Articles

https://doi.org/10.1038/s41564-020-00805-8

The histone demethylase KDM6B fine-tunes the

host response to Streptococcus pneumoniae
Michael G. Connor 1, Tiphaine M. N. Camarasa1,2, Emma Patey1,8, Orhan Rasid 1, Laura Barrio 3,4
,
Caroline M. Weight 5, Daniel P. Miller6, Robert S. Heyderman 5, Richard J. Lamont 7,
Jost Enninga3,4 and Melanie A. Hamon 1 ✉

Streptococcus pneumoniae is a natural colonizer of the human respiratory tract and an opportunistic pathogen. Although epi-
thelial cells are among the first to encounter pneumococci, the cellular processes and contribution of epithelial cells to the host
response are poorly understood. Here, we show that a S. pneumoniae serotype 6B ST90 strain, which does not cause disease
in a murine infection model, induces a unique NF-κB signature response distinct from an invasive-disease-causing isolate of
serotype 4 (TIGR4). This signature is characterized by activation of p65 and requires a histone demethylase KDM6B. We show,
molecularly, that the interaction of the 6B strain with epithelial cells leads to chromatin remodelling within the IL-11 promoter in
a KDM6B-dependent manner, where KDM6B specifically demethylates histone H3 lysine 27 dimethyl. Remodelling of the IL-11
locus facilitates p65 access to three NF-κB sites that are otherwise inaccessible when stimulated by IL-1β or TIGR4. Finally, we
demonstrate through chemical inhibition of KDM6B with GSK-J4 inhibitor and through exogenous addition of IL-11 that the host
responses to the 6B ST90 and TIGR4 strains can be interchanged both in vitro and in a murine model of infection in vivo. Our
studies therefore reveal how a chromatin modifier governs cellular responses during infection.

T
he pathobiont Streptococcus pneumoniae is a heterogeneous
species with over 90 serotypes, and naturally colonizes the
upper respiratory tract (URT) of humans1–6. Globally, dis-
ease attributed to S. pneumoniae infection is a priority due to
potential lethality resulting from pneumonia, sepsis or menin-
gitis4,7. Importantly, colonization of the human URT forms the
natural reservoir for this obligate pathobiont. It either remains in
the nasopharynx and is eventually cleared, or it may progress to
cause disease2,3,5,8,9.
At these initial stages, pneumococci interact with the host naso-
pharyngeal epithelial barrier and the innate immune system. Recent
studies using the experimental human pneumococcal carriage
model revealed the role of NF-κB driven responses for susceptibil-
ity, pathogenesis and transmission of pneumococcus8,10–13. However,
it is unclear how these cellular processes are shaped molecularly14.
NF-κB is a master transcriptional regulator of pro- and
anti-inflammatory host responses15–25 upon cellular sensing of exter-
nal stimuli. NF-κB activation occurs through post-translational
modifications (PTM) to NF-κB subunits, which in turn bind con-
sensus sequence sites to initiate transcription of NF-κB-dependent
genes26–28. However, cellular signalling alone is insufficient and chro-
matin remodelling is required to support a full NF-κB response15–25.
Chromatin, a highly ordered structure of DNA wrapped around
histone proteins, dynamically shifts between open and closed states
that influence gene accessibility and transcription29–33, as a result of
chromatin remodelling enzymes writing and erasing PTM upon
histone tails. One enzyme suggested to influence NF-κB transcrip-
tional responses is KDM6B (JMJD3), a histone demethylase15,34.
KDM6B belongs to the Jumonji C-domain family (JMJD) of his-
tone demethylases23,35. Mounting evidence, mainly in macrophages,
suggests KDM6B is essential for modulating inflammatory gene
expression during wound healing, upon lipopolysaccharide (LPS)
stimulation and for immunological tolerance to anthrax toxin19–22,36.
However, the role of KDM6B in bacteria-induced host responses
remains understudied.
We demonstrate that a pneumococcal isolate of serotype 6B ST90
CC156 lineage F, which in a murine infection model is contained
to the URT and does not cause invasive disease, activates a unique
NF-κB signature distinct from an invasive-disease-causing serotype
4 isolate (TIGR4). This signature includes upregulation of KDM6B
and the cytokine IL-11 in human epithelial cells. We show that the
promoter of IL-11 is remodelled upon 6B challenge via KDM6B
demethylation of H3K27me2, which allows p65 binding upstream
of the IL-11 transcriptional start site. We establish the importance
of this in regulating epithelial cell integrity and show, in vivo and
in vitro, that chemical inhibition of KDM6B causes the 6B strain,
which is otherwise non-invasive, to cause disease. Conversely, exog-
enous addition of recombinant IL-11 during challenge with an inva-
sive TIGR4 isolate leads to better containment of this strain to the
URT in a murine model of infection.
Results
Serotype 6B ST90 and TIGR4 display different disease outcomes
in a murine model of intranasal infection. We performed compar-
ative intranasal challenge studies of pneumococcal isolates TIGR4
(serotype 4) and 6B ST90 CC156 lineage F (serotype 6B; hereafter
referred to as 6B). Infection of mice with TIGR4 (Extended Data
Fig. 1a) induced weight loss and morbidity within 4 days
post-challenge (Fig. 1a,b). In contrast, 6B did not cause symptom-
atic pneumococcal disease, even at an infection dose one log higher

1
G5 Chromatin and Infection, Institut Pasteur, Paris, France. 2Université Paris Diderot, Sorbonne Paris Cité, Paris, France. 3Dynamics of Host–Pathogen
Interactions Unit, Institut Pasteur, Paris, France. 4UMR CNRS, Paris, France. 5Division of Infection and Immunity, University College London, London,
UK. 6Department of Microbiology and Immunology, Virginia Commonwealth University School of Medicine, Richmond, VA, USA. 7Department of Oral
Immunology and Infectious Diseases, School of Dentistry, University of Louisville, Louisville, KY, USA. 8Present address: University of Glasgow,
Scotland, UK. ✉e-mail: melanie.hamon@pasteur.fr

Nature Microbiology | www.nature.com/naturemicrobiology

Articles Nature Microbiology

than that of TIGR4. Since murine infection outcomes were differ-
ent, we chose to study the underlying molecular processes of infec-
tion with these strains of 6B or TIGR4.
6B actively induces a unique cellular response. To dissect the host
processes differentially regulated by the strains 6B and TIGR4, we
completed an exploratory microarray of human A549 epithelial
cells 2 h post-challenge (Extended Data Fig. 1b). In comparison to
uninfected cells, 6B differentially influenced 388 transcripts (200
upregulated and 188 downregulated) while TIGR4 modulated the
expression of 1,205 (143 upregulated and 1,062 downregulated;
Supplementary Table 1). Examination of the 6B dataset showed that
a larger portion contained NF-κB binding sites in contrast to TIGR4
(12% 6B versus 3% TIGR4; Extended Data Fig. 1c), suggesting an
early NF-κB response to 6B. Therefore, a panel of 41 genes (31 direct
targets of NF-κB) were tested by RT–qPCR (quantitative PCR with
reverse transcription). With 6B challenge, a significant increase in
the expression of CXCL2, IL-11, KDM6B and RELA occurred in
comparison to TIGR4 (Fig. 1c). This cellular response to 6B was not
restricted to A549 cells but was also observed with nasopharyngeal
Detroit 562 and non-cancer Beas2b epithelial lines, as KDM6B and
IL-11 expression levels clustered across cell lines (Extended Data
Fig. 1d). We cross-compared the same panel under stimulation
by IL-1β, a known activator of p65/RelA, and performed princi-
pal component analysis (PCA) for 6B, TIGR4 and IL-1β (Fig. 1d).
Comparative analysis of the first two components demonstrated that
the cellular response to 6B was characterized by upregulation of a sub-
set of NF-κB-regulated genes distinct from both TIGR4 and IL-1β.
In agreement with the expression data, upregulation of KDM6B
occurred in the nucleus of epithelial cells challenged with 6B in
comparison to TIGR4 and uninfected cells. Similarly, a modest
increase in IL-11 was observed in Detroit 562 cell supernatants
6 hours post-challenge with 6B or TIGR4 (Extended Data Fig. 1f).
Interestingly, no increase in KDM6B occurred following challenge
with paraformaldehyde-killed bacteria (Fig. 1e and Extended Data
Fig. 1e), suggesting that nuclear KDM6B levels increase only upon
interaction of live pneumococci with epithelial cells. Finally, we
show that KDM6B is the only JMJD demethylase in its family to be
upregulated by 6B (Extended Data Fig. 1g). Together, these results
show that 6B induces a differential transcriptional response in epi-
thelial cells that is characterized, in part, by upregulation of KDM6B
and the cytokine IL-11.
6B cellular response requires p65 activation and catalytically
active KDM6B. Serine 536 (S536) phosphorylation of the p65 sub-
unit is known to correlate with NF-κB activation27, for which we
probed (Fig. 2a). Compared to uninfected cells, both IL-1β and 6B
induced p65 S536 phosphorylation at 120 minutes, whereas TIGR4
did not (Fig. 2b). Although the expression of certain genes showed
dynamic differences between 6B and TIGR4, the expression of
IL-11 and KDM6B was upregulated by 6B in a temporal manner,
whereas the TIGR4 profile suggested dampening of these tran-
scripts (Extended Data Fig. 2a), indicating differences in NF-κB
activation between the strains.
To determine whether p65 activation was required for
6B-dependent expression of IL-11 and KDM6B, we used BAY
11-7082 to chemically inhibit p65 activation37–39. BAY 11-7082
alone did not affect viability or gene expression (Extended
Data Fig. 2b,c), aside from the expected reduction in PTGS2
(p65-activation-dependent positive control; Fig. 2c). Expression of
KDM6B and IL-11 was dampened during 6B challenge with inhibi-
tor in comparison to cells treated with dimethylsulfoxide (DMSO)
(Fig. 2c). Therefore, 6B-dependent expression of KDM6B and
IL-11 depends on p65 activation. To determine whether KDM6B has
an active role in 6B-mediated expression of IL-11 and KDM6B, we
used GSK-J4, an inhibitor of the catalytic JMJ domain of KDM6B40.
GSK-J4 treatment alone did not affect cell viability or gene expres-
sion in comparison to untreated cells (Fig. 2c and Extended Data
Fig. 2b,c). However, a threefold loss of expression for both IL-11 and
KDM6B was observed during 6B challenge in comparison to DMSO
(Fig. 2c). These data demonstrated that KDM6B catalytic activity
was required for IL-11 expression.
Chromatin is remodelled within the IL-11 promoter upon 6B
challenge. To address whether 6B induced p65 recruitment to
KDM6B and chromatin was remodelled within the IL-11 promoter,
we used chromatin immunoprecipitation (ChIP)-qPCR. Multiple
kappa-binding sites (κbs) upstream of the KDM6B and IL-11 tran-
scriptional start sites were predicted (Fig. 3a and Extended Data
Fig. 3c). We first confirmed p65 recruitment to κbs within KDM6B
(Extended Data Fig. 3a,b) and IL-11 promoters (Fig. 3b; 6B dark
blue, uninfected white) during 6B challenge in comparison to unin-
fected cells. For the IL-11 promoter a significant (P ≤ 0.001) recov-
ery of p65 at κbs P6 (25%), P3 (20%) and P2 (10%) occurred, along
with recovery of KDM6B across the same κbs in 6B-challenged cells
(Fig. 3c). In contrast, no recruitment of p65 or KDM6B occurred dur-
ing IL-1β or TIGR4 challenge (Extended Data Fig. 3d,e). Recovery
of p65 and KDM6B was abolished with GSK-J4 treatment (Fig. 3b,c;
6B light blue; uninfected grey). Thus, the promoter of IL-11 is rear-
ranged in a manner requiring the catalytic activity of KDM6B.
Because H3K27me3 is the primary enzymatic target of
KDM6B41,42, we determined its level across κbs within the IL-11 pro-
moter. Surprisingly, H3K27me3 increased in comparison to unchal-
lenged cells (Fig. 3d), and no enrichment at these κbs occurred with
IL-1β or TIGR4 challenge (Fig. 3e and Extended Data Fig. 3g).
This suggested that KDM6B enzymatic activity was not targeting
H3K27me3; thus, we tested H3K27me2, another proposed sub-
strate of KDM6B43. Strikingly, H3K27me2 was reduced upon 6B
challenge in comparison to uninfected cells, and GSK-J4 treat-
ment inhibited demethylation across all κbs (Fig. 3f). Interestingly,

