Letter https://doi.org/10.1038/s41586-019-1521-8
Salmonella persisters promote the spread of
antibiotic resistance plasmids in the gut
Erik Bakkeren1
, Jana S. Huisman2,3
, Stefan A. Fattinger1,4
, Annika Hausmann1
, Markus Furter1
, Adrian Egli5,6
, Emma Slack1,7
,
Mikael E. Sellin4
, Sebastian Bonhoeffer2
, Roland R. Regoes2
, Médéric Diard1,8,9
* & Wolf-Dietrich Hardt1,9
*
The emergence of antibiotic-resistant bacteria through
mutations or the acquisition of genetic material such as resistance
plasmids represents a major public health issue1,2
. Persisters are
subpopulations of bacteria that survive antibiotics by reversibly
adapting their physiology3–10
, and can promote the emergence of
antibiotic-resistant mutants11
. We investigated whether persisters
can also promote the spread of resistance plasmids. In contrast
to mutations, the transfer of resistance plasmids requires the cooccurrence
of both a donor and a recipient bacterial strain. For our
experiments, we chose the facultative intracellular entero-pathogen
Salmonella enterica serovar Typhimurium (S. Typhimurium)
and Escherichia coli, a common member of the microbiota12
.
S. Typhimurium forms persisters that survive antibiotic therapy
in several host tissues. Here we show that tissue-associated
S. Typhimurium persisters represent long-lived reservoirs of
plasmid donors or recipients. The formation of reservoirs of
S. Typhimurium persisters requires Salmonella pathogenicity
island (SPI)-1 and/or SPI-2 in gut-associated tissues, or SPI-2 at
systemic sites. The re-seeding of these persister bacteria into the
gut lumen enables the co-occurrence of donors with gut-resident
recipients, and thereby favours plasmid transfer between various
strains of Enterobacteriaceae. We observe up to 99% transconjugants
within two to three days of re-seeding. Mathematical modelling
shows that rare re-seeding events may suffice for a high frequency
of conjugation. Vaccination reduces the formation of reservoirs
of persisters after oral infection with S. Typhimurium, as well as
subsequent plasmid transfer. We conclude that—even without
selection for plasmid-encoded resistance genes—small reservoirs of
pathogen persisters can foster the spread of promiscuous resistance
plasmids in the gut.
S. enterica and E. coli strains contain numerous resistance plas-
mids13,14
, and different strains frequently colonize the same host15–19
.
High cell densities in the gut lumen allow high rates of plasmid transfer
within and between species20–22
. S. Typhimurium strains can colonize
the gut lumen and survive within the tissues of their host for prolonged
periods5–7,15–17,23
. In the absence of suitable resistance genes, the gutluminal
population of S. Typhimurium is eliminated by antibiotics
within a few hours, whereas tissue-associated S. Typhimurium persister
cells survive for more than ten days5–8,24
(Supplementary Discussion).
After withdrawal of antibiotics, surviving cells can migrate to the gut
lumen and resume growth24
. We hypothesized that this promotes the cooccurrence
of donors with gut-luminal recipients, and thereby fuels the
transfer of resistance plasmids in vivo (Extended Data Fig. 1a).
Wild-type S. Typhimurium SL1344 naturally carries P2, a wellcharacterized
conjugative plasmid of the IncI1 incompatibility
group25
in Enterobacteriaceae20–22
(Extended Data Fig. 1b); this strain
of S. Typhimurium served as the donor in our initial experiments.
We labelled P2 with a chloramphenicol-resistance marker (P2cat
) to
monitor plasmid transfer by plating (Extended Data Fig. 1c, d). Controls
excluded cross-resistance for any of the resistance markers used in this
study (Extended Data Fig. 2). In the ‘oral model’, the donor (that is,
S. Typhimurium P2cat
) initially colonized the gut lumen and invaded
host tissues (Fig. 1a, b, Extended Data Fig. 3). Ciprofloxacin cleared
the gut-luminal bacteria, whereas tissue-associated S. Typhimurium
P2cat
persisters survived5–8
. The recipient S. Typhimurium ATCC
14028S strain, which naturally lacks P2, was introduced at day 821,22
.
Transconjugants (that is, recipients that obtained the plasmid) replaced
the recipients within one to three days of donor re-seeding (>99%
median; day 11–12 of the experiment) (Fig. 1c, Extended Data Fig. 3a).
Control experiments refuted that the replacement by transconjugants
is attributable to P2-mediated fitness benefits over the recipients
(Extended Data Fig. 3d, e). Further control experiments verified that
P2cat
transfer occurs in the gut lumen (Extended Data Fig. 3f, g) and that
P2cat
spreads by conjugation (Extended Data Fig. 3h, i). Non-invasive
donor mutants indicated that the donor cells originated from reservoirs
of tissue-associated S. Typhimurium P2cat
persisters (Fig. 1c–e,
Extended Data Figs. 3b, 4). However, this cannot definitively rule out
that rare persisters could also exist in the gut lumen.
To verify the importance of reservoirs of tissue-associated
S. Typhimurium persisters, we used an ‘intravenous infection model’
(Fig. 2a, b). Consistent with previous work23
, S. Typhimurium P2cat
formed large populations of persisters that survived in the spleen and
the liver after intraperitoneal treatment with ceftriaxone (Fig. 2c, d,
Extended Data Fig. 5a). In contrast to the oral model, gut-luminal colonization
by S. Typhimurium P2cat
is observed only very rarely by day
5 (Extended Data Fig. 5). Nevertheless, when donors were detected in
the gut lumen, transconjugants were formed with high efficiency within
one to three days (Fig. 2e, Extended Data Fig. 5b). Donors that lacked a
key virulence factor that promotes both systemic growth of pathogens
in vivo26
and the survival of persisters27
yielded much smaller reservoirs
of persisters than did wild-type S. Typhimurium P2cat
(Fig. 2d), and
did not produce transconjugants (Fig. 2e, Extended Data Fig. 5c). We
conclude that tissue-associated S. Typhimurium persisters can serve as
reservoirs that promote the spread of conjugative plasmids.
We next identified the rate-limiting process in transconjugant formation
in vivo, focusing on the oral model. Three key parameters may dictate
plasmid transfer dynamics: (1) the rate at which plasmid-carrying
donors re-enter the gut lumen and perform the initial conjugation event
(that is, re-seeding), which is dependent on the size of the reservoir
of persisters; (2) the rate of plasmid transfer from transconjugants to
recipients; and (3) the relative growth rate of transconjugants compared
to recipients. To estimate these rates, we used donor mixtures
that carried five variants of DNA-tagged P2cat
(S. Typhimurium P2cat tag
)
(Extended Data Fig. 6). Whereas all tags were present in roughly
equivalent abundance in the inoculum, the faeces (on day 1) and the
mucosa (on day 15), most transconjugant populations contained just
1
Institute of Microbiology, Department of Biology, ETH Zurich, Zurich, Switzerland. 2
Institute of Integrative Biology, Department of Environmental Systems Science, ETH Zurich, Zurich, Switzerland.
3
Swiss Institute of Bioinformatics, Lausanne, Switzerland. 4
Science for Life Laboratory, Department of Medical Biochemistry and Microbiology, Uppsala University, Uppsala, Sweden. 5
Division of
Clinical Microbiology, University Hospital Basel, Basel, Switzerland. 6
Applied Microbiology Research, Department of Biomedicine, University of Basel, Basel, Switzerland. 7
Institute of Food, Nutrition
and Health, Department of Health Sciences and Technology, ETH Zurich, Zurich, Switzerland. 8
Biozentrum, University of Basel, Basel, Switzerland. 9
These authors contributed equally: Médéric
Diard, Wolf-Dietrich Hardt. *e-mail: mederic.diard@unibas.ch; hardt@micro.biol.ethz.ch
2 7 6 | N A T U RE | V O L 5 7 3 | 1 2 S E P TE M B ER 2 0 1 9
Letter RESEARCH
one or two of the five tags (Fig. 3a, b, Extended Data Fig. 6b, c). In
the intravenous-infection experimental setup (Fig. 2b, e), transconjugants
were also dominated by only one tag (Extended Data Fig. 6d–f).
Transconjugant populations therefore arise from very few donorto-recipient
conjugation events, followed by transconjugant-to-recipient
spread.
To quantify the relative contribution of donor re-seeding and plasmid
conjugation, we developed a mathematical model (Extended Data
Fig. 6g). This model assessed the interdependence of the rate of donor
re-seeding (η, the sum of donor re-seeding plus the donor-to-recipient
conjugation) and the rate of transconjugant-to-recipient conjugation
(γ) (Supplementary Information). Fitting our mathematical model to
the distributions of the plasmid tags confirmed that η is rate-limiting,
and identified the most-likely parameters; these were per-recipient rates
of η = 3.16 × 10−10
re-seeding events per day and γ = 3.16 × 10−8
conjugation events per colony-forming unit (CFU) per gram of
–1 0 1 2 3 4 5 6 7 8 9
Streptomycin
Oral donor
(SmR)
Ciprofloxacin
Ampicillin in
drinking water
Oral recipient
(AmpR)
Donor col. Clearance with antibiotics Recipient colonization and plasmid transfer
10 11 12 13 14 15
a
b
c Oral wild-type (invasive) donor
Oral non-invasive donor
Oral wild-type (invasive) donor
Oral non-invasive donor
***NS NS NS
***
8 9 10 11 12 13 14 15
100 105
104
103
102
101
100
10–2
10–4
10–6
10–8
Day of experiment
Oral
S.Tm recipient
Proportiontransconjugants
(ofrecipients)infaeces
>99%
****
Mucosa-associated population
(oral model, day 15)
CFUdonorspercaecaltissue
d
***
e Oral non-invasive donor (day 8)Oral wild-type (invasive) donor (day 8)
Day:
Lu
Ep
LP
Lu
Ep
LP
Lu
Ep
LP
NucleiActinS.TmICAM-1
Oral model
?
Gut lumen
Epithelium
Lamina propria/mLN
?
?
Donor
colonization (oral)
Clearance with
antibiotics
Recipient
colonization (oral)
Plasmid seeding
by donor
Amplification?
S.Tm donor
S.Tm persisters
Recipient Transconjugant
?
?
?
?
Euthanize
Fig. 1 | S. Typhimurium persisters associated with gut tissue are a
reservoir for conjugative plasmids. a, Plasmid-bearing S. Typhimurium
(S. Tm) (blue lozenge) form persisters (shown as smaller lozenges) in
gut tissues and the mesenteric lymph nodes (mLN). These can survive
antibiotic treatment, re-seed the gut lumen and transfer their plasmid
to recipients (green). Transconjugants (red) can further amplify the
spread of plasmids. b, Oral infection mouse model. Col., colonization.
c, Donor mucosa invasion is required for persister-promoted plasmid
transfer. Invasive (streptomycin-resistant (SmR
) and CmR
) or non-invasive
S. Typhimurium SL1344 P2cat
strain (TTSS-1-negative, SmR
and CmR
) was
used as a donor, and S. Typhimurium 14028S aphT (free of P2) (KanR
and
AmpR
) was used as a recipient, as shown in b. Transconjugant proportions
in faeces from experiments with invasive (solid black circles; grey bars
indicate median; n = 15 mice; 5 independent experiments; n = 9 mice
pooled from data in Fig. 3) or non-invasive donors (open black circles;
white bars indicate median; n = 6 mice, 2 independent experiments).
Dashed lines connect data points from the same mice. d, Donor reservoirs
in the caecal mucosa from mice, as infected in c, were quantified by tissue
gentamicin protection. Solid blue circles represent invasive donors; open
blue circles represent non-invasive donors; lines indicate the median.
