are commonly isolated from a range of animals, for example, ST11 from poultry, ST15 from companion animals, ST23 from non-human primates and horses and ST25 from pigs
could be a single sporadic with no onward transmission and little risk
alternatively, inter host transmission might result in the bacteria spreading within the new host
greater levels of host specialism may reduce the chance of successful spillover
transmission opportunity, ecology, use of antibiotics. setting-specific drivers (eg. animal husbandry, wastewater treatment processes)
sporadic transmission/infection of environmental strains
cycling/enrichment of known high-risk AMR strains
emergence and spread of novel AMR strains in the environment (most esoteric risk)
how do we interpret data on the presence/absence of clinically important genes/strains in non clinical settings, with respect to risk, or the likelihood of these outcomes?
review of AMR and one health with a focus on carbapenemase producing genes
carbapenemase genes have been reported in the environment on almost every continent
current reservoirs
hospital and municipal wastewater
drinking water
natural waterways
sediments
recreational waters
companion animals
wildlife
agricultural environemnts
food animals
retail food products
ESBLs and the plasmid mediated AmpC gene in e coli
considering te contribution of various sources to asymptomatic carriage of these genes within the community
source attribution model on a large aggregated dataset on the prevalence of these genes in the Netherlands from 2005-2017
conclusion that two thirds of the cases of acquisition of these AMR genes by the community reflects spread from within the community itself - do not dismiss the contribution of other sources such as food, animal contact and the enivornemnt (particularly water)
without input from non human sources the rate of transmission (R0) would not be sufficiently high within the community for these genes to be maintained. this implies that it should in principle be possible to mitigate the seemingly inexorable spread of AMR genes
WGS to compare e coli isolates from human bloodstream infections in the UK with isolates collected from farms and the food chain in the east of the country
limited evidence of transmission of strains between animals and humans
limited evidence of the transfer of resistance genes between animals and humans
substantial evidence of clonal spread within single farms
however the livestock and human samples were not exact like fro like - they were sampled from different places at different times
no carbapenemase genes isolated outside of healthcare settings and very few in species other than k pneumoniae (isolates were not selected for resistance)
detecting transmission by phylogenetic/clustering analysis: example of cow to human transmission
are environmental and animal niches important in the transmission of st307 and other high risk clones of klebsiella
most acquisition by humans of klebsiella is derived from other humans
AMR AND VIRULENCE ARE CONVERGING IN K PNEUMONIAE
virulence and resistance genes in kp tend to be harboured on large conjugative incFIB or incFII plasmids
high risk clones have historically tended to be either resistant (hospital acquired) or virulent (community acquired)
many convergent clones have recently been described that combine both AMR and virulence traits
the majority of reports of convergence come from clinical settings in Asia, particularly china
convergence of virulent and resistance can arise through the presence of virulence and resistance plasmids in the same celll, or by the emergence of hybrid plasmids that carry both traits
is it the strains or the genes that matter - one health implications
whether for multi host bacteria such as s aueus, horizontally transferred genes follow the distribution of the bacterial cell that harbor them or create their own distribution
should we be considering the overlap in strains between different hosts or the overlap of specific genes? in this case antibiotic resistance and virulence genes
WIDESPREAD GENE SHARING BETWEEN BACTERIAL STRAINS COLONIZING DIFFERENT ANIMAL HOSTS
recombination can play a key role in host adaptation
GWAS describes a number of statistical approaches to identifying the genetic changes responsible for notable bacterial phenotypes
virulence
toxicity
adhesion
transmissibility
immune evasion
tissue specificity
antibiotic resistance
host specificity/adaptation
bottom up approach
starts with DNA sequence (genes) and tests the effect on the phenotype
top down approach
starts with phenotype and associates it with particular genomic elements
genotype can refer to any genetic change
snp in core genome
kmers
gene presence/absence in accessory genome
genome rearrangement
changes in gene expression
why have only two diseases been eradicated
smallpox
rindepest
how did rinderpest vaccination provide immunity against PPRV