Peptidoglycan induces loss of a nuclear peptidoglycan recognition protein during host tissue development in a beneficial animal-bacterial symbiosis.

Abstract

Peptidoglycan recognition proteins (PGRPs) are mediators of innate immunity and recently have been implicated in developmental regulation. To explore the interplay between these two roles, we characterized a PGRP in the host squid Euprymna scolopes (EsPGRP1) during colonization by the mutualistic bacterium Vibrio fischeri. Previous research on the squid-vibrio symbiosis had shown that, upon colonization of deep epithelium-lined crypts of the host light organ, symbiont-derived peptidoglycan monomers induce apoptosis-mediated regression of remote epithelial fields involved in the inoculation process. In this study, immunofluorescence microscopy revealed that EsPGRP1 localizes to the nuclei of epithelial cells, and symbiont colonization induces the loss of EsPGRP1 from apoptotic nuclei. The loss of nuclear EsPGRP1 occurred prior to DNA cleavage and breakdown of the nuclear membrane, but followed chromatin condensation, suggesting that it occurs during late-stage apoptosis. Experiments with purified peptidoglycan monomers and with V. fischeri mutants defective in peptidoglycan-monomer release provided evidence that these molecules trigger nuclear loss of EsPGRP1 and apoptosis. The demonstration of a nuclear PGRP is unprecedented, and the dynamics of EsPGRP1 during apoptosis provide a striking example of a connection between microbial recognition and developmental responses in the establishment of symbiosis.

Full text

Page 1

VOLUME 11 NUMBER 7 JULY 2009
cellular
microbiology
www.cell-micro.com ISSN 1462-5814
Thematic Reviews: Modelling Pathogenesis
Resistance to non-oxidative killing
Virulence determinants of Francisella

