LUBAC-synthesized linear ubiquitin chains restrict … · LUBAC-synthesized linear ubiquitin chains...

LUBAC-synthesized linear ubiquitin chains restrictcytosol-invading bacteria by activatingautophagy and NF-κBJessica Noad1, Alexander von der Malsburg1, Claudio Pathe1, Martin A. Michel1, David Komander1

and Felix Randow1,2*

Cell-autonomous immunity relies on the ubiquitin coat surrounding cytosol-invading bacteria functioning as an ‘eat-me’signal for xenophagy. The origin, composition and precise mode of action of the ubiquitin coat remain incompletelyunderstood. Here, by studying Salmonella Typhimurium, we show that the E3 ligase LUBAC generates linear (M1-linked)polyubiquitin patches in the ubiquitin coat, which serve as antibacterial and pro-inflammatory signalling platforms. LUBACis recruited via its subunit HOIP to bacterial surfaces that are no longer shielded by host membranes and are alreadydisplaying ubiquitin, suggesting that LUBAC amplifies and refashions the ubiquitin coat. LUBAC-synthesized polyubiquitinrecruits Optineurin and Nemo for xenophagy and local activation of NF-κB, respectively, which independently restrictbacterial proliferation. In contrast, the professional cytosol-dwelling Shigella flexneri escapes from LUBAC-mediatedrestriction through the antagonizing effects of the effector E3 ligase IpaH1.4 on deposition of M1-linked polyubiquitin andsubsequent recruitment of Nemo and Optineurin. We conclude that LUBAC-synthesized M1-linked ubiquitin transformsbacterial surfaces into signalling platforms for antibacterial immunity reminiscent of antiviral assemblies on mitochondria.

Mammalian cells maintain a sterile cytosol by deployinggalectins to survey endomembrane damage and subsequentlycoat invading bacteria as well as damaged membranes with

polyubiquitin1. The bacterial ubiquitin coat comprises multiplelinkage types, synthesized by several E3 ligases such as LRSAM1,Parkin, Smurf1 and RNF166 (refs 2–5). Galectin-8 and thepolyubiquitin coat provide ligands (‘eat-me’ signals) for NDP52,Optineurin and other cargo receptors that induce antibacterialautophagy (‘xenophagy’) and restrict bacterial proliferation1,6–10.Autophagy cargo receptors are not evenly dispersed aroundcytosol-invading bacteria7,9,11. Such a biased distribution of cargoreceptors, sometimes referred to as ‘microdomains’, reflects theavailability of their ligands, suggesting that recruitment signalsand the substrate specificity of individual E3 ligases are crucial tosuccessful xenophagy.

ResultsLUBAC synthesizes M1-linked ubiquitin chains on cytosolicS. Typhimurium to restrict bacterial proliferation. Tounderstand microdomain origin we investigated the detailedtopology of the ubiquitin coat of Salmonella enterica serovarTyphimurium (S. Typhimurium) in infected murine embryonicfibroblasts (MEFs) using structured illumination microscopy(SIM), a super-resolution technique. Co-staining with antibodies1E3 and FK2, which recognize M1-linked ubiquitin or all linkagetypes except K6, respectively (Supplementary Fig. 1), revealed thatM1-linked ubiquitin is not homogenously distributed throughoutthe bacterial ubiquitin coat (Fig. 1a) and that it is synthesizedlater than other chain types (Fig. 1b and Supplementary Fig. 2a).The existence of M1-linked ubiquitin patches suggests thatspecific E3 ligases, the availability of their substrates and/orde-ubiquitinating enzymes ultimately determine the distribution

of the ubiquitin-sensing cargo receptors. We next tested whetherthe synthesis of M1-linked ubiquitin chains on S. Typhimuriumrequires LUBAC, a multimeric E3 ubiquitin ligase composed ofHOIP, HOIL-1 and Sharpin, which is known to specifically formthis linkage type12–15. Super-resolution microscopy revealedextensive co-localization of M1-linked polyubiquitin with LUBAC(Fig. 1c). Depletion of HOIP or HOIL-1 or a defect in Sharpindue to a spontaneous mutation (Sharpincpdm) substantiallyreduced the fraction of S. Typhimurium coated with M1-linkedubiquitin chains while not affecting the fraction of FK2-labelledbacteria (Fig. 1d and Supplementary Figs 2b and 3a–c). Weconclude that LUBAC contributes to the ubiquitin coat of cytosol-invading S. Typhimurium by synthesizing M1-linked ubiquitinchains in a manner requiring all three LUBAC subunits.

Because the ubiquitin coat of cytosol-invading bacteria activatesantibacterial autophagy, we tested whether LUBAC is required torestrict the proliferation of S. Typhimurium (Fig. 1e andSupplementary Figs 3a–c and 4). Bacteria proliferated significantlymore in cells derived from Sharpincpdm mice, in MEFs depleted ofHOIP or HOIL-1, as well as in human HOIP−/− epithelial cells,suggesting that all three subunits are required for LUBAC toprotect cells against invading bacteria.

LUBAC is recruited to the surface of S. Typhimurium after escapefrom Salmonella-containing vacuoles (SCVs). The bacterialubiquitin coat comprises both ubiquitylated bacterial surfaceproteins as well as host proteins on damaged Salmonella-containing vacuoles16–18, so we used galectin-8 as a marker ofdamaged endomembranes7 to investigate the localization of LUBACon cytosol-invading S. Typhimurium relative to the vacuolarmembrane remnants (Fig. 1f). Super-resolution microscopyrevealed that LUBAC is recruited to the bacterial surface and that

1MRC Laboratory of Molecular Biology, Divison of Protein and Nucleic Acid Chemistry, Francis Crick Avenue, Cambridge CB2 0QH, UK. 2University ofCambridge, Department of Medicine, Addenbrooke’s Hospital, Cambridge CB2 0QQ, UK. *e-mail: [email protected]

ARTICLESPUBLISHED: 8 MAY 2017 | VOLUME: 2 | ARTICLE NUMBER: 17063

NATURE MICROBIOLOGY 2, 17063 (2017) | DOI: 10.1038/nmicrobiol.2017.63 | www.nature.com/naturemicrobiology 1

© 2017 Macmillan Publishers Limited, part of Springer Nature. All rights reserved.

mailto:[email protected]

http://dx.doi.org/10.1038/nmicrobiol.2017.63

http://www.nature.com/naturemicrobiology

1E3 HOIL Salmonella

Rela

tive

fluor

esce

nce

HOIL

S. Tm

1E3 0

1

Pear

son’

s co

rrel

atio

n co

effici

ent

M1 vs HOIL

siCon siHOIP#94

0

14*

M1-

posi

tive

S. T

m (%

)

M1-

posi

tive

S. T

m (%

)

siCon siHOIL#12

0

14 **

M1-

posi

tive

S. T

m (%

)

WT cpdm0

14*

Sharpin

d

a

b

0

50

Time p.i. (h)1 2 4

* *

NSFK2 (total)1E3 (linear)

Ub-

posi

tive

S. T

m (%

)

00:08 h 00:14 h

00:08 h 00:14 h 00:28 h 00:50 h 01:08 h 00:34 h

00:28 h 00:50 h 01:08 h 00:34 h

Salm

onel

laH

OIL

/Gal

ectin

-8

eWT MEFcpdm

0

30

Time p.i. (h)

**

2 4 6 8

Salm

onel

la(f

old

repl

icat

ion)

0

40

Time p.i. (h)

siControl

siHOIP #94siHOIP #93siHOIP #92

******

2 4 6 8

Salm

onel

la(f

old

repl

icat

ion)

siControl

siHOIL #12siHOIL #11siHOIL #10

0

15

Time p.i. (h)Sa

lmon

ella

(fol

d re

plic

atio

n) **

NS**

2 4 6 8

g

hf

01 2 3 4

20

Time p.i. (h)

