1
Staphylococcal
secreted cytotoxins are competition sensing signals for
Pseudomonas
aeruginosa
Grace Z. Wang
1
, Elizabeth A. Warren
2
, Allison L. Haas
3
, Andrea Sánchez Peña
2
, Megan R. Kiedrowski
3
, Brett
Lomenick
4
, Tsui
-
Fen Chou
4
, Jennifer M.
Bomberger
3,5*
, David A. Tirrell
1
, Dominique H. Limoli
2*
1
Division of Chemistry and Chemical Engineering, California Institute of Technology, Pasadena, CA, USA
2
Department of Microbiology and Immunology, Carver College of Medicine, University of Iowa, Io
wa City, IA, USA
3
Department of Microbiology and Molecular Genetics, University of Pittsburgh, Pittsburgh, PA, USA
4
Beckman Institute, California Institute of Technology, Pasadena, CA, USA
5
Geisel School of Medicine at Dartmouth, Hanover, NH, USA
*
Corr
esponding author
ABSTRACT
Coinfection with two notorious opportunistic pathogens, the Gram
-
negative
Pseudomonas aeruginosa
and Gram
-
positive
Staphylococcus aureus,
dominates chronic pulmonary infections. While coinfection is
associated with poor patient
outcomes, the interspecies interactions responsible for such decline remain
unknown. Here, we dissected molecular mechanisms of interspecies sensing between
P. aeruginosa
and
S. aureus.
We discovered that
P. aeruginosa
senses
S. aureus
secreted peptides and, counterintuitively,
moves towards these toxins.
P. aeruginosa
tolerates such a strategy through “competition sensing”,
whereby it preempts imminent danger/competition by arming cells with type six secretion (T6S) and iron
acquisitio
n systems. Intriguingly, while T6S is predominantly described as weaponry targeting Gram
-
negative and eukaryotic cells, we find that T6S is essential for full
P. aeruginosa
competition with
S.
aureus,
a previously undescribed role for T6S. Importantly, com
petition sensing was activated during
coinfection of bronchial epithelia, including T6S islands targeting human cells. This study reveals critical
insight into both interspecies competition and how antagonism may cause collateral damage to the host
environ
ment
.
(which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
The copyright holder for this preprint
this version posted January 29, 2023.
;
https://doi.org/10.1101/2023.01.29.526047
doi:
bioRxiv preprint
2
INTRODUCTION
1
The future of microbiome research lies in our ability to manipulate polymicrobial interactions toward
2
improved human health outcomes, which requires a fundamental molecular
understanding of how
3
microbial species sense and respond to ecological competition. Chronic respiratory infections in people
4
with cystic fibrosis (CF) consist of diverse and heterogeneous microbial communities
1
. Nonetheless,
5
Pseudomonas aeruginosa
and
Stap
hylococcus aureus
are the most prevalent
2,3
. Critically, coinfection
6
with these pathogens correlates with worsened clinical outcome and altered antibiotic efficacy
2
-
4
, urging
7
the need for molecular dissection of their interspecies crosstalk.
8
9
We previously
reported that
P. aeruginosa
is attracted to
S. aureus
resulting in invasion of
S. aureus
10
colonies
5
; however, what, if any, selective benefit
P. aeruginosa
achieves by adopting this behavior
11
remains unknown. A potential role for such a strategy may be to b
ridge cellular distances for contact
-
12
dependent mechanisms of antagonism. The type six secretion system (T6SS), widely found in Gram
-
13
negative bacteria, such as
P. aeruginosa
, equips cells with a versatile nanomachinery that functions as
14
an
interspecies weapon capable of targeting both eukaryotic and prokaryotic cells
6
-
8
.
P. aeruginosa
15
typically maintains low basal T6SS activity but is capable of rapid reciprocal firing following T6SS attack
16
by other Gram
-
negative species
9,10
. However, whethe
r an analogous response may occur in response to
17
Gram
-
positive competitors lacking T6SS remains unknown.
A greater fundamental understanding of
18
interspecies pathogen sensing and resulting competition, particularly between Gram negative and
19
positive pathoge
ns common during coinfection, is necessary to develop interventions directed at
20
interspecies interactions.
21
22
Here, we report the discovery that
P. aeruginosa
rapidly activates T6SS after an encounter with the
23
Gram
-
positive pathogen
S. aureus
. We present a “
competition sensing” model uncovered by a
24
combination of genetics, microscopy and multi
-
omics approaches whereby secreted
Staphylococcal
25
peptides are key interspecies signals that trigger
P. aeruginosa
antagonism.
P. aeruginosa
was found to
26
sense
S. aureus
via secreted peptides at a distance, subsequently increasing directional motility and
27
activating T6SS antagonism. Surprisingly, such activation allowed for T6SS
-
dependent competition with
28
S. aureus,
extending the functional role of T6SS to not only compet
ition between Gram
-
negatives, but
29
also between Gram
-
negative and positive bacteria. Furthermore, we examined coinfection on fully
30
differentiated CF
-
derived bronchial epithelia, the gold standard model of
in vivo
CF airway infection, and
31
found
P. aeruginosa
T6SS was activated, including host
-
targeting T6SS islands. Overall, these results
32
broaden our mechanistic understanding of interspecies antagonism between distantly related species,
33
(which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
The copyright holder for this preprint
this version posted January 29, 2023.
;
https://doi.org/10.1101/2023.01.29.526047
doi:
bioRxiv preprint
3
reveal interspecies pathways that might be targeted therapeutically, and
lend insight into the mechanism
34
of increased patient decline during coinfection with
P. aeruginosa
and
S. aureus.
35
36
RESULTS
37
PSM
ɑ
peptides are necessary and sufficient for
P. aeruginosa
attraction toward
S. aureus
38
We previously reported that
P. aeruginosa
tr
avels up a gradient of
S. aureus
secreted factors using type
-
39
IV pilus (TFP)
-
dependent motility
5
.