Statistics were performed using a two-tailed Mann–Whitney U-test.
NS, not significant (P ≥ 0.05), *P < 0.05, **P < 0.01, ***P < 0.001,
****P < 0.0001. Dotted line denotes the detection limit. e, Persisters in the
caecum lamina propria (LP) (oral model, day 8). S. Typhimurium (yellow,
anti-LPS O5 and anti-LPS O12 staining), nuclei (blue), actin (green) and
lamina propria (red, ICAM-1) are stained. Ep, epithelium; lu, lumen.
Scale bars, 20 μm. White arrows highlight S. Typhimurium (magnified in
inset). Representative of three independent experiments. Quantification is
provided in Extended Data Fig. 4.
d
e
Proportiontransconjugants
(ofrecipients)infaeces
Oral
S.Tm recipient
>99%
8 9 10 11 12 13 14 15
Day of experiment
7 16 17 1918 20 21 22 23 24 25
NSNS NS
** * **NS NS NS NS NS
* * * * * * *
CFUdonorsperorgan
Caecal
tissue
mLN Gall
bladder
Tissue-associated population
(i.v. model, day 25)
** ** *** NS
CFUdonorsperorgan
Spleen SpleenLiver Liver
Tissue-associated population
(i.v. model, day 5)
******
c
Untreated
3× ceftriaxone i.p.
0 1 2 3 4 5 6 7 8 9
Intravenous
donor
Ceftriaxone
Oral
recipient
Donor
spread
Treatment with
antibiotics
Recipient colonization and
plasmid transfer
10 11 12 13 14 15
a
b
Day:
i.v. model
?
Gut lumen
Epithelium
Systemic sites
?
Donor
spread (i.v.)
Treatment with
antibiotics
Recipient
colonization (oral)
Plasmid seeding
by donor
Amplification?
S.Tm donor S.Tm persisters
Recipient Transconjugant
?
?
?
25...
i.v. wild-type donor
i.v. SPI-2-deficient donor
Euthanize
105
104
103
102
101
100
100
10–2
10–4
10–6
10–8
105
104
103
102
101
100
Fig. 2 | Persisters at systemic sites are a reservoir for plasmid transfer
in the gut. a, As in Fig. 1a, but donors are introduced by intravenous
infection. b, Intravenous (i.v.) infection mouse model. c, Persisters in
the spleens and livers. In two independent experiments, n = 8 mice
(grey circles) were infected intravenously with an equal mix of five
S. Typhimurium P2cat tag
strains (SL1344 strain; SmR
and CmR
). Black
circles indicate control mice (n = 8) infected for 5 days without ceftriaxone
treatment. Extended Data Figure 5a shows counts for additional organs.
d, e, SPI-2 promotes donor reservoir formation at systemic sites. In two
independent experiments, mice were infected as described in b with
donors; that is, with mixtures of five wild-type (n = 6; closed circles) or
SPI-2-deficient (n = 6; open circles) S. Typhimurium P2cat tag
strains (SmR
and CmR
). d, Donor populations in internal organs from mice infected
with wild-type (solid blue) or SPI-2-deficient donors (open blue). Lines
indicate the median. e, Faecal populations from mice from d. Proportions
of transconjugants in faeces, from experiments with wild-type (solid black
circles; grey bars indicate median) or non-invasive donors (open black
circles; white bars indicate median). Dashed lines connect data points
from the same mice. Statistics were performed using a two-tailed Mann–
Whitney U-test. NS, not significant (P > 0.05), *P < 0.05, **P < 0.01,
***P < 0.001. Dotted lines denote detection limits.
1 2 S E P TE M B ER 2 0 1 9 | V O L 5 7 3 | N A T U RE | 2 7 7
LetterRESEARCH
faeces per day (red in Fig. 3c; marginal posterior densities are listed in
Supplementary Table 4). This corresponds to approximately 1.6 donor
re-seeding events (plus the initial plasmid transfer) per day into the gut
lumen. By contrast, transconjugant-to-recipient plasmid conjugation
rates were ≥32 events per day for the initial conjugations, and increased
exponentially thereafter (Supplementary Information).
We then used our mathematical model to predict the effect of reducing
the rate of donor re-seeding. The transconjugant-to-recipient conjugation
parameter was fixed at γ = 3.16 × 10−8
events per CFU per
gram of faeces per day (Fig. 3c), and we evaluated the probability of
conjugation events occurring within the course of our 15-day experiment
(Fig. 3d). Reducing η by more than 100-fold diminished conjugation
(Fig. 3d). To test this prediction, we used oral vaccination with
killed S. Typhimurium cells28
, a procedure that is known to reduce
S. Typhimurium invasion of the gut tissue. Consistent with previous
work22
, vaccinated mice contained 10–500-fold smaller reservoirs of
persisters than did naive mice (Fig. 3e). Indeed, this vaccination prevented
plasmid transfer in the vast majority of mice (Fig. 3f, Extended
Data Fig. 7a–c). Moreover, the transconjugants detected in vaccinated
mice appeared later than the transconjugants in non-vaccinated controls
(Fig. 3f). Control experiments with reduced fractions of invasive
donors verified this observation (Extended Data Fig. 7d–g). Thus, the
size of the tissue-associated reservoir of persisters is a main driver of the
ensuing rise of transconjugants. These data confirm the predictions of
our mathematical model (Fig. 3d) and suggest that vaccination might
provide a means to prevent the formation of mucosa-associated reservoirs
of conjugative plasmids.
We next addressed whether strains of the microbiota could also
acquire these conjugative plasmids. First, we analysed P2cat
transfer
to the commensal E. coli strain 817820,22
. For this purpose, we created
an ampicillin-resistant derivative of E. coli 8178 that was free of the P2
plasmid20
, and used it as a recipient in the oral model (Fig. 1b). This
yielded efficient plasmid transfer (Fig. 4a, b, Extended Data Fig. 8a).
We tested whether tissue-associated S. Typhimurium persisters could
also serve as recipients. In the oral model (Fig. 1b), mice were first
infected with a P2-cured variant of S. Typhimurium SL1344 (kanamycin-resistant
(KanR
)). Then, we treated the mice with ciprofloxacin
and ampicillin, and introduced the donor (E. coli 8178 P2cat
;
ampicillin-resistant (AmpR
) and chloramphenicol-resistant (CmR
)).
S. Typhimurium re-seeded the gut lumen and transconjugants were
formed (>38% median) (Extended Data Fig. 8b–d). Thus, tissueassociated
S. Typhimurium persisters can serve as donors or as recipients
of plasmid transfer between different species of Enterobacteriaceae.
To assess whether mucosa-associated S. Typhimurium can also
serve as a reservoir for clinically relevant plasmids that carry genes of
the extended-spectrum β-lactamase (ESBL) family, we used pESBL15
(Extended Data Fig. 1b). S. Typhimurium SL1344 pESBL15 was used
as a donor. We modified our oral model from Fig. 1b in two ways:
by replacing ampicillin with kanamycin in the drinking water, and by
using an ampicillin-sensitive (AmpS
), but KanR
, variant of our standard
recipient strain (S. Typhimurium 14028S aphT (KanR
and AmpS
)). We
observed tissue-associated persisters, re-seeding of the gut lumen and
efficient plasmid transfer (Fig. 4c, d, Extended Data Fig. 8e). Thus, our
findings apply to clinically important resistance plasmids. The choice of
the antibiotics used for the different phases of our oral model had little
effect on these fast plasmid-transfer kinetics (Extended Data Fig. 9a–d).
Finally, we validated the transfer of antibiotic-resistance plasmids
to E. coli in the intravenous model. Intravenous infection of
S. Typhimurium P2cat
led to populations of persisters in the internal
organs (Fig. 4e); these populations could re-seed the gut lumen
and transferred the plasmid into the luminal recipient population,
forming >25% E. coli transconjugants in the absence of antibiotic
selection (Fig. 4f, Extended Data Fig. 9e). Thus, tissue-associated
S. Typhimurium persisters can serve as a reservoir for resistance
plasmids that can be efficiently transferred to different species of
Enterobacteriaceae.
Our results uncover a mechanism by which antibiotic persistence
promotes the spread of antibiotic-resistance plasmids, specifically by
promoting the co-occurrence of donors and recipient bacteria in the
gut-luminal niche. This reveals a role for persistence in clinical bacterial
infection: not only can bacterial persistence lead to relapse of disease
in chronic infections, but also it can facilitate the spread of antibiotic
resistance. In the case discussed here, the transfer rate is independent
fe
**NS NS
>99%
****
Mucosa-associated population
(oral model, day 15)
CFUdonorspercaecaltissue
**********
8 9 10 11 12 13 14 15
Day of experiment
Oral
S.Tm recipient
Proportiontransconjugants
(ofrecipients)infaeces
*
0
20
40
60
80
100
M
ost
2nd
3rd
4th
Least
Mucosa-associated donor population
at euthanasia (oral model, day 15)
Faecal transconjugant population
at euthanasia (oral model, day 15)M
ost
2nd
3rd
4th
Least
100
10–1
10–2
10–2
10–4
10–6
105
104
103
102
101
100
100
10–2
10–4
10–6
10–8
10–8
10–10
10–12
10–12 10–10 10–8 10–6 10–4 10–2
10–3
10–4
10–5
10–6
100
10–1
10–2
10–3
10–4
10–5
10–6
Relativeplasmidproportion
a b
c
Donor re-seeding rate
( )
10–12 10–10 10–8 10–6 10–4 10–2
Donor re-seeding rate
( )
Transconjugant-to-recipient
transferrate(γ)
Successful
simulations (%)
d
Simulationswith
conjugation(%) 100
80
60
40
20
0
Conservative detection limit
Conservative
detection limit
Estimation of parameters fitting
data (oral model)
= 3.16 × 10–10
Dependence of conjugation
on donor re-seeding rate
(oral model)
Tag abundance (relative rank) Tag abundance (relative rank)
Relativeplasmidproportion
Naive mice (oral wild-type donor)
Vaccinated mice (oral wild-type donor)
Fig. 3 | Plasmid transfer is initiated by rare re-seeding events from
donors and can be prevented by vaccination. Mice were orally infected
(Fig. 1b) with tagged donor (five S. Typhimurium P2cat tag
SL1344 strains;
SmR
and CmR
) and recipient mixtures (five tagged S. Typhimurium
14028S strains; KanR
and AmpR
). n = 8 mice, 3 independent experiments.
Donors, recipients and transconjugants were enumerated by plating
(Extended Data Fig. 7a) and plasmid-tag distributions were analysed by
quantitative PCR (qPCR) (raw data shown in Extended Data Fig. 6b).
a, b, Plasmid-tag distribution in mucosa-associated donors and faecal
transconjugants at day 15. Dotted lines indicate conservative qPCR
detection limits (2.9 × 10−3
). Line indicates mean; error bars indicate
s.d. c, Modelling results. Simulations are described in the Supplementary
Information. A rate of donor re-seeding and donor-to-recipient
conjugation of η = 3.16 × 10−10
 events per recipient per day, and a rate
of transconjugant-to-recipient conjugation of γ = 3.16 × 10−8
 events
per recipient per CFU per gram of faeces per day, optimally fit the data
(red). d, Effect of η on transconjugant detection. Analysis as described in
the Supplementary Information. Reducing η by 100-fold below the value
estimated for naive mice (black dotted line; η = 3.16 × 10−10
 events)
diminishes transconjugant detection (red dotted line). e, f, Vaccination
diminishes conjugation. Mice vaccinated with killed S. Typhimurium
(open circles; n = 14 mice, 4 independent experiments) or PBS (naive;
closed circles; n = 9 mice, 3 independent experiments) were infected with
tagged donor and recipient mixtures as in a and b. Donors, recipients and
transconjugants were enumerated by plating. e, Mucosa-associated donor
populations at day 15. f, Faecal transconjugant proportions in naive (black
circles; grey bars indicate median) or vaccinated mice (light grey circles;
light grey bars indicate median) from e. Dashed lines connect data points
from the same mice. Statistics were performed using a two-tailed Mann–
Whitney U-test. NS, not significant (P ≥ 0.05), *P < 0.05, **P < 0.01,
***P < 0.001, ****P < 0.0001. Dotted lines denote detection limits.
2 7 8 | N A T U RE | V O L 5 7 3 | 1 2 S E P TE M B ER 2 0 1 9
Letter RESEARCH
of selective pressures for the functions encoded on the plasmid. Thus,
in the absence of antibiotic-driven selection, resistance genes on promiscuous
plasmids can spread from a very small number of donor cells
that co-occur with dense recipient-cell populations.