HOIP positiveGalectin-8 positiveDouble positive

Mar

ker-

posi

tive

S.Tm

(%)

c

FK2 1E3 Salmonella

Rela

tive

fluor

esce

nce

1E3

Salmonella

FK2

HOIP Galectin-8 Salmonella

Rela

tive

fluor

esce

nce

Galectin-8

Salmonella

HOIP

Figure 1 | LUBAC synthesizes M1-linked ubiquitin chains on cytosolic S. Typhimurium and restricts their proliferation. a,c,f, Structured illuminationmicrographs. MEFs infected with blue fluorescent protein (BFP)-expressing S. Typhimurium, PFA-fixed at 90 min (a,c) or 1 h (f) p.i. and stained for total(FK2, red) and M1-linked (1E3, green) polyubiquitin (a), co-expressing GFP:HOIL-1, Flag:HOIP and Flag:Sharpin and stained for M1-linked (1E3, red)polyubiquitin (c) or expressing mCherry:HOIP and stained for galectin-8 (false-coloured green) (f). The open arrowhead indicates a region with no galectin-8and the filled arrowhead indicates retracting galectin-8+ membrane. Line graphs show fluorescence plots along the indicated dashed white lines. Data arerepresentative of n > 5 images from at least three independent experiments. Scale bars, 2 µm. In c, the far right graph presents the Pearson’s correlationcoefficient of M1-linked polyubiquitin and HOIL (mean ± standard deviation (s.d.) of n = 5 images). b,d,h, Percentage of marker-positive S. Typhimurium ascounted by eye using wide-field microscopy at 90 min p.i. or at the indicated time points in PFA-fixed MEFs stained with antibodies 1E3 for M1-linked andFK2 for total polyubiquitin (b,d), or expressing GFP:HOIP and mCherry:galectin-8 (h) (mean ± standard error of the mean (s.e.m.) of triplicate coverslips fromthree independent repeats, n > 100 bacteria per coverslip). *P < 0.05, **P < 0.01, Student’s t-test. e, Fold replication of S. Typhimurium in siRNA-treated orSharpincpdm MEFs. Bacteria were counted based on their ability to grow on agar plates (mean ± s.d. of triplicate MEF cultures and duplicate colony counts,representing three independent repeats). NS, not significant, **P < 0.01, one-way analysis of variance (ANOVA) with Dunnett’s multiple comparisons test(siRNA-treated MEFs) or Student’s t-test (cpdm MEFs). g, Selected frames from live imaging on a confocal spinning disk microscope (Supplementary Video 1)of MEFs co-expressing mCherry:galectin-8, GFP:HOIL-1, Flag:HOIP and Flag:Sharpin, infected with BFP-expressing S. Typhimurium and imaged every 2 min.The shown event is representative of more than nine videos from three independent experiments. Scale bars, 10 µm. S. Tm, S. Typhimurium.

ARTICLES NATURE MICROBIOLOGY

NATURE MICROBIOLOGY 2, 17063 (2017) | DOI: 10.1038/nmicrobiol.2017.63 | www.nature.com/naturemicrobiology2




galectin-8+ membrane remnants1,7, rather than attracting LUBAC,shield bacteria from it. Only where the membrane has beendestroyed (open arrowhead in Fig. 1f) or where the membraneremnants retract (filled arrowhead in Fig. 1f) can LUBAC gainaccess to the bacterial surface.

Live microscopy directly confirmed that membrane damage andgalectin-8 accumulation preceded LUBAC recruitment, whichinitially was confined to a discrete area in the gap of the galectincoat before spreading around the bacterium (Fig. 1g andSupplementary Video 1). Notably, after bacterial division, bothdaughter bacteria retained their LUBAC+ status, suggesting that thebacterial ubiquitin coat provides a more durable signal thangalectin-8+ membranes. Consistent with this notion, the fraction ofLUBAC+ bacteria increased between 1 and 4 h post infection (p.i.),while the fraction of galectin-8+ bacteria decreased (Fig. 1h).Membrane remnants and LUBAC therefore provide two complemen-tary signals to the cell—an immediate but relatively short-lived cuethat arises from the membrane remnant, and a delayed but moresustainable one that originates from the bacterial surface. These arecharacterized by galectin-8 andM1-linked polyubiquitin, respectively.

HOIP recruits the LUBAC complex to pre-existing ubiquitin oncytosol-invading Salmonella. To understand how LUBAC isrecruited to cytosol-invading S. Typhimurium, we expressedindividual LUBAC subunits. Only GFP-HOIP, and not GFP-HOIL-1or GFP-Sharpin, accumulated on S. Typhimurium (Fig. 2a,b andSupplementary Fig. 5a). Recruitment of GFP-HOIP required thepresence of endogenous HOIL-1, consistent with the need forHOIL-1 to stabilize HOIP (Supplementary Fig. 5b)13,14. BecauseGFP-HOIL-1 and GFP-Sharpin were recruited to S. Typhimuriumonly if co-expressed with FLAG-HOIP (Fig. 2b,c and SupplementaryFig. 5c), we conclude that HOIP recruits HOIL-1 and Sharpin tocytosol-invading S. Typhimurium.

To further investigate how HOIP senses cytosol-invadingS. Typhimurium, we generated truncation mutants. HOIPN-termand HOIPC-term accumulated independently aroundS. Typhimurium, thus revealing the existence of two independentrecruitment signals for HOIP (Fig. 2d). Within HOIPN-term, thedouble NZF (dNZF) domain was sufficient for recruitment tobacteria in cells. The purified dNZF also bound S. Typhimuriumin vitro, but only if bacteria had been extracted from MEFs, not ifthey had been grown in Luria Bertani (LB) medium (Fig. 2e andSupplementary Fig. 6). The dNZF domain may therefore sense ahost-derived ligand such as ubiquitin or Nemo, for which dNZFcomprises binding sites19. Treating host cell-extractedS. Typhimurium in vitro with the deubiquitinase USP21 toremove ubiquitin precluded binding of dNZF, as did a mutationin the ubiquitin binding site of NZF1 (T360A)19, which preventedbinding of dNZFT360A and HOIPN-termT360A to bacteria in vitroand in cells, respectively (Fig. 2e,f ). In contrast, HOIPN-termR375A,deficient in binding to Nemo19, was recruited at wild-type levels(Fig. 2f). We therefore conclude that the dNZF domain recruitsHOIP to cytosol-invading S. Typhimurium in a ubiquitin-dependentmanner. Within the C-terminal fragment, both the ubiquitin-associ-ated (UBA) and ring-between-ring (RBR) domain are required forrecruitment to S. Typhimurium, as deletion of either domain pre-vented accumulation of HOIPC-term around bacteria (Fig. 2d). TheRBR domain provides E3 ligase activity; it contains the catalyticcysteine C885 and binding sites for the acceptor and donor ubiqui-tins of theM1-linked polyubiquitin chain under synthesis, which areinactive in HOIPR935A and HOIPD983A, respectively

1,6–9,20. We foundthat HOIPC-term C885A, R935A and D983A were not recruited toS. Typhimurium (Fig. 2g). Catalytic activity is therefore essentialfor HOIPC-term to accumulate on cytosol-invading bacteria.

We next tested the effect of the inactivating point mutations onfull-length HOIP (Fig. 2h). Individually inactivating the ubiquitin

binding domain (HOIPT360A) or the catalytic domain (HOIPC885A,HOIPR935A or HOIPD983A) had partial effects on HOIP recruitment,whereas T360A in combination with the catalytic mutationscompletely abrogated HOIP accumulation on S. Typhimurium.We conclude that the efficient recruitment of HOIP relies on twocooperating modi operandi, namely the ability of HOIP to bindand synthesize ubiquitin chains via its N and C terminus, respectively.