The
S. aureus
attractants identified are secreted
S. aureus
peptides,
40
referred to as phenol soluble modulin (PSMs).
S. aureus
produces five alpha peptides: PSM
ɑ1
-
4 and
41
PSM
δ (δ toxin)
and two β
-
peptides
:
PSMβ1 and 2
(
Supplementary Fig. 1
)
. Here, we first asked if
P.
42
aeruginosa
possesses
specificity in attraction towards individual peptides in a macroscopic TFP
43
chemotaxis assay (
Fig.1a
). PSM
ɑ peptides were examine
d for initial characterization given that the
ɑ
44
peptides have known roles in neutrophil chemoattraction
11
and cytotoxicity to mammalian host cells
12
.
P.
45
aeruginosa
traveled further towards an increasing gradient of WT
S. aureus
supernatant (
Fig. 1b
),
46
whereas directional motility towards supernatant derived from a double
psm
ɑ1
-
4 and
psm
δ mutant
47
(Δ
psm
ɑ1
-
4 δATG
-
ATT,
Supplementa
ry
Fig
.
1
) was eliminated, suggesting that at least one
ɑ
-
peptide is
48
necessary for attracting
P. aeruginosa
(
Fig. 1b, c
). The m
agnitude of attraction towards Δ
psm
ɑ1
-
4 was
49
between WT
S. aureus
and the double Δ
psm
ɑ1
-
4 δATG
-
ATT mutants, suggesting PSMδ, along with the
50
other
ɑ peptides,
is
necessary for
P. aeruginosa
directional
motility.
We then determined if PSMs are
51
sufficient to a
ttract
P. aeruginosa
and the specificity of individual PSM peptide’s contribution. Pure
52
synthetic PSM
ɑ3 and δ
-
toxin strongly attracted
P. aeruginosa
in a dose
-
dependent manner (
Fig. 1d
).
53
These data demonstrate that PSMδ and PSM
ɑ3 are necessary and sufficie
nt for
S. aureus
to attract
P.
54
aeruginosa
.
55
56
It has been widely acknowledged that
P. aeruginosa
outcompetes
S. aureus
in vitro
13
–
16
, though the exact
57
mechanisms of cellular death are poorly elucidated. Curiously, when PSM
-
deficient
S. aureus
were
58
coculture
d
with
P. aeruginosa
, a moderate increase in
S. aureus
survival was observed (
Fig. 1e
). Addition
59
of PSM peptides to
coculture
with the Δ
psm
mutant restored
S. aureus
survival to the reduced level seen
60
with WT strains, raising the possibility that there exis
t unknown PSM
-
dependent killing mechanisms
61
between
P. aeruginosa
and
S. aureus
. These factors further led us to investigate the roles of PSMs in
62
mediating
P. aeruginosa
responses to
S. aureus
, and the cellular events occurring after
P. aeruginosa
63
cells are recruited to the site of
S. aureus.
64
65
P. aeruginosa
undergoes immediate, systematic proteome remodeling in response to PSM
66
peptide pulse
-
in and
coculture
with
S. aureus
67
(which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
The copyright holder for this preprint
this version posted January 29, 2023.
;
https://doi.org/10.1101/2023.01.29.526047
doi:
bioRxiv preprint
4
To gain insight into
the effects PSMs have on
P. aeruginosa
cellular functions, we took advantage of the
68
precise temporal resolution afforded by
B
io
O
rthogonal
N
on
-
C
anonical
A
mino acid
T
agging (BONCAT)
17
69
to monitor
P. aeruginosa
immediate protein synthesis in response either to
direct addition of PSMs or to
70
coculture
with
S. aureus
cells (
Fig. 2a
).
P. aeruginosa
cells constitutively expressing an engineered
71
mutant methionyl
-
tRNA synthetase (NLL
-
MetRS) allow for selective metabolic labeling of newly
72
synthesized proteins by the az
ide
-
bearing methionine (Met) analog: azidonorleucine (Anl)
73
(
Supplementary Fig. 2
). Downstream chemical enrichment
18
of labeled proteins enables targeted
74
analysis of nascent
P. aeruginosa
protein synthesis during the Anl labeling period.
75
76
We identified 3348
and 3365 total proteins newly synthesized by
P. aeruginosa
during the 30
-
min labeling
77
period immediately following PSM pulse
-
in and
coculture
with
S. aureus
, respectively (
Fig. 2b,
c
), and
78
quantified differentially expressed proteins in each condition. We
found 60
P. aeruginosa
proteins with
79
statistically significant and greater than 2
-
fold increase and 98 with greater than 2
-
fold decrease in
80
abundances in response to PSMs pulse
-
in. For
coculture
with
S. aureus
compared to
monoculture
, 178
81
proteins with si
gnificant increase (>2
-
fold) in abundances and 124 with significant decrease (>2
-
fold) in
82
abundances
(
Supplementary
T
able 2
).
Candidates were then grouped by their annotated functional
83
categories, which include the following: T6SS, pyoverdine biosynthesis,
cyclic di
-
GMP sensing/regulating
84
enzymes, chemotaxis/motility, cellular response to envelope stress, and DNA damage/stress response
85
(
Fig. 2
d
, e
).
Strikingly,
PSMs alone are sufficient to promote the synthesis of proteins in each category.
86
87
P.
aeruginosa
activates T6SS in response to PSMs and
S. aureus
cells
88
Notably, proteins involved in T6SS are over
-
represented among the total significantly up
-
regulated hits
89
in
P. aeruginosa
global proteomic response to PSM pulse
-
in and
S. aureus
coculture
(
Fig. 3,
90
Supplementary Fig.
3
).
P. aeruginosa
T6SS is a speargun
-
shap
ed secretory apparatus that loads and
91
injects toxic cargo into prey cells. We detected significantly increased synthesis of various components
92
of the T6SS structural architecture, including core, accessory, bacteriophage
-
like subunits, and
93
membrane
-
associa
ted components
19,20
(
Fig. 3a, c
), suggesting the T6SS apparatus is being
94
systematically assembled during the 30
-
min labeling period following introduction of PSMs or
S. aureus
.