Mathematical modelling showed that re-seeding is the rate-limiting
aspect of this process, and suggested that reducing the tissue-associated
population of S. Typhimurium (for example, through vaccination)
could reduce plasmid re-seeding to negligible levels. As re-seeding
events are rare compared to the number of tissue-associated persisters,
such persister-promoted spread of plasmids may last for weeks
or months after an acute infection. We do not yet know whether this
translates to infections in livestock or humans, during which tissueassociated
reservoirs of persisters might be smaller.
Associations between persistence and chronic infections occur in
a variety of clinical contexts29
and in biofilms30
. Therefore, reservoirs
for conjugative plasmids could exist in a multitude of populations of
persisters. In the two mouse models discussed here, S. Typhimurium
persisters associated with different organs appear to be capable of
re-seeding the gut lumen. The intracellular environment in classical
dendritic cells7
or macrophages6
can induce persistence. In contrast
to infected epithelial cells31,32
, these phagocytes are long-lived, which
suggests that they hold most of the donor persisters in both the oral
and the intravenous model. The difference between the location of
the reservoir of persister cells in the oral (lamina propria and mesenteric
lymph nodes) and the intravenous (spleen, liver and gall bladder)
models, and the differential requirement for SPI-1- and SPI-2-encoded
virulence factors, indicate that there are diverse mechanisms through
which tissue-associated S. Typhimurium persisters can establish a reservoir,
survive and—ultimately—re-seed the gut lumen to engage in
plasmid transfer.
The link between persistence and plasmid-mediated evolution of
antibiotic resistance may be particularly relevant in the farming industry,
in which S. enterica has a high prevalence and animals are often
co-colonized by strains of Salmonella and E. coli. There can be two
Salmonella strains within the same host, one within lymphoid tissues
(for example, mesenteric lymph nodes) and the other in the intestinal
content17
. This is also consistent with the isolation of Salmonella
spp. from intestinal biopsies in swine that have chronic subclinical
Salmonella infections, indicating that reservoirs of pathogen persisters
are common in the gut mucosa of these animals15,16
. Our results suggest
that this may promote the spread of resistance plasmids in animal herds.
Strategies to reduce bacterial persistence are of general importance33
.
This is relevant not only for the clinical treatment of infected indi-
viduals29
, but also to minimize the effect of persistence on the global
evolution of antibiotic resistance via both resistance mutations11
and
resistance plasmid spread, as shown here. Inactivated vaccines, such as
the Salmonella cells killed by peracetic acid that we use here, should be
easily and safely applicable28
. Our data show that vaccination efficiently
prevents tissue invasion and disease, and minimizes the size of persister
reservoirs—and thereby reduces the subsequent spread of resistance
plasmids, such as those encoding ESBL proteins (Fig. 3).
Online content
Any methods, additional references, Nature Research reporting summaries,
source data, extended data, supplementary information, acknowledgements, peer
review information; details of author contributions and competing interests; and
statements of data and code availability are available at https://doi.org/10.1038/
s41586-019-1521-8.
Received: 18 October 2018;Accepted: 5August 2019;
Published online 4 September 2019.
	1.	 Parisi, A. et al. Health outcomes from multidrug-resistant Salmonella infections
in high-income countries: a systematic review and meta-analysis. Foodborne
Pathog. Dis. 15, 428–436 (2018).
	2.	 Wright, G. D. The antibiotic resistome: the nexus of chemical and genetic
diversity. Nat. Rev. Microbiol. 5, 175–186 (2007).
	3.	 Brauner, A., Fridman, O., Gefen, O. & Balaban, N. Q. Distinguishing between
resistance, tolerance and persistence to antibiotic treatment. Nat. Rev. Microbiol.
14, 320–330 (2016).
	4.	 Fridman, O., Goldberg, A., Ronin, I., Shoresh, N. & Balaban, N. Q. Optimization of
lag time underlies antibiotic tolerance in evolved bacterial populations. Nature
513, 418–421 (2014).
	5.	 Claudi, B. et al. Phenotypic variation of Salmonella in host tissues delays
eradication by antimicrobial chemotherapy. Cell 158, 722–733 (2014).
	6.	 Helaine, S. et al. Internalization of Salmonella by macrophages induces
formation of nonreplicating persisters. Science 343, 204–208 (2014).
	7.	 Kaiser, P. et al. Cecum lymph node dendritic cells harbor slow-growing bacteria
phenotypically tolerant to antibiotic treatment. PLoS Biol. 12, e1001793 (2014).
	8.	 Dolowschiak, T. et al. IFN-γ hinders recovery from mucosal inflammation during
antibiotic therapy for Salmonella gut infection. Cell Host Microbe 20, 238–249
(2016).
b
8 9 10 11 12 13 14 15
Day of experiment
Oral
E. coli recipient
Proportiontransconjugants
(ofrecipients)infaeces
a
Mucosa-associated population
(oral model, day 15)
Mucosa-associated population
(oral model, day 15)
8 9 10 11 12 13 14 15
Day of experiment
Oral
S.Tm recipient
ProportionESBL+transconjugants
(ofrecipients)infaeces
>99%
dc
>86%
CFUdonorsper
caecaltissue
CFUdonorsper
caecaltissueDonor (oral): S.Tm P2cat
Recipient (oral): E. coli 8178 P2cured
Donor (oral): S.Tm pESBL15
Recipient (oral): S.Tm 14028S
f
Proportiontransconjugants
(ofrecipients)infaeces
Oral
E. coli recipient
>25%
8 9 10 11 12 13 14 15
Day of experiment
7 16 17 1918 20 21 22 23 24 25
e
CFUdonorsperorgan
Caecal
tissue
mLN Spleen Liver Gall
bladder
Tissue-associated population
(i.v. model, day 25)
Donor (i.v.): S.Tm P2cat
Recipient (oral): E. coli 8178 P2cured
105
104
103
102
101
100
105
104
103
102
101
100
105
104
103
102
101
100
100
10–2
10–4
10–6
10–8
100
10–2
10–4
10–6
10–8
100
10–2
10–4
10–6
10–8
Fig. 4 | Tissue-associated persisters promote the transfer of resistance
plasmids between different species of Enterobacteriaceae. a, b, Plasmid
transfer to E. coli. Five mice (one experiment) were infected orally
as described in Fig. 1b (except that ampicillin was removed from the
drinking water on day 8), with donors (S. Typhimurium P2cat
SL1344
strain, SmR
and CmR
) and recipients (E. coli 8178 P2cured
strain, AmpR
).
Mucosa-associated donors and faecal transconjugants were enumerated
by plating. Median indicated by solid line or bars. c, d, Transfer of
plasmids containing ESBL genes between Salmonella spp. The experiment
used S. Typhimurium SL1344 pESBL15 (SmR
and AmpR
) as the donor
and S. Typhimurium 14028S aphT (KanR
) as the recipient (n = 5 mice,
1 experiment). This experiment was performed and evaluated as in a,
b, except that kanamycin was used in the drinking water from day 3 to
day 8 (instead of ampicillin, as pESBL15 confers ampicillin resistance).
e, f, Plasmid transfer from S. Typhimurium to E. coli using the intravenous
model. Six mice (two independent experiments) were infected as described
in Fig. 2b, with donors (S. Typhimurium P2cat
SL1344 strain (SmR
and
CmR
), intravenous) and recipients (E. coli 8178 P2cured
(AmpR
), oral).
Tissue-associated donors and faecal transconjugants were enumerated by
plating. Median indicated by solid line or bars. a–f, Dotted lines indicate
detection limit by selective plating. b, d, f, Dashed lines connect data
points from the same mice to illustrate the progression of plasmid spread
after initial detection.
1 2 S E P TE M B ER 2 0 1 9 | V O L 5 7 3 | N A T U RE | 2 7 9
LetterRESEARCH
	9.	 Balaban, N. Q. et al. Definitions and guidelines for research on antibiotic
persistence. Nat. Rev. Microbiol. 17, 441–448 (2019).
	10.	 Balaban, N. Q., Merrin, J., Chait, R., Kowalik, L. & Leibler, S. Bacterial persistence
as a phenotypic switch. Science 305, 1622–1625 (2004).
	11.	 Levin-Reisman, I. et al. Antibiotic tolerance facilitates the evolution of resistance.
Science 355, 826–830 (2017).
	12.	 Wotzka, S. Y. et al. Microbiota stability in healthy individuals after single-dose
lactulose challenge–a randomized controlled study. PLoS ONE 13, e0206214
(2018).
	13.	 Coque, T. M., Baquero, F. & Canton, R. Increasing prevalence of ESBL-producing
Enterobacteriaceae in Europe. Eurosurveillance 13, 19044 (2008).
	14.	 Crump, J. A., Sjölund-Karlsson, M., Gordon, M. A. & Parry, C. M. Epidemiology,
clinical presentation, laboratory diagnosis, antimicrobial resistance, and
antimicrobial management of invasive Salmonella infections. Clin. Microbiol.
Rev. 28, 901–937 (2015).
	15.	 Wilcock, B. P., Armstrong, C. H. & Olander, H. J. The significance of the serotype
in the clinical and pathological features of naturally occurring porcine
salmonellosis. Can. J. Comp. Med. 40, 80–88 (1976).
	16.	 Wood, R. L., Pospischil, A. & Rose, R. Distribution of persistent Salmonella
typhimurium infection in internal organs of swine. Am. J. Vet. Res. 50,
1015–1021 (1989).
	17.	 San Román, B. et al. Relationship between Salmonella infection, shedding and
serology in fattening pigs in low–moderate prevalence areas. Zoonoses Public
Health 65, 481–489 (2018).
	18.	 Tenaillon, O., Skurnik, D., Picard, B. & Denamur, E. The population genetics of
commensal Escherichia coli. Nat. Rev. Microbiol. 8, 207–217 (2010).
	19.	 Apperloo-Renkema, H. Z., Van der Waaij, B. D. & Van der Waaij, D. Determination
of colonization resistance of the digestive tract by biotyping of
Enterobacteriaceae. Epidemiol. Infect. 105, 355–361 (1990).
	20.	 Stecher, B. et al. Gut inflammation can boost horizontal gene transfer between
pathogenic and commensal Enterobacteriaceae. Proc. Natl Acad. Sci. USA 109,
1269–1274 (2012).
	21.	 Diard, M. et al. Inflammation boosts bacteriophage transfer between Salmonella
spp. Science 355, 1211–1215 (2017).
	22.	 Moor, K. et al. High-avidity IgA protects the intestine by enchaining growing
bacteria. Nature 544, 498–502 (2017).