Page 2

Peptidoglycan induces loss of a nuclear peptidoglycan
recognition protein during host tissue development in
a beneficial animal-bacterial symbiosiscmi_1315 1114..1127
Joshua V. Troll,1Dawn M. Adin,2Andrew M. Wier,1
Nicholas Paquette,3Neal Silverman,3
William E. Goldman,4Frank J. Stadermann,5
Eric V. Stabb2and Margaret J. McFall-Ngai1*
1Department of Medical Microbiology and Immunology,
University of Wisconsin-Madison, Madison, WI 53706,
USA.
2Department of Microbiology, University of Georgia,
Athens, GA 30602, USA.
3Division of Infectious Diseases, Department of
Medicine, University of Massachusetts Medical School,
Worcester, MA 01605, USA.
4Department of Microbiology and Immunology,
University of North Carolina, Chapel Hill, NC 27599,
USA.
5Department of Physics, Laboratory of Space Sciences,
Washington University, St. Louis, MO 63130, USA.
Summary
Peptidoglycan recognition proteins (PGRPs) are
mediators of innate immunity and recently have
been implicated in developmental regulation. To
explore the interplay between these two roles, we
characterized a PGRP in the host squid Euprymna
scolopes (EsPGRP1) during colonization by the
mutualistic bacterium Vibrio fischeri. Previous
research on the squid-vibrio symbiosis had shown
that, upon colonization of deep epithelium-lined
crypts of the host light organ, symbiont-derived
peptidoglycanmonomers
mediated regression of remote epithelial fields
involved in the inoculation process. In this study,
immunofluorescence microscopy revealed that
EsPGRP1 localizes to the nuclei of epithelial cells,
and symbiont colonization induces the loss of
EsPGRP1fromapoptoticnuclei.Thelossofnuclear
EsPGRP1 occurred prior to DNA cleavage and
breakdown of the nuclear membrane, but followed
chromatin condensation, suggesting that it occurs
induceapoptosis-
during late-stage apoptosis. Experiments with
purified peptidoglycan monomers and with V. fis-
cheri mutants defective in peptidoglycan-monomer
release provided evidence that these molecules
trigger nuclear loss of EsPGRP1 and apoptosis.
The demonstration of a nuclear PGRP is unpre-
cedented, and the dynamics of EsPGRP1 during
apoptosisprovideastriking
connection between microbial recognition and
developmental responses in the establishment of
symbiosis.
example ofa
Introduction
Peptidoglycan recognition proteins, or PGRPs, are a
protein family of innate immune system effectors and
receptors conserved across the animal kingdom (Werner
et al., 2000; Goodson et al., 2005; Dziarski and Gupta,
2006b; Coteur et al., 2007; Zhang et al., 2007). PGRPs
mediate host responses to the bacterial cell envelope
component, peptidoglycan (PGN). In-depth research on
the PGRPs in the last 10 years has principally focused on
biochemical characterizations of the activity and binding
specificity of the purified proteins, as well as the genetics
of their in vivo function in Drosophila melanogaster
(Kaneko et al., 2004; 2006; Filipe et al., 2005; Dziarski
and Gupta, 2006a; Lim et al., 2006; Lu et al., 2006; Park
et al., 2007). For example, in vitro studies have demon-
strated that the characteristic PGRP domain binds spe-
cifically to PGN ligands (Chang et al., 2006; Swaminathan
et al., 2006) and in some isoforms also cleaves the sub-
strate (Steiner, 2004). Genetic analyses have implicated
these behaviours in the regulation of downstream
response pathways (Michel et al., 2001; Kaneko et al.,
2004; Bischoff et al., 2006).
These studies have defined similarities and differences
among the vertebrates and invertebrates. In both animal
groups, some PGRPs can act to attenuate the inflamma-
tory response by degrading the PGN through N-acetyl-
muramyl-L-alanine amidase activity (Gelius et al., 2003;
Bischoff et al., 2006; Zaidman-Remy et al., 2006), which
requires a cysteine residue in the PGN-binding pocket. In
this enzymatic behaviour, the stem peptide is cleaved from
Received 17 December, 2008; revised 20 February, 2009; accepted
23 February, 2009. *For correspondence. E-mail mjmcfallngai@
wisc.edu; Tel. (+1) 608 262 2393; Fax (+1) 608 262 8418.
Cellular Microbiology (2009) 11(7), 1114–1127doi:10.1111/j.1462-5822.2009.01315.x
First published online 27 March 2009
© 2009 Blackwell Publishing Ltd
cellular microbiology