The nature of the two recruitment mechanisms for HOIPrevealed above suggests the potential existence of a feedforwardloop, which inspired us to investigate whether the accumulation ofeither of HOIP’s termini is independent of other LUBAC subunits.Using siRNA-resistant HOIP fragments, we observed that therecruitment of HOIPN-term occurred equally efficiently in controlcells, cells from Sharpincpdm mice, or cells depleted of HOIL-1 orHOIP (Fig. 2i). We conclude that the ubiquitin-mediated recruit-ment of HOIPN-term is not dependent on LUBAC activity, thusrevealing the existence of an upstream E3-ubiquitin ligase mediatingLUBAC recruitment. Based on the binding specificity of HOIPdNZFto ubiquitin chain types in vitro (Supplementary Fig. 7), theupstream E3 ligase may produce K63 chains, but depletion ofligases previously implicated in LUBAC activation during NF-κBsignalling (cIAP1, cIAP2, XIAP, Traf3 and Traf6) or ubiquitin-coating of cytosol-invading bacteria (LRSAM1 and Parkin) affectedneither the recruitment of HOIPN-term nor the deposition ofM1-linked ubiquitin chains (Supplementary Figs 3d,e and 8). Incontrast to HOIPN-term, the recruitment of HOIPC-term requiredHOIL-1 and Sharpin (Fig. 2i,j), as well as catalytic activity inHOIP (Fig. 2g). Both HOIL-1 and Sharpin contain ubiquitin-binding domains (UBDs)13,21,22 that were individually required forthe efficient recruitment of HOIPC-term to S. Typhimurium(Fig. 2k,l). Taken together, our data reveal two ubiquitin-dependentrecruitment mechanisms for LUBAC, namely (1) to pre-existingubiquitin chains via HOIPdNZF and (2) to LUBAC-synthesizedM1-linked ubiquitin chains via HOIL-1 and Sharpin, therebystabilizing existing and, importantly, recruiting new LUBACcomplexes in a feedforward mechanism. A ubiquitin-centred feed-forward loop also occurs during mitophagy, where the kinasePINK1 activates the E3 ligase Parkin by phosphorylating Parkinas well as Parkin-synthesized ubiquitin chains, an allosteric activatorof Parkin23. However, in contrast to CCCP-treated, depolarizedmitochondria, no phospho-Ser65-ubiquitin accumulated onS. Typhimurium (Supplementary Fig. 9). Taken together, wepropose a model in which HOIP is recruited to cytosol-invadingS. Typhimurium via two cues: the primary provided by an upstreamE3 ligase and sensed by the ubiquitin-binding dNZF domain ofHOIP, and the secondary comprising a feedforward loop ofLUBAC synthesizing and binding M1-linked polyubiquitin.

LUBAC-synthesized M1-linked polyubiquitin recruits Nemo andOptineurin to cytosol-invading bacteria. To investigate howLUBAC restricts bacterial proliferation and affects host cellbehaviour, we investigated the LUBAC-dependent recruitment ofcytosolic ubiquitin receptors. Super-resolution microscopyrevealed that NDP52 and p62 only partially overlapped withLUBAC on bacteria and were also recruited to LUBAC−,galectin-8+ areas (Fig. 3a and Supplementary Fig. 10a). Incontrast, the linear ubiquitin binder Nemo24 and the relatedubiquitin-binding protein Optineurin coincided selectively withLUBAC and were, like LUBAC, excluded from galectin-8+ areas.Nemo, Optineurin and HOIP were present continuously atS. Typhimurium over the course of an infection (Fig. 3b). UsingHOIP−/− human epithelial cells we found that Nemo andOptineurin—but not galectin-8, NDP52 and p62—requiredLUBAC for their recruitment to S. Typhimurium (Fig. 3c andSupplementary Fig. 3f). Nemo and Optineurin are recruited viaubiquitin, because alleles deficient in ubiquitin binding (NemoD311N

NATURE MICROBIOLOGY ARTICLES





and OptnD474N)25 failed to accumulate on S. Typhimurium (Fig. 3d).

In contrast, direct contacts of Nemo with HOIP are not essential,because HOIPR375A, an allele deficient in Nemo binding19,

complemented HOIP−/− cells (Fig. 3e). In murine cells, conjugationof LC3 to the membrane of the damaged SCV occursindependently of macro-autophagy, thus providing an alternative

e

NZF1+2

LBMEFs+Usp21

Cou

nt

ControlWTT360A

GFP

-HO

IP

GFP

-HO

IL

GFP

-Sha

rpin

HO

IL

Shar

pin

HO

IP

Shar

pin

HO

IP

HO

IL

0

14

HOIP HOIL Sharpin

Flag

GFP

GFP

-pos

itive

S.

Tm

(%)

WT

T360A

C885AR935A

D983A

C885AR935A

D983A0

16

+T360A

**

**

NS1 NS2

# #

# H

OIP

-pos

itive

S.

Tm

(%)

§§ §§ §§

siContro

l

siHOIP

siHOIL

0

12

**

HO

IP-p

ositi

ve

S. T

m (%

)

HOIP C-term

WT

cpdm0

12

*

siContro

l

siHOIP

siHOIL

0

12

HO

IP-p

ositi

ve

S. T

m (%

)

WT

cpdm0

12

HOIP N-term

a cb

d

h

i j k

HOIP C-term

HOIPW

TC885A

R935A

D983A0

14 ****

**

HO

IP-p

ositi

ve

S. T

m (%

)

HOIPW

TT360A

R375A0

14800

0

**

HOIP N-term

NS

HO

IP-p

ositi

ve

S. T

m (%

)

f g

PUB UBA RBRNZF NZFZFPUB NZF NZFZF

UBA RBRPUB

NZF NZFZFNZFZFNZF NZFNZF

NZFUBA

RBR

HOIPN-termC-term

PUB3xZnF

ZF+NZF1NZF1+2

NZF1NZF2UBARBR

HOIP-positive S.Tm (%)

0 14

l

WT

Control

+Sharpin

+T358L/

F359V

0

8

cpdm

HO

IP C

-ter

m-p

ositi

ve

S. T

m (%

)NS

** **

Sharpin

HOIL

HOIPS. Tm/DAPI

HO

IL/S

. Tm

Shar

pin/

S. T

m

HOIL

HOIP

HOIL

HOIP

0

15

NS

siContro

l

Control

+HOIL

+T203A

siHOIL

HO

IP C

-ter

m-p

ositi

ve

S. T

m (%

)

**

**

Cou

nt

800

0−103 103 104 1050

NZF1+2−103 103 104 1050

Figure 2 | HOIP senses and amplifies the ubiquitin coat of S. Typhimurium. a,c, Confocal micrographs of PFA-fixed MEFs infected with mCherry-expressing(a) or 4′,6-diamidino-2-phenylindole (DAPI)-stained (c) S. Typhimurium (blue) and expressing GFP-tagged LUBAC subunits only (a) or co-expressingFlag-tagged LUBAC subunits (red) (c) at 1 h p.i. Data are representative of three independent experiments. Scale bars, 20 µm. Split channels are displayed inSupplementary Fig. 2a,c. b,d,f–l, Percentage of marker-positive S. Typhimurium as counted by eye using wide-field microscopy at 1 h p.i. in PFA-fixed MEFsexpressing the indicated GFP- and FLAG-tagged LUBAC subunits (b) or GFP-tagged HOIP alleles (d,f–l). NS, not significant. In i–l, MEFs from wild-type orSharpincpdm mice were treated with control siRNA or siRNA against the indicated murine LUBAC components and complemented with Flag-tagged humanHOIP, HOIL-1 or Sharpin alleles as indicated. Data are presented as mean ± s.d. of triplicate coverslips, representing two independent repeats (b) and asmean ± s.e.m. of triplicate coverslips from three independent repeats (d,f–l), n > 100 bacteria per coverslip. *P < 0.05, **P < 0.01, one-way ANOVA withDunnett’s multiple comparisons test (f,g,h,l) or Student’s t-test (h,j). In h, # or NS1, compared to WT; §§, compared to T360A. e, Flow cytometry of S.Typhimurium grown in LB as indicated or extracted from MEFs (all other samples), treated with Usp21 as indicated, and stained with recombinant HOIPNZF1+2 (WT, blue), mutant allele (T360A, red) or secondary reagent only (Control, grey). term, terminal.