Fig. 1 | Serotype 6B induces a unique inflammatory signature. C57B6 mice (8–9 weeks of age) were challenged intranasally with TIGR4 (3 × 106 to
4 × 106 colony forming units (c.f.u.); n = 8), 6B (3 × 106 to 4 × 106 c.f.u.; n = 6) or 6B (3 × 107 to 4 × 107 c.f.u.; n = 6). Survival and weight were monitored
for 14 days post-infection. a, Percent survival curve. Cox–Mantel test for significance, ***P ≤ 0.001. For TIGR4 versus 6B (low), P = 0.0005. For TIGR4
versus 6B (high) P = 0.0005. “ indicates that the animal was euthanized but did not succumb to disease. b, Percent initial body weight. Solid lines with
shading (grey indicates TIGR4, blue indicates 6B (low), and pink indicates 6B (high)) representing ±1 s.d. Dashed line indicates the 20% weight loss
threshold. c, RT–qPCR validation of inflammatory genes from cells collected 2 h post-challenge with either TIGR4 (multiplicity of infection (MOI) 20) or
6B (MOI 20). IL-11, KDM6B (JMJD3), PTGS2, CXCL8 (IL-8), FOS and JUNB are hits from the microarray included in the panel. The heat map represents fold
change to uninfected condition (n = 5 biological replicates). Two-tailed multiple t-test 6B to TIGR4 with significant genes in red. d, PCA of inflammatory
RT–qPCR panel (n = 5 biological replicates) comparing IL-1β (10 ng ml−1; red), TIGR4 (MOI 20; grey) and 6B (MOI 20; blue). Bioplot of the mean-centred
and log2-transformed expression data using the first two components (74.9% of the total variance) and 95% confidence ellipses around each group.
e, Quantification of immunofluorescence microscopy of A549 cells 2 h post-challenge with either TIGR4 (MOI 20), 6B (MOI 20) or paraformaldehyde
(PFA)-fixed 6B or TIGR4 (MOI 20) for nuclear staining of KDM6B normalized to DAPI (n = 4 biological replicate; 30–100 cells per replicate and group).
Violin plots with median denoted in red. One-way analysis of variance (ANOVA) non-parametric Kruskal–Wallis test with Dunn’s multiple comparison
post-hoc test; ****P ≤ 0.0001; NS, not significant.

Nature Microbiology | www.nature.com/naturemicrobiology

Nature Microbiology Articles
although H3K27me2 levels decreased at the IL-11 locus, globally, Together, these data show that (1) upon 6B challenge, the pro-
both H3K27me2 and H3K27me3 levels decreased upon 6B chal- moter of IL-11 is remodelled through the cooperative role of
lenge, and in the presence of GSK-J4 the decrease was not observed KDM6B and p65, (2) KDM6B enzymatic activity appears to be
(Extended Data Fig. 3h–j). directed toward H3K27me2 and independent of H3K27me3 at

a b 120
“
100

110

Percent body weight

80
Percent survival

TIGR4 (~3 × 106 to 4 × 106)

60 100
*** 6B (~3 × 106 to 4 × 106) TIGR4
40
90 6B
(~3 × 107 to 4 × 107)
***
6B
20 6B
80
0
0 2 4 6 8 10 12 14 0 2 4 6 8 10 12 14
Time (days) Time (days)

c A549 cells d
6.0
6B TIGR4

–0.24 –0.19 BCL2

CCL2 TIGR4
3.39 2.10
1.57 –0.05 CCL3 3.0
0.10 1.27 CCL4
CCL5 6B
0.73 0.39
PC2 (18.8%)

6
4.44 2.42 CSF2
3.02 1.56 CXCL1 0

0.49 0.61 CXCL10

0.40 –1.17 CXCL11
3.41 1.92 CXCL2 (P = 0.026)
–3.0 IL-1β
3.38 2.38 CXCL3
Average relative expression to uninfected (log2)

4 CXCL5
0.25 –0.29
–0.21 –1.00 EGFR
5.73 6.23 FOS –5.0 –2.5 0 2.5 5.0 7.5
–0.05 0.49 GADD45B PC1 (56.1%)
1.16 –0.27 ICAM1
0.33 –0.24 IL10 e 4 ****
2 IL11 (P = 0.001)
2.62 0.51
****
IL15
–0.05 –0.60 ****
–0.85 0.27 IL17C
NS
3
IL17RE
Ratio of KDM6B to DAPI

0.12 0.80
4.23 3.54 IL18
2.45 2.23 IL1A
0 IL1B 2
0.39 –0.42
–0.02 –0.16 IL25
–0.14 0.12 IL33
4.50 3.36 IL6 1
5.09 4.43 IL8
1.83 1.26 JUNB
–2
2.61 0.12 KDM6B (P = 0.007)
0
1.31 0.63 NLRP3
Uninfected 6B PFA TIGR4 PFA
–0.49 –0.38 PTAFR 6B TIGR4
2.37 2.05 PTGS2
–0.17 –0.67 RELA (P = 0.026)
1.19 0.02 RELB
–0.17 –0.93 TLR2
–0.57 –1.04 TLR9
1.06 1.09 TNF
–0.55 –0.75 TNFRSF1A
0.10 –0.25 TNFSF9
0.31 –0.71 TRAF1

Nature Microbiology | www.nature.com/naturemicrobiology

Articles Nature Microbiology

a b

Lane marker
1.5 ** *

Uninfected

Actin-normalized ratio of phos-p65

**

TIGR4
IL-1β

6B

S536 to total p65

1.0
GFP-p65 (~95 kDa)

Actin (~37 kDa)

0.5

GFP-phos-p65
(S536)
0

TIGR4
Uninfected

IL-1β

6B
Actin

c
6B TIGR4 IL-1β
6 6 6
GAPDH normalized relative expression

4 4 4

**
to uninfected (log2)

2 2 2

0 0 0

–2 –2 * –2
**

–4 * –4 –4
*
***

–6 * –6 –6
PTGS2 IL-11 KDM6B PTGS2 IL-11 KDM6B PTGS2 IL-11 KDM6B

No inhibitor DMSO 10 µM BAY 11-7082 10 µM GSK-J4

Fig. 2 | Expression of KDM6B and IL-11 is specific to 6B and requires p65 activation. Immunoblot of stable HeLa GFP-p65-expressing cells. Whole cell
lysates 2 hours post-challenge with either IL-1β (10 ng/ml), TIGR4 (MOI 20) or 6B (MOI 20) and probed for actin, p65 and phosphorylated p65 at Serine
536. Full blots provided in Supplementary Information 1. a, Representative image of immunoblot. b, Actin-normalized ratio of phos-p65 S536 to total p65
(n = 7 biological replicates). Dot plot with mean denoted in red. One-way ANOVA with Tukey’s multiple comparison post-hoc test, *P ≤ 0.05, **P ≤ 0.01.
For uninfected versus 6B, P = 0.0322. For uninfected versus IL-1β, P = 0.0059. For TIGR4 versus uninfected, P = 0.6358. For 6B versus TIGR4, P = 0.0025.
c, Total RNA 2 hours post-challenge with 6B (MOI 20), TIGR4 (MOI 20) or IL-1β was harvested from A549 cells treated with either 10 µM BAY 11-7082
(n = 3 biological replicates), 10 µM GSK-J4 (n = 4 biological replicates), DMSO vehicle control (n = 4 biological replicates) or untreated (no inhibitor; n = 4
biological replicates). For all biological replicates, 2–3 technicals were performed. Transcript levels for PTGS2, IL-11 and KDM6B were determined by
RT–qPCR. Data are shown as mean ± s.d. Two-tailed unpaired Student’s t-test treatment to untreated (no inhibitor), *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001.

these kappa-binding sites and (3) 6B infection induces global loss of uninfected (untreated) and 6B groups (Fig. 4d and Extended Data
both H3K27 di- and trimethyl. Fig. 4c). These data suggest that KDM6B and IL-11 are involved in
moderating host transcription, particularly for a subset of NF-κB
KDM6B and IL-11 shape the cellular response. Since targets, and could potentially be used to alter the host response dur-
KDM6B-dependent chromatin remodelling is necessary for IL-11 ing pneumococcal infection.
expression, we hypothesized that it influenced the host response
to 6B and, conversely, that exogenously adding IL-11 would alter KDM6B and IL-11 contribute to epithelial cell integrity. Previous
the host response to TIGR4. 6B challenge with GSK-J4 significantly work demonstrated that KDM6B and p65 were required for kera-
dampened expression of 8/41 transcripts (Fig. 4a), and exogenous tinocyte wound healing22. We hypothesized that KDM6B was
IL-11 altered the expression of 13/41 transcripts upon TIGR4 chal- involved in maintaining epithelial cell integrity during pneumococ-
lenge. Importantly, KDM6B and IL-11 influenced general NF-κB cal infection. We first tested plasma membrane integrity of epithe-
responses. Upon IL-1β stimulation with GSK-J4 or IL-11, important lial cells challenged with 6B or TIGR4 using trypan blue exclusion,
modulations of cellular transcription were observed. In particular, as damaged membranes are permissive to dye accumulation. Upon
inhibiting KDM6B enzymatic activity led to a threefold repression 6B challenge, epithelial integrity was comparable to uninfected cells,
of RelB, a negative mediator of RelA inflammatory responses44, while TIGR4 challenge resulted in 60% epithelial membrane dam-
while exogenous addition of IL-11 regulated specific inflammatory age (Fig. 5a,b). Strikingly, GSK-J4 treatment induced a significant
mediators, such as TNF (fivefold reduction; Extended Data Fig. 4a). 20% (P ≤ 0.001) increase in trypan blue staining upon 6B chal-
Finally, we used PCA to compare how GSK-J4 or IL-11 treat- lenge, while GSK-J4 alone had no effect on cell viability or integ-
ments influenced inflammatory gene transcription. For the GSK-J4 rity (Fig. 5b and Extended Data Fig. 5a). We tested if integrity loss
PCA, 6B+GSK-J4 overlapped with TIGR4 in the second dimension was attributed to differential activity or expression of pneumolysin,
(Fig. 4c and Extended Data Fig. 4b), while the IL-11 PCA showed a well-documented pore-forming cholesterol-dependent cytolysin
that TIGR4+IL-11 grouped and partially overlapped with the conserved across all pneumococcal isolates3,45. No difference in

Nature Microbiology | www.nature.com/naturemicrobiology

Nature Microbiology Articles
a b P6 P3 P2
P6 P3 P2
50 *** *** *** ** *** **

– 2,077

Percent recovery
– 774

– 406
40

+ 83
30

p65
1 2 3 4 IL-11 locus 20
κB 10
site 0
0
TSS – + – + – + – +

+ GSK-J4

Untreated

+ GSK-J4
No inhibitor
Uninfected 6B Uninfected 6B

Uninfected 6B

c P6 P3 P2 d

30 *** *** * ** * ** 2.5 * ** NS

* NS NS

Percent recovery
Percent recovery

H3K27me3/H3
2.0
20
KDM6B

1.5
1.0
10
0.5
0 0.0
– + – + – + – + – + – + – + – + – + – + – + – +

Uninfected 6B Uninfected 6B Uninfected 6B Uninfected 6B Uninfected 6B Uninfected 6B

e f NS P = 0.06
NS * NS * NS NS ** *** ** ***
100 2.5
Percent recovery
Percent recovery

H3K27me2/H3

80 2.0
60 1.5
H3

40 1.0
20 0.5
0 0.0
– + – + – + – + – + – + – + – + – + – + – + – +
Uninfected 6B Uninfected 6B Uninfected 6B Uninfected 6B Uninfected 6B Uninfected 6B

Fig. 3 | 6B induces IL-11 promoter rearrangement. Chromatin was obtained from untreated A549 cells (blue) and A549 cells treated with 10 µM GSK-J4
(light blue) 2 hours post-challenge with 6B (MOI 20) in comparison to uninfected cells (untreated white; treated grey). 10 µg chromatin input was used
for ChIP of p65, KDM6B, H3K27me3, H3K27me2 and histone H3 (H3), followed by ChIP-qPCR at locations (P6, P3 and P2) spanning the NF-κB sites
upstream of the transcriptional start site (TSS). a, Schematic of IL-11 promoter with ChIP-qPCR primer locations (P6, P3 and P2) and NF-κB sites as
predicted by AliBaba2-curated eukaryotic transcription factor DNA-binding motifs from the TRANSFAC database74. b–f, Per cent recovery of input for p65
(b), KDM6B (c), H3K27me3 normalized to H3 (d), H3K27me2 normalized to H3 (f) or H3 (e) bound at P6, P3 and P2 in untreated and GSK-J4-treated
samples (n = 3 untreated; n = 3 GSK-J4 treated). Tukey box and whisker plot with defined box boundaries being the upper and lower interquartile range
(IQR), ‘whiskers’ (fences) being ± 1.5 times IQR and the median depicted by the middle solid line. Dots represent outliers. Two-tailed unpaired Student’s
t-test comparisons for untreated to GSK-J4-treated or 6B-infected to uninfected, *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001.