95
In particular, the expression levels of two proteins
—
the hemolysin coregulated
protein (Hcp, T6SS
96
“sheath”) and the valine
-
glycine repeat protein G (VgrG, T6SS “tip”)
—
are often used to determine
97
whether T6SS is functional
5,6,20,21
. Their relative fold
-
changes are the highest among other T6SS proteins
98
that showed significantly changed
abundances in response to PSM pulse
-
in or
coculture
with
S. aureus
99
(
Fig. 3a
). Additionally, proteins encoded by all three known
P. aeruginosa
T6SS loci, denoted HSI
-
I
100
(PA00
-
), HSI
-
II (PA16
-
) and HSI
-
III (PA23
-
)
5,19,20
were increased (
Fig. 3a, c
), further
supporting that
P.
101
(which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
The copyright holder for this preprint
this version posted January 29, 2023.
;
https://doi.org/10.1101/2023.01.29.526047
doi:
bioRxiv preprint
5
aeruginosa
systematically up
-
regulates T6SS after encountering
S. aureus
via sensing of
Staphylococcal
102
secreted PSMs.
103
104
We next ranked the nascent
P. aeruginosa
proteome by individual protein raw abundances quantified by
105
l
abel
-
free quanti
fication (LFQ) via mass spectrometry (
Supplementary Fig. 4
) to examine cellular
106
allocation of protein synthesis resources following PSM pulse
-
in and
S. aureus
coculture
challenge.
107
Remarkably, most T6SS proteins appeared in the top quartile with significant
ly elevated average
108
abundances in PSM
-
treated and
S. aureus
coculture
samples compared to untreated/
monoculture
109
controls, further indicating T6SS antagonism is prioritized by
P. aeruginosa
in responding to interspecies
110
stress.
111
112
Although T6SS apparatus asse
mbly does not necessarily indicate firing of T6SS effectors, significantly
113
higher abundances of the AAA+ ATPases ClpV (
Fig. 3a, c
) suggest increased sheath contraction and
114
propulsion of effectors
22
-
24
. To examine
P. aeruginosa
deployment of T6SS, single
-
ce
ll microscopy using
115
a fluorescent reporter of ClpV1 activity was employed (ClpV1
-
GFPmut3
9
) and confirmed that PSMs are
116
sufficient to induce
P. aeruginosa
deployment of T6SS (
Fig. 3d
). PSM
-
treated cells exhibited both
117
significantly increased GFPmut3
puncta formation (
Fig. 3e
) as well as overall fluorescence intensity (
Fig.
118
3f
) per cellular total area, further supporting that PSMs induce interspecies antagonistic T6SS attacks by
119
P. aeruginosa
.
120
121
Staphylococcal
secreted PSM peptides increase siderophore
biosynthesis
122
We also observed that the T6SS induction in
P. aeruginosa
is accompanied by systematic upregulation
123
of the pyoverdine biosynthesis cluster (
Supplementary Fig. 3
), which produces a siderophore that binds
124
to extracellular Fe
3+
with high affinity
25,26
. Iron starvation is a major stress response pathway evolutionarily
125
conserved in bacteria. Proteins encoded by five pyoverdine operons (
Fig. 3b
) entirely covering the
126
complex cellular biosynthesis machinery for pyoverdine siderophore were found to be
significantly up
-
127
regulated in response to PSM pulse
-
in or
S. aureus
coculture
—
including the extracytoplasmic
128
function
iron starvation σ factor PvdS, which
positively regulates pyoverdine biosynthesis and
129
secretion
27
, and PvdR, which controls transport of
pyoverdine out of the cell
28
, indicating that
130
siderophores are being increasingly synthesized and dispatched out of the cell during the 30
-
min labeling
131
period.
Thus, we simultaneously monitored pyoverdine production and induction of gene expression
132
using
a fluorescent reporter P’
pvdG
-
mScarlet
29
(
Supplementary Fig. 5
)
. Consistent with the proteomic
133
results, we observed significantly increased pyoverdine production by
P. aeruginosa
following PSMs
134
treatment, as well as
significant induction of
pvdG
promoter a
ctivity.
135
(which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
The copyright holder for this preprint
this version posted January 29, 2023.
;
https://doi.org/10.1101/2023.01.29.526047
doi:
bioRxiv preprint
6
136
Pathogens face intense competition for iron with host and other microbial species due to the essentiality
137
of iron as a nutrition source, and siderophore production is often reported to be involved in exploitive
138
interspecies competition
13,30,31
. I
nterestingly, a recent study reported upregulation of siderophore
139
biosynthesis in
P. aeruginosa
when treated with
Staphylococcal
culture supernatant
29
, though the
140
molecular signals responsible for the observed upregulation remained elusive. Here, we show t
hat
141
Staphylococcal
secreted PSM peptides alone could trigger increased pyoverdine biosynthesis and export,
142
further suggesting that PSMs play important roles in mediating interspecies competition between
P.
143
aeruginosa
and
S. aureus
.
144
145
PSMs may activate compe
tition sensing via induction of transient membrane stress
146
We
next
probed the molecular mechanism of PSM
-
induced T6SS activation
.
Previous literature suggests
147
P. aeruginosa
T6SS could be induced via kin cell lysis
9
and/or envelope stress
32
. In particular, the pore
-
148
forming antibiotic polymyxin B induces T6SS in
P. aeruginosa
via endogenous membrane stress
32
.
149
150
S. aureus
secreted PSMs are virulence factors with hemolytic activity toward mammalian cells
11
. While
151
PSMs ge
nerally exhibit low activity towards bacterial membranes
33
, we asked whether PSMs could
152
permeabilize the
P. aeruginosa
membrane, cause kin cell lysis, and/or cause envelope stress in
P.