HOIP NEMO OPTN

Galectin-8

Galectin-8

Galectin-8

Galectin-8

HOIL OPTNSalmonella

HOIL NEMOSalmonella

HOIL NDP52 L374ASalmonella

HOIL p62Salmonella

a

Rela

tive

fluor

esce

nce

NDP52

S. Tm

HOIL

Gal-8

NEMO

S. Tm

HOIL

Gal-8

OPTN

S. Tm

HOIL

Gal-8

p62

S. Tm

HOIL

Gal-8

b c ed

**

Gal8

NDP52p62

Optn

NEMO

0

20

Mar

ker-

posi

tive

S.Tm

(%)

** **

WTHOIP−/−

1 2 4 1 2 4 1 2 40

18

Mar

ker-

posi

tive

S.Tm

(%)

Time p.i. (h)Contro

l

Control

+HOIP

+R375A0

10

NEM

O-p

ositi

ve

S.Tm

(%)

HOIP−/−

**

WTHOIP−/−

WT

D474N

OPTN

** **

WT

D311N0

10

NEMO

Mar

ker-

posi

tive

S.Tm

(%)

**** **

WTHOIP−/−

**

Control +Shigella(co-infection)

0

40 1E3 (linear)FK2 (total)

Ub-

posi

tive

S.Tm

(%)

1 40

50

Time p.i. (h)

Ub-

posi

tive

S.Tm

(%)

1E3 (linear)FK2 (total)

1 20

50

Time p.i. (h)

Ub-

posi

tive

S.fl

(%)


Salmonella Shigella

Con 1.4 9.80

15

+IpaHCon 1.4 9.8

+IpaH

Mar

ker-

posi

tive

S.Tm

(%)

NEMOOPTN F178S

**

g if h

WT 1.4 9.8

0

25

**

NS

+IpaH

Ub-

posi

tive

S.Tm

(%)


Figure 3 | LUBAC recruits Optineurin and Nemo to S. Typhimurium. a, Structured illumination micrographs. MEFs infected with BFP-expressingS. Typhimurium co-expressing the indicated GFP-tagged proteins and mCherry:HOIL-1, Flag:HOIP and Flag:Sharpin were PFA-fixed at 1 h p.i. and stainedfor galectin-8 (white). NDP52 constructs lack the SKICH domain to prevent aggregation. Graphs show fluorescence plots along indicated dashedwhite lines. Data are representative of n > 5 images from three independent experiments. Scale bars, 2 µM. b–e, Percentage of marker-positiveS. Typhimurium as counted by eye using wide-field microscopy at 1 h p.i. in PFA-fixed MEFs (b) or wild-type or HOIP−/− HCT116 cells (c–e) expressingGFP-tagged alleles of HOIP, Nemo, Optn or antibody-stained for galectin-8, NDP52 or p62. e, HOIP expression was complemented with Flag-taggedHOIP alleles as indicated. Data are presented as mean ± s.e.m. of triplicate coverslips from three independent repeats, n > 100 bacteria per coverslip.*P <0.05, Student’s t-test. f–i, Percentage of marker-positive S. flexneri or S. Typhimurium as counted by eye using wide-field microscopy at 1 h p.i.in PFA-fixed MEFs stained with antibodies 1E3 for M1-linked and FK2 for total polyubiquitin (f–h) or expressing the indicated GFP-tagged genes (i).Stable expression of Flag-tagged IpaH genes as indicated. In co-infections, S. Typhimurium were identified by expression of mCherry. Data presented asmean ± s.e.m. of triplicate coverslips from three independent repeats, n > 100 bacteria per coverslip. *P < 0.05, **P < 0.01, Student’s t-test.NS, not significant.






ligand for Optn (ref. 26). Revealing the essential contribution ofLUBAC to the recruitment of Optn therefore required inactivationof its LC3-binding LIR motif (OptnF178S)

9, while p62 and NDP52recruitment was independent of LUBAC, even if their LIR motifsand the NDP52 galectin-8 binding site were destroyed(p62DDDW335–338AAAA and NDP52V136S,L374A)

27–29 (SupplementaryFig. 10b–e). We conclude that Optineurin and Nemo are effectorproteins of LUBAC, which, via M1-linked ubiquitin chains,provides the critical signal for their recruitment to cytosol-invadingS. Typhimurium.

The Shigella-encoded E3 ligase IpaH1.4 antagonizes LUBAC-mediated recruitment of Nemo and Optineurin to cytosol-invading bacteria. We next tested whether Shigella flexneri, aGram-negative bacterium highly adapted to the cytosolicenvironment, has evolved countermeasures against LUBAC-mediated deposition of M1-linked ubiquitin chains and restrictionof proliferation. Similar to S. Typhimurium, S. flexneri recruitedLUBAC in a HOIP-dependent manner (Supplementary Fig. 11a,b).However, in contrast to S. Typhimurium, depletion of HOIP didnot significantly affect proliferation of S. flexneri (SupplementaryFig. 11c), suggesting that S. flexneri indeed antagonizes LUBAC-mediated restriction. We therefore compared the deposition ofM1-linked ubiquitin chains on both bacteria and found that a muchsmaller fraction of S. flexneri was labelled with M1-linked ubiquitinchains (Fig. 3f). Note that S. flexneri enters the cytosol almostquantitatively, while most S. Typhimurium remain membrane-enclosed. To distinguish whether S. flexneri avoids recognition bythe ubiquitylation machinery or whether it actively antagonizesubiquitylation, we performed co-infection experiments. Thedeposition of M1-linked ubiquitin chains on S. Typhimurium waslower in cells co-infected with S. flexneri, revealing that S. flexneriactively antagonizes the accumulation of M1-linked ubiquitinchains even in trans on cytosol-invading S. Typhimurium(Fig. 3g), possibly by secreting LUBAC-targeting effector proteins.Consistent with a recent report on S. flexneri suppressing NF-κBsignalling by targeting LUBAC for degradation via IpaH1.4, asecreted E3 ubiquitin ligase30, we found that in cells transducedwith IpaH1.4 but not IpaH9.8, fewer S. Typhimurium becamepositive for M1-linked ubiquitin chains and recruited Nemo orOptineurin (Fig. 3h,i). Active antagonism of M1-linkedpolyubiquitin coating of bacterial surfaces by S. flexneriindicates the evolutionary importance of the LUBAC pathway forcell-autonomous defence against cytosol-invading bacteria.

Bacteria coated with M1-linked polyubiquitin activate NF-κB.The generation of M1-linked polyubiquitin on cytosol-invadingbacteria by LUBAC and the subsequent recruitment of Nemosuggest that such bacteria may directly induce NF-κB. To testwhether the IκB kinase (IKK) complex is recruited to andactivated on cytosol-invading bacteria, we stained Salmonella-infected cells with phospho-Ser176/180-IKKα, a marker ofactivated IKKα. Phospho-IKKα was specifically observed ongalectin-8+ bacteria, that is, those exposed to the cytosol, andlabelled 16.6 ± 1.4% of intracellular S. Typhimurium at 1 h p.i.(Fig. 4a). To investigate whether S. Typhimurium induces NF-κBspecifically upon entering the host cytosol, we evaluated thenuclear translocation of the NF-κB subunit p65 in cells exposedto extracellular, vacuole-contained or cytosol-invading bacteria. At1 h p.i., 68 ± 5% of cells containing at least one galectin-8+ (thatis, cytosolic) bacterium had activated NF-κB, compared toonly 39 ± 7.2% of cells where bacteria had remained in SCVs and8 ± 2.4% of cells without intracellular bacteria (Fig. 4b). Treatmentwith siRNA against HOIP abrogated the induction of NF-κB.In contrast, lack of RipK2, an essential component of thepeptidoglycan-sensing nucleotide-binding oligomerization domain

pathway31, had no effect on the recruitment of HOIP, thedeposition of M1-linked ubiquitin chains, the recruitment ofphospho-Ser176/180-IKKα or the proliferation of S. Typhimurium(Supplementary Figs 3g,h and 12). We conclude that, by depositingM1-linked polyubiquitin, LUBAC transforms the bacterial surfaceinto a pro-inflammatory signalling platform for the local activationof NF-κB.