	23.	 Monack, D. M., Bouley, D. M. & Falkow, S. Salmonella typhimurium persists within
macrophages in the mesenteric lymph nodes of chronically infected Nramp1+/+
mice and can be reactivated by IFNγ neutralization. J. Exp. Med. 199, 231–241
(2004).
	24.	 Diard, M. et al. Antibiotic treatment selects for cooperative virulence of
Salmonella Typhimurium. Curr. Biol. 24, 2000–2005 (2014).
	25.	 Sampei, G. et al. Complete genome sequence of the incompatibility group I1
plasmid R64. Plasmid 64, 92–103 (2010).
	26.	 Hensel, M. et al. Simultaneous identification of bacterial virulence genes by
negative selection. Science 269, 400–403 (1995).
	27.	 Stapels, D. A. C. et al. Salmonella persisters undermine host immune defenses
during antibiotic treatment. Science 362, 1156–1160 (2018).
	28.	 Moor, K. et al. Peracetic acid treatment generates potent inactivated oral
vaccines from a broad range of culturable bacterial species. Front. Immunol. 7,
34 (2016).
	29.	 Fauvart, M., De Groote, V. N. & Michiels, J. Role of persister cells in chronic
infections: clinical relevance and perspectives on anti-persister therapies.
J. Med. Microbiol. 60, 699–709 (2011).
	30.	 Roberts, M. E. & Stewart, P. S. Modelling protection from antimicrobial agents in
biofilms through the formation of persister cells. Microbiology 151, 75–80
(2005).
	31.	 Knodler, L. A. et al. Noncanonical inflammasome activation of caspase-4/
caspase-11 mediates epithelial defenses against enteric bacterial pathogens.
Cell Host Microbe 16, 249–256 (2014).
	32.	 Sellin, M. E. et al. Epithelium-intrinsic NAIP/NLRC4 inflammasome drives
infected enterocyte expulsion to restrict Salmonella replication in the intestinal
mucosa. Cell Host Microbe 16, 237–248 (2014).
	33.	 Defraine, V., Fauvart, M. & Michiels, J. Fighting bacterial persistence: current and
emerging anti-persister strategies and therapeutics. Drug Resist. Updat. 38,
12–26 (2018).
Publisher’s note: Springer Nature remains neutral with regard to jurisdictional
claims in published maps and institutional affiliations.
© The Author(s), under exclusive licence to Springer Nature Limited 2019
2 8 0 | N A T U RE | V O L 5 7 3 | 1 2 S E P TE M B ER 2 0 1 9
Letter RESEARCH
Methods
No statistical methods were used to predetermine sample size. The experiments
were not randomized and investigators were not blinded to allocation during
experiments and outcome assessment. This was necessary to avoid crosscontamination
of pathogens and plasmid donors between the different experimental
groups as far as possible.
Strains and plasmids used in this study. Supplementary Table 1 contains all
strains and plasmids used in this study. For cultivation of bacteria, lysogeny
broth (LB) medium containing the appropriate antibiotics (50 μg/ml streptomycin
(AppliChem), 6 μg/ml chloramphenicol (AppliChem), 50 μg/ml kanamycin
(AppliChem) or 100 μg/ml ampicillin (AppliChem)) were used. To create P2cat tag
strains (using previously published neutral genetic barcodes34
) or gene-deletion
mutants (for example, oriT::aphT on P2), the λ red system was used, as previously
described35
. If desired, antibiotic-resistance cassettes were removed using the temperature-inducible
FLP recombinase encoded on pCP2035
. Mutations or sequence
tags coupled to antibiotic-resistance cassettes were transferred into the desired
genetic background using transduction with P22 HT105/1 int-20136
. Primers
used for strain construction or verification of genetic background are listed in
Supplementary Table 2.
In vitro plasmid-transfer kinetics. Overnight cultures with appropriate antibiotics
of donor (SL1344 P2cat
) and recipient (14028S aphT) strains carrying pM975 to
confer ampicillin resistance were subcultured 1:20 in LB without antibiotics, and
grown for 4 h. Approximately 102
CFU of each were added (sequentially as 25-μl
volumes) to 450 μl LB with 100 μg/ml ampicillin. Samples were incubated at 37 °C
mixing at 1,000 r.p.m. for 24 h. Aliquots (10 μl) were taken every hour, diluted, and
plated on selective MacConkey agar for enumeration.
In vitro antibiotic-resistance profiling. Flat-bottom transparent 96-well plates
were filled with 100 μl of LB containing twofold dilutions of the specified antibiotic
(streptomycin (AppliChem), chloramphenicol (AppliChem), kanamycin
(AppliChem), ampicillin (AppliChem), ciprofloxacin (ciprofloxacin hydrochloride
monohydrate, Sigma-Aldrich), gentamicin (AppliChem) or ceftriaxone (ceftriaxone
disodium salt hemi(heptahydrate), Sigma-Aldrich). Each plate contained
eleven twofold dilution steps of each antibiotic, plus a no antibiotic control and
a no-bacteria sterility control. Overnight cultures of each bacterial strain tested
were grown in the presence of appropriate antibiotics, subcultured (1:20 dilution)
for 4 h at 37 °C in LB without antibiotics, and diluted in PBS. Cells were seeded in
each well at a final density of 105
CFU/ml. Ninety-six-well plates were incubated
at 37 °C at 120 r.p.m. for 16 h and the optical density at 600 nm (OD600 nm) was
measured. The no-bacteria sterility control was used for background subtraction.
Infection experiments. Oral infection model. For in vivo plasmid-transfer
experiments, 8–12-week-old 129/SvEv mice were used. These mice are
Nramp1+/+
(Nramp1 is also known as Slc11a1) and therefore allow for long-term
S. Typhimurium infections37
. They carry a complex specified-pathogen-free
microbiota without E. coli. The experiment has three phases: (1) donor colonization;
(2) clearance with antibiotics; and (3) recipient colonization and conjugative
transfer. In phase (1), mice were pre-treated with 25 mg streptomycin
orally to allow for robust colonization of S. Typhimurium38
. Donor (that is, plasmid-bearing)
S. Typhimurium was cultivated overnight at 37 °C in LB containing
the appropriate antibiotics, subcultured (1:20 dilution) for 4 h at 37 °C in LB without
antibiotics, washed with sterile PBS, and 5 × 107
CFU were introduced into
mice via oral gavage38
. In phase (2), 2 days after infection, 3 mg of ciprofloxacin
(ciprofloxacin hydrochloride monohydrate, Sigma-Aldrich) dissolved in 100 μl
sterile dH2O was administered by oral gavage for 3 consecutive days. Mice were
transferred to fresh cages after each ciprofloxacin treatment to minimize gut
recolonization from the environment. Ampicillin (2 g/l) or kanamycin (1 g/l) was
added to the drinking water starting at day 3 after infection, and maintained until
either day 8 or day 15. This prevented premature donor re-seeding after the cessation
of ciprofloxacin treatment. In phase (3), mice were kept individually after
ciprofloxacin treatment to prevent cross-contamination owing to coprophagy.
Recipient bacteria (that is, plasmid-free S. Typhimurium or E. coli; 5 × 107
CFU)
were orally introduced into mice on day 8 after infection (culture and subculture
conditions as above) and populations were monitored for 7 days until they were
euthanized (day 15).
Intravenous infection model. The in vivo plasmid-transfer experiment had three
phases: (1) donor colonization; (2) clearance with antibiotics; and (3) recipient
colonization and conjugative transfer. In phase (1), the donor strain was injected
intravenously into the tail vein of 8–12-week-old 129/SvEv mice (103
CFU). In
phase (2), 2 days after intravenous infection of donors, 1.5 mg of ceftriaxone (ceftriaxone
disodium salt hemi(heptahydrate) dissolved in 100 μl PBS, Sigma-Aldrich)
was intraperitoneally injected for 3 consecutive days. After the third treatment,
mice were transferred to fresh cages and kept individually to prevent cross-
contaminationowingtocoprophagy.Inphase(3),therecipientwasintroducedonday
7 after donor infection (108
CFU by gavage) and faecal populations were monitored
for 18 days until the mice were euthanized (day 25).
In both infection models, faeces were collected daily, homogenized in PBS with
a steel ball at 25 Hz for 1 min and diluted; selective plating on MacConkey agar
(supplemented with the appropriate antibiotics) was used to enumerate populations
of donors, recipients or transconjugants. The proportion of transconjugants
was calculated by dividing the transconjugant population (CmR
and KanR
) by the
sum of transconjugants and plasmid-free recipients (KanR
). Lipocalin-2 enzymelinked
immunosorbent assay (ELISA) (R&D Systems kit; protocol according to
manufacturer) was performed on faeces to determine the inflammatory state of
the gut. When the mice were euthanized (at day 5, 8, 15 or 25, as specified in figure
legends), the mesenteric lymph nodes, spleen, liver and gall bladder were collected,
homogenized in PBT at 25 Hz for 2 min, and bacteria were enumerated by selective
plating. The caecum was removed, opened longitudinally, washed 3 times in PBS,
and then placed in 400 μg/ml gentamicin (AppliChem) for 30 min at room temperature.
Nine consecutive washing steps in PBS (45 s each) ensured the removal
of gentamicin; caecal tissue was subsequently homogenized and mucosa-associated
bacteria were enumerated as for the other organs (gentamicin protection modified
from a previously published protocol24
). In Extended Data Fig. 5a, additional
organs were collected as indicated in the figure legend. All organs were processed as
indicated above. For the jejunum, ileum and colon, 1 cm of intestine was collected,
washed briefly in PBS and the tissue (including the majority of the content) was
analysed. For analysis of bacteria in the blood, 100 μl of blood was aspirated from
the heart immediately after euthanasia and mixed in PBS containing 2% BSA and
1 mM EDTA to prevent coagulation.
For competition experiments, 8–12-week-old 129/SvEv mice were pre-treated
with 20 mg ampicillin orally, and ampicillin (2 g/l) was maintained in the drinking
water throughout the experiment. The two competitor strains were cultured
separately in LB containing the appropriate antibiotics, subcultured and washed
with PBS, as above. Strains were mixed at a 1:1 ratio immediately before gavaging
5 × 107
CFU of the mixture into the ampicillin-pre-treated mice. Faeces were
monitored daily, homogenized and enumerated by selective plating. The competitive
index was calculated using the ratio of population sizes of competitors at the
indicated time. Mice were euthanized after seven days.
All mouse infection experiments were approved by the responsible authority
(Tierversuchskommission, Kantonales Veterinäramt Zürich, license 193/2016).
Sample size was not predetermined. Mouse age and gender were matched between
treatment groups and animals were randomly distributed among groups. In four
cases, data from mice were excluded because the mice needed to be euthanized
prematurely, owing to disease or symptom severity.
Confocal microscopy. To visualize persisters in the caecum lamina propria,
caecum tissues were fixed in PBS/4% paraformaldehyde, saturated in PBS/20%
sucrose and embedded in optimal cutting temperature medium (OCT, TissueTek)
before being flash-frozen in liquid nitrogen. Ten-micrometre cryosections
were air-dried, rehydrated with PBS, permeabilized with PBS/0.5%Triton X-100
and blocked with PBS/10% normal goat serum. Anti-S. Typhimurium LPS O5
(Difco), anti-S. Typhimurium LPS O12 (STA5, ref. 22
), anti-ICAM-1/CD54 (BD
Biosciences), appropriate secondary antibodies, DAPI (Sigma Aldrich) and AlexaFluor-488-conjugated
phalloidin (Santa Cruz) were used for the staining. A Zeiss
Axiovert 200m microscope with 10–100× objectives, a spinning-disc confocal-laser
unit (Visitron), and two Evolve 512 EMCCD cameras (Photometrics)
were used for acquiring images. Images were processed using Visiview (Visitron).