153
aeruginosa
. Live imaging of
P. aeruginosa
with propidium iodide +/
-
PS
Ms did not show evidence of kin
154
lysis, inner membrane permeability (
Fig. 4a
), or altered
P. aeruginosa
growth rate (
Supplementary Fig.
155
6a
). In comparison, polymyxin B significantly inhibited
P. aeruginosa
growth (
Supplemental Fig. 6a
)
and
156
induced a moderat
e uptake of propidium iodide (
Fig. 4a
)
. Further analysis of outer membrane
157
permeability by uptake of 1
-
N
-
phenylnaphthylamine (NPN) also did not reveal significant permeability
158
with PSM treatment, while polymyxin significantly induced outer membrane permeab
ility (
Fig. 4b
).
159
160
Given these differences in membrane activity between polymyxin B and PSMs, we revisited T6SS
161
activation by polymyxin B with ClpV1 fluorescent reporter
9
u
nder the current study conditions for
162
comparison. Polymyxin B
-
treated cells displaye
d distinct ClpV1 puncta induction (
Fig. 4c
) but yielded low
163
mean fluorescence intensity per cell (
Fig. 4d
), suggesting potential molecular differences between
164
mechanisms of T6SS induction by polymyxin B versus by PSMs treatment. Polymyxin B can be inserted
165
into the membrane, causing cell lysis by creating pores in the envelope
34
. In contrast, PSMs are cationic,
166
amphipathic small helical peptides with membrane perturbing and cell surface
-
adhering properties
35
.
167
168
(which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
The copyright holder for this preprint
this version posted January 29, 2023.
;
https://doi.org/10.1101/2023.01.29.526047
doi:
bioRxiv preprint
7
While we were unable to detect significant memb
rane damaging activity by PSMs, we hypothesize that
169
non
-
lethal membrane perturbations may explain PSM
-
induced T6SS activation in
P. aeruginosa
. Several
170
factors contributed to this hypothesis: first, global differential proteomic profiling revealed signific
ant and
171
systematically decreased production of electron transport chain (ETC) enzymes in response to PSMs
172
pulse
-
in and
coculture
with
S. aureus
(
Fig. 4e, f, Supplementary Fig. 3
), a characterized cellular
173
response to envelope stress evolutionarily conserve
d in
E. coli
and other Gram
-
negative bacteria
36
–
39
. In
174
addition, while we did not detect increased protein synthesis of classic regulators of membrane stress,
175
such as σ
E
and CpxAR
39
, we observed significant up
-
regulation of a subset of proteins involved in
176
membrane stress responses, most notably protein encoded by
PA3731
, a close homologue of the phage
177
shock protein PspA in
E. coli
and member of a family of proteins characterized to play crucial roles in the
178
cellular response to and protection against envel
ope stress in
E. coli
and other Gram
-
negative species
40,41
.
179
Therefore, we hypothesize that PSMs provoke
P. aeruginosa
T6SS firing via induction of cell envelope
180
stress via short
-
term perturbations.
181
182
Significantly increased
P. aeruginosa
T6SS activity in
coculture
with
S. aureus
on CF patient
-
183
derived bronchial epithelial cells
184
Previous studies reported that Hcp1 is detected at high levels in chronic CF
sputum
5
, and H
SI
-
II
and
III
185
T6SS are required for and induced upon
P. aeruginosa
infect
ion of epithelial cells
7,42
, suggesting that
186
differential regulation of any of the three T6SS loci in polymicrobial infections may have implications for
187
the host. Prompted by the fact that all three HSI
-
T6SS loci in
P. aeruginosa
have previously characteri
zed
188
roles in CF pathogenicity, we further investigated
P. aeruginosa
and
S. aureus
interactions in a host
-
189
derived environment to explore interspecies virulence factor crosstalk in a clinically relevant context.
190
191
For this purpose, we obtained the health car
e
-
associated
m
ethicillin
-
resistant
Staphylococcus aureus
192
(HA
-
MRSA) strain USA100, a highly antibiotic resistant clinical isolate and a leading cause of invasive
193
infections by MRSA
43,44
, and
P. aeruginosa
strain PAO1, a laboratory derivative more closely re
lated than
194
PA14 to most clinical isolates of CF
45
. We performed RNA
-
sequencing to examine
P. aeruginosa
195
transcriptomic changes that contribute to interspecies interactions in a coinfection model with
S. aureus
196
using polarized, fully differentiated CF bronc
hial epithelial cells (CFBE41o
-
,
Fig. 5a
). This model closely
197
mimics the CF host environment by recapitulating approximately 84% of
P. aeruginosa
gene expression
198
in human expectorated CF sputum, outperforming both laboratory media and the acute mouse
199
pneum
onia model of infection
46
.
200
201
(which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
The copyright holder for this preprint
this version posted January 29, 2023.
;
https://doi.org/10.1101/2023.01.29.526047
doi:
bioRxiv preprint
8
We identified 1,325 differentially expressed genes during
coculture
with
S. aureus
(fold change >2 or <
-
202
2,
P
<0.05). Of these, we detected increased transcription of T6SS genes from all three HSI
-
T6SS clusters
203
(
Fig. 5b, c
). Trans
cription of
hcp
,
vgrG
, and sheath genes was significantly increased,
suggesting
that
204
the T6SS apparatus is functional. In addition, we observed significantly increased transcription of several
205
effector genes including
tse1
(
PA1844
), a peptidoglycan amidase
47
,
pldB
(
PA5089
), a phospholipase
48
,
206
and
tseF
(
PA2374
)
49
, a known facilitator of iron uptake in
P. aeruginosa
48
.
207
208
P. aeruginosa
T6SS mediates the killing of
S. aureus
209
A wealth of existing literature shows that T6SS
-
delivered effectors target and
kill Gram
-
negative
210
bacteria
9,10,20,47
. However, until a recent study that demonstrated T6SS secreted effectors by
211
Acinetobacter baumannii
could kill Gram
-
positive bacteria
50
, it had previously been assumed that Gram
-
212
positive species are not susceptible to
T6SS
-
mediated killing
51
.