expression of pneumolysin between 6B and TIGR4 nor in haemo- with symptomatic disease and uncommonly carried. We compared
lytic activity in the presence of GSK-J4 inhibitor occurred (Extended isolates of serotypes 19A and 19F56, which are associated with car-
Data Fig. 5b,c). Importantly, a lactate dehydrogenase (LDH) assay riage, and two serotype 1 isolates harbouring either a haemolytic or
showed that not all trypan blue positive cells were dead (Extended non-haemolytic pneumolysin allele, which are less commonly car-
Data Fig. 5a), which is in line with previous toxin studies46,47. These ried but cause disease states. The 19A and 19F isolates upregulated
results suggest KDM6B plays a role in mediating cell integrity dur- IL-11 expression in epithelial cells in comparison to uninfected
ing 6B challenge. cells, whereas the serotype 1 isolates did not (Fig. 5e). We further
Respiratory epithelial cells can produce IL-11 (refs. 48–50), and tested five additional oral microbiome constituents, all of which
IL-11 regulates epithelial cell plasma membrane proteins51 while upregulated IL-11 expression in immortalized gingival keratino-
modulating inflammatory, healing and mucosa responses52–55. We cytes (Extended Data Fig. 5d). Together, our data show that pneu-
therefore tested if exogenously adding the cytokine IL-11 could mococcal isolates associated with carriage and commensal bacteria
mitigate cell integrity loss during TIGR4 challenge. Exogenous increase IL-11 expression in vitro.
IL-11 had no effect on uninfected or 6B-challenged cells (Fig. 5d).
However, addition of IL-11 during TIGR4 challenge raised cell KDM6B and IL-11 are essential for localized containment of
integrity by 20% (Fig. 5d) and lowered TIGR4 cytotoxicity by pneumococcus in vivo. Given our in vitro data, we hypothesized
15% (Extended Data Fig. 5a) in comparison to untreated controls. that local inhibition of KDM6B could promote the escape of 6B
Together, these data suggest that IL-11 contributes to maintaining from the nasopharynx. Animals were challenged with either
epithelial cell integrity during pneumococcal challenge. 6B or TIGR4 supplemented with DMSO or GSK-J4. No signifi-
Our results raised the possibility that epithelial integrity regula- cant effect of DMSO or GSK-J4 on bacterial viability occurred
tion through IL-11 could be a common mechanism shared by com- (Extended Data Fig. 6d). Bacterial burden in the nasal lavage,
monly carried pneumococcal isolates but not by isolates associated bronchoalveolar lavage fluid (BALF), lungs and spleens of mice

Nature Microbiology | www.nature.com/naturemicrobiology

Articles Nature Microbiology

a b c
6B TIGR4
GSK-J4 comparison
GSK-J4 _ + IL-11 _ +
5.0
n=5 n=5 n=5 n=4
FOS 5.7305 6.4462 6.2287 4.8920 FOS
2.5
IL8 5.0880 2.9652 4.4259 2.7926 IL8

PC2 (21.3%)
IL6 4.4963 2.3131 3.5369 –2.6739 IL18
0
CSF2 4.4448 0.3422 3.3635 1.6475 IL6
IL18 4.2329 –0.3248 2.4190 0.2306 CSF2
−2.5
CXCL2 3.4115 1.7814 2.3796 1.1403 CXCL3
CCL2 3.3946 0.3872 2.2319 3.5441 IL1A
−5.0
CXCL3 3.3836 1.8896 2.0990 –0.1975 CCL2
−5.0 −2.5 0 2.5 5.0 7.5
CXCL1 3.0221 1.4691 2.0470 0.3610 PTGS2
PC1 (32.5%)
IL11 2.6165 0.3954 1.9220 0.3976 CXCL2
KDM6B 2.6110 0.4471 1.5597 0.3264 CXCL1 6B+GSK-J4 (ΔCT)

IL1A 2.4536 1.2240 1.2650 –0.4132 CCL4 TIGR4 (ΔCT)

PTGS2 2.3725 1.8650 1.2611 0.7436 JUNB Uninfected (ΔCT)

JUNB 1.8350 1.7408 1.0946 –0.0925 TNF

CCL3 1.5706 –0.2809 0.7984 –0.7338 IL17RE
NLRP3 1.3096 –0.6824 0.6311 –0.2850 NLRP3
RELB 1.1917 0.0944 0.6146 1.4877 CXCL10 d IL-11 comparison
ICAM1 1.1599 0.8040 0.5078 –0.3797 IL11
6.0
TNF 1.0606 0.7379 0.4934 0.8217 GADD45B
CCL5 0.7330 1.1383 0.3851 0.2217 CCL5
3.0
CXCL10 0.4905 –2.7024 0.2738 –0.1238 IL17C
PC2 (19.3%)

CXCL11 0.3976 2.1452 0.1243 –1.6180 KDM6B

0
IL1B 0.3900 –0.3738 0.1189 –0.2018 IL33
IL10 0.3336 1.7568 0.0187 –2.0117 RELB
−3.0
TRAF1 0.3053 0.5105 –0.0497 –1.9036 CCL3
CXCL5 0.2462 0.3847 –0.1648 –0.0335 IL25
−6.0
IL17RE 0.1236 0.3613 –0.1938 0.4221 BCL2
−5.0 −2.5 0 2.5 5.0 7.5
CCL4 0.1036 0.7957 –0.2355 –1.4032 IL10 PC1 (31.2%)
TNFSF9 0.0962 –0.0005 –0.2536 –0.1897 TNFSF9
–0.0228 0.3723 –0.2676 –2.6437 TIGR4+IL-11(ΔCT)
IL25 ICAM1
–0.0465 0.0891 –0.2900 –0.9126 6B (ΔCT)
GADD45B CXCL5
Uninfected (ΔCT)
IL15 –0.0538 –0.4745 –0.3784 –0.6111 PTAFR
IL33 –0.1424 –0.7470 –0.4207 0.2427 IL1B
RELA –0.1707 –0.6153 –0.5970 –0.1632 IL15
TLR2 –0.1748 0.1244 –0.6653 –2.4968 RELA
EGFR –0.2101 0.3309 –0.7121 –1.7825 TRAF1
BCL2 –0.2442 0.1640 –0.7469 –1.2192 TNFRSF1A
PTAFR -0.4887 0.2336 –0.9266 –0.3750 TLR2
TNFRSF1A –0.5532 –0.7146 –1.0041 –1.5796 EGFR
TLR9 –0.5749 0.0461 –1.0434 –1.4538 TLR9
IL17C –0.8525 0.4319 –1.1656 0.2890 CXCL11

6 4 2 0 –2
Average relative expression to matched uninfected (log2)

Fig. 4 | GSK-J4 and IL-11 treatment alters the host response to 6B and TIGR4. A549 cells untreated or treated with 10 µM GSK-J4 or 100 ng/ml
recombinant human IL-11 prior to 2-h challenge with either 6B or TIGR4. Total RNA was harvested and RT–qPCR completed for 41 inflammatory genes.
a,b, Heat-mapped expressions of 6B (MOI 20)-challenged cells with (+) or without (−) GSK-J4 (a, n = 5 biological replicates) or TIGR4 (MOI 20)-
challenged cells with or without IL-11 (b, n = 4 biological replicates) to their respective uninfected conditions. For all data and every biological replicate
an average of two technicals are displayed. Two-tailed multiple t-test 6B to treatment and TIGR4 to treatment with significant genes highlighted in red.
c,d, PCA of inflammatory RT–qPCR panel using ΔCT comparing either TIGR4 (grey; n = 5 biological replicates), 6B + 10 µM GSK-J4 (blue; n = 5 biological
replicates) and uninfected (orange; n = 5 biological replicates) (c) or TIGR4 + 100 ng ml−1 IL-11 (grey; n = 4 biological replicates), 6B (blue; n = 5 biological
replicates) and uninfected (orange; n = 5 biological replicates) (d). Bioplot of the mean-centred ΔCT data using the first two components and 60%
concentration ellipses around each group.

post-inoculation were quantified by conventional c.f.u. (24 h total had decreased, while the burden in the lung and spleen increased
c.f.u. shown in Extended Data Fig. 6a and 48 h total c.f.u. shown compared to 24 h. However, 6B c.f.u. numbers remained either
in Fig. 6a). By 48 h, TIGR4 c.f.u. in the nasal lavage and BALF constant or decreased in all samples (Fig. 6b; 6B light blue).

Nature Microbiology | www.nature.com/naturemicrobiology

Nature Microbiology Articles
a Uninfected TIGR4 6B b NS ***

100 *** *** NS

DMSO ***

80

Percent trypan-positive cells

60

+ GSK-J4
40
NS

20
c
Uninfected TIGR4 6B
0

DMSO uninfected

GSK-J4 uninfected

DMSO TIGR4

GSK -J4 TIGR4

DMSO 6B

GSK-J4 6B
–

d NS
100 **
+ IL-11

80 NS NS

Percent trypan-positive cells

*** ***

e 5 60
IL-11 relative expression to uninfected

40
4
NS
20
3

0
2
+ IL-11 (–) (+) (–) (+) (–) (+)

Uninfected TIGR4 6B
1

0
6B 19F 19A
)

)
Ply

Ply
tic

tic
oly

ly
mo
em

ae
ha

n-h
1(

no
1(

Fig. 5 | KDM6B and IL-11 contribute to epithelial cell integrity in response to pneumococcus. A549 cells untreated or treated with 10 µM GSK-J4 or
100 ng ml−1 recombinant human IL-11 prior to 2-h challenge with either TIGR4 (MOI 20) or 6B (MOI 20). Post-challenge cells were incubated with trypan
blue for cell integrity, fixed with 2.5% paraformaldehyde and imaged with a bright-field microscope. a, Representative image of GSK-J4-treated A549 cells
after the 2-h challenge. Scale bars, 100 µm. b, Percent trypan-positive cells between untreated and GSK-J4-treated (pink) (n = 6 biological replicates;
200–300 cells per replicate and group). Uninfected white, TIGR4 grey and 6B blue. Tukey box and whisker plot of the average per biological replicate per
group with defined box boundaries being the upper and lower IQR, whiskers (fences) being ± 1.5 times IQR and the median depicted by the middle solid
line. c, Representative image of IL-11-treated A549 cells after the 2-h challenge. Scale bars, 100 µm. d, Percent trypan-positive cells between untreated
and IL-11 treated (green) (untreated n = 6 and IL-11-treated n = 4 biological replicates; 200–300 cells per replicate and group). Uninfected white, TIGR4
grey and 6B blue. Tukey box and whisker plot of the average per biological replicate per group with defined box boundaries being the upper and lower IQR,
whiskers (fences) being ± 1.5 times IQR and the median depicted by the middle solid line. e, Total RNA harvested from A549 cells 2 h post-challenge with
isolates of serotypes 6B, 19F, 19A and two isolates of 1 (haemolytic and non-haemolytic Ply; n = 3 biologicals per isolate; 3 technicals per biological per
isolate). Relative expression of IL-11 to uninfected cells. Graphed as mean ± s.d. of all data. All data analysed by one-way ANOVA with Tukey’s multiple
comparison post-hoc test. **P ≤ 0.01; ***P ≤ 0.001.