LPS-positive (O5-positive and/or O12-positive) S. Typhimurium was manually
enumerated blindly in 8–12 nonconsecutive sections per mouse. Phalloidin and
ICAM-1 staining were used to differentiate the lamina propria and epithelium. All
data represent averages per section.
Mucus fixation and staining. Caecal tissue samples were fixed with freshly prepared
Methacarn solution (60% methanol, 30% chloroform and 10% glacial acetic
acid) for 24 h at room temperature39
. The samples were transferred to methanol
for 2 h and processed over night with a LogosJ tissue processor (Milestone) using
the following program: 30 min, 37 °C EtOH; 30 min, 37 °C EtOH; 60 min, 37 °C
EtOH; 60 min, 40 °C isopropanol, 15 min for heating up; 60 min, 45 °C isopropanol,
20 min for heating up; 180 min, 68 °C isopropanol, 20 min for heating up; and
240 min, 82 °C paraffin, 75 min for heating up.
The paraffinized tissue was then embedded as paraffin blocks for further storage.
Ten-micrometre sections were deparaffinized in Xylene substitute solution
(Sigma-Aldrich) for 20 min. The sections were rehydrated in sequential baths of
100%, 95%, 70%, 50% and 30% ethanol for 5 min each, and subsequently incubated
for 10 min in PBS. For mucus visualization, the sections were stained with
DAPI, phalloidin–FITC, wheat germ agglutinin AF647 conjugate (Invotrogen,
cat. no. W32466), rabbit polyclonal anti-Salmonella O5 (Becton Dickinson,
cat. no. 226601) and goat polyclonal anti-rabbit Fab Cy3 conjugate (Jackson
ImmunoResearch Laboratories, cat. no. 111-167-003; RRID: AB_2313593) anti-
bodies.
Analysis of plasmid-transfer dynamics. Mice were orally or intravenously
infected with a mixture of 5 isogenic SL1344 P2cat tag
strains (tags at a 1:1:1:1:1
LetterRESEARCH
ratio; inoculum made of approximately 107
CFU of each tagged strain for oral
infection and 2 × 102
CFU of each tagged strain for intravenous infection) using
the model protocol described in ‘Oral infection model’ and ‘Intravenous infection
model’ (Figs. 1b, 2b). On day 8 (oral model), mice were gavaged with a mixture of
5 recipient strains (14028S tag, KanR
and AmpR
; tags at 1:1:1:1:1 ratio; inoculum
approximately 107
CFU for each tag). Barcode analysis of the recipient chromosome
tags could not be performed because of technical issues with kanamycin
enrichments and subsequent qPCR. In the intravenous model, the recipient used
was not tagged, but was introduced on day 7 (108
CFU by oral gavage). The plasmid
and recipient tags can be easily distinguished, as the primer pairs used for qPCR
are unique (that is, WITSX-R paired with Cat_internal for P2 tags and WITSX-R
paired with Kan_internal for recipient tags; in the intravenous model in which
recipient tags were not used, WITSX was paired with ydgA for qPCR as there
was no need for antibiotic-resistance-specific qPCR primers). For the oral model,
the inocula, faeces at days 1, 9 and 15, as well as caecal tissue after gentamicin
treatment, was enriched overnight in 5 ml LB containing the appropriate antibiotics
(donors = streptomycin + chloramphenicol; recipients = kanamycin; and
transconjugants = chloramphenicol + kanamycin) in parallel with selective plating
to enumerate bacterial population sizes. For the intravenous model, the donor
inoculum, as well as faeces, gall bladder, liver, spleen, mesenteric lymph nodes
and caecal tissue at euthanasia (day 25) were enriched in LB with the appropriate
antibiotics.
Enrichments were concentrated and genomic DNA was extracted using a
QIAamp DNA Mini Kit (Qiagen). qPCR analysis was performed using temperature
conditions as previously described34
. qPCR primers are listed in Supplementary
Table 2. For each sample, 5 primer pairs were used to amplify P2 tags (in the oral
model WITSX-R with Cat_internal, in which X denotes 2, 11, 13, 19 or 21; in the
intravenous model WITSX with ydgA in which X denotes 2, 11, 13, 19 or 21).
Primers were modified from the original, previously described WITS primers34
; tag
loci remain the same as previously described34
. Relative proportion was determined
by dividing the DNA copy number (calculated from the CT value; a dilution standard
of purified chromosomal DNA allowed for a correlation between DNA copy
number and CT value) of each tag detected by the sum of all tags in the sample. The
detection limit was determined by the CT value of the most-diluted DNA standard
(linearity of the standard curve decreases markedly below this value) for each qPCR
run. The least-precise detection limit constitutes the conservative detection limit
used for fitting our mathematical model (2.9 × 10–3
). Plasmid tags were ranked
according to frequency (Extended Data Fig. 6c) as the strains are isogenic (except
for the tag), and each mouse can be treated as an independent realization of the
stochastic population dynamics. The data of plasmid-tag frequencies and the bacterial
population counts were used to parameterize a mechanistic model of the plasmid
re-seeding dynamics, and infer the most-likely rates of donor re-seeding (including
initial plasmid transfer) and transconjugant-to-recipient transfer (Extended Data
Fig. 6g; a detailed description is given in Supplementary Information).
Oral vaccination. Oral vaccines of S. Typhimurium inactivated by peracetic acid
were prepared as previously described28
. S. Typhimurium was grown overnight and
concentrated to 1010
CFU/ml in PBS. Peracetic acid (Sigma-Aldrich) was added to
a final concentration of 0.4% v/v, mixed vigorously and incubated at room temperature
for 2 h. Traces of acid were removed after inactivation by washing bacteria
with sterile PBS three times. Inactivated bacteria were resuspended at 1011
particles
per ml in sterile PBS. Complete inactivation was ensured by inoculating a 100-μl
dose of vaccine into 50 ml LB and checking for sterility. Vaccines were stored at
4 °C for up to 2 months. Mice received a 100-μl dose of vaccine by oral gavage
once per week for 5 weeks. Naive control mice were mock-vaccinated with PBS.
Statistical analysis. Statistical tests on experimental data were performed using
GraphPad Prism 7 for Windows. Fitting of the mathematical model to experimental
values was performed using an approximate Bayesian computation approach40
.
Transferrateswerevariedonagridfrom10−12
to10−1
;thesummarystatisticsconsist
of the skew of the plasmid-tag abundance distribution, the fraction of tags above
the detection limit on day 15, the total size of the transconjugant population on day
15 and the time at which the transconjugant population size first exceeds 106
CFU
per gram of faeces. A simulation is called ‘successful’ if all the summary statistics
are within three s.d. of the experimentally observed mean of these statistics.
Ethical statement. All mouse experiments were ethically approved by the responsible
authorities (Tierversuchskommission, Kantonales Veterinäramt Zürich,
license 193/2016). In four cases, data from mice were excluded because the mice
needed to be euthanized prematurely, owing to disease or symptom severity.
Reporting Summary. Further information on research design is available in
the Nature Research Reporting Summary linked to this article.
Data availability
The genome and plasmid sequence of E. coli ESBL 15 have been deposited
in GenBank under accession numbers CP041678–CP041681 (Biosample
SAMN12275742). Numerical Source Data for all figures are provided with the
paper. Source images are available upon request to the corresponding authors.
Code availability
Code for the stochastic simulation of plasmid-transfer dynamics and parameter
estimation from the experimental data are provided with the paper. The R code
follows the notation used in the Supplementary Information.
	34.	 Grant, A. J. et al. Modelling within-host spatiotemporal dynamics of invasive
bacterial disease. PLoS Biol. 6, e74 (2008).
	35.	 Datsenko, K. A. & Wanner, B. L. One-step inactivation of chromosomal genes in
Escherichia coli K-12 using PCR products. Proc. Natl Acad. Sci. USA 97,
6640–6645 (2000).
	36.	 Sternberg, N. L. & Maurer, R. Bacteriophage-mediated generalized transduction
in Escherichia coli and Salmonella typhimurium. Methods Enzymol. 204, 18–43
(1991).
	37.	 Stecher, B. et al. Chronic Salmonella enterica serovar Typhimurium-induced
colitis and cholangitis in streptomycin-pretreated Nramp1+/+
mice. Infect.
Immun. 74, 5047–5057 (2006).
	38.	 Barthel, M. et al. Pretreatment of mice with streptomycin provides a Salmonella
enterica serovar Typhimurium colitis model that allows analysis of both
pathogen and host. Infect. Immun. 71, 2839–2858 (2003).
	39.	 Johansson, M. E. & Hansson, G. C. Preservation of mucus in histological
sections, immunostaining of mucins in fixed tissue, and localization of bacteria
with FISH. Methods Mol. Biol. 842, 229–235 (2012).
	40.	 Marjoram, P., Molitor, J., Plagnol, V. & Tavare, S. Markov chain Monte Carlo
without likelihoods. Proc. Natl Acad. Sci. USA 100, 15324–15328 (2003).
	41.	 Zankari, E. et al. Identification of acquired antimicrobial resistance genes. J.
Antimicrob. Chemother. 67, 2640–2644 (2012).
Acknowledgements We thank the members of the Hardt, Slack, Bonhoeffer,
Stadler and Ackermann labs for helpful discussion, and the staff at the RCHCI
and EPIC animal facilities for their excellent support. This work has been funded,
in part, by grants from the Swiss National Science Foundation (SNF; 310030B-
173338), the Promedica Foundation, Chur and the Helmut Horten Foundation
to W.-D.H., and from the SNF NFP 72 (407240-167121) to W.-D.H., S.B. and
A.E. M.D. is funded by an SNF professorship grant (PP00PP_176954), E.B. by a
Boehringer Ingelheim Fonds PhD fellowship, and M.E.S. and S.A.F. (in part) by
the Swedish Research Council (2015-00635, 2018-02223). R.R.R. is funded by
SNF grant number 31003A_179170. E.S. is supported by grant GRS 073/17
from the ‘Microbials’ programme of the Gebert Rüf Foundation and the SNF
Bridge Discover Grant 20B2-1 180953.
Author contributions E.B. (Figs. 1, 2, 3a, b, e, f, 4, Extended Data Figs. 1–3,
4a–c, e, f, 5–9), S.A.F. (Fig. 1e, Extended Data Fig. 4a–d), A.H. (Fig. 2, Extended
Data Figs. 5, 6e, f, 9e), M.F. (Extended Data Fig. 4e–h) and E.S. (Fig. 3e, f,
Extended Data Fig. 7b) performed the experiments shown in the indicated
figures. J.S.H., S.B. and R.R.R. (Fig. 3c, d, Extended Data Figs. 1b, 6g, 10)
performed mathematical modelling shown in the indicated figures. A.E.
(Extended Data Fig. 1b) provided E. coli strain Z2115. E.B., M.D. and W.-D.H.
designed the experiments. S.A.F., M.F. and M.E.S. designed the microscopybased
experiments and analysis. E.B., M.D. and W.-D.H. conceived the project
and wrote the manuscript. All authors read, commented and approved this
manuscript.
Competing interests The authors declare no competing interests.
Additional information
Supplementary information is available for this paper at https://doi.org/
10.1038/s41586-019-1521-8.
Correspondence and requests for materials should be addressed to M.D. or
W.-D.H.
Peer review information Nature thanks Sophie Helaine and the other,
anonymous, reviewer(s) for their contribution to the peer review of this work.
Reprints and permissions information is available at http://www.nature.com/
reprints.