LUBAC activates autophagy and NF-κB, which independentlyrestrict proliferation of cytosolic S. Typhimurium. We nextinvestigated the contribution of the LUBAC effectors Optineurinand Nemo to LUBAC-mediated restriction of Salmonellaproliferation. We found that Optineurin, as reported previously9,is required to antagonize S. Typhimurium and that, unexpectedly,cells depleted of Nemo also failed to restrict bacterial growth(Supplementary Figs 3i,j and 13a,b). We therefore tested whetherNemo-mediated restriction of bacterial growth depends on theother subunits of the IKK complex and found that in cellsdepleted of IKKα or IKKβ, S. Typhimurium also hyperproliferated(Supplementary Figs 3k,l and 13c,d). We therefore conclude thatOptineurin and Nemo, as well as IKKα and IKKβ, protect cellsagainst bacterial proliferation. To investigate whether the IKKcomplex, like Optineurin9, restricts bacterial proliferation in anautophagy-dependent manner, the IKK subunits were depleted inautophagy-proficient and -deficient cells. Consistent withOptineurin protecting cells via autophagy, depleting Optineurinincreased bacterial proliferation only in wild-type cells, not inautophagy-deficient ATG5−/− cells (Fig. 4c). In contrast, depletingNemo, IKKα or IKKβ enhanced bacterial proliferation in bothwild-type and autophagy-deficient ATG5−/− cells (Fig. 4d–f ),revealing that the IKK complex protects cells againstS. Typhimurium in an autophagy-independent fashion. To testwhether the antibacterial activity of the IKK complex ismanifested via NF-κB, we deployed I-κBαSS32,36AA, a dominantnegative inhibitor of NF-κB (Supplementary Fig. 14). Expressionof I-κBαSS32,36AA enhanced proliferation of S. Typhimurium inautophagy-proficient and -deficient ATG5−/− cells (Fig. 4g),revealing that NF-κB, like the IKK complex, protects cells in anautophagy-independent fashion. Taken together, our data suggestthat LUBAC coordinates two pathways of cell-autonomousanti-bacterial defence, namely Optineurin-induced antibacterialautophagy and Nemo-controlled NF-κB activation. If correct,LUBAC should appear non-epistatic to either pathway. To testthis hypothesis, we depleted HOIP from wild-type, autophagy-deficient ATG5−/− and I-κBαSS32,36AA-expressing cells (Fig. 4h,i).HOIP antagonized bacterial proliferation in either scenario,consistent with its proposed role at the branching point ofOptineurin-induced antibacterial autophagy and Nemo-controlledNF-κB activation.

DiscussionOur data reveal an essential role for LUBAC in coordinating cell-autonomous defence against cytosol-invading S. Typhimurium.By synthesizing M1-linked polyubiquitin on bacteria, LUBACtransforms the bacterial surface into a multivalent signallingplatform that coordinates two cellular defence pathways, namelyantibacterial autophagy and NF-κB signalling, for which wedemonstrate a direct antibacterial function independent of autophagy.

The polyubiquitin coat on the surface of cytosol-invadingbacteria, first described in 2004 (ref. 6), has so far been considereda homogeneous entity. Although different linkage types areknown to occur within the bacterial ubiquitin coat4,5,32,33,information on the spatial distribution of specific linkage types isnot available. Our data reveal that M1-linked polyubiquitin localizesto the bacterial surface, but not to membrane remnants of the burstSCV, and that M1-linked ubiquitin chains can occur in distinct






0

150

Salm

onel

la (f

old

repl

icat

ion)

Atg5+ siNEMO #76Atg5+ siControl

Atg5−/− siNEMO #76Atg5−/− siControl

**

**

2 4 6 8Time p.i. (h)

a

0

150

Salm

onel

la (f

old

repl

icat

ion) Atg5+ siHOIP #94

Atg5+ siControl

Atg5−/− siHOIP #94Atg5−/− siControl

**

**

2 4 6 8Time p.i. (h)

0

200

Salm

onel

la (f

old

repl

icat

ion) Atg5+ siIKKβ #73

Atg5−/− IKKβ #73

Atg5+ siControl

Atg5−/− siControl**

**

2 4 6 8

Time p.i. (h)

0

200

Salm

onel

la (f

old

repl

icat

ion) Atg5+ siIKKα #61

Atg5−/− siIKKα #61

Atg5+ siControl

Atg5−/− siControl**

**

2 4 6 8

Time p.i. (h)

02 4 6 8

150

Time p.i. (h)

Salm

onel

la (f

old

repl

icat

ion) Atg5+ siOPTN #30

Atg5+ siControl

Atg5−/− siOPTN #30Atg5−/− siControl

**

NS

0

250

Salm

onel

la (f

old

repl

icat

ion)

Atg5−/− IKB DN

Atg5+ IKB DNAtg5−/−

Atg5+

**

**

2 4 6 8Time p.i. (h)

0

60

Salm

onel

la (f

old

repl

icat

ion)

IKB DN siHOIP #94IKB DN siControlMEF siHOIP #94MEF siControl

2 4 6 8Time p.i. (h)

**

**

LPS

LPS

Phospho-IKKα

Phospho-IKKα Galectin-8

LPS

Phospho-IKKαGalectin-8

Galectin-8

b c

d e f

g h i

ConUninf

VesCyto Con

UninfVes

Cyto0

100

+Salmonella+Salmonella

*

siControlsiHOIP

p65-

posi

tive

nucl

ei (%

)

Figure 4 | LUBAC activates autophagy and NF-κB, which independently restrict cytosolic S. Typhimurium. a, Confocal micrographs of methanol-fixedMEF cells at 1 h p.i. with S. Typhimurium stained with antibodies against LPS (red), phosphorylated IKKα (green) and galectin-8 (blue). Data arerepresentative of three independent experiments. Scale bar, 20 µm. b, Percentage of p65-positive nuclei as counted by eye using wide-field microscopyat 1 h p.i. in siRNA-treated PFA-fixed MEFs infected with GFP-expressing S. Typhimurium and stained for p65 and galectin-8. Cells in the infected sample(+Salmonella) were classified by bacterial status: no bacteria (Uninfected, ‘Uninf’), galectin-8− bacteria only (Vesicular, ‘Ves’) or ≥1 galectin-8+

bacteria (Cytosolic, ‘Cyto’). Data are presented as mean ± s.e.m. of triplicate coverslips from three independent repeats. n > 100 bacteria counted percoverslip. c–i, Fold replication of S. Typhimurium in MEFs treated with indicated siRNAs against Optn (c), Nemo (d), IKKα (e), IKKβ (f), HOIP (h,i) orexpressing I-κBαSS32,36AA (I-κBα DN) (g,i). Bacteria were counted based on their ability to form colonies on agar plates. Data are presented as mean ± s.d.of triplicate MEF cultures and duplicate colony counts, representing two (e) or three (c,d,f–i) independent repeats. *P < 0.5, **P < 0.01, Student’s t-test (b–i),calculated for 8 h timepoint.






islands within the polyubiquitin coat. The origin of M1-linked poly-ubiquitin islands remains unknown, but is most probably caused bythe localized recruitment of LUBAC (as seen in SupplementaryVideo 1), the accessibility of LUBAC substrates (for example,when SCV remnants shield the bacterial surface as seen in Fig. 1f)and/or the antagonizing effects of Otulin (see the accompanyingArticle by van Wijk et al.).

LUBAC recruitment to Gram-negative bacteria is mediated bythe N-terminal dNZF domain of HOIP. HOIPdNZF binds ubiquitinon the bacterial surface and accumulates on cytosol-invadingbacteria, even in cells lacking LUBAC function, indicating thatLUBAC refashions an existing ubiquitin coat and that an upstreamE3 ligase may function as a novel pattern recognition receptor todirect LUBAC to cytosol-invading bacteria. In addition to recruitmentvia HOIPdNZF, LUBAC also accumulates in a manner requiringthe catalytic activity of HOIP and the ubiquitin-binding domainsin Sharpin and HOIL-1, suggesting that LUBAC-synthesizedM1-linked ubiquitin chains anchor LUBAC in the bacterial vicinityand, importantly, may recruit further LUBAC molecules. Theresulting amplification mechanism is functionally reminiscent ofParkin-mediated ubiquitylation of damaged mitochondria leadingto mitophagy23, although no phospho-Ser65-ubiquitin was detectedon cytosol-invading S. Typhimurium.

LUBAC synthesized M1-linked ubiquitin chains on the bacterialsurface activate antibacterial autophagy and NF-κB signalling. Wetherefore propose that, by synthesizing M1-linked polyubiquitin,LUBAC transforms the bacterial surface into an antibacterial andpro-inflammatory signalling platform, which by its polyvalentnature, efficiently recruits adaptor proteins and initiates cellularsignalling pathways. M1-linked ubiquitin chains are essential forthe recruitment of Optineurin and Nemo, which activate selectiveautophagy and recruit the IKK complex, respectively. TheLUBAC-dependent occurrence of phosphorylated and hencecatalytically active IKK on the surface of cytosol-invading bacteriareveals a previously unrecognized principle of IKK activationdistinct from receptor-mediated IKK activation. Further work willreveal whether upstream components typically required forreceptor-mediated NF-κB activation, for example TAK1, are alsorecruited and activated on cytosol-invading S. Typhimurium. Inlight of the evolutionary relationship between bacteria andmitochondria, the functional parallels between signalling platformson bacteria and antiviral assemblies on mitochondria seemintriguing and worthy of further comparison34,35.