213
214
To examine if T6SS activity provides
P. aeruginosa
a competitive advantage in
coculture
with
S. aureus
215
in association with airway epithelial cells, we constructed clean deletions
of each HSI T6SS sheath gene,
216
∆
tssB1
(H1),
∆
hsi
B2
(H2), and ∆
hsiB3
(H3)
, and surprisingly found that each mutant exhibited decreased
217
competitive index compared to WT in
coculture
with
S. aureus
in the
airway cell model (
Fig. 6a, b
). We
218
next focused on a H
SI
-
III
T6SS
-
encoded effector TseF for several
reasons. First,
tseF
(
PA2374
) was the
219
most significantly up
-
regulated effector in coinfection with
S. aureus
on host cells (
Fig. 5c
). Secondly,
220
TseF facilitates
P. aeruginosa
iron uptake
49
, a functional role likely to affect polymicrobial competition.
221
Further, TseF was characterized to be coregulated with the
Pseudomonas
q
uinolone
s
ignal
49
, a quorum
-
222
sensing system in
P. aeruginosa
with known roles in competition against
S. aureus
15
. Interes
tingly,
223
deletion of
tseF
alone significantly reduced
P. aeruginosa
competitive fitness against
S. aureus
and
224
rescued
S. aureus
survival in
coculture
(
Fig. 6a, b
).
225
226
DISCUSSION
227
The “competition
-
sensing” hypothesis states that bacteria adapt evolutionarily co
nserved stress response
228
pathways to directly detect and respond to ecological competition
52
. The results presented here provide
229
empirical evidence for this hypothesis, which predicts increased bacterial toxin production in response to
230
stress caused by comp
etitors. We propose a model in accordance, whereby
P. aeruginosa
senses
231
transient cellular stress caused by secreted competitor signals and swiftly responds by moving towards
232
the signals and activating antagonistic responses (
Fig. 7
). Activation of membran
e stress and iron
233
(which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
The copyright holder for this preprint
this version posted January 29, 2023.
;
https://doi.org/10.1101/2023.01.29.526047
doi:
bioRxiv preprint
9
starvation responses observed in
P. aeruginosa
further supports that “competition sensing” is manifested
234
in several stress response pathways.
235
236
P. aeruginosa
is attracted to diverse bacterial species and moves towards the site of competiti
on
5
; thus,
237
a close analogy can be drawn between
P. aeruginosa
and the notorious predacious bacterium,
238
Myxococcus xanthus
, characterized to coordinate group responses to invade and lyse prey
53
.
P.
239
aeruginosa
displays incipient multicellularity via complex c
ollective behaviors, including ones of a
240
predatory nature as described here. We propose that upon sensing interspecies signals,
P. aeruginosa
241
cells move to “trap” a
S. aureus
colony, further enabling contact
-
dependent invasion
and/or local
242
concentration of
secreted antimicrobials.
243
244
One potential mechanistic model of competition sensing is that
P. aeruginosa
closely monitors cell
245
envelope integrity to detect environmental and/or interspecies insults. While PSMs do not affect the
P.
246
aeruginosa
membrane suffic
iently to allow permeabilization, even transient envelope stress may induce
247
T6SS assembly and firing.
Interestingly, it has been recently reported that
P. aeruginosa
chemotaxis
248
towards, instead of away from, antibiotics and releases bacteriocins before dyi
ng
54
. While PSMs did not
249
reduce
P. aeruginosa
viability, we found induction of two pyocins in
P. aeruginosa
in response to both
250
PSMs pulse
-
in and co
-
culture with
S. aureus
, potentially supporting a similar “suicidal chemotaxis” model.
251
PSMs alone are sufficient to trigger TFP
-
mediated motility, synthesis and transport of siderophores,
252
activation of T6SS antagonism and envelope stress responses, suggesting that PSMs are important
253
interspecies signals that help
P.
aeruginosa
sense and resp
ond to imminent danger/competition.
254
Interestingly, T6SS, pyoverdine production, chemotaxis and cellular response to envelope stress in
P.
255
aeruginosa
are all known to be regulated by cyclic di
-
GMP
5
5
–
5
9
. We propose herein that secondary
256
messengers signaling
networks mediate “competition sensing” and global bacterial responses to
257
interspecies insults. In support of this hypothesis, we observed up
-
regulation of multiple c
-
di
-
GMP
258
metabolizing enzymes, suggesting
several
c
-
di
-
GMP mediated signaling networks are a
ctivated and are
259
involved in
P. aeruginosa
response to PSMs and
S. aureus
(
Supplementary Fig. 7
).
260
261
Further, proteomic analysis detected significantly increased abundance of PA1611, a known inhibitor of
262
RetS and activator of the
g
lobal
a
ctivation of antibi
otic and
c
yanide synthesis/regulator of
s
econdary
263
m
etabolism (Gac/Rsm) pathway
59
both in response to PSM treatment and
S. aureus
coculture
. Gac/Rsm
264
post
-
transcriptionally regulates all three T6SS loci in
P. aeruginosa
6
1
and mediates
P
.
a
eruginosa
265
r
esponse
to
a
ntagonism (PARA)
10
. Also consistent with previous reports that
P. aeruginosa
T6SS and
266
T3SS are inversely regulated via RetS
5
6
, we detected systematic repression of T3SS and simultaneously
267
(which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
The copyright holder for this preprint
this version posted January 29, 2023.
;
https://doi.org/10.1101/2023.01.29.526047
doi:
bioRxiv preprint
10
increased T6SS activity during coinfection with
S. aureus
(
Suppl
ementary Fig. 8
). Intriguingly, the
268
Gac/Rsm pathway and c
-
di
-
GMP signaling networks both regulate T6SS and iron uptake
20,58,59
. Future
269
work will be dedicated to studying overlap in signal transduction pathways and potential coordination of
270
interspecies phe
notypes reported in this study, including
P. aeruginosa
TFP
-
mediated directional motility,
271
downstream antagonistic attacks and exploitive iron scavenging.