Letter RESEARCH
Extended Data Fig. 1 | See next page for caption.
LetterRESEARCH
Extended Data Fig. 1 | Emergence and spread of antibiotic resistance in
bacteria using P2 as a model for conjugation. a, Antibiotic resistance in
bacteria can emerge through mutation, or be acquired via horizontal gene
transfer. Plasmid transfer is an important driver of the spread of antibiotic
resistance. Tolerance increases the abundance of bacteria that survive
antibiotic exposure, allowing for a higher probability of the emergence
of mutations that lead to resistance11
. We hypothesize that antibiotic
resistance can also spread through the formation of reservoirs of persisters
that contain plasmids; here, we hypothesize that the gut mucosa of the
host can serve as a reservoir for persisters. The formation of long-term
reservoirs, followed by re-seeding of bacteria from this reservoir into a
niche occupied by other bacteria (for example, the gut lumen occupied by
the microbiota following antibiotic therapy) increases the chance that two
different strains interact with each other, leading to plasmid transfer (that
is, increased strain co-occurrence). The representation in the bottom right
panel is an example of donor persisters boosting co-occurrence. Note that
tissue-associated recipient persisters may also increase co-occurrence.
b, P2 shares homology with resistance plasmids. An alignment is shown
between S. Typhimurium SL1344 P2 (GenBank sequence identifier
HE654725.1), S. Typhimurium plasmid R64 (GenBank sequence identifier
AP005147.1), and pESBL15 of E. coli Z2115 (strain isolated from a rectal
swab of a patient at the University Hospital Basel) using the Artemis
comparison tool (https://www.sanger.ac.uk/science/tools/artemiscomparison-tool-act).
Red fill indicates high sequence identity (>85%
sequence identity), blue fill indicates inversions and no fill indicates
no sequence identity. For each plasmid, open reading frames
(in each of the six translational frames) are shown by white regions
(detected by the Artemis comparison tool). Antibiotic resistances
(for example, streptomycin and tetracycline resistances on R64 and
CTX-M-1 on pESBL15) are labelled, shown by light-blue directed
rectangles (found by a Basic Local Alignment Search Tool (NCBI) search
against the ResFinder antibiotic-resistance-gene database41
). In P2,
the locus for insertion of the chloramphenicol-resistance cassette and
neutral sequence tags is shown. For each alignment, the percentage of
the sequence that aligns to P2 is shown, as well as the average sequence
similarity for these regions. c, Model strains for addressing evolution by
conjugation in S. Typhimurium. SL1344 contains P2cat
(chloramphenicolresistance
cassette (cat) that allows the enumeration of plasmidbearing
strains by selective plating) that can be conjugated to 14028S
(kanamycin-resistance cassette (aphT) used for selective plating) to form
a transconjugant (CmR
and KanR
). Transconjugants can then transfer
P2cat
to additional recipients. d, P2cat
transfer kinetics in vitro. P2cat
transfer is dependent on the density of donors and recipients. Donor and
recipient strains were inoculated into LB (n = 2, 1 experiment) at a 1:1
ratio and selective plating was performed every hour. CFUs per millilitre
are reported for each population (donors in blue, recipients in green and
transconjugants in red). Solid lines connect medians. The dotted line
indicates the detection limit by selective plating.
Letter RESEARCH
Extended Data Fig. 2 | Antibiotic-resistance profile of key strains.
Antibiotic susceptibility testing was performed in LB in 96-well plates. Six
strains (indicated on the figure axes) were tested against seven antibiotics,
grown at 37 °C at 120 r.p.m. for 16 h, at which point the OD600 nm was
measured. For each antibiotic, the highest concentration used was based
on the working concentration in this study. For example, the concentration
of antibiotic used for selective plating in the case of streptomycin,
kanamycin, ampicillin and chloramphenicol (the highest concentration
of chloramphenicol was fivefold-higher than the concentration used
for selective plating, because this was already close to the minimum
inhibitory concentration), or the concentration of antibiotic used for
the gentamicin protection assay. Importantly, a very low minimum
inhibitory concentration was observed for antibiotics used in vivo to
enrich for persisters (that is, ciprofloxacin and ceftriaxone). The mean
of three experiments is presented on a blue–white colour gradient, in
which blue indicates a large amount of bacterial growth. This is calculated
by subtracting the OD600 nm measured for each sample well from the
background generated by the medium.
LetterRESEARCH
Extended Data Fig. 3 | See next page for caption.
Letter RESEARCH
Extended Data Fig. 3 | Controls for conjugation after antibiotic
treatment in the oral infection model. a–c, Faecal bacterial-population
sizes and inflammatory status of mice in Fig. 1c, d (comparison of invasive
versus non-invasive donors in the oral model). Faecal loads of donors
(blue, SmR
and CmR
), recipients (green, KanR
), and transconjugants (red,
CmR
and KanR
) were determined by selective plating on MacConkey agar.
Black dotted line indicates detection limits for donors and transconjugants.
Green dotted line indicates detection limit for recipients. The detection
limit is higher for recipients once transconjugants reach a density of >108
CFU per gram of faeces. Before this happens, recipients can be found
below the detection limit; the black dotted line should then be considered
as the detection limit. Blue lines connect medians of donor populations;
red lines connect medians of transconjugant populations. a, Mice infected
with invasive S. Typhimurium donors (solid circles; n = 15 singly housed
mice from 5 independent experiments). b, Mice infected with noninvasive
S. Typhimurium donors (open circles; n = 6 singly housed mice
from 2 independent experiments). c, Inflammatory status was determined
by lipocalin-2 ELISA. Statistics were performed using a two-tailed
Mann–Whitney U-test. NS, not significant (P ≥ 0.05), ****P < 0.0001,
comparing mice infected with an invasive donor (solid black circles;
n = 15 singly housed mice from 5 independent experiments) to mice
infected with a non-invasive donor (open black circles; n = 6 singly
housed mice from 2 independent experiments) at each time point.
Medians are shown (solid red line for invasive donors; dotted red line
for non-invasive donors). Dotted line indicates the detection limit.
d, e, Carrying P2cat
does not lead to a measurable fitness cost or benefit.
d, A ‘locked’ transconjugant (14028S P2aphT ΔoriT
(KanR
and AmpR
), in
which conjugation is blocked by removing the origin of transfer) was
competed against a recipient (14028S cat (CmR
and AmpR
)). e, To ensure
that removing the origin of transfer did not affect fitness, the locked
transconjugant was competed against a transconjugant with a normal P2cat
plasmid (that is, mobile transconjugant) (14028S P2cat
(CmR
and AmpR
)).
In d, e, both strains were introduced at a 1:1 ratio (total inoculum size
about 5 × 107
 CFU per os) and faeces were monitored daily by selective
plating (n = 6 singly housed mice from 2 independent experiments for
both experiments). The competitive index is calculated by dividing the
population size of one competitor by the other. Lines indicate medians.
The dotted line indicates no competitive advantage for either strain. This
is consistent with previously published data20
. These data indicate that
it is the plasmid-encoded conjugation efficiency (not the effects of the
plasmid on host bacterial fitness) that drives the rise of transconjugants
(Fig. 1c). f, g, Plasmid transfer in the oral model does not require an
invasive recipient. Invasive donors (SL1344 P2cat
(SmR
and CmR
)) were
orally infected into pre-treated mice. After ciprofloxacin treatment, a
non-invasive mutant of S. Typhimurium 14028S was used as a recipient
(non-invasive 14028S aphT (KanR
and AmpR
)). n = 5 mice. f, Selective
plating determined faecal loads of donors (blue, SmR
and CmR
), recipients
(green, KanR
), and transconjugants (red, CmR
and KanR
). Black dotted
line indicates the detection limit for donors and transconjugants. Green
dotted line indicates the detection limit for recipients. The detection limit
is higher for recipients once transconjugants reach a density of >108
CFU
per gram of faeces. Before this happens, recipients can be found below
the detection limit; the black dotted line should then be considered as the
detection limit. Blue lines connect medians of donor populations; red lines
connect medians of transconjugant populations. g, Donor populations
enumerated after a gentamicin protection assay on caecal tissue of mice
shown in f. Median indicated by solid line. Dotted line indicates the
detection limit. h, i, Conjugation is required for plasmid transfer after
antibiotic treatment. Mice were infected with invasive S. Typhimurium
that lacks the origin of transfer in P2cat
(SL1344 P2cat ΔoriT
(SmR
and
CmR
)) as a donor, and 14028S aphT (KanR
and AmpR
) as a recipient,
after antibiotic treatment. n = 5 mice. h, Selective plating determined
the faecal loads of donors (blue, SmR
and CmR
), recipients (green, KanR
)
and transconjugants (red, CmR
and KanR
). Black dotted line indicates the
detection limit for donors and transconjugants. Green dotted line indicates
detection limit for recipients. The detection limit is higher for recipients
once transconjugants reach a density of >108
CFU per gram of faeces.
Before this happens, recipients can be found below the detection limit; the
black dotted line should then be considered as the detection limit. Blue
lines connect medians of donor populations; red lines connect medians
of transconjugant populations. i, Donor populations enumerated after a
gentamicin protection assay on caecal tissue of mice shown in h. Median
indicated by solid line. Dotted line indicates the detection limit.
LetterRESEARCH
Extended Data Fig. 4 | See next page for caption.
Letter RESEARCH
Extended Data Fig. 4 | Quantification and localization of
S. Typhimurium in the host mucosa after antibiotic treatments in
the oral model. a–d, Mice were orally infected with either an invasive
(SL1344 P2cat
) (blue solid circles, n = 7 mice) or non-invasive (noninvasive
SL1344 P2cat
, TTSS-1 negative) (blue open circles, n = 7 mice)
donor and treated with antibiotics. Mice were euthanized at day 8 after
infection (at which point recipients were normally added), and organs
were analysed. Dotted lines indicate the detection limits. a, Faecal
populations were monitored daily by selective plating on MacConkey agar.
Blue lines connect medians. b, Organ loads were determined by selective
plating. c, Population size of donors in the caecal mucosa determined by
selective plating after a gentamicin protection assay, or microscopy of
tissue sections (same mice for each quantification method). Each data
point is the average of 12 sections (10-μm thick). b, c, Statistics were
performed using a two-tailed Mann–Whitney U-test. NS, not significant
(P ≥ 0.05), *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001,
comparing mice infected with invasive or non-invasive donors for each
organ. d, Localization of S. Typhimurium detected in the caecal tissue by
microscopy, reported as a percentage of bacteria detected in c in either the
lamina propria (l.p.) or epithelium (ep.). Bar indicates the median from
five mice. e–h, Analysis of reservoirs of persisters in Carnoy-fixed caecal
tissue sections. Mice were orally infected with an invasive (SL1344 P2cat
,
n = 5 mice) donor and treated with antibiotics. Mice were euthanized at
day 8 after infection (at which point recipients are normally added) and
organs were analysed. Dotted lines indicate the detection limits. e, Faecal
populations were monitored daily by selective plating on MacConkey agar.
Blue lines connect medians. f, Organ loads were determined by selective
plating. Line indicates median. g, A Carnoy fixation was performed on
caeca of mice to preserve the mucus structure. Ten-micrometre sections
were stained to visualize S. Typhimurium (yellow, anti-LPS O5), actin
(green, phalloidin–FITC), the mucus (red, wheat germ agglutinin AF647
conjugate) and nuclei (blue, DAPI). Mu., mucus. Scale bars, 20 μm. White
arrows highlight S. Typhimurium (magnified in inset). Representative
images shown from two independent experiments. h, Localization of
S. Typhimurium detected in the caecal tissue by microscopy, reported
as a percentage of bacteria detected each section (12 sections per mouse
caecum) in the lamina propria, epithelium, mucus or lumen. Bar indicates
the median from five mice.