With this data, we concluded 6B was primarily contained within GSK-J4-treated animals showed increased burdens at 24 and
the nasal cavity, whereas TIGR4 escapes the URT and dissemi- 48 h, with a significantly (P ≤ 0.05) higher c.f.u. in the BALF and
nates from the lungs. spleen compared to 6B-DMSO-treated animals at 48 h (Fig. 6a and

Nature Microbiology | www.nature.com/naturemicrobiology

Articles Nature Microbiology

a
Nasal lavage BALF Lung Spleen

NS NS NS NS

NS * ** *
9
8 8 NS *
P = 0.07
10 *
7 NS 9 8
7
48 h total c.f.u. (log10)

6 *** 8 7
6
5 7 P = 0.08 6
5 6
4 5 **
4 5
3 4
X 4
X X
2 X LD LD LD
LD
1 2 2 2
DMSO TIGR4

GSK-J4 TIGR4

DMSO 6B

GSK-J4 6B

DMSO TIGR4

GSK-J4 TIGR4

DMSO 6B

GSK-J4 6B

DMSO TIGR4

GSK-J4 TIGR4

DMSO 6B

GSK-J4 6B

DMSO TIGR4

GSK-J4 TIGR4

DMSO 6B

GSK-J4 6B
b
** **
7 5 7 7
** P = 0.07
6
6
48 h total c.f.u. (log10)

6
5
4
4 5 5

3 4 4
3
2
3 3
LD LD
LD LD
0 2 2 2
TIGR4 TIGR4+IL-11 TIGR4 TIGR4+IL-11 TIGR4 TIGR4+IL-11 TIGR4 TIGR4+IL-11

c 110

* ***
Relative body weight (%)

100

90

TIGR4
80
TIGR4+IL-11

70
0 1 2
Time (days)

Fig. 6 | KDM6B is required for host response to serotype 6B. C57B6 mice (8–9 weeks of age) were challenged intranasally with 3 × 106 to 4 × 106 c.f.u.
of TIGR4 or 6B supplemented with either DMSO (vehicle control), 5 mM GSK-J4 or 0.15 μg recombinant mouse IL-11. At indicated endpoints bacterial
load was enumerated by conventional c.f.u. counts on 5 µg ml−1 gentamicin Columbia blood agar selection plates from the nasal lavage, BALF, lungs
and spleen of infected animals. a, 6B and TIGR4±GSK-J4 with matched DMSO controls. c.f.u. burden of indicated samples 48 h post-inoculation (n = 9
animals; geometric mean of two technical counts per lavage or organ). DMSO TIGR4 white, GSK-J4 TIGR4 grey, DMSO 6B light blue, GSK-J4 6B dark
blue. b, TIGR4±IL-11-challenged animals 48 h post-inoculation. c.f.u. burden of indicated samples (n = 9 animals; geometric mean of two technical counts
per lavage or organ). TIGR4 (grey; n = 9) and TIGR4+IL-11 (green; n = 9). All graphed as Tukey box and whisker plots with defined box boundaries being
the upper and lower IQR, whiskers (fences) being ± 1.5 times IQR and the median depicted by the middle solid line. Dots represent outliers. One-way
ANOVA non-parametric Kruskal–Wallis with Dunn’s multiple comparison post-hoc test, *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001. Exact P are reported in Fig. 6
source data. Dotted lines show the limits of detection (LD). The LD for nasal lavage was 50 c.f.u. The LD for BALF, lung and spleen were 1,000 c.f.u. for
DMSO and GSK-J4 and 250 c.f.u. for TIGR4±IL-11. X indicates that the animal died prior to the endpoint. c, Percent initial body weight TIGR4 (grey) and
TIGR4+IL-11 (green). Solid lines with shading represent s.d. The dashed line indicates the 20% weight-loss threshold. Repeated measures of one-way
ANOVA with Tukey’s multiple comparison post-hoc test, **P ≤ 0.01, ***P ≤ 0.001. TIGR4 versus TIGR+IL-11 day 1 P = 0.01444. TIGR4 versus TIGR4+IL-11
day 2 P < 0.0001.

Extended Data Fig. 6a; 6B DMSO light blue, 6B GSK-J4 dark blue). responses. Altogether, these data show KDM6B activity is required
Importantly, bacteria were recovered from the spleen of 6B GSK-J4 for localized containment of 6B during infection of the murine nasal
animals, an organ where c.f.u. were largely undetected in control cavity and is potentially a negative regulator of TIGR4 dissemination.
animals. The TIGR4 GSK-J4 group also showed an increase in From our in vitro studies, we hypothesized that localized admin-
c.f.u. at 48 h, suggesting a basal role of KDM6B in regulating NF-κB istration of IL-11 in vivo might delay translocation of TIGR4 into the