Page 3

the sugar backbone of PGN, resulting in products that are
less immunogenic (Bischoff et al., 2006). Although all
mammalian PGRPs described thus far are reported to be
secreted proteins, the invertebrate homologues can act as
receptors for PGN, a feature reflected in their localization.
The invertebrate PGRPs are predicted to be either intra-
cellular, membrane-associated or secreted (Royet and
Dziarski, 2007), although few studies have defined precise
cellular localization (Dziarski et al., 2003; Kaneko et al.,
2006;Gupta,2008).Themorevariedpredictedlocalization
of invertebrate PGRPs correlates with broader function. In
Drosophila, PGRPs have been implicated in the regulation
ofapoptosisanddevelopment(Bischoffetal.,2006;Maillet
et al., 2008), activation or dampening of the immune
response (Yoshida et al., 1996; Takehana et al., 2002;
Kaneko et al., 2004; 2006; Maillet et al., 2008), control of
phagocytosis (Ramet et al., 2002) and as direct immune
effectors (Mellroth and Steiner, 2006).
The PGRPs have been implicated not only in mediating
responses to microbial pathogens, but also in the dynam-
ics of mutualistic symbioses in invertebrates (Anselme
et al., 2006; 2008; Bischoff et al., 2006). The binary asso-
ciation between the Hawaiian bobtail squid Euprymna
scolopes and the marine bacterium Vibrio fischeri has
been used for the last 20 years as a model for the study of
mutualistic symbioses (for reviews see Nyholm and
McFall-Ngai, 2004; Visick and Ruby, 2006). The analysis
of an EST database constructed from the symbiotic
tissues of juvenile E. scolopes revealed the expression of
four host PGRPs, EsPGRP1–4, in these bacteria-
containing tissues (Goodson et al., 2005). This finding
suggested the squid-vibrio system could be a model for
the study of these proteins in mutualistic associations. In
this naturally occurring, binary symbiosis, the partners
can be experimentally manipulated. This feature facili-
tates the discovery of the molecular underpinnings of
the cellular interactions between host and symbiont.
In addition, molecular genetics have been developed in
V. fischeri, providing the opportunity to study changes in
the host in response to genetic alterations in key symbiont
characters (Visick and Ruby, 2006).
A key feature of this symbiosis is its exclusivity: only
V. fischeri is capable of colonizing host tissues and
inducing morphogenesis (McFall-Ngai and Ruby, 1991).
The exclusivity is achieved through the multiple steps
required for colonization, including competitive domi-
nance during aggregation in host mucus, locomotion to
the light-organ pores and resistance to the oxidatively
stressful environment of the ducts (Nyholm et al., 2004).
V. fischeri cells appear to signal morphogenesis from the
light-organ interior, which is several cell layers away from
the tissue layer that regresses (Fig. 1) (Montgomery and
McFall-Ngai, 1994; Doino and McFall-Ngai, 1995). Thus,
either the morphogenetic signals are sensed by the inter-
nal epithelia and responses are transduced through the
tissues to the superficial epithelium, or the bacterial
signal molecules are transported through host tissues to
the sites of morphogenesis where they would act directly
on host cells. In either case, the development is irrever-
sibly triggered at 12 h and completed by 96 h (Doino
et al., 1995).
Studies of these early developmental events implicated
the cell envelope constituents of V. fischeri, specifically
derivatives of lipopolysaccharide (LPS) and PGN, in the
triggering of host tissue morphogenesis. V. fischeri is
one of the few bacterial species known to release the
tetrapeptide PGN monomer, ‘tracheal cytotoxin’ (TCT)
(Rosenthal et al., 1987; Cloud-Hansen et al., 2006), which
mediates the destruction of epithelial tissues in certain
pathogenic associations (Cookson et al., 1989; Cloud and
Dillard,2002).Further,asstatedabove,onlyV. fischerican
enter the crypts of E. scolopes to present these morph-
ogens. Light-organ development can be triggered in the
absence of V. fischeri by exposing juvenile squid to the
synergisticactivityofTCTandthelipidAcomponentofLPS
(Koropatnick et al., 2004). LPS alone fails to drive light-
organ development but will induce chromatin condensa-
tion without progression into the later stages of apoptosis,
such as DNA fragmentation (Foster et al., 2000). When
added as pharmacological agents, which are readily taken
upintothecryptspaces,TCTsynergizeswithLPStomimic
symbiont-induced chromatin condensation and epithelial
regression (Koropatnick et al., 2004). While the effects of
TCT and LPS on late-stage apoptosis were not investi-
gated, this study led to the model that these biomolecules
are sufficient to induce light-organ morphogenesis.
The finding that V. fischeri PGN is critical for morpho-
genesis, coupled with the discovery of expression of
PGRPs in the juvenile light organ, suggested that the
EsPGRPs are good candidates to mediate the host
response to bacterial PGN products. EsPGRP1 is of par-
ticular interest because its expression is upregulated
during early morphogenesis (Chun et al., 2008). The
derived amino acid sequence and lack of a putative signal
peptide predicted that EsPGRP1 is a 23.5 kDa intra-
cellular protein (Goodson et al., 2005). The presence
of conserved amino acid residues required for N-acetyl-
muramyl-L-alanineamidase
EsPGRP1 has PGN amidase activity.
In this study, we sought to describe the role of
EsPGRP1 during early postembryonic development of the
light organ. We observed this protein in host tissues of
uncolonized animals and during colonization by wild-type
V. fischeri or by TCT-production mutants, as well as
during the host response to pharmacological exposure to
PGN and LPS derivatives. The data presented here
provide evidence that PGRPs are components of the host
response to mutualistic microbial associations