LetterRESEARCH
Extended Data Fig. 5 | See next page for caption.
Letter RESEARCH
Extended Data Fig. 5 | Determination of reservoirs of tissue-associated
persisters after intravenous infection, and subsequent plasmid transfer.
a, An equal mix of five SL1344 P2cat tag
strains (SmR
and CmR
) were
intravenously infected into mice (103
CFU). Treatments were as described
in Fig. 2b (grey circles, 3 intraperitoneal doses of ceftriaxone; n = 8 singly
housed mice from 2 independent experiments) or mice were left untreated
(black circles; n = 8 singly housed mice from 2 independent experiments).
Mice were euthanized on day 5 of the experiment (after the final dose
of ceftriaxone was given). The tissue-associated populations in organs
were enumerated by selective plating. Detection limit by selective plating
is shown as a dotted line. Lines indicate median. b, c, Faecal bacterial
population sizes of mice in Fig. 2d, e (comparison of wild-type and
SPI-2-negative donors in the intravenous model). Faecal loads of donors
(blue, SmR
and CmR
), recipients (green, KanR
) and transconjugants
(red, CmR
and KanR
) were determined by selective plating on
MacConkey agar. Black dotted line indicates the detection limits for
donors and transconjugants. Green dotted line indicates the detection
limit for recipients. The detection limit is higher for recipients once
transconjugants reach a density of >108
CFU per gram of faeces. Before
this happens, recipients can be found below the detection limit; the
black dotted line should then be considered as the detection limit.
Blue lines connect medians of donor populations; red lines connect
medians of transconjugant populations. b, Mice infected with wild-type
S. Typhimurium donors (solid circles). c, Mice infected with SPI-2deficient
S. Typhimurium donors (open circles).
LetterRESEARCH
Extended Data Fig. 6 | See next page for caption.
Letter RESEARCH
Extended Data Fig. 6 | Experimental strategy for assessing population
dynamics and tag frequencies. Relates to Fig. 3a–d. a–c, Tags introduced
in the oral model. a, Tags coupled to a chloramphenicol-resistance cassette
were introduced in P2. qPCR primers are specific to the chloramphenicolresistance
cassette and the specific tag (shown as one-sided arrows).
Five tagged donors were pooled and orally infected as a 1:1:1:1:1 mix
into mice. b, Relative plasmid-tag proportion detected by qPCR in the
initial donor population, the donor population persisting in the caecal
mucosa and the transconjugant population is shown for eight mice (three
independent inocula). Dotted line indicates the detection limit. Each
tag is given a unique colour. c, Scheme illustrating how tags were sorted
and recoloured to yield the plots in Fig. 3a, b. Two mice are shown as
examples. For the mucosa-associated donor populations (top) and for the
faecal transconjugant populations (bottom), tags were sorted according
to relative frequency. These tags were re-coloured (darker colour
indicates higher frequency) to visualize the trends shown in Fig. 3a, b.
These re-ordered tags were used as the experimental data for fitting the
mathematical model. d–f, Experimental strategy to assess plasmid-transfer
dynamics in the intravenous model. d, Tags coupled to a chloramphenicolresistance
cassette were introduced in P2. qPCR primers are specific to
ydgA (a pseudogene flanking the specific tags) and the specific tag (shown
as one-sided arrows). Five tagged donors were pooled and intravenously
infected as a 1:1:1:1:1 mix into mice. e, Tags from the faecal transconjugant
population at day 25 are sorted by abundance. Note that each abundance
rank can consist of any tag (ranking and re-colouring scheme in c and raw
tag data in f). n = 6 singly housed mice from 2 independent experiments.
Dotted line indicates the conservative detection limit by qPCR. This
detection limit refers to the most-conservative detection limit of any qPCR
run (2.9 × 10−3
; that is, tags can appear below this detection limit if the
qPCR run yielded a lower detection limit). Line indicates the mean; error
bars indicate the s.d. f, Relative plasmid-tag proportion detected by qPCR
in the inoculum, the donor population persisting in the internal organs
and the transconjugant population in the faeces is shown for six mice
(three independent inocula; faeces and organ population data in Fig. 2d).
Dotted line indicates the detection limit. Each tag is given a unique colour.
g, Representation of key parameters in the mathematical model. Donors
form a reservoir of persisters in host tissues. These cells can interact
with other bacteria that colonize the gut lumen (that is, recipients), and
transfer plasmids. Our mathematical model summarizes these plasmidtransfer
dynamics using a few key parameters. Donors re-seed the gut
lumen and transfer plasmids at rate η, transconjugants transfer plasmids
to recipients (without a plasmid) at rate γ and the turn-over of each
bacterial population is given by r − c, in which r is the birth rate and c is
the clearance rate.
LetterRESEARCH
Extended Data Fig. 7 | See next page for caption.
Letter RESEARCH
Extended Data Fig. 7 | Faecal bacterial-population sizes and
inflammatory status of mice (comparison of vaccinated versus naive
mice in the oral model), and validation by mixtures of invasive and
non-invasive donors. Relates to Fig. 3e, f. a, b, Faecal loads of donors
(blue, SmR
and CmR
), recipients (green, KanR
), and transconjugants (red,
CmR
and KanR
) were determined by selective plating on MacConkey
agar. Barcode analysis of the recipient chromosome tags could not be
performed because of technical issues with kanamycin enrichments and
subsequent qPCR. Black dotted lines indicate the detection limits for
donors and transconjugants. Green dotted line indicates the detection
limit for recipients. The detection limit is higher for recipients once
transconjugants reach a density of >108
CFU per gram of faeces. Before
this happens, recipients can be found below the detection limit; the
black dotted line should then be considered as the detection limit. Blue
lines connect medians of donor populations; red lines connect medians
of transconjugant populations. a, Naive mice infected with invasive
S. Typhimurium donors (solid circles). b, Vaccinated mice infected
with invasive S. Typhimurium donors (open circles with light-grey
fill). c, Inflammatory status to determine the success of vaccination
was determined by lipocalin-2 ELISA. Statistics were performed using
a two-tailed Mann–Whitney U-test. NS, not significant (P ≥ 0.05),
*P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001, comparing
naive mice (black circles, n = 9 singly housed mice from 3 independent
experiments) to vaccinated mice (open circles, n = 14 singly housed mice
from 4 independent experiments) at each time point. Medians are shown
(solid red line for naive mice and dotted red line for vaccinated mice).
Dotted line the indicates detection limit. d–g, Experimentally reducing
the population size of mucosa-associated persisters by mixing invasive
and non-invasive donors reduces plasmid transfer in the oral model.
Mice were orally infected with a mixture of invasive (SL1344 P2cat
, SmR
and CmR
) and non-invasive (non-invasive SL1344 P2cat
, SmR
and CmR
)
donors at two ratios: 1:10 (n = 6 singly housed mice) or 1:500 (n = 5 singly
housed mice). In both cases, an excess of non-invasive donors was used
to experimentally reduce the number of cells that can establish a reservoir
of persisters in the intestinal mucosa (two independent experiments).
S. Typhimurium 14028S aphT (KanR
and AmpR
) was used as a recipient
after antibiotic treatment. d, Donor populations enumerated after a
gentamicin protection assay on caecal tissue of mice infected with a 1:10
ratio (blue circles with dark grey fill; line indicates median) or 1:500
ratio (blue circles with light grey fill; line indicates median) of invasive
to non-invasive donors. Statistics were performed using a two-tailed
Mann–Whitney U-test **P < 0.01. e, Proportion of transconjugants
(transconjugant population size divided by the sum of recipients and
transconjugants) in the faeces is shown for each day, for mice infected with
a 1:10 ratio (black circles with dark grey fill, grey bars indicate median)
or a 1:500 ratio (black circles with light grey fill, light grey bars indicate
median) of invasive to non-invasive donors. f, g, Faecal loads of donors
(blue, SmR
and CmR
), recipients (green, KanR
) and transconjugants (red,
CmR
and KanR
) were determined by selective plating on MacConkey
agar. f, Mice infected with a 1:10 ratio of invasive to non-invasive
donors. g, Mice infected with a 1:500 ratio of invasive to non-invasive
donors. In d–g, the black dotted line indicates the detection limit for
donors and transconjugants. Green dotted line indicates the detection
limit for recipients. The detection limit is higher for recipients once
transconjugants reach a density of >108
CFU per gram of faeces. Before
this happens, recipients can be found below the detection limit; the black
dotted line should then be considered as the detection limit.
LetterRESEARCH
Extended Data Fig. 8 | See next page for caption.
Letter RESEARCH
Extended Data Fig. 8 | Faecal bacterial-population sizes of mice and
the role of persistence and conjugation in host tissues if the recipient
constitutes the reservoir. Relates to Fig. 4. a, Faecal loads of donors (blue,
SmR
and CmR
), recipients (green, AmpR
), and transconjugants (red, CmR
and AmpR
) were determined by selective plating on MacConkey agar
(same mice as in Fig. 4a, b; S. Typhimurium donor and E. coli recipient
in the oral model). Black dotted line indicates the detection limit for
donors and transconjugants. Green dotted line indicates the detection
limit for recipients. The detection limit is higher for recipients once
transconjugants reach a density of >108
CFU per gram of faeces. Before
this happens, recipients can be found below the detection limit; the black
dotted line should then be considered as the detection limit. Blue lines
connect medians of donor populations; red lines connect medians of
transconjugant populations. b–d, Persistence in host tissues also promotes
plasmid transfer if the recipient constitutes the reservoir. Pre-treated mice
were orally infected with a S. Typhimurium recipient (SL1344 P2cured
aphT, SmR
and KanR
) and treated with ciprofloxacin and ampicillin as
in the oral model (Fig. 1b). On day 8, ampicillin was removed from the
drinking water and an E. coli donor was introduced orally (E. coli 8178
P2cat
, CmR
and AmpR
). b, Recipient populations enumerated after a
gentamicin protection assay on caecal tissue (n = 5 singly housed mice
from 2 independent experiments). Line indicates median. c, Proportion
of transconjugants (transconjugant population size divided by sum of
recipients and transconjugants) in the faeces is shown for each day. Grey
bars indicate median. d, Faecal loads of recipients (green, SmR
and KanR
),
donors (blue, CmR
and AmpR
), and transconjugants (red, SmR
, CmR
and KanR
) were determined by selective plating on MacConkey agar. In
b–d, dotted lines indicate detection limits by selective plating. e, Faecal
bacterial-populations sizes of mice in Fig. 4c, d (S. Typhimurium ESBL
donor in the oral model). Faecal loads of donors (blue, SmR
and AmpR
),
recipients (green, KanR
) and transconjugants (red, KanR
and AmpR
) were
determined by selective plating on MacConkey agar. Black dotted line
indicates the detection limit for donors and transconjugants. Green dotted
line indicates the detection limit for recipients. The detection limit is
higher for recipients once transconjugants reach a density of >108
CFU
per gram of faeces. Before this happens, recipients can be found below
the detection limit; the black dotted line should then be considered as the
detection limit. Blue lines connect medians of donor populations; red lines
connect medians of transconjugant populations.
LetterRESEARCH
Extended Data Fig. 9 | See next page for caption.