The potent antibacterial function of LUBAC-synthesizedM1-linked ubiquitin chains against cytosol-invading S. Typhimuriumsuggests that professional cytosol-dwelling bacteria may haveevolved efficient means to evade or antagonize the pathway. Weobserved that S. flexneri, a Gram-negative enterobacterium likeS. Typhimurium, is not restricted by LUBAC and that S. flexnerideploys IpaH1.4, a member of the multi-gene family of secretedIpaH E3 ligases, to antagonize the accumulation of M1-linkedubiquitin chains on bacterial surfaces, as well as the recruitmentof Optineurin and Nemo. The effect of IpaH1.4 on the coating ofbacteria with M1-linked ubiquitin chains is most probably relatedto its recently discovered ability to target LUBAC for degradation30.Importantly, protection against the coating of bacterial surfaces withM1-linked ubiquitin provided by IpaH1.4 extends in trans toco-infecting bacteria, as demonstrated here for S. Typhimurium,suggesting that S. flexneri profoundly cripples LUBAC-dependentcellular defence mechanisms with potentially far reachingconsequences for the outcome of co-infections.

MethodsAntibodies. Antibodies were from Enzo Life Science (ubiquitin FK2), Millipore(linear ubiquitin 1E3, phospho-ubiquitin Ser65), R&D Systems (galectin-8,cIAP2 for western blots), AbD Serotec (lipopolysaccharide, LPS), Invitrogen

(Phospho-IKKa Ser176 Ser180 and all Alexa-conjugated anti-mouse and anti-rabbitantisera), Abcam (HOIP, β-actin, Park2 for western blots), Santa Cruz (RipK2and HOIL-1, for western blots), Cayman Chemicals (Optineurin, for western blots),BD Biosciences (p62 and Nemo, for western blots), Imgenex (IKKα, IKKβ,for western blots) and Dabco (horseradish peroxidase-conjugated reagents).NDP52 antiserum was a gift from J. Kendrick-Jones.

Plasmids. M5P or closely related plasmids were used to produce recombinantmurine leukaemia virus for the expression of proteins in mammalian cells36. Openreading frames encoding human LUBAC subunits, galectin-8, NDP52, p62,Optineurin, Nemo and IκBα were amplified by PCR. Mutations were generated byPCR and verified by sequencing. For truncations, the following domain boundarieswere used: HOIP N-term (amino acids (aa) 1–438), HOIP C-term (aa 480–1072),HOIP PUB (aa 1–294), HOIP 3xZnF (aa 295–438), HOIP ZF+NZF1 (aa 295–379),HOIP NZF1+2 (aa 349–438), HOIP NZF1 (aa 330–379), HOIP NZF2 (aa 380–439),HOIP UBA (aa 439–636) and HOIP RBR (aa 693–1072). For imaging purposes,NDP52 lacking the SKICH domain (ΔN127) was used to prevent unwantedaggregation. For IκBα-DN, mutations S32A and S36A were introduced.

For recombinant protein production, HOIP NZF1+2 was amplified by PCR andcloned into a pETM30 vector encoding an N-terminal 6xHis-GST tag. BiotinylatedHOIP NZF1+2 was cloned using primers encoding a C-terminal BirA biotinylationsite into a modified pET11 vector with a ribosomal binding site followed by BirA.

Protein purification. Proteins were expressed in Escherichia coli strain BL21.Bacteria were grown to an optical density at 600 nm (OD600) of 0.7 at 37 °C beforeovernight induction at 16 °C in the presence of 100 µM isopropyl β-D-1-thiogalactopyranoside (for biotinylated proteins, 1 mM biotin was added). Cells weremechanically lysed in lysis buffer (20 mM Tris pH 8.0, 150 mM NaCl, 1 mMdithiothreitol (DTT), protease inhibitors (Roche)) and cleared by centrifugation beforeincubation with equilibrated Ni-nitrilotriacetic acid (NTA) resin for 2 h at 4 °C.Protein was then washed (20 mM Tris pH 8.0, 300 mM NaCl, 30 mM imidazole and1 mM DTT) and eluted with elution buffer (20 mM Tris pH 8.0, 150 mM NaCl,300 mM imidazole and 1 mM DTT) before dialysis overnight in the presence of6xHis-TEV protease in gel filtration buffer (20 mM Tris pH 8.0, 150 mM NaCl and1 mM DTT). To remove the 6xHis-TEV protease and the cleaved 6xHis-GST, thesample was passed through a 5 ml Ni-NTA column. Flow-through was concentratedto 1 ml and loaded onto a Superdex 75 column equilibrated in gel filtration buffer.Protein-containing fractions were pooled, concentrated and stored at −80 °C.

Cell culture. Cells were grown in Iscove’s modified Dulbecco’s medium (IMDM)supplemented with 10% FCS at 37 °C in 5% CO2. All cell lines tested negative formycoplasma. Stable cell lines were generated by retroviral transduction and selectedfor by antibiotic resistance. Where applicable, individual proteins carried uniqueresistance markers to ensure successful co-expression. Wild-type and Sharpincpdm

MEFs were obtained from Henning Walczack, ATG5−/− MEFs37 fromN. Mizushima, wild-type, HOIP−/− and RipK2−/− HCT116s from M. Gyrd-Hansen.Verification of genotypes is shown in Supplementary Fig. 3.

Bacteria. S. Typhimurium (strain 12023), provided by D. Holden, was grownovernight in Luria broth (LB) and subcultured (1:33) in fresh LB for 3.5 h beforeinfection. Such cultures were further diluted (1:5) in antibiotic-free IMDM plus 10%FCS immediately before 10 µl was used to infect MEF cells in 24-well plates for10 min at 37 °C. Following two washes with warm PBS, cells were cultured in100 µg ml−1 gentamycin for 2 h and 20 µg ml−1 gentamycin thereafter. For p65translocation assays, bacteria were washed once in warm antibiotic-free IMDM,resuspended in the same volume, and 10 µl was used to infect cells. To enumerateintracellular bacteria, cells from triplicate wells were lysed in 1 ml cold PBScontaining 0.1% Triton-X-100. Serial dilutions were plated in duplicate on TYE agar.

Extraction of S. Typhimurium from infected cells. A 10 cm dish of MEFs wasinfected with mCherry-expressing S. Typhimurium. After 3 h, cells were washedtwice with PBS and detached with trypsin. Cells were washed twice with PBS + 2%FCS and lysed in 1 ml lysis buffer (20 mM Tris pH 7.4, 150 mM NaCl, 1% TritonX-100 and 1 mM iodoacetamide) for 30 min at 4 °C. Bacteria were recovered bycentrifugation (5 min at 13,000 r.p.m. at 4 °C) and washed twice with reaction buffer(20 mM Tris pH 7.4, 150 mMNaCl and 2 mMDTT). DNase I (Sigma) was added toa final concentration of 100 µg ml–1 before incubation for 30 min at 37 °C. Bacteriawere washed with reaction buffer and incubated at 37 °C in the presence of buffer orUsp21 (0.2 mg ml–1) for 30 min at 37 °C. After washing with PBS + 2% FCS, thebacteria were stained with primary antibodies and/or recombinant, biotinylatedHOIP NZF1+2 (5 µg ml–1) for 30 min on ice.

After washing with PBS + 2% FCS, bacteria were stained with fluorescentlylabelled secondary antibodies and/or Alexa488 labelled streptavidin for 30 min onice. Bacteria were fixed in 2% paraformaldehyde for 30 min at room temperature andanalysed on an LSRII flow cytometer.