272
273
Interestingly, we observed inverse regulation of siderophore biosynthesis in
coculture
with
S.
aureus
274
using global proteomics analysis performed
in vitro
versus transcriptomic analysis performed in a host
275
environment.
P. aeruginosa
down
-
regulates pyoverdine biosynthesis during coinfection with
S. aureus
276
on CF
-
derived epithelial cells (
Supplementary
Fig. 8
). We attribute this to the differences in temporal
277
resolution of the experiments
—
while chemo
-
selective proteomic analysis captured immediate
278
“competition sensing” responses, global RNA
-
sequencing investigated long
-
term coinfection phenotypes.
279
These
results highlight
P. aeruginosa
versatile genetic plasticity in regulating iron scavenging behaviors
280
during short
-
term
versus
long
-
term competition and underline the importance of studying and comparing
281
polymicrobial interactions both
in vitro
and
in vivo
.
282
283
Numerous studies have reported that
P. aeruginosa
produces diverse secondary metabolites known to
284
be toxic to
S. aureus
14,15
, but insufficient to account for total
S. aureus
cellular death in
285
coculture
16
.
Nonetheless, when embarking on this study, we p
resumed that
P. aeruginosa
T6SS would
286
neither be activated by, nor effective in competition with
S. aureus.
Several reasons contributed to this
287
initial assumption
51
. First, Gram
-
positive bacteria lack a conjugative pilus, and therefore cannot provoke
288
P. ae
ruginosa
reciprocal firing. Further, the Gram
-
positive cell wall constitutes a thicker peptidoglycan
289
(PG) layer in comparison to that of Gram
-
negative species, which was thought to prohibit penetration by
290
the T6SS apparatus and effective delivery of toxic
effectors. The discovery here that
P. aeruginosa
T6SS
291
is both induced by and mediates the killing of a Gram
-
positive pathogen, challenges our prior
292
assumptions, and expands the role of T6SS during infection, opening a wealth of new opportunities to
293
study,
inhibit, or co
-
opt interspecies competition.
294
295
How does
P. aeruginosa
T6SS kill
S. aureus
? Intriguingly, proteomic, transcriptional, and mutational
296
analyses suggest that all three HSI loci have a role in facilitating the killing of
S. aureus
. While we focu
sed
297
on the HSI
-
III T6SS effector TseF for further study due to its significantly increased transcript level
298
revealed by RNA
-
sequencing analysis, future work will be dedicated to defining the scope and specificity
299
of functionality for
P. aeruginosa
antagoni
sm against
S. aureus
mediated by T6SS effectors. Moreover,
300
the global proteomics study was only performed on the
P. aeruginosa
intracellular lysate fraction, which
301
(which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
The copyright holder for this preprint
this version posted January 29, 2023.
;
https://doi.org/10.1101/2023.01.29.526047
doi:
bioRxiv preprint
11
did not include most secreted protein effectors found in the extracellular fraction; thus, i
t is possible
S.
302
aureus
induces the secretion of T6SS effectors not identified here.
A recent study demonstrated that
303
Tse4, a T6SS muramidase effector of
A. baumannii
, exhibits promiscuous PG
-
degrading activity and kills
304
Gram
-
positive species, including
S
. aureus
50
. While previous literature indicated T6SS
-
exported
305
muramidases generally cannot effectively lyse Gram
-
positive cells
62
, the possibility remains that certain
306
PG
-
targeting T6SS effectors can impact cellular functions of Gram
-
positive bacteria, not
limited to
307
causing cellular death or lysis. Beyond cell wall degrading toxins, developing evidence that suggests the
308
T6SS apparatus can inject and deliver effectors into the Gram
-
positive cell wall
50
points to the emerging
309
possibility that diverse T6SS ef
fectors could have bacteriostatic and bactericidal potential towards both
310
Gram
-
negative and Gram
-
positive bacteria. For instance, studies analyzing differential regulation for
S.
311
aureus
in coculture with
P. aeruginosa
consistently reported up
-
regulation of
SOS response
and oxidative
312
stress response pathways
15,40
, but it remained unclear how
P. aeruginosa
triggers these responses in
S.
313
aureus
. It is therefore curious to speculate that these effects could be due to previously unknown attacks
314
by
P. aeruginosa
antibacterial T6SS nuclease toxins
63
and NAD(P)+ glycohydrolases effectors
64
.
315
316
Cumulatively, our findings provide a new model of T6SS
-
medi
ated interspecies interactions for Gram
-
317
negative and Gram
-
positive species. Our results revealed complex polymicrobial virulence factors
318
crosstalk and highlight the importance of leveraging a comprehensive molecular understanding of
319
polymicrobial competiti
on while studying the host
-
pathogen interface. Considering both
Staphylococcal
320
PSMs and
P. aeruginosa
T6SS have well
-
characterized functions in modulating host immune responses,
321
their interactivity uncovered by our study could have detrimental implications
on the host.
322
323
Acknowledgements
:
324
This work was supported by the Jacobs Institute for Molecular Engineering for Medicine and the Center
325
for Environmental Microbial Interactions at Caltech, and by the Institute for Collaborative Biotechnologies
326
through coop
erative agreement W911NF
-
19
-
2
-
0026 from the U.S. Army Research Office, the Cystic
327
Fibrosis Foundation (LIMOLI19R3 to DHL and BOMBER18G0 to JMB), and the National Institutes of
328
Health (1R35GM142760
-
01 to DHL and 1R01HL142587 to JMB). We thank Drs. Megan Ber
gkessel
329
(University of Dundee), Melanie Spero (University of Oregon), Alex Horswill (University of Colorado
330
Denver), Mike Schurr (University of Colorado Denver), and Li Wu (University of Iowa) for helpful
331
discussions and valuable insight. We also thank mem
bers of the Limoli and Tirrell Labs for careful editing
332
of the manuscript and helpful discussions. We thank Dr. Jeff Jones (Caltech) for an in
-
house pipeline for
333
proteomics data processing, Dr. J. Muse Davis for the use of the stereoscope, and Drs. Joseph
Mougous
334
(which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
The copyright holder for this preprint
this version posted January 29, 2023.