Letter RESEARCH
Extended Data Fig. 9 | Exchanging ampicillin for kanamycin to limit
gut-luminal growth of the donor does not affect the overall plasmidtransfer
kinetics, and faecal bacteria-population sizes of intravenously
infected mice. Relates to Fig. 4. a–d, Mice were orally infected with
SL1344 P2cat
(SmR
and CmR
) as a donor, and 14028S aphT (KanR
and
AmpR
) as a recipient after antibiotic treatment. Mice were either treated
with ampicillin in the drinking water until day 15 (normal protocol as in
Fig. 1b), day 8 (when the recipient is added) or kanamycin until day 8.
a, Donor populations enumerated after a gentamicin protection assay on
caecal tissue of mice, in which ampicillin is maintained until day 15 (solid
blue circles, n = 15 singly housed mice from 5 independent experiments;
data taken from Fig. 1d), ampicillin is removed on day 8 (blue circles with
pink fill, n = 3 singly housed mice from 1 experiment) or kanamycin is
used until day 8 (blue circles with yellow fill, n = 3 singly housed mice
from 1 experiment). Median indicated by solid line. b, Proportion of
transconjugants (transconjugant population size divided by the sum
of recipients and transconjugants) is shown for the groups receiving
ampicillin treatment until day 15 (solid black circles, grey bars indicate
median; n = 15 singly housed mice from 5 independent experiments;
data taken from Fig. 1c), ampicillin treatment until day 8 (black circles
with pink fill, pink bars indicate median; n = 3 singly housed mice from
1 experiment), and kanamycin treatment until day 8 (black circles with
yellow fill, yellow bars indicate median; n = 3 singly housed mice from
1 experiment). c, d, Faecal loads of donors (blue, SmR
and CmR
), recipients
(green, KanR
) and transconjugants (red, CmR
and KanR
) were determined
by selective plating on MacConkey agar. c, Mice treated until day 8 with
ampicillin. d, Mice treated until day 8 with kanamycin. In a–d, the black
dotted line indicates the detection limit for donors and transconjugants.
Green dotted line indicates the detection limit for recipients. The detection
limit is higher for recipients once transconjugants reach a density of
>108
 CFU per gram of faeces. Before this happens, recipients can be found
below the detection limit; the black dotted line should then be considered
as the detection limit. e, Faecal bacterial-populations sizes of mice in
Fig. 4e, f (S. Typhimurium donor and E. coli recipient in the intravenous
model). Faecal loads of donors (blue, SmR
and AmpR
), recipients (green,
KanR
) and transconjugants (red, KanR
and AmpR
) were determined
by selective plating on MacConkey agar. Black dotted line indicates
the detection limit for donors and transconjugants. Green dotted line
indicates the detection limit for recipients. The detection limit is higher
for recipients once transconjugants reach a density of >108
CFU per gram
faeces. Before this happens, recipients can be found below the detection
limit; the black dotted line should then be considered as the detection
limit. Blue lines connect medians of donor populations; red lines connect
medians of transconjugant populations.
LetterRESEARCH
Extended Data Fig. 10 | See next page for caption.
Letter RESEARCH
Extended Data Fig. 10 | Increasing growth rate at carrying capacity
to model inflammation, or running simulations on a finer parameter
grid, does not affect overall simulation trends. a, b, Simulations were
run with identical parameters to Fig. 3c, d, but an increased birth and
death rate at carrying capacity to simulate cases in which inflammation
is present (Supplementary Information). The trends of the simulations
remain the same as in Fig. 3c, d. a, Likelihood of the model as a function
of the donor re-seeding (including donor-to-recipient conjugation) rate,
and the rate of transconjugant-to-recipient plasmid transfer. All other
parameter values are given in the Supplementary Information. The mostlikely
set of parameters is shown in red (η = 1 × 10−9
 events per recipient
per day; γ = 3.16 × 10−8
 events per recipient per CFU per gram of faeces
per day). b, The fraction of simulations with plasmid re-seeding (defined
as a final transconjugant population size above 5 × 108
CFU per gram of
faeces) is shown as a function of η. Here, γ is fixed at its most-likely value
of 3.16 × 10−8
 events per recipient per CFU per gram of faeces per day.
The black vertical dotted line at η = 1 × 10−9
 events per recipient per day
indicates the estimated most-likely value from a. The red vertical dotted
line at η = 1 × 10–11
 events per recipient per day indicates a hypothetical
100-fold decrease of η (shown by a red arrow) (as might be caused, for
example, by vaccination). c–f, Running simulations on a finer parameter
grid does not affect the overall simulation trends. Supplementary Table 4
provides details of the differences between specific simulation results.
c, d, Simulations were run on a grid of η = 10−12
to 10−6
and γ = 10−10
to 10−1
with 0.25-log increments, instead of the η = 10−12
to 10−1
and γ = 10−12
to 10−1
with 0.5-log increments used in Fig. 3c, d. e, f,
Simulations were run with parameters identical to a, b, but on a grid of
η = 10−12
to 10−6
and γ = 10−10
to 10−1
with 0.25-log increments, instead
of η = 10−12
to 10−1
and γ = 10−12
to 10−1
with 0.5-log increments.
c, e, Likelihood of the model as a function of the donor re-seeding
(including donor-to-recipient conjugation) rate, and the rate of
transconjugant-to-recipient plasmid transfer. All other parameter values
are given in the Supplementary Information. The most-likely set of
parameters is shown in red. d, f, The fraction of simulations with plasmid
re-seeding (defined as a final transconjugant population size above 5 × 108
CFU per gram of faeces) is shown as a function of η. Here γ is fixed at its
most likely value. The black vertical dotted line indicates the estimated
most-likely value (from c or e). The red vertical dotted line indicates a
hypothetical 100-fold decrease of η (shown by a red arrow) (as might be
caused, for example, by vaccination).
1
natureresearch|reportingsummaryApril2018
Corresponding author(s):
Reporting Summary
Nature Research wishes to improve the reproducibility of the work that we publish. This form provides structure for consistency and transparency
in reporting. For further information on Nature Research policies, see Authors & Referees and the Editorial Policy Checklist.
Statistical parameters
When statistical analyses are reported, confirm that the following items are present in the relevant location (e.g. figure legend, table legend, main
text, or Methods section).
n/a Confirmed
The exact sample size (n) for each experimental group/condition, given as a discrete number and unit of measurement
An indication of whether measurements were taken from distinct samples or whether the same sample was measured repeatedly
The statistical test(s) used AND whether they are one- or two-sided
Only common tests should be described solely by name; describe more complex techniques in the Methods section.
A description of all covariates tested
A description of any assumptions or corrections, such as tests of normality and adjustment for multiple comparisons
A full description of the statistics including central tendency (e.g. means) or other basic estimates (e.g. regression coefficient) AND
variation (e.g. standard deviation) or associated estimates of uncertainty (e.g. confidence intervals)
For null hypothesis testing, the test statistic (e.g. F, t, r) with confidence intervals, effect sizes, degrees of freedom and P value noted
Give P values as exact values whenever suitable.
For Bayesian analysis, information on the choice of priors and Markov chain Monte Carlo settings
For hierarchical and complex designs, identification of the appropriate level for tests and full reporting of outcomes
Estimates of effect sizes (e.g. Cohen's d, Pearson's r), indicating how they were calculated
Clearly defined error bars
State explicitly what error bars represent (e.g. SD, SE, CI)
Our web collection on statistics for biologists may be useful.
Software and code
Policy information about availability of computer code
Data collection
Data analysis
For manuscripts utilizing custom algorithms or software that are central to the research but not yet described in published literature, software must be made available to editors/reviewers
upon request. We strongly encourage code deposition in a community repository (e.g. GitHub). See the Nature Research guidelines for submitting code & software for further information.
Wolf-Dietrich Hardt
Mederic Diard
Data collection is described in detail in Materials and Methods. The mathematical model is described in detail in the supplementary
materials. All code needed to reproduce the mathematical modelling is available as supplementary data. The custom code was written in
R, and tested on MacOS 10.13.4 with R 3.4.4, as well as a computing cluster running CentOS 7 with R 3.4.0. There is an essential
dependence on the R package adaptivetau (v2.2.1); and optional dependencies on the packages ggplot2 (v2.2.1), reshape2 (v1.4.3),
gridExtra (v2.3), wesanderson (v0.3.6), parallel (v3.4.4), and tictoc (v1.0).
Statistical analysis and computation of the plots was performed using Graph Pad Prism, version 7; Fitting of the mathematical model to
experimental values was performed using an Approximate Bayesian Computation (ABC) approach (the reference Marjoram et al. 2003,
PNAS describes this in detail).
2
natureresearch|reportingsummaryApril2018
Data
Policy information about availability of data
All manuscripts must include a data availability statement. This statement should provide the following information, where applicable:
- Accession codes, unique identifiers, or web links for publicly available datasets
- A list of figures that have associated raw data
- A description of any restrictions on data availability
Field-specific reporting
Please select the best fit for your research. If you are not sure, read the appropriate sections before making your selection.
Life sciences Behavioural & social sciences Ecological, evolutionary & environmental sciences
For a reference copy of the document with all sections, see nature.com/authors/policies/ReportingSummary-flat.pdf
Ecological, evolutionary & environmental sciences study design
All studies must disclose on these points even when the disclosure is negative.
Study description
Research sample
Sampling strategy
Data collection
Timing and spatial scale
Data exclusions
Reproducibility
Randomization
Blinding
Did the study involve field work? Yes No
Reporting for specific materials, systems and methods
All data from experiments and from the mathematical model calculations are shown. The sequence data is available at the indicated repositories (accession number
is provided in the data availability statement).
We performed mouse infection experiments. The numbers of mice per group are indicated in each figure. In figures where statistical
comparisons were performed, we used a minimum of 5 mice that were singly housed per group (to ensure that each mouse is truly
independent) per group from at least two independent experiments. In other experiments, we used a minimum of three singly
housed mice per group. We sampled feces at the indicated times post infection and analyzed pathogens loads in tissues by plating
and by fluorescence microscopy.
We used defined clones of the indicated S. Typhimurium and E. coli strains. 8-12 weeks old 129 SvEv (Nramp+/+) inbred mice were
used. The Nramp+/+ genotype confers resistance against long-term S. Typhimurium infection. The mice harbor a specified pathogen
free microbiota. This resident microbiota does not feature any E. coli or Salmonella strains.
We sampled feces at the indicated times post infection and analyzed pathogens loads in tissues by plating and by fluorescence
microscopy. Sample sizes were not predetermined. As large differences were expected, 5-7 mice were used in general for statistical
comparisons. In some cases where the underlying data was central to the main hypothesis, we used as many as 15 mice to ensure
claims were accurate and that we encompasses as large variability as possible.
We sampled feces at the indicated times post infection and analyzed pathogens loads in tissues by plating and by fluorescence
microscopy. Data was recorded by the experimenter (indicated in the author contributions section).
The time course of the sampling is shown in detail in Figure 1B and Figure 2B. Mice were infected and feces were monitored daily for
either 5, 7, 8, 15, or 25 days (indicated for each figure specifically). Antibiotic treatments were given on days 2, 3, and 4 (where
applicable). Organs were collected at defined endpoints in the experiment.
All experimental data from all repeat of each experiment is shown. We did not exclude any data, except in the case of four mice
where animals needed to be sacrificed prematurely due to disease or symptom severity.
The number of animals used in the independent experiments are indicated in the figure legends. Mice were singly housed to ensure
that each sample is completely independent. Experiments were very reproducible in each independently housed mouse. In addition,
in the bulk majority of experiments, at least two independent experiments (with separate inocula) were used to validate
reproducibility.
Mice were randomly assigned to the different experimental groups.
The data analysis was not performed in a blinded fashion. This is explained by the objective nature of the plating data (colony present
= yes/no), which excludes enumeration bias. For microscopy, slides were blinded prior to analysis.