RNA interference. A total of 5 × 104 cells per well were seeded in 24-well plates.The following day, cells were transfected with 6 pmol siRNA (Invitrogen) usinglipofectamine RNAiMAX (Invitrogen) in Optimemmedium (Invitrogen). Optimem






was replaced with complete IMDM medium after 4 h, and experiments wereperformed after 3 days. For ATG5−/− MEFs, 5 × 103 cells were seeded initially andsiRNA transfection was repeated 2 days after first siRNA treatment.

siRNAs targeted the following sequences:siHOIP #92 5′-CGGCAUUGACUGUCCGAAAsiHOIP #93 5′-ACUUCACCAUUGCCCUGAAsiHOIP #94 5′-GACCCUAACUGCAAGGUGAsiHOIL-1 #10 5′-CUGCGAAUGUUGGAAGAUUsiHOIL-1 #11 5′-GGGUGCAAGUAAAACCCGAsiHOIL-1 #12 5′-ACACGUCACUCAACCCACAsiSharpin #84 5′-GCUCAAAUACCUAAAGCAAsiSharpin #85 5′-CCACCCACAUUGCUCCAUUsiSharpin #86 5′-CUUUCAUCAAUGCCUCAAAsiOptn #29 5′-GAAGCUAAAUAAUCAAGCUsiOptn #30 5′-GCCUCGCAGUAUUCCGAUUsiOptn #31 5′-CAAUUGAAGAACUAACCAAsiNemo #76 5′-GGAUUGAGGAUAUGAGGAAsiNemo #77 5′-GGAUUCGAGCAGUUAGUGAsiNemo #78 5′-AGGCCUCUGUGAAAGCUCAsiIKKα #61 5′-GAAAGAUCCAAAGUGUAUAsiIKKα #62 5′-CUAUUGAUCUUACUUUGAAsiIKKβ #73 5′-CGUUGGACAUGGAUCUUGUsiIKKβ #75 5′-GAAGAUCGCCUGUAGCAAAsiLRSAM1 #93 5′-GAAUAAGAAUGGAGCAGUUsiLRSAM1 #94 5′-GAACCAGAUUAGGCUAAUAsiPARK2 #27 5′-GGAACAACAGAGUAUUGUAsiPARK2 #28 5′-GGAGGAUGUAUGCACAUGAsiTRAF3 #28 5′-GGUCUACUGUCGGAAUGAAsiTRAF3 #29 5′-GGUACAAACCAGCAGAUCAsiTRAF6 #200 5′-GUGAGGAACUUUACUCUUAsiTRAF6 #208 5′-UGUGGAAUGUGGAGAGGAAsiRipK2 #49 5′-GUGUGAAGCAUGAUAUAUAsiRipK2 #66 5′-GGAUUUAUCGCUAAACAUAsicIAP1 #57 5′-GAAGACUUCUCAUCAAGGAsicIAP1 #58 5′-GGACAAAGGAGAGCGAAGAsicIAP2 #54 5′-GAAGAGUGCUGACACCUUUsicIAP2 #55 5′-CAGCCCGUAUUAGAACAUUsiXIAP #60 5′-GAUCGUUACUUUUGGAACAsiXIAP #61 5′-CCAGGGUGCAAAUACCAUThe non-targeting negative control No. 1 (Invitrogen) was used as control.

Microscopy. MEF cells were grown on glass coverslips before infection. Afterinfection, cells were washed twice with warm PBS and fixed in 4% paraformaldehydeor ice-cold methanol (phospho-IKKα stained samples only) for 20 min. Cells werewashed twice in PBS and then simultaneously permeabilized and blocked in PBSB(PBS, 0.01% saponin, 2% BSA). Coverslips were incubated with primary followed bysecondary antibodies for 1 h in PBSB. Samples were mounted in mounting mediumwith DAPI (Vector Laboratories) for confocal or Prolong Antifade mountingmedium (Invitrogen) for super-resolution microscopy. Marker positive bacteriawere counted by eye among at least 100 bacteria per coverslip using a wide-fieldmicroscope. Confocal images were taken with a ×63, 1.4 numerical aperture (NA)objective on either a Zeiss 710 or Zeiss 780 microscope. Live imaging was performedon a Nikon Eclipse Ti equipped with an Andor Revolution XD system and aYokogawa CSU-X1 spinning disk unit. Super-resolution images were acquired usingan Elyra S1 structured illumination microscope (Carl Zeiss Microscopy). The systemhas four laser excitation sources (405, 488, 561 and 640 nm) with fluorescenceemission filter sets matched to these wavelengths. SIM images were obtained using a63 × 1.4 NA oil immersion lens with grating projections at three rotations and fivephases in accordance with the manufacturer’s instructions. The number of Z planesvaried with sample thickness. Super-resolution images were calculated from the rawdata using Zeiss ZEN software. Pearson’s correlation coefficients were calculatedfrom n > 5 images using Imaris software.

Western blot. Cells were lysed in lysis buffer (20 mM Tris pH 7.4, 150 mM NaCl,0.1% Triton X-100, 10% glycerol plus 1 mM phenylmethylsulfonyl fluoride (PMSF),1 mM benzamidine, 2 µg ml–1 aprotinin, 5 µg ml–1 leupeptin, 1 mM DTT) beforeclearing by centrifugation and addition of SDS loading buffer. Samples were thenseparated on 4–12% denaturing gels (Invitrogen) and visualized by immunoblottingusing enhanced chemiluminescence detection reagents (Amersham Bioscience).

Reverse transcriptase PCR. Total RNA from siRNA-treated MEF cells was extractedusing the RNeasy PlusMini Kit (Qiagen) followed by conversion into cDNA using theSuperScript III reverse transcriptase kit (Invitrogen) according to the manufacturer’sprotocol. Gene expression was quantified using specific primer pairs with a PowerSYBR qPCR green kit (Applied Biosystems) by following the manufacturer’sprotocol. Relative amounts of cDNA were calculated using the ddCt method andnormalized to β-actin cDNA levels in each sample. The following primers were used:

IKBKG (Nemo)5′-CAGGTCCCATAAGGCTACAAG

5′-TTCGTTCAGGCATACACAGGSharpin5′-AGAAGGAGTATTTGCAGGAGC5′-TGGAGCAGGGAGTAAAGGAGACTB (β-actin)5′-TGACAGCATTGCTTCTGTGTAAATT5′-ATTGGTCTCAAGTCAGTGTACAGGCTraf35′-GCACCTGTAGTTTTAAGCGC5′-TGAAGGATCTTGGACTCGTTGTraf65′-GAACTGAGACATCTCGAGGATC5′-AGAGGACAGCTTTGATCATGGXIAP5′-CTGAAAAAACACCACCGCTAAC5′-CTAAATCCCATTCGTATAGCTTCTTGLRSAM15′-CTCGAGAATGAGGTCCTTGG5′-GCTGACAGCAGCAGACGTGRipK25′-CTCGTGTTCCTTGGCTGTAA5′-CAATGGCTTCCCTCTTACTCTG

Biacore. SPR experiments were performed on a Biacore 2000 instrument(GE Healthcare) as described previously38. CM5 chips (GE Healthcare) werefunctionalized with a 1:1 mix of 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide/N-hydroxysuccinimide and one flow channel per chip was immediately quenchedwith 1 M ethanolamine pH 8.0 to serve as reference. For the other channels, thedifferently linked diUb were diluted to 100 ng µl–1 in 20 mM sodium acetate pH 5.0and injected until a response of ∼2,000 RU was reached. Protein samples at differentconcentrations were prepared in SPR buffer (20 mM Tris pH 7.4, 150 mM NaCl,1 mM DTT) and injected for 60 s followed by 150 s of dissociation in SPR buffer at20 °C. Dissociation constant Kd was determined from fitting triplicate experimentsusing the reference-corrected equilibrium response in Prism 6 (GraphPad Software).

Data availability. The data that support the findings of this study are available fromthe corresponding author upon request.

Received 18 February 2017; accepted 23 March 2017;published 8 May 2017

References1. Deretic, V., Saitoh, T. & Akira, S. Autophagy in infection, inflammation and

immunity. Nat. Rev. Immunol. 13, 722–737 (2013).2. Cadwell, K. et al. A key role for autophagy and the autophagy gene Atg16l1 in

mouse and human intestinal Paneth cells. Nature 456, 259–263 (2008).3. Heath, R. J. et al. RNF166 determines recruitment of adaptor proteins during

antibacterial autophagy. Cell Rep. 17, 2183–2194 (2016).4. Manzanillo, P. S. et al. The ubiquitin ligase parkin mediates resistance to

intracellular pathogens. Nature 501, 512–516 (2013).5. Franco, L. H. et al. The ubiquitin ligase Smurf1 functions in selective autophagy

of Mycobacterium tuberculosis and anti-tuberculous host defense. Cell HostMicrobe 21, 59–72 (2017).