Nature Microbiology | www.nature.com/naturemicrobiology

Nature Microbiology
lower respiratory tract by promoting containment within the URT.
We challenged animals intranasally with TIGR4 supplemented with
recombinant mouse IL-11 at the time of infection (Extended Data
Fig. 6c). Supplementation of IL-11 had no effect on TIGR4 bacte-
rial growth (Extended Data Fig. 6d). At 48 h, there was substantial
recovery of TIGR4 in the BALF, lung and spleen, along with a 10%
loss of starting weight (Fig. 6b,c), which indicates symptomatic
pneumococcal disease. In contrast, IL-11 treatment resulted in five-
fold more c.f.u. in the nasal lavage, a one-log reduction in c.f.u. from
internal organs and a 5–10% preservation of weight in contrast to
the TIGR4 group (Fig. 6b,c). Overall, these data demonstrate that
during TIGR4 pneumococcal infection, a localized IL-11 response
contributes to reduce bacterial dissemination to the lower respira-
tory tract.
Discussion
Bacterial–epithelial interaction is an essential process leading to
pneumococcal carriage and precedes symptomatic pneumococcal
disease3,5. However, the molecular and transcriptional processes
across the spectrum of host interaction of nasal microbiome con-
stituent to potentially lethal disease are not completely understood.
Herein, we first demonstrated the pneumococcal strain 6B does not
cause symptomatic disease or lethality, in contrast to a serotype 4
(TIGR4) strain in a murine infection model. Using these two strains,
we show that the 6B isolate differentially regulated 388 genes, with
enrichment for NF-κB associated genes in comparison to the TIGR4
strain. In fact, 6B induces a unique transcriptional response requir-
ing phosphorylation of p65 and expression of KDM6B and IL-11.
Molecularly, the 6B isolate induced KDM6B-dependent remodel-
ling of the IL-11 promoter to reveal three NF-κB sites that were
inaccessible during IL-1β or TIGR4 stimulation. Together, this is
the first demonstration that a 6B strain remodels chromatin within
epithelial cells.
The lysine demethylase KDM6B, mainly characterized in cellular
development, has been shown to fine-tune inflammatory responses
and wound healing downstream of p65 through largely unknown
mechanisms20–23,57–59. We report a biological and molecular role for
KDM6B in regulation of a specific gene locus (IL-11) during bac-
terial challenge. Although KDM6B was shown to primarily target
H3K27me3 and, to a lesser extent, H3K27me241–43, our results sug-
gest KDM6B is selectively demethylating H3K27me2 at the IL-11
promoter. This is consistent with previous observations of KDM6B
gene regulation independent of H3K27me320. We hypothesize
KDM6B could differentially regulate inflammatory gene expression
through selective demethylation of H3K27me2 via an unknown
regulatory element or PTM on KDM6B. Future ChIP-seq and pro-
teomic studies using biological stimuli, such as pneumococcus, may
reveal KDM6B complexes and dynamics of H3K27me3/2 in epigen-
etic control of inflammatory signalling cascades.
Our findings support the idea that pneumococcal-epithelial
interactions may actively induce URT responses that promote
confinement of bacteria, leading eventually to clearance. Host
inflammatory response is a known contributor to the commensal
and pathogenic lifestyles of pneumococcus through roles in clear-
ance, transmission and establishment of a replicative niche3,13,60.
Our results suggest KDM6B and regulation of IL-11 transcription
are key components modulating localized epithelial responses and
host-driven containment. Since KDM6B differentially regulates
multiple genes, we cannot rule out the possibility that other fac-
tors are regulating gene expression concurrently or synergistically
with IL-11. However, our IL-11 rescue experiments with TIGR4
suggest a role for IL-11 in locally maintaining host-limiting/tolero-
genic epithelial-pneumococcal responses. Interestingly, IL-11 is
known to influence mucus production, cell membrane components,
wound healing of gastric ulcers and resistance of endothelial cells
to immune mediated injury51,53,54,61,62. Promoting localized responses
Nature Microbiology | www.nature.com/naturemicrobiology
Articles
within the nasopharynx is also reflected in our microarray data, as
52 genes associated with wound healing were upregulated by 6B in
comparison to TIGR4 (Supplementary Table 1). Additionally, 39 of
52 upregulated genes were also associated with KDM6B, as deter-
mined by ChIP-seq studies from macrophages (Supplementary
Table 1). Within this principle, we propose that some pneumococcal
strains, along with some oral microflora constituents, actively influ-
ence their localization within the host through a subset of NF-κB
driven ‘wound healing’ genes as a means of countering deleterious
pro-inflammatory responses. How the localized inflammatory pro-
cesses and host-mediated confinement influence carriage duration,
clearance, transmission or the impact of microinvasion13 remains to
be defined.
Our molecular studies provide evidence that 6B stimulation
recruits/induces additional p65 interacting partners/PTM, aside
from phosphorylation of serine 536, to trigger KDM6B and IL-11
expression. These findings biologically support the ‘NF-κB barcode’
hypothesis, whereby a signature barcode of PTM could mediate
specific gene expression patterns to pneumococcus63,64. One could
hypothesize that KDM6B is a chromatin regulator that fine-tunes
inflammatory signalling in conjunction with a p65 barcode across a
unique ‘p65–KDM6B’ axis. Here, KDM6B would function to modu-
late host responses based upon the degree of inflammatory signal
input. Additional studies identifying PTM and interacting partners
of p65 and KDM6B will advance our understanding of not only
pneumococcal induced host responses, but also the regulation of
tolerogenic inflammatory responses to pathobionts.
Overall, our data demonstrate a biological role of KDM6B in
moderating the host response to bacteria. We reveal that catalyti-
cally active KDM6B is required for host-mediated tolerogenic con-
finement of a 6B strain within a murine model of infection. We
further show that exogenous addition of IL-11 in the same murine
model promotes confinement of an otherwise lethal infection with a
TIGR4 strain. Future studies characterizing the molecular interplay
of chromatin and cellular processes across the spectrum of pneu-
mococcal host response will not only identify new means to combat
pneumococcal disease, but may also reveal the processes exploited
by other pathobionts as well.
Methods
Bacterial strains, growth conditions and c.f.u. enumeration. Clinical isolates
of serotypes 6B ST90 CC156 lineage F (ST90; CNRP no. 43494), 19A (ST276;
CNRP no. 45426) and 1 (non-haemolytic; ST306; CNRP no. 43810) were obtained
from the Centre National de Référence des Pneumocoques (CNRP), serotype 19F
from the Karolinska Institutet56 (BHN100; ST162), serotype 4 TIGR4 from the
Universität Greifswald and serotype 1 (ST304 haemolytic) from the Université
de Tours. Experimental starters were prepared from frozen master stocks struck
on 5% Columbia blood agar plates (Biomerieux reference no. 43041) and grown
overnight at 37 °C with 5% CO2 prior to outgrowth in Todd–Hewitt (BD) broth
supplemented with 50 mM HEPES (Sigma) (TH+H) at 37°C with 5% CO2 in
closed falcon tubes. Midlog bacteria were pelleted and diluted to 0.6 OD600 per ml
in TH+H media supplemented with Luria–Bertani (BD) and 15% glycerol final
concentration. Aliquots were made and frozen at −80 °C for experiments. All
experiments were performed with frozen experimental starters of S. pneumoniae
less than 14 days old. For experiments, starters were grown to midlog phase in
TH+H broth at 37 °C with 5% CO2 in closed falcon tubes, pelleted at 1,500 × G
for 10 min at room temperature, washed in DPBS and concentrated in 1 ml DPBS
prior to dilution at desired c.f.u. ml−1 using 0.6 OD600 per ml conversion factors in
either cell culture media or DPBS for animal studies (conversion factors shown
in Supplementary Table 2). For bacteria killed with paraformaldehyde (PFA), the
concentrated bacteria, prior to dilution, were incubated in 4% PFA for 30 min at
room temperature, washed in DPBS and diluted to the desired c.f.u. ml−1 using 0.6
OD600 per ml conversion factors. PFA fixation not only inactivates pneumococcus
but is known to maintain bacterial morphology, including pili and extracellular
polymeric substances such as capsules65. Bacterial c.f.u. enumeration was
determined by serial dilution plating on 5% Columbia blood agar plates.
Cell culture and in vitro challenge. A549 human epithelial cells (ATCC reference
no. CCL-185) were maintained in F12K media (Gibco) supplemented with 1X
GlutaMax (Gibco) and 10% heat-inactivated fetal calf serum at 37 °C with 5%
CO2. Detroit 562 nasopharyngeal epithelial cells (ATCC reference no. CCL-138)
Articles
and BEAS-2B epithelial cells (ATCC reference no. CRL-9609) were maintained in
DMEM supplemented with 1X sodium pyruvate (Gibco) and 1X GlutaMax (Gibco)
10% heat-inactivated fetal calf serum. Stable HeLa GFP-p65 were generated
using the transposon-based sleeping beauty system and maintained in DMEM
supplemented with 1X sodium pyruvate (Gibco), 1X GlutaMax (Gibco) and 10%
heat inactivated fetal calf serum66. A549, BEAS-2B, Detroit 562 or HeLa GFP-p65
cells were used until passage 15. For in vitro challenge studies, A549, BEAS-2B,
Detroit 562 or HeLa GFP-p65 cells were plated in tissue culture treated plates at
2 × 105 cells (in 6-well plates for 72 h), 5 × 104 cells (in 24-well plates for 48 h) or
1 × 104 cells (in 96-well plates for 48 h). Cells were used at 90% confluency for all
in vitro studies. Bacterial inocula diluted in cell culture media were added to cells,
and bacterial–epithelial cell contact was synchronized by centrifugation at 200 × G
for 10 min at room temperature, then the samples were moved to 37 °C with 5%
CO2 for 2 h. For inhibitor studies, cell culture media was aspirated and replaced
with filter-sterilized culture media containing, as inhibitor, volume-matched
DMSO (Sigma), GSK-J4 (Sigma reference no. SML0701) or BAY 11-7082 (Sigma
reference no. B5556) at 10 µM final concentration for either 24 h or 3 h prior to
bacterial addition. Human IL-11 (Miltenyi Biotec reference no. 130-094-623)
and human IL-1β (Enzo Life Sciences reference no. ALX-522-056) were used at
100 ng ml−1 and 10 ng ml−1 final concentrations, respectively, in cell culture media.
Immunofluorescence and trypan blue bright-field microscopy. To quantify
nuclear KDM6B, A549 cells were seeded on acid-washed and ultraviolet-treated
coverslips in 24-well plates as described above. Two hours post-challenge, the
medium was aspirated and cells were washed in DPBS and fixed with 2.5% PFA
for 10 min at room temperature. Coverslips were blocked and permeabilized
overnight in 5% BSA 0.5% Tween20. Coverslips were incubated for 1 h at room
temperature with KDM6B antibody (1:500; Abcam reference no. ab38113) diluted
in 5% BSA 0.5% Tween20, washed three times in 0.5% Tween20 and incubated for
1 h with Alexa Fluor 488 secondary antibody. After secondary antibody incubation,
coverslips were washed three times in 0.5% Tween20 and mounted using
ProLong Gold with 4,6-diamidino-2-phenylindole (DAPI) (Invitrogen). Confocal
microscopy images were acquired on a Zeiss Axio Observer spinning disk confocal
microscope using MetaMorph acquisition software. Nuclear KDM6B intensity per
cell was quantified within a region of interest generated from the DAPI signal in
Fiji67. For trypan exclusion microscopy, A549 cells were seeded in 96-well plates as
described above. Two hours post-challenge, the culture medium was aspirated, cells
were washed in DPBS and trypan blue (Thermo) was added for 10 min at room
temperature. Next, trypan blue was removed, and cells were fixed with 2.5% PFA for
10 min at room temperature. Finally, PFA was removed and fixed cells were washed
in DPBS prior to imaging on an EVOS FL imaging system (Thermo). Trypan
positive cells were scored manually as per cent of total cells in an imaged field.
A549 epithelial microarray. A549 cells were infected as described above, and total
RNA was harvested using an RNeasy kit (Qiagen). RNA quality was confirmed
using a Bioanalyzer (Agilent). Affymetrix GeneChip human transcriptome
array 2.0 was processed as per manufacturer’s instructions. Data were analysed
using TAC 4.0 (Applied Biosystems). The dataset is deposited with ArrayExpress
accession no. E-MTAB-9603.
LDH assay. LDH assays were performed on cell culture supernatants as per
manufacturer’s instructions (Pierce LDH cytotoxicity kit reference no. 88953).
LDH absorbance was read using Cytation 5 (BioTek) at the manufacturer’s
recommended excitation and emission wavelengths.
ChIP and ChIP-qPCR. In brief, 8 × 106 A549 cells were cross-linked in tissue
culture plates with 1% formaldehyde for 10 min at room temperature, then
quenched with 130 mM glycine for 5 min at room temperature. Cells were washed
in DPBS, gently scraped, and transferred to an Eppendorf. Harvested cells were
pelleted at 200 × G, the supernatant was aspirated and the pellets were frozen at
−20°C. To obtain chromatin, cell pellets were thawed on ice and lysed for 30 min
on ice in nuclear isolation buffer (15 mM Tris (pH 7.5), 60 mM KCl, 15 mM NaCl,
250 mM sucrose, 1 mM CaCl2, 5 mM MgCl2) supplemented with 0.2% Triton
X-100 and inhibitor cocktail (1X PhosSTOP, 10 mM sodium butyrate, 0.2 mM
PMSF, 1X cOmplete protease cocktail). Nuclei were pelleted, and re-suspended
in chromatin shearing buffer (1% SDS, 10 mM Tris HCl (pH 8.0), 1 mM EDTA,
0.5 mM EGTA) supplemented with inhibitor cocktail for sonication (10 cycles of
15 s ‘on’ and 30 s ‘off ’) with a Bioruptor (Diagenode) to 200–900 bp size. Sheared
chromatin was cleared by centrifugation, then sampled for size using 2% agarose
gel electrophoresis and quantified using Pico488 (Lumiprobe reference no. 42010).
ChIP grade antibodies top65 (L8F6) (CST reference no. 6956), KDM6B (Abcam
reference no. ab38113), H3K27me3 (Abcam reference no. ab6002), H3 (Abcam
reference no. ab195277) or H3K27me2 (Diagenode reference no. C15410046-
10) were used at the manufacturer’s recommended concentrations and bound to
DiaMag beads (Diagenode reference no. C03010021-150) overnight with gentle
rotation. Quantified chromatin was diluted to 10 µg per immunoprecipitation
condition in SDS dilution buffer (0.6% Triton X-100, 0.06% NaDOC, 150 mM
NaCl, 12 mM Tris HCl (pH 8.0), 1 mM EDTA, 0.5 mM EGTA) supplemented with
inhibitor cocktail and added to antibody-bound DiaMag beads overnight with
Nature Microbiology
gentle rotation; 8% of input was reserved. Beads were washed with buffers 1–6
(buffer 1 (isotonic) components: 1% Triton X-100, 0.1% NaDOC, 150 mM NaCl,
10 mM Tris HCl (pH 8.0); buffer 2 (isotonic, ionic charge change) components:
0.5% NP40, 0.5% Triton X-100, 0.5% NaDOC, 150 mM NaCl, 10 mM Tris HCl
(pH 8.0); buffer 3 (high salt dilution) components: 0.7% Triton X-100, 0.1%
NaDOC, 250 mM NaCl, 10 mM Tris HCl (pH 8.0); buffer 4 (high salt dilution)
components: 0.5% NP40, 0.5% Triton X-100, 250 mM LiCl, 1 mM EDTA, 20 mM
Tris HCl (pH 8.0); buffer 5 (salt dilution) components: 0.1% NP40, 150 mM
NaCl, 1 mM EDTA, 20 mM Tris HCl (pH 8.0); buffer 6 (TE) components: 1 mM
EDTA, 20 mM Tris HCl (pH 8.0)), de-cross-linked by boiling for 10 min with
10% Chelex, treated with RNase and proteinase K, then purified using phenol–
chloroform extraction followed by isopropanol precipitation. Recovered DNA was
suspended in molecular grade water, and 1 µl was used for Sybr Green reactions
as per manufacturer’s instructions on a BioRad CFX384 (BioRad). ChIP-qPCR
primers (50–150 bp; 60 °C maximum melting temperature) were designed to
span the NF-κB sites of interest within the promoters of IL-11 and KDM6B and
blasted against the human reference genome for specificity. Primers were tested
for efficiency using serial dilutions of A549 genomic DNA. Per cent recovery was
calculated as 2 raised to the adjusted input CT minus IP CT multiplied by 100. For
histone marks, H3K27me3/2, the per cent recovery was normalized to the per cent
recovery of H3. ChIP-qPCR primers are listed in Supplementary Table 2.
RNA isolation and RT–qPCR. Total RNA isolated using the TRIzol (Life
Technologies reference no. 15596-026) extraction method as per manufacturer’s
recommendations. Recovered RNA was suspended in molecular grade water,
quantified spectrophotometrically using a NanoDrop and a 5 µg aliquot was
converted to cDNA using Super Script IV as per manufacturer’s instructions.
cDNA was diluted to 20 ng µl−1 in molecular grade water and 1 µl aliquots
were used for Sybr Green reactions in technical duplicate or triplicate as per
manufacturer’s instructions on a BioRad CFX384 (BioRad). Relative expression
was calculated by the ΔΔCT method to GapDH68. RT–qPCR primers are listed in
Supplementary Table 2.
Immunoblots and quantification. Cell culture medium was removed, cells
were washed in DPBS and whole cell lysates were harvested with Laemmli
buffer69. Lysates were boiled 10 min and frozen at −20 °C. Whole cell lysates were
run on 8% polyacrylamide SDS PAGE gels, transferred to PVDF membranes
(BioRad TransBlot), blocked for 1 h in 5% BSA TBST, then probed for p65 (CST
reference no. 6956), p65 phosphorylation at serine 536 (CST reference no. 3033)
or actin AC-15 monoclonal (Sigma reference no. A5441) as per manufacturer’s
recommendations. Appropriate secondary-HRP conjugated antibodies were used
with clarity ECL (BioRad) developing reagents. Cell fraction samples (18 µl input,
18 µl cytoplasmic and 8 µl nuclear fraction) were run on 15% polyacrylamide SDS
PAGE gels, transferred to PVDF membrane (BioRad TransBlot), blocked for 1 h in
5% BSA TBST or 5% milk TBST, then probed for H3K27me2 (Abcam reference no.
ab24684), H3K27me3 (Abcam reference no. ab6002) or GapDH (Abcam reference
no. ab9485) as per manufacturer’s recommendations. Appropriate secondary-HRP
conjugated antibodies were used with clarity ECL (BioRad) developing reagents.
After probing cell fraction immunoblots the PVDF membranes were stained with
Coomassie Brilliant Blue R-250 (Sigma reference no. 1125530025) for 30 min at
room temperature prior to destaining for 30 min (5% v/v ethanol, 7.5% acetic acid).
Membranes were allowed to dry completely and imaged on ChemiDoc Touch
(BioRad). Band and lane intensity were quantified using Image Lab (BioRad).
In vivo animal studies. All protocols for animal experiments were reviewed and
approved by the CETEA (Comité d’Ethique pour l’Expérimentation Animale) of
the Institut Pasteur under approval number Dap170005 and were performed in
accordance with national laws and institutional guidelines for animal care and
use. Wild-type C57BL/6 female 8–9 week old mice were purchased from Janvier
Labs. Wild-type C57BL/6 male 10–12 week old mice were used for survival studies
(Institut Pasteur). Animals were anaesthetized with a ketamine and xylazine
cocktail prior to intranasal challenge with 20 µl of 3 × 106 to 4 × 106 c.f.u. of TIGR4
or 6B. Bacterial inocula were made as described above with minor modifications.
In brief, after bacteria were concentrated to 0.6 OD600, they were diluted in
either filter-sterilized DMSO/DPBS or 5 mM GSK-J4/DPBS. For IL-11 studies,
recombinant mouse IL-11 (Enzo; ENZ-PRT243-0010) was added to bacterial
inoculate to a final concentration of 7,500 ng ml−1 (0.15 μg per mouse). Either 24 h
or 48 h post-inoculation, animals were euthanized by CO2 affixation. The nasal
lavage was obtained by blocking the oropharynx to avoid leakage into the oral
cavity and lower airway, and nares were flushed with 500 µl DPBS. BALF, lungs and
spleens were collected and placed in 1 ml DPBS supplemented with 2X protease
inhibitor cocktail (Sigma reference no. P1860). c.f.u. were enumerated as described
above on 5 µg ml−1 gentamicin Columbia blood agar selection plates. For survival
studies, animals were monitored daily for weight loss and disease signs. Animals
were euthanized by CO2 asphyxiation at ≥20% loss of initial weight or at body
conditioning score (BCS) ≤ 2 (ref. 70).
Human IL-11 ELISA. Detroit 562 cells were infected with S. pneumoniae for 6 h,
and supernatants were collected for IL-11 analysis using an IL-11 Human Elisa
Nature Microbiology | www.nature.com/naturemicrobiology
Nature Microbiology