;
https://doi.org/10.1101/2023.01.29.526047
doi:
bioRxiv preprint
12
and Anupama Khare for the generous gifts of the ClpV1
-
GFPmut3 and P’
pvdA
-
mScarlet
reporters,
335
respectively.
336
Competing interests
: The authors declare no competing interests.
337
338
References
339
1.
O’Sullivan, B. P. & Freedman, S. D. Cystic fibrosis.
Lancet (London, England)
373
, 1891
–
1904
340
(2009).
341
2.
Fischer, A. J.
et al.
Sustained coinfections with
Staphylococcus aureus
and
Pseudomonas
342
aeruginosa
in cystic fibrosis.
Am. J. Respir. Crit. Care Med
.
203
, 328
–
338 (2021).
343
3.
Limoli, D. H.
et al.
Staphylococcus aureus
and
Pseudomonas aeruginosa
coinfection is associated
344
with cystic fibrosis
-
related diabetes and poor clinical outcomes.
Eur. J. Clin. Microbiol. Infect. Dis.
345
35
, 947
–
953 (2016).
346
4.
Hubert, D.
et al.
Association between
Staphylococcus aureus
alone or combined with
347
Pseudomonas aeruginosa
and the clinical condition of patients with cystic fibrosis.
J. Cyst.
348
Fibros.
Off. J. Eur. Cyst. Fibros.
Soc.
12
, 497
–
503 (2013).
349
5.
Limoli
, D. H.
et al.
Interspecies interactions induce exploratory motility in
Pseudomonas
350
aeruginosa
.
e
life
8
, (2019).
351
6.
Mougous, J. D.
et al.
A virulence locus of
Pseudomonas aeruginosa
encodes a protein secretion
352
apparatus.
Science
312
, 1526
–
1530 (2006).
353
7.
Pukatzki, S., Ma, A. T., Revel, A. T., Sturtevant, D. & Mekalanos, J. J. Type VI secretion system
354
translocates a phage tail spike
-
like protein into target cells where it cross
-
links actin.
Proc. Natl.
355
Acad. Sci. U. S. A.
104
, 15508
–
15513 (2007).
356
8.
Sana, T. G.
et al.
The second type VI secretion system of
Pseudomonas aeruginosa
strain PAO1
357
is regulated by quorum sensing and Fur and modulates internalization in epithelial cells.
J. Biol.
358
Chem.
287
, 27095
–
27105 (2012).
359
9.
Basler, M., Ho, B. T.
& Mekalanos, J. J. Tit
-
for
-
tat: Type VI secretion system counterattack during
360
bacterial cell
-
cell interactions.
Cell
152
, 884
–
894 (2013).
361
10.
LeRoux, M.
et al.
Kin cell lysis is a danger signal that activates antibacterial pathways of
362
Pseudomonas aerug
inosa
.
e
life
4
, e05701 (2015).
363
11.
Peschel, A. & Otto, M. Phenol
-
soluble modulins and
Staphylococcal
infection.
Nat. Rev. Microbiol.
364
11
, 667
–
673 (2013).
365
12.
Surewaard, B. G. J.
et al.
Staphylococcal
alpha
-
phenol soluble modulins contribute to neutrophi
l
366
lysis after phagocytosis.
Cell. Microbiol.
15
, 1427
–
1437 (2013).
367
(which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
The copyright holder for this preprint
this version posted January 29, 2023.
;
https://doi.org/10.1101/2023.01.29.526047
doi:
bioRxiv preprint
13
13.
Mashburn, L. M., Jett, A. M., Akins, D. R. & Whiteley, M.
Staphylococcus aureus
serves as an iron
368
source for
Pseudomonas aeruginosa
during
in vivo c
oculture.
J. Bacteriol.
187
,
554
–
566 (2005).
369
14.
Hoffman, L. R.
et al.
Selection for
Staphylococcus aureus
small
-
colony variants due to growth in
370
the presence of
Pseudomonas aeruginosa
.
Proc. Natl. Acad. Sci.
103
, 19890 LP
–
19895 (2006).
371
15.
Filkins, L. M.
et al.
Coculture of
Sta
phylococcus aureus
with
Pseudomonas aeruginosa
drives
S.
372
aureus
towards fermentative metabolism and reduced viability in a cystic fibrosis model.
J.
373
Bacteriol.
197
, 2252 LP
–
2264 (2015).
374
16.
Limoli, D. H.
et al.
Pseudomonas aeruginosa
alginate overprodu
ction promotes coexistence with
375
Staphylococcus aureus
in a model of cystic fibrosis respiratory infection.
m
Bio
8
, (2017).
376
17.
Dieterich, D. C., Link, A. J., Graumann, J., Tirrell, D. A. & Schuman, E. M. Selective identification
377
of newly synthesized prot
eins in mammalian cells using bioorthogonal noncanonical amino acid
378
tagging (BONCAT).
Proc. Natl. Acad. Sci.
103
, 9482 LP
–
9487 (2006).
379
18.
Landgraf, P., Antileo, E. R., Schuman, E. M. & Dieterich, D. C. BONCAT: metabolic labeling, click
380
chemistry, and
affinity purification of newly synthesized proteomes.
Methods Mol. Biol.
1266
, 199
–
381
215 (2015).
382
19.
Silverman, J. M., Brunet, Y. R., Cascales, E. & Mougous, J. D. Structure and regulation of the type
383
VI secretion system.
Annu. Rev. Microbiol.
66
, 453
–
472
(2012).
384
20.