6. Perrin, A., Jiang, X., Birmingham, C., So, N. & Brumell, J. Recognition of bacteriain the cytosol of mammalian cells by the ubiquitin system. Curr. Biol. 14,806–811 (2004).

7. Thurston, T. L. M., Wandel, M. P., von Muhlinen, N., Foeglein, A. & Randow, F.Galectin 8 targets damaged vesicles for autophagy to defend cells againstbacterial invasion. Nature 482, 414–418 (2012).

8. Thurston, T. L. M., Ryzhakov, G., Bloor, S., Muhlinen von, N. & Randow, F.The TBK1 adaptor and autophagy receptor NDP52 restricts the proliferation ofubiquitin-coated bacteria. Nat. Immunol. 10, 1215–1221 (2009).

9. Wild, P. et al. Phosphorylation of the autophagy receptor optineurin restrictsSalmonella growth. Science 333, 228–233 (2011).

10. Boyle, K. B. & Randow, F. The role of ‘eat-me’ signals and autophagy cargoreceptors in innate immunity. Curr. Opin. Microbiol. 16, 339–348 (2013).

11. Cemma, M., Kim, P. K. & Brumell, J. H. The ubiquitin-binding adaptor proteinsp62/SQSTM1 and NDP52 are recruited independently to bacteria-associatedmicrodomains to target Salmonella to the autophagy pathway. Autophagy 7,341–345 (2011).

12. Kirisako, T. et al. A ubiquitin ligase complex assembles linear polyubiquitinchains. EMBO J. 25, 4877–4887 (2006).

13. Ikeda, F. et al. SHARPIN forms a linear ubiquitin ligase complex regulatingNF-κB activity and apoptosis. Nature 471, 637–641 (2011).

14. Tokunaga, F. et al. SHARPIN is a component of the NF-κB-activating linearubiquitin chain assembly complex. Nature 471, 633–636 (2011).

15. Gerlach, B. et al. Linear ubiquitination prevents inflammation and regulatesimmune signalling. Nature 471, 591–596 (2011).






16. Fiskin, E., Bionda, T., Dikic, I. & Behrends, C. Global analysis of host andbacterial ubiquitinome in response to Salmonella typhimurium infection.Mol. Cell 62, 967–981 (2016).

17. Collins, C. A. et al. Atg5-independent sequestration of ubiquitinatedmycobacteria. PLoS Pathogens 5, e1000430 (2009).

18. Fujita, N. et al. Recruitment of the autophagic machinery to endosomes duringinfection is mediated by ubiquitin. J. Cell Biol. 203, 115–128 (2013).

19. Fujita, H. et al. Mechanism underlying I B kinase activation mediated by thelinear ubiquitin chain assembly complex. Mol. Cell Biol. 34, 1322–1335 (2014).

20. Stieglitz, B. et al. Structural basis for ligase-specific conjugation of linearubiquitin chains by HOIP. Nature 503, 422–426 (2013).

21. Haas, T. L. et al. Recruitment of the linear ubiquitin chain assembly complexstabilizes the TNF-R1 signaling complex and is required for TNF-mediated geneinduction. Mol. Cell 36, 831–844 (2009).

22. Sato, Y. et al. Specific recognition of linear ubiquitin chains by the Npl4 zincfinger (NZF) domain of the HOIL-1L subunit of the linear ubiquitin chainassembly complex. Proc. Natl Acad. Sci. USA 108, 20520–20525 (2011).

23. Swatek, K. N. & Komander, D. Ubiquitin modifications. Cell Res. 26,399–422 (2016).

24. Rahighi, S. et al. Specific recognition of linear ubiquitin chains by NEMO isimportant for NF-κB activation. Cell 136, 1098–1109 (2009).

25. Bloor, S. et al. Signal processing by its coil zipper domain activates IKK gamma.Proc. Natl Acad. Sci. USA 105, 1279–1284 (2008).

26. Kageyama, S. et al. The LC3 recruitment mechanism is separate fromAtg9L1-dependent membrane formation in the autophagic response againstSalmonella. Mol. Biol. Cell 22, 2290–2300 (2011).

27. Muhlinen von, N. et al. LC3C, bound selectively by a noncanonical LIR motif inNDP52, is required for antibacterial autophagy. Mol. Cell 48, 329–342 (2012).

28. Li, S. et al. Sterical hindrance promotes selectivity of the autophagy cargoreceptor NDP52 for the danger receptor galectin-8 in antibacterial autophagy.Sci. Signal. 6, ra9 (2013).

29. Ichimura, Y. et al. Structural basis for sorting mechanism of p62 in selectiveautophagy. J. Biol. Chem. 283, 22847–22857 (2008).

30. de Jong, M. F., Liu, Z., Chen, D. & Alto, N. M. Shigella flexneri suppressesNF-κB activation by inhibiting linear ubiquitin chain ligation. Nat. Microbiol. 1,16084 (2016).

31. Caruso, R., Warner, N., Inohara, N. & Núñez, G. NOD1 and NOD2: signaling,host defense, and inflammatory disease. Immunity 41, 898–908 (2014).

32. Huett, A. et al. The LRR and RING domain protein LRSAM1 is an E3 ligasecrucial for ubiquitin-dependent autophagy of intracellular Salmonellatyphimurium. Cell Host Microbe 12, 778–790 (2012).

33. van Wijk, S. J. L. et al. Fluorescence-based sensors to monitor localization andfunctions of linear and K63-linked ubiquitin chains in cells. Mol. Cell 47,797–809 (2012).

34. Randow, F. & Youle, R. J. Self and nonself: how autophagy targets mitochondriaand bacteria. Cell Host Microbe 15, 403–411 (2014).

35. Wu, J. & Chen, Z. J. Innate immune sensing and signaling of cytosolic nucleicacids. Annu. Rev. Immunol. 32, 461–488 (2014).

36. Randow, F. & Sale, J. E. Retroviral transduction of DT40. Subcell. Biochem. 40,383–386 (2006).

37. Kuma, A. et al. The role of autophagy during the early neonatal starvationperiod. Nature 432, 1032–1036 (2004).

38. Michel, M. A. et al. Assembly and specific recognition of k29- and k33-linkedpolyubiquitin. Mol. Cell 58, 95–109 (2015).

AcknowledgementsThis work was supported by the Medical Research Council (U105170648 to F.R. andU105192732 to D.K.) and theWellcome Trust (WT104752MA) and Boehringer IngelheimFonds PhD fellowships (to C.P. and M.A.M.). The authors thank N. Mizushima forATG5−/− MEFs, H. Walczack for Sharpincpdm MEFs, M. Gyrd-Hansen for RipK2−/− andHOIP−/− HCT116s and E. Werner for technical help.

Author contributionsJ.N. performed and analysed all experiments, except those in Fig. 2e (A.v.d.M., D.K.),Fig. 3h (C.P.), Fig. 3i (C.P., J.N.), Supplementary Fig. 1 (A.M., D.K.), Supplementary Fig. 2a(A.M.), Supplementary Fig. 6 (A.M.) and Supplementary Fig. 7 (A.M., M.A.M., D.K.).J.N. and F.R. wrote the manuscript.

Additional informationSupplementary information is available for this paper.

Reprints and permissions information is available at www.nature.com/reprints.

Correspondence and requests for materials should be addressed to F.R.

How to cite this article: Noad, J. et al. LUBAC-synthesized linear ubiquitin chains restrictcytosol-invading bacteria by activating autophagy and NF-κB. Nat. Microbiol. 2, 17063 (2017).

Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims inpublished maps and institutional affiliations.

Competing interestsThe authors declare no competing financial interests.





http://www.nature.com/reprints



LUBAC-synthesized linear ubiquitin chains restrict … · LUBAC-synthesized linear ubiquitin chains...

Documents

Transcript of LUBAC-synthesized linear ubiquitin chains restrict … · LUBAC-synthesized linear ubiquitin chains...