Duo Set (R&D Systems, DY218 lot number 5225707), according to manufacturers’
instructions.
AlamarBlue cytotoxicity. A549 cell viability was determined using alamarBlue
(Thermo reference no. DAL1025) as previously described71. AlamarBlue
absorbance was read using a Cytation 5 (BioTek) at manufacturer’s recommended
excitation and emission wavelengths.
Crude haemolytic assay. Red blood cells (RBC) from horse blood (Thermo
reference no. SR0050C) were washed three times in DPBS and suspended at a
final concentration of 1% v/v, with 100 µl moved to individual wells of a 96-well
U-bottom plate and stored at 4 °C until use. Cell culture supernatants were
collected from infected A549 cells 2 h post-infection, centrifuged and filtered prior
to 100 µl being combined with RBCs. For positive lysis control, volume matched
water and 2% triton-X100 were used; the negative control was RBCs with cell
culture media alone. The plate was incubated for 30 min at 37 °C with 5% CO2,
then centrifuged for 5 min at 200 × g. Supernatants (100 µl) were transferred to a
new clear 96-well plate and absorbance (OD540) was read using Cytation 5 (BioTek).
Per cent lysis was calculated as absorbance of sample adjusted for media alone over
positive lysis sample adjusted for media alone multiplied by 100.
Oral commensal in vitro infection and RT–qPCR. Streptococcus gordonii,
Streptococcus oralis and Streptococcus sanguinis were all cultivated in brain
heart infusion (BHI) broth, Fusobacterium nucleatum was grown in BHI broth
supplemented with 5 μg ml−1 hemin and 1 μg ml−1 menadione, and Eikenella
corrodens was grown in TYGVS broth. All bacteria were cultured anaerobically
until midlog phase. Human telomerase immortalized gingival keratinocytes
(TIGK) were maintained at 37 °C with 5% CO2 in Dermalife-K serum-free
culture medium (Lifeline Cell Technology)72. Epithelial cells at 80% confluency
were challenged with the different bacteria species at an MOI of 100 for 3 h, were
washed and fresh media was added for another 18 h. Total RNA was isolated using
the Qiagen RNeasy mini kit (Qiagen). RNA concentrations were determined
spectrophotometrically using a NanoDrop 2000 and quality was determined
using the Qubit 4 fluorimeter. cDNA from total RNA was synthesized (500 ng
RNA per reaction volume) using a High Capacity cDNA reverse transcription
kit (Applied Biosystems). RT-qPCR) was performed using IL-11 and GAPDH
primers in Supplementary Table 2. RT–qPCR was performed on an Applied
Biosystems QuantStudio 3 cycler and the auto-calculated threshold cycle selected.
The cycle threshold (Ct) values were determined, and mRNA expression levels
were normalized to GAPDH and expressed relative to controls following the 2−ΔΔCT
method. RT–qPCR primers listed in Supplementary Table 2.
Cell fractionation. A549 cells were fractionated using a modified REAP method73.
In brief, cells were directly lysed in the well using 500 µl nuclear isolation buffer
(NIB) supplemented with inhibitors and 0.2% Triton X 100 on ice for 30 min.
Lysates were collected to a fresh Eppendorf tube via gentle scraping; 25% (v/v) was
retained as input and 5X Laemmli buffer added. Lysates were centrifuged for 1 min
at 2,000 × G at 4 °C, and ~500 µl supernatant (cytoplasmic fraction) was collected
into a new tube and 5X Laemmli buffer was added. The nuclear pellet was washed
three times in NIB with inhibitors by gentle resuspension and centrifugation
for 1 min at 10,000 × G at 4 °C. The nuclear pellet was suspended in 85 µl NIB
supplemented with inhibitors and 0.2% Triton X 100, vortexed and 5X Laemmli
buffer was added. Samples were boiled 10 min and frozen at −20 °C.
Pneumococcus growth curve with recombinant murine IL-11. Midlog bacterial
inoculate for both TIGR4 and TIGR4 supplemented with IL-11, as described in
in vivo animal studies, were used for growth analysis with the final dilution being
in TH+H. Bacterial aliquots (100 µl each) were distributed into wells of a clear
96-well plate. Plates were incubated at 37 °C with 5% CO2 for 2 h, and absorbance
(OD600) was taken at 20 minute intervals using a Cytation5 (BioTek). Absorbance
was calculated after normalizing to TH+H growth media alone.
Statistical analysis. All experiments, unless otherwise noted, were biologically
repeated 3–5 times with a minimum of 2–3 technical replicates per biological
and the statistical test is reported in the figure legend. Data normality was tested
by Shapiro-Wilk test, and appropriate parametric or non-parametric tests were
performed. P values were calculated using GraphPad Prism software and the exact
values are in source data. For RT–qPCR all statistics were calculated on either
the ΔΔCT or ΔCT depending on the comparison, with two or three technical
replicates per biological for reproducibility. PCA plots using the ‘prcomp’ function
of the base stats package in R on scaled and mean-centred log2-transformed data,
or ΔCT for cross-treatment comparisons. Microscopy data was collected from
analysis of 30–100 cells for nuclear staining, or 200–300 cells for bright-field per
biological replicate per group. ChIP-qPCR studies used four technical replicates
per biological for reproducibility and all data graphed. Animal studies used the
minimum number of animals required for power calculated by post-hoc analysis of
c.f.u. using G*Power software.
Reporting Summary. Further information on research design is available in the
Nature Research Reporting Summary linked to this article.
Nature Microbiology | www.nature.com/naturemicrobiology
Articles
Data availability
All data in the present study is available upon request from the corresponding
author. The complete microarray dataset (ArrayExpress accession number
E-MTAB-9603), PCA dataset and RT–qPCR ΔCT values are contained within
Supplementary Table 1 and respective source data files. Full immunoblots are in
Supplementary Figs. 1–3. Source data are provided with this paper.
Received: 18 September 2019; Accepted: 28 September 2020;
Published: xx xx xxxx
References
1. Donkor, E. S. Understanding the pneumococcus: transmission and evolution.
Front. Cell. Infect. Microbiol. https://doi.org/10.3389/fcimb.2013.00007 (2013).
2. Shak, J. R., Vidal, J. E. & Klugman, K. P. Influence of bacterial interactions on
pneumococcal colonization of the nasopharynx. Trends Microbiol. 21,
129–135 (2013).
3. Weiser, J. N., Ferreira, D. M. & Paton, J. C. Streptococcus pneumoniae:
transmission, colonization and invasion. Nat. Rev. Microbiol. 16,
355–367 (2018).
4. World Health Organization. Pneumococcal vaccines: WHO position
paper—2012. Wkly Epidemiol. Rec. 87, 129–144 (2012).
5. Bogaert, D., De Groot, R. & Hermans, P. W. Streptococcus pneumoniae
colonisation: the key to pneumococcal disease. Lancet Infect. Dis. 4,
144–154 (2004).
6. Henriques-Normark, B., Blomberg, C., Dagerhamn, J., Battig, P. & Normark,
S. The rise and fall of bacterial clones: Streptococcus pneumoniae. Nat. Rev.
Microbiol. 6, 827–837 (2008).
7. Antibiotic Resistance Threats in the United States, 2013 (Centers for Disease
Control and Prevention, 2013).
8. Ferreira, D. M. et al. Controlled human infection and rechallenge with
Streptococcus pneumoniae reveals the protective efficacy of carriage in healthy
adults. Am. J. Respiratory Crit. Care Med. 187, 855–864 (2013).
9. Weiser, J. N. The pneumococcus: why a commensal misbehaves. J. Mol. Med.
88, 97–102 (2010).
10. Henriques-Normark, B. & Tuomanen, E. I. The pneumococcus: epidemiology,
microbiology, and pathogenesis. Cold Spring Harb. Perspect. Med. 3,
a010215 (2013).
11. Jochems, S. P., Weiser, J. N., Malley, R. & Ferreira, D. M. The immunological
mechanisms that control pneumococcal carriage. PLoS Pathog. 13,
e1006665 (2017).
12. Robson, R. L., Reed, N. A. & Horvat, R. T. Differential activation of
inflammatory pathways in A549 type II pneumocytes by Streptococcus
pneumoniae strains with different adherence properties. BMC Infect. Dis. 6,
71 (2006).
13. Weight, C. M. et al. Microinvasion by Streptococcus pneumoniae induces
epithelial innate immunity during colonisation at the human mucosal surface.
Nat. Commun. 10, 3060 (2019).
14. Quinton, L. J. & Mizgerd, J. P. Dynamics of lung defense in pneumonia:
resistance, resilience, and remodeling. Annu. Rev. Physiol. 77, 407–430 (2015).
15. Ghosh, S. & Hayden, M. S. New regulators of NF-κB in inflammation. Nat.
Rev. Immunol. 8, 837–848 (2008).
16. Bhatt, D. & Ghosh, S. Regulation of the NF-κB-mediated transcription of
inflammatory genes. Front. Immunol. https://doi.org/10.3389/fimmu.
2014.00071 (2014).
17. Dong, J., Jimi, E., Zhong, H., Hayden, M. S. & Ghosh, S. Repression of gene
expression by unphosphorylated NF-κB p65 through epigenetic mechanisms.
Genes Dev. 22, 1159–1173 (2008).
18. Fujioka, S. et al. NF-κB and AP-1 connection: mechanism of
NF-κB-dependent regulation of AP-1 activity. Mol. Cell. Biol. 24,
7806–7819 (2004).
19. Das, N. D., Jung, K. H. & Chai, Y. G. The role of NF-κB and H3K27me3
demethylase, JMJD3, on the anthrax lethal toxin tolerance of RAW 264.7
cells. PLoS ONE 5, e9913 (2010).
20. De Santa, F. et al. JMJD3 contributes to the control of gene expression in
LPS-activated macrophages. EMBO J. 28, 3341–3352 (2009).
21. De Santa, F. et al. The histone H3 lysine-27 demethylase JMJD3 links
inflammation to inhibition of polycomb-mediated gene silencing. Cell 130,
1083–1094 (2007).
22. Na, J. et al. Histone H3K27 demethylase JMJD3 in cooperation with
NF-κB regulates keratinocyte wound healing. J. Invest. Dermatol. 136,
847–858 (2016).
23. Salminen, A., Kaarniranta, K., Hiltunen, M. & Kauppinen, A. Histone
demethylase Jumonji D3 (JMJD3/KDM6B) at the nexus of epigenetic
regulation of inflammation and the aging process. J. Mol. Med. 92,
1035–1043 (2014).
24. Diermeier, S. et al. TNFα signalling primes chromatin for NF-κB binding and
induces rapid and widespread nucleosome repositioning. Genome Biol. 15,
536 (2014).
Articles
25. Lone, I. N. et al. Binding of NF-κB to nucleosomes: effect of translational
positioning, nucleosome remodeling and linker histone H1. PLoS Genet. 9,
e1003830 (2013).
26. Wong, D. et al. Extensive characterization of NF-κB binding uncovers
non-canonical motifs and advances the interpretation of genetic functional
traits. Genome Biol. 12, R70 (2011).
27. Christian, F., Smith, E. L. & Carmody, R. J. The regulation of NF-κB subunits
by phosphorylation. Cells 5, 12 (2016).
28. Oeckinghaus, A. & Ghosh, S. The NF-κB family of transcription factors and
its regulation. Cold Spring Harb. Perspect. Biol. 1, a000034 (2009).
29. Avvakumov, N., Nourani, A. & Côté, J. Histone chaperones: modulators of
chromatin marks. Mol. Cell 41, 502–514 (2011).
30. Bannister, A. J. & Kouzarides, T. Regulation of chromatin by histone
modifications. Cell Res. 21, 381–395 (2011).
31. Black, J. C. & Whetstine, J. R. Chromatin landscape: methylation beyond
transcription. Epigenetics 6, 13–19 (2011).
32. Hammond, C. M., Stromme, C. B., Huang, H., Patel, D. J. & Groth, A.
Histone chaperone networks shaping chromatin function. Nat. Rev. Mol. Cell
Biol. 18, 141–158 (2017).
33. Kouzarides, T. Chromatin modifications and their function. Cell 128,
693–705 (2007).
34. Lorch, Y. & Kornberg, R. D. Chromatin-remodeling and the initiation of
transcription. Q. Rev. Biophys. 48, 465–470 (2015).
35. Markolovic, S. et al. Structure–function relationships of human JMJC
oxygenases—demethylases versus hydroxylases. Curr. Opin. Struct. Biol. 41,
62–72 (2016).
36. Satoh, T. et al. The JMJD3-Irf4 axis regulates M2 macrophage polarization
and host responses against helminth infection. Nat. Immunol. 11,
936–944 (2010).
37. Pierce, J. W. et al. Novel inhibitors of cytokine-induced IκBα phosphorylation
and endothelial cell adhesion molecule expression show anti-inflammatory
effects in vivo. J. Biol. Chem. 272, 21096–21103 (1997).
38. Karin, M. The beginning of the end: IκB kinase (IKK) and NF-κB activation.
J. Biol. Chem. 274, 27339–27342 (1999).
39. Gilmore, T. D. The Rel/NF-kappaB signal transduction pathway: introduction.
Oncogene 18, 6842–6844 (1999).
40. Kruidenier, L. et al. A selective jumonji H3K27 demethylase inhibitor
modulates the proinflammatory macrophage response. Nature 488,
404–408 (2012).
41. Xiang, Y. et al. JMJD3 is a histone H3K27 demethylase. Cell Res. 17,
850–857 (2007).
42. Agger, K. et al. UTX and JMJD3 are histone H3K27 demethylases involved in
HOX gene regulation and development. Nature 449, 731–734 (2007).
43. Hong, S. et al. Identification of JMJC domain-containing UTX and JMJD3 as
histone H3 lysine 27 demethylases. Proc. Natl Acad. Sci. USA 104,
18439–18444 (2007).
44. Millet, P., McCall, C. & Yoza, B. RelB: an outlier in leukocyte biology.
J. Leukoc. Biol. 94, 941–951 (2013).
45. Kadioglu, A., Weiser, J. N., Paton, J. C. & Andrew, P. W. The role of
streptococcus pneumoniae virulence factors in host respiratory colonization
and disease. Nat. Rev. Microbiol. 6, 288–301 (2008).
46. Walev, I. et al. Resealing of large transmembrane pores produced by
streptolysin O in nucleated cells is accompanied by NF-κB activation and
downstream events. FASEB J. 16, 237–239 (2002).
47. Tran, S.-L., Puhar, A., Ngo-Camus, M. & Ramarao, N. Trypan blue dye enters
viable cells incubated with the pore-forming toxin HlyII of Bacillus cereus.
PLoS ONE 6, e22876 (2011).
48. Elias, J. A. et al. Epithelial interleukin-11. Regulation by cytokines,
respiratory syncytial virus, and retinoic acid. J. Biol. Chem. 269,
22261–22268 (1994).
49. Du, X. & Williams, D. A. Interleukin-11: review of molecular, cell biology,
and clinical use. Blood 89, 3897–3908 (1997).
50. Traber, K. E. et al. Roles of interleukin-11 during acute bacterial pneumonia.
PLoS ONE 14, e0221029 (2019).
51. Yap, J., Foo, C. F. H., Lee, M. Y., Stanton, P. G. & Dimitriadis, E. Proteomic
analysis identifies interleukin 11 regulated plasma membrane proteins in
human endometrial epithelial cells in vitro. Reprod. Biol. Endocrinol. 9,
73 (2011).
52. Han, Y. et al. Repairing effects of interleukin 11 (IL-11) towards high
dose methotrexate-induced rat small intestinal mucositis and its
impacts on T-lymphoblastic leukemia cell line. Iran. J. Basic Med. Sci. 19,
837–843 (2016).
53. Lee, C. G. et al. Endogenous IL-11 signaling is essential in Th2- and
IL-13-induced inflammation and mucus production. Am. J. Respir. Cell Mol.
Biol. 39, 739–746 (2008).
54. Wen, C.-Y. et al. IL-11 up-regulates Tie-2 expression during the healing of
gastric ulcers in rats. World J. Gastroenterol. 9, 788–790 (2003).
55. Orazi, A., Du, X., Yang, Z., Kashai, M. & Williams, D. A. Interleukin-11
prevents apoptosis and accelerates recovery of small intestinal mucosa in
Nature Microbiology
mice treated with combined chemotherapy and radiation. Lab. Invest. 75,
33–42 (1996).
56. Sandgren, A. et al. Virulence in mice of pneumococcal clonal types with
known invasive disease potential in humans. J. Infect. Dis. 192, 791–800 (2005).
57. Ohguchi, H. et al. KDM6B modulates MAPK pathway mediating multiple
myeloma cell growth and survival. Leukemia 31, 2661–2669 (2017).
58. Chen, S. et al. The histone H3 Lys 27 demethylase JMJD3 regulates gene
expression by impacting transcriptional elongation. Genes Dev. 26,
1364–1375 (2012).
59. Young, M. D. et al. ChIP-seq analysis reveals distinct H3K27me3
profiles that correlate with transcriptional activity. Nucleic Acids Res. 39,
7415–7427 (2011).
60. Zafar, M. A., Wang, Y., Hamaguchi, S. & Weiser, J. N. Host-to-host
transmission of streptococcus pneumoniae is driven by its inflammatory
toxin, pneumolysin. Cell Host Microbe 21, 73–83 (2017).
61. Nishina, T. et al. Interleukin-11 links oxidative stress and compensatory
proliferation. Sci. Signal 5, ra5 (2012).
62. Mahboubi, K., Biedermann, B. C., Carroll, J. M. & Pober, J. S. IL-11 activates
human endothelial cells to resist immune-mediated injury. J. Immunol. 164,
3837–3846 (2000).
63. Moreno, R., Sobotzik, J.-M., Schultz, C. & Schmitz, M. L. Specification of the
NF-κB transcriptional response by p65 phosphorylation and TNF-induced
nuclear translocation of IKKε. Nucleic Acids Res. 38, 6029–6044 (2010).
64. Collins, P. E., Mitxitorena, I. & Carmody, R. J. The ubiquitination of NF-κB
subunits in the control of transcription. Cells 5, 23 (2016).
65. Chao, Y. & Zhang, T. Optimization of fixation methods for observation of
bacterial cell morphology and surface ultrastructures by atomic force
microscopy. Appl. Microbiol. Biotechnol. 92, 381–392 (2011).
66. Kowarz, E., Loscher, D. & Marschalek, R. Optimized sleeping beauty
transposons rapidly generate stable transgenic cell lines. Biotechnol. J. 10,
647–653 (2015).
67. Schindelin, J. et al. Fiji: an open-source platform for biological-image analysis.
Nat. Methods 9, 676–682 (2012).
68. Livak, K. J. & Schmittgen, T. D. Analysis of relative gene expression data
using real-time quantitative PCR and the 2 −ΔΔCT method. Methods 25,
402–408 (2001).
69. Laemmli, U. K. Cleavage of structural proteins during the assembly of the
head of bacteriophage T4. Nature 227, 680–685 (1970).
70. Ullman-Cullere, M. H. & Foltz, C. J. Body condition scoring: a rapid and
accurate method for assessing health status in mice. Lab. Anim. Sci. 49,
319–323 (1999).
71. Connor, M. G. et al. Yersinia pestis targets the host endosome recycling
pathway during the biogenesis of the Yersinia-containing vacuole to avoid
killing by macrophages. mBio 9, e01800-17 (2018).
72. Moffatt-Jauregui, C. E. et al. Establishment and characterization of a
telomerase immortalized human gingival epithelial cell line. J. Periodontal
Res. 48, 713–721 (2013).
73. Nabbi, A. & Riabowol, K. Rapid isolation of nuclei from cells in vitro.
Cold Spring Harb. Protoc. 2015, 769–772 (2015).
74. Wingender, E. Compilation of transcription regulating proteins. Nucleic Acids
Res. 16, 1879–1902 (1988).
75. Metsalu, T. & Vilo, J. ClustVis: a web tool for visualizing clustering of
multivariate data using principal component analysis and heatmap. Nucleic
Acids Res. 43, W566–W570 (2015).
Acknowledgements
We thank E. Varon, B. H. Normark, M. M. Si-Tahar and T. Kohler for their generous gifts
of S. pneumoniae strains. We are thankful to G. Dore (Institut Pasteur) for processing
the microarray Affymetrix GeneChips. Biostatistics and R discussions with J. Nabuco
(Institut Pasteur) and J. Camp (University of Veterinary Medicine Vienna) were greatly
appreciated. M.G.C. is supported by the Pasteur Foundation Fellowship. T.M.N.C. is
supported by Foundation pour la Recherche Médicale, grant no. PMJ201810007628
(Prix MARIANE JOSSO), and a donation from Credit Agricole. M.A.H. is supported by
the Institut Pasteur, and the Agence National de la Recherche (ANREpiBActIn), and is a
member of the IBEID LabEx. C.M.W. was supported by the Medical Research Council
(MR/T016329/1). We thank R. S. Heyderman (UCL) for in-depth discussion on the
manuscript, and he is supported by the MRC (MR/T016329/1). J.E. and L.B., members
of the Dynamics of Host–Pathogen interactions Unit (Institut Pasteur), are supported by
the European Commission (ERC-CoG-Endosubvert) and the ANR-HBPsensing, and are
members of the IBEID and Milieu Interieur LabExes. R.J.L. is funded by NIH/NIDCR
nos. DE01111, DE012505, DE017921 and DE023193.
Author contributions
M.G.C. and M.A.H. conceived and designed all experiments. M.G.C. performed
experiments. E.P. performed KDM6B microscopy and the HeLa GFP-p65 western blot.
T.C. and O.R. performed the animal studies. C.M.W. performed the IL-11 ELISA. D.P.M.
and R.J.L. performed all oral commensal infections and RT–qPCR. M.G.C. analysed
data. L.B. generated the stable HeLa GFP-p65 cell line under the supervision of J.E.
Nature Microbiology | www.nature.com/naturemicrobiology
Nature Microbiology Articles
M.G.C. wrote the original manuscript draft. M.G.C. and M.A.H. edited and reviewed the Supplementary information is available for this paper at https://doi.org/10.1038/
manuscript. M.A.H. supervised the research. All authors approved the final manuscript. s41564-020-00805-8.
Correspondence and requests for materials should be addressed to M.A.H.
Competing interests
The authors declare no competing interests. Reprints and permissions information is available at www.nature.com/reprints.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in
Additional information published maps and institutional affiliations.
Extended data is available for this paper at https://doi.org/10.1038/s41564-020-00805-8. © The Author(s), under exclusive licence to Springer Nature Limited 2020