Chen, L., Zou, Y., She, P. & Wu, Y. Composition, function, and regulation of T6SS in
385
Pseudomonas aeruginosa
.
Microbiol. Res.
172
, 19
–
25 (2015).
386
21.
Stefan, P.
et al.
Identification of a conserved bacterial protein secretion system in
Vibrio
cholerae
387
using the
Dictyostelium
host model system.
Proc. Natl. Acad. Sci.
103
, 1528
–
1533 (2006).
388
22.
Brodmann, M., Dreier, R. F., Broz, P. & Basler, M.
Francisella
requires dynamic type VI secretion
389
system and ClpB to deliver effectors for phagosomal e
scape.
Nat. Commun.
8
, 15853 (2017).
390
23.
Mougous, J. D., Gifford, C. A., Ramsdell, T. L. & Mekalanos
, J. J. Threonine phosphorylation post
-
391
translationally regulates protein secretion in
Pseudomonas aeruginosa
.
Nat. Cell Biol.
9
, 797
–
803
392
(2007).
393
24.
Silverman, J. M.
et al.
Haemolysin coregulated protein is an exported receptor and chaperone of
394
type VI s
ecretion substrates.
Mol. Cell
51
, 584
–
593 (2013).
395
25.
Braud, A., Hoegy, F., Jezequel, K., Lebeau, T. & Schalk, I. J. New insights into the metal specificity
396
of the
Pseudomonas aeruginosa
pyoverdine
-
iron uptake pathway.
Environ. Microbiol.
11
, 1079
–
397
109
1 (2009).
398
26.
Ehud, B., L., V. M. & Peter, G. E. Iron and
Pseudomonas aeruginosa
biofilm formation.
Proc. Natl.
399
Acad. Sci.
102
, 11076
–
11081 (2005).
400
(which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
The copyright holder for this preprint
this version posted January 29, 2023.
;
https://doi.org/10.1101/2023.01.29.526047
doi:
bioRxiv preprint
14
27.
Leoni, L., Orsi
, N., de Lorenzo, V. & Visca, P. Functional analysis of PvdS, an iron starvation sigma
401
factor of
Pseudomonas aeruginosa
.
J. Bacteriol.
182
, 1481
–
1491 (2000).
402
28.
Wandersman, C. & Delepelaire, P. Bacterial iron sources: from siderophores to hemophores.
An
nu.
403
Rev. Microbiol.
58
, 611
–
647 (2004).
404
29.
Zarrella, T. M. & Khare, A. Systematic identification of molecular mediators of interspecies sensing
405
in a community of two frequently coinfecting bacterial pathogens.
PLoS Biol.
20
, e3001679 (2022).
406
30.
Togno
n, M., Köhler, T., Luscher, A. & van Delden, C. Transcriptional profiling of
Pseudomonas
407
aeruginosa
and
Staphylococcus aureus
during
in vitro
coculture
.
BMC Genomics
20
, 30 (2019).
408
31.
Leinweber, A., Weigert, M. & Kümmerli, R. The bacterium
Pseudomonas a
eruginosa
senses and
409
gradually responds to interspecific competition for iron.
Evolution (N. Y).
72
, 1515
–
1528 (2018).
410
32.
Stolle, A.
-
S., Meader, B. T., Toska, J. & Mekalanos, J. J. Endogenous membrane stress induces
411
T6SS activity in
Pseudomonas aerugino
sa
.
Proc. Natl. Acad. Sci. U. S. A.
118
, (2021).
412
33.
Cheung, G. Y. C., Joo, H.
-
S., Chatterjee, S. S. & Otto, M. Phenol
-
soluble modulins
—
critical
413
determinants of
Staphylococcal
virulence.
FEMS Microbiol. Rev.
38
, 698
–
719 (2014).
414
34.
Ulhuq
, F. R. & Mariano, G. Bacterial pore
-
forming toxins.
Microbiology
168
, (2022).
415
35.
Kizaki, H.
et al.
Cell
-
surface phenol soluble modulins regulate
Staphylococcus aureus
colony
416
spreading.
PLoS One
11
, e0164523 (2016).
417
36.
Tsviklist, V., Guest, R. L. & R
aivio, T. L. The Cpx stress response regulates turnover of respiratory
418
chain proteins at the inner membrane of
Escherichia coli
.
Frontiers in Microbiology
12
, (2022).
419
37.
Guest, R. L., Wang, J., Wong, J. L. & Raivio, T. L. A bacterial stress response reg
ulates respiratory
420
protein
c
omplexes to control envelope stress adaptation.
J. Bacteriol.
199
, (2017).
421
38.
Raivio, T. L., Leblanc, S. K. D. & Price, N. L. The
Escherichia coli
Cpx envelope stress response
422
regulates genes of diverse function that impact a
ntibiotic resistance and membrane integrity.
J.
423
Bacteriol.
195
, 2755
–
2767 (2013).
424
39.
Hews, C. L., Cho, T., Rowley, G. & Raivio, T. L. Maintaining integrity under stress: Envelope stress
425
response regulation of pathogenesis in Gram
-
negative bacteria.
Fron
t. Cell. Infect. Microbiol.
9
,
426
313 (2019).
427
40.
Manganelli, R. & Gennaro, M. L. Protecting from envelope stress: Variations on the phage
-
shock
-
428
protein theme.
Trends Microbiol.
25
, 205
–
216 (2017).
429
41.
Brissette, J. L., Russel, M., Weiner, L. & Model, P.
Phage shock protein, a stress protein of
430
Escherichia coli
.
Proc. Natl. Acad. Sci.
87
, 862
–
866 (1990).
431
42.
Chugani, S. & Greenberg, E. P. The influence of human respiratory epithelia on
Pseudomonas
432
aeruginosa
gene expression.
Microb. Pathog.
42
, 29
–
35 (20
07).
433
(which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
The copyright holder for this preprint
this version posted January 29, 2023.
;
https://doi.org/10.1101/2023.01.29.526047
doi:
bioRxiv preprint