Nature Microbiology | www.nature.com/naturemicrobiology

Articles Nature Microbiology

Extended Data Fig. 1 | See next page for caption.

Nature Microbiology | www.nature.com/naturemicrobiology

Nature Microbiology Articles
Extended Data Fig. 1 | Human microarray, KDM6B microscopy and demethylase RT-qPCR. a, TIGR4 and 6B inocula for animal intranasal challenge
model (n = 4 biological replicates; Geometric mean of 2 technical counts). Graphed as mean ± s.d. Analyzed by One way-ANOVA with Tukey’s multiple
comparison post-hoc test, ***= pV≤0.001, ns=not significant. Exact pV are reported in in vivo source data. b, Human microarray of epithelial A549 cells
2 hrs post-challenge (MOI 20) with either 6B (blue) or TIGR4 (gray). All genes differentially regulated by ± 1.5 fold-change to uninfected condition.
Circled 6B genes are NF-κB associated. c, Comparison of genes identified by microarray containing NF-κB sites between 6B and TIGR4. d, Clustering heat
map, using ClustVis75, (Euclidean for genes, Correlation for cell line) of the median log2 relative expression RT-qPCR data for indicated genes from A549
(n = 5 biological replicates; 2–3 technicals per biological replicate), Detroit 562 (n = 1 biological replicate with 2 technicals per replicate) and Beas2b (n = 1
biological replicate with 2 technicals per replicate) cell lines. KDM6B and IL-11 cluster highlighted in red. e, Representative images of KDM6B (green)
merged with nucleus stained with DAPI (false colored red for visual display). Scale = 10 µm. f, Quantified IL-11 ELISA (n = 5 biological replicates; 4–6
technicals per biological) 6 hrs post-infection of Detroit 562 cells. Median centered (solid red line) dot plot of all values. Dashed red line is the assay limit
of detection. g, Total RNA 2 hrs post-challenge with 6B or TIGR4 from A549 cells. Histone demethylase panel RT-qPCR shown as ΔCt (n = 4 biological
replicates; 2–3 technicals per biological replicate per group). Graphed as mean ± s.d. Two-tailed multiple T-test using Holm-Sidak correction.
***= pV≤0.001, ns=not significant. Exact pV are reported in Extended Data Fig. 1 source data.

Nature Microbiology | www.nature.com/naturemicrobiology

Articles Nature Microbiology

Extended Data Fig. 2 | NF-κB transcript profiles and inhibitor treatment cell viability. a, RT-qPCR of NF-κB transcript profiles for IL-1β, 6B, or TIGR4
challenged A549 cells over a 2 hr time course (n = 3 biological replicates; 2 technicals per replicate). Displayed as a truncated violin plot with each
data point. Graphed as the relative expression of each indicated transcript to matched uninfected/unstimulated control per time point. Gray line, when
applicable, denotes uninfected/unstimulated level. b, A549 cells untreated or treated with 10 µM BAY 11-7082, 10 µM GSK-J4, or DMSO vehicle control
(n = 4 biological replicates; 2-3 technicals per replicate). Cell viability determined by AlamarBlue. Expressed as % of untreated cells and graphed as mean
± s.d. One way-ANOVA with Dunnett’s multiple comparison post-hoc test to untreated, no significant difference observed (ns). Exact pV are reported in
Fig. 2 and Extended Data Fig. 2 source data. c, Transcript levels for IL-11, KDM6B and PTGS2 determined by RT-qPCR. Displayed as ΔCt for comparison to
untreated (n = 4 biological replicates; average of 2 ΔCt technical values per replicate). Graphed as mean ± s.d. All data analyzed by One way-ANOVA with
Tukey’s multiple comparison post-hoc test, no significant difference observed (ns). Exact pV are reported in Extended Data Fig. 2 source data.

Nature Microbiology | www.nature.com/naturemicrobiology

Nature Microbiology Articles

Extended Data Fig. 3 | See next page for caption.

Nature Microbiology | www.nature.com/naturemicrobiology

Articles Nature Microbiology
Extended Data Fig. 3 | ChIP-PCR KDM6B locus and IL-11 locus for TIGR4 and IL-1β. Chromatin obtained from untreated and 10 µM GSK-J4 treated
A549 cells 2 hrs post-challenge with 6B (MOI 20), TIGR4 (MOI 20) or IL-1β. 10 µg chromatin input used for ChIP of p65, KDM6B, H3K27me3 and
histone H3 (H3), followed by ChIP-qPCR at primer locations (P6, P3 & P2) spanning the NF-κB sites upstream of the transcriptional start site (TSS).
a, Schematic of KDM6B promoter with ChIP-qPCR primer locations (P4 & P3) and the NF-κB sites. b, % recovery of input for p65 2 hrs post-infection
with 6B (n = 3 untreated; n = 3 GSK-J4 treated). Two-tailed unpaired Student’s T-Test comparisons for untreated to GSK-J4 treated or 6B infected to
uninfected, *= pV≤0.05, ***= pV≤0.001, ns=not significant. c, Schematic of IL-11 promoter with ChIP-qPCR primer locations (P6, P3 & P2) and the
NF-κB sites. d–g, % recovery of input for p65, KDM6B, H3K27me3 normalized to H3, or H3 bound at P6, P3 & P2 in untreated and GSK-J4 treated samples
(n = 3 untreated; n = 3 GSK-J4 treated). Graphed as Tukey box and whisker plot. Box boundaries are the upper and lower interquartile range (IQR),
“whiskers” are ± 1.5 times IQR and the median (solid line). Dots represent outliers. One way-ANOVA with Tukey’s multiple comparison post-hoc test,
no significant difference observed in comparison to uninfected/untreated or 6B samples from Fig. 3. Cell fractions from A549 cells 2 hrs post-challenge
±10 µM GSK-J4. h, Representative example of fraction purity. Full blots provided in Supplementary Information 2. i, Representative images of nuclear
fractions and coomassie stained PVDF membranes. Full blots provided in Supplementary Information 2. j, Signal intensity for H3K27me2 or H3K27me3
for IL-1β, TIGR4 or 6B normalized to total protein as a ratio to uninfected/unstimulated control (n = 4; biological replicates). Bar graph with s.d. and each
individual value displayed. Significance calculated between ±GSK-J4 groups by two-tailed Student’s T-test, p-values displayed. *= pV≤0.05.

Nature Microbiology | www.nature.com/naturemicrobiology

Nature Microbiology Articles

Extended Data Fig. 4 | IL-1β RT-qPCR with GSK-J4 or IL-11 treatments. a, Heat map represents fold change to either untreated or treatment matched
controls per condition (n = 5 biologicals for untreated, and n = 3 biologicals per treatment; average 2 technicals per biological per group). Heat mapped
difference between relative expression of treatments to untreated. Two-tailed uncorrected multiple T-Test of treatment groups to untreated with significant
genes highlighted with “*”. IL-1β vs. IL-1β + GSK-J4: ICAM1 pV= 0.013, RELB pV= 0.150, TNF pV= 0.032, IL-6 pV=0.035, CXCL2 pV= 0.047. IL-1β vs.
IL-1β + IL-11: CXCL2 pV=0.007, TNF pV=0.030. b & c, PCA variable correlation plots for GSJ-J4 PCA and IL-11 PCA. KDM6B and IL-11 highlighted in red.
Exact pV are reported in Extended Data Fig. 4 source data.

Nature Microbiology | www.nature.com/naturemicrobiology

Articles Nature Microbiology

Extended Data Fig. 5 | LDH cytotoxicity, pneumolysin activity and IL-11 RT-qPCR with oral commensals. a, % Cytotoxicity (LDH release) from A549
supernatants 2 hrs post-challenge with TIGR4 (MOI 20), 6B (MOI 20) or uninfected ± indicated treatments (n = 4 biological replicates; 2-3 technicals
per replicate). UT = untreated. Tukey box and whisker plot with defined box boundaries being the upper and lower interquartile range (IQR), “whiskers”
(fences) being ± 1.5 times IQR and the median depicted in the middle (solid line). All data was analyzed by One way-ANOVA with Tukey’s multiple
comparison post-hoc test, **= pV≤0.01, ***= pV≤0.001, ns=not significant. Exact pV are reported in Extended Data Fig. 5 source data. b, Hemolytic
activity of A549 cell culture supernatants ± 10 µM GSK-J4 2 hrs post-infection with either 6B (MOI 20) or TIGR4 (MOI 20) with uninfected control. All
samples n = 4 biologicals with 2-3 technicals per biological. Graphed as mean ± s.d. Analyzed by One way-ANOVA with Tukey’s multiple comparison
post-hoc test, ns=not significant. Exact pV are reported in Extended Data Fig. 5 source data. c, A549 cell protein lysates 2 hrs post-infection with
TIGR4 (MOI 20; n = 7 biological replicates) or 6B (MOI 20; n = 7 biological replicates) probed for pneumolysin. Relative band intensity quantified with
representative blot images. ΔPlyTIGR4 is a control for immunoblot. Scatter plot of all replicates with mean denoted in red. Data analyzed by two-tailed
unpaired Student’s T-Test, ns=not significant. Full blots provided in Supplementary Information 3. d, Total RNA harvested from immortalized gingival
keratinocytes 21 hrs post-challenge with Streptococcus gordonii, Streptococcus sanguinis, Streptococcus oralis, Eikenella corrodens and Fusobacterium nucleatum,
at an MOI of 100. Transcript levels for IL-11 determined by RT-qPCR and represented as relative expression to uninfected (n = 1; average of two technicals).

Nature Microbiology | www.nature.com/naturemicrobiology

Nature Microbiology Articles

Extended Data Fig. 6 | TIGR4 and 6B intranasal animal challenge. a, 6B and TIGR4 ± GSK-J4 with matched DMSO controls. 24 hrs post-inoculation
bacterial burden from the nasal lavage (NL), bronchoalveolar lavage fluid (BALF), lungs, and spleen of infected animals by conventional CFU counts on
5 µg/mL Gentamicin Columbia Blood agar selection plates (n = 9 animals; Geometric mean of 2 technical counts per lavage or organ). DMSO TIGR4
white, GSK-J4 TIGR4 gray, DMSO 6B light blue, GSK-J4 6B dark blue. All graphed as Tukey box and whisker plot with defined box boundaries being the
upper and lower interquartile range (IQR), “whiskers” (fences) being ± 1.5 times IQR and the median depicted in the middle (solid line). Dots represent
outliers. One way-ANOVA non-parametric Kruskal-Wallis with Dunn’s multiple comparison post-hoc test, *= pV≤0.05, **= pV≤0.01, ***= pV≤0.001,
ns=not significant. Exact pV are reported in Extended Data Fig. 6 source data. CFU = colony forming unit. Dotted lines =Limit of detection (LD). LD for
each organ: NL (50 CFU); BALF, Lung and Spleen (1000 CFU). b, TIGR4 and 6B inocula for animal intranasal challenge model in DMSO (vehicle control) or
5 mM GSK-J4 (n = 3 biological replicates; Geometric mean of 2 technical counts). Conventional CFU enumeration shows no significant difference. Graphed
as mean ± s.d. Analyzed by One way-ANOVA with Tukey’s multiple comparison post-hoc test, ns=not significant. c, TIGR4 ± IL-11 inocula for animal
intranasal challenge model (n = 3 biological replicates; 2 technical counts). Conventional CFU enumeration shows no significant difference. Graphed as
mean ± s.d. Analyzed by two-tailed Student’s T-Test, ns=not significant. d, Absorbance (OD600) growth curve for mid-log TIGR4 ± IL-11 mouse inocula over
2 hrs (n = 4 biological replicates). No significant difference in growth observed. Mean (dot) ± s.d. Analyzed by two-tailed multiple T-Test with Holm-Sidak
correction at each 20 min interval, ns=not significant. Exact F are reported in Extended Data Fig. 6 source data.

Nature Microbiology | www.nature.com/naturemicrobiology

 nature research | reporting summary
Corresponding author(s): Dr. Melanie Hamon
Last updated by author(s): Sep 20, 2020

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.

Statistics
For all statistical analyses, confirm that the following items are present in the 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
A statement on 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 statistical parameters 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
Our web collection on statistics for biologists contains articles on many of the points above.

Software and code

Policy information about availability of computer code
Data collection Gen5 BioTek, Applied Biosystems QuantStudio 3, BioRad CFX384 and MetaMorph. Software was used generically.

Data analysis Image Lab, GraphPad Prism 8, GPower 3.1, R studio, FIJI, TAC 4.0 and Microsoft Excel 2016. No custom code was used in the manuscript.
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.
We strongly encourage code deposition in a community repository (e.g. GitHub). See the Nature Research guidelines for submitting code & software for further information.

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

All data in the present study is available upon request from the corresponding author. The complete microarray dataset (accession# E-MTAB-9603), PCA dataset,
and RT-PCR ΔCt values are contained within Table 1. Full immunoblots are in supplementary information 1-3.
October 2018

Field-specific reporting
Please select the one below that is 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

1
For a reference copy of the document with all sections, see nature.com/documents/nr-reporting-summary-flat.pdf

nature research | reporting summary

Life sciences study design
All studies must disclose on these points even when the disclosure is negative.
Sample size In vivo experimental sample size was determined and based on power calculations using lung and spleen 48 hr CFU geometric means
obtained under wildtype challenge conditions. For in vitro experiments no predetermined sample size calculations were conducted. For these
experiments were repeated 3+ biological times with 2-4 technical repeats per biological.

Data exclusions No data was excluded.

Replication All experiments, unless otherwise noted, were biologically repeated 3-5 times with a minimum of 2-3 technical replicates per biological group.
Selected in vitro experiments were repeated by multiple scientists for replication of findings. Oral commensal in vitro infection and RT-PCR
was completed by Dr. Daniel P. Miller and Dr. Richard J. Lamont at the University of Louisville. Detroit 562 cell line infection and IL-11 ELISA
were completed by Dr. Caroline M. Weight within the laboratory of Dr. Robert S. Heyderman at the University College London using gifted
bacterial strains from the G5 Chromatin and Infection group at the Institut Pasteur.

Randomization Animals were randomized into treatment groups prior to in vivo experiments.

Blinding Microscopy was blinded at analysis to the scientist who completed image acquisition. In vivo CFU counts were blinded to the scientist who
conducted the animal challenge.

Reporting for specific materials, systems and methods

We require information from authors about some types of materials, experimental systems and methods used in many studies. Here, indicate whether each material,
system or method listed is relevant to your study. If you are not sure if a list item applies to your research, read the appropriate section before selecting a response.

Materials & experimental systems Methods

n/a Involved in the study n/a Involved in the study
Antibodies ChIP-seq
Eukaryotic cell lines Flow cytometry
Palaeontology MRI-based neuroimaging
Animals and other organisms
Human research participants
Clinical data

Antibodies
Antibodies used p65 (L8F6) (CST ref #6956), p65 phosphorylation at serine 536 (CST ref# 3033), actin AC-15 monoclonal (Sigma ref# A5441),
H3K27me2 (Abcam ref# ab24684), H3K27me3 (Abcam ref# ab6002), GapDH (Abcam ref# ab9485), KDM6B (abcam ref#
ab38113), H3K27me3 (abcam ref# ab6002), H3 (abcam ref# ab195277), H3K27me2 (diagenode ref# C15410046-10)

Validation All antibodies have previously been validated by the source providers with citations, and data present on the manufacturer's
website.

Eukaryotic cell lines

Policy information about cell lines
Cell line source(s) A594, HeLa, and Beas2b cell lines were obtained from ATCC. Detroit 562 were a gift from Dr. Robert S. Heyderman. Human
telomerase immortalized gingival keratinocytes (TIGK) were used by Dr. Daniel P. Miller and Dr. Richard J. Lamont at the
University of Louisville.

Authentication None of these cell lines were authenticated.

October 2018

Mycoplasma contamination All cell lines were PCR tested for mycoplasma contamination and found to be negative within the respected laboratories.

Commonly misidentified lines None present in manuscript.

(See ICLAC register)

2
Animals and other organisms

nature research | reporting summary

Policy information about studies involving animals; ARRIVE guidelines recommended for reporting animal research
Laboratory animals Wildtype C57BL/6 female 8-9 or Wildtype C57BL/6 male 10-12 week old animals were used for studies.

Wild animals Studies did not involve wild animals.

Field-collected samples Studies did not involve field-collected samples.

Ethics oversight All protocols for animal experiments were reviewed and approved by the CETEA (Comité d'Ethique pour l'Expérimentation
Animale - Ethics Committee for Animal Experimentation) of the Institut Pasteur under approval number Dap170005 and were
performed in accordance with national laws and institutional guidelines for animal care and use.
Note that full information on the approval of the study protocol must also be provided in the manuscript.

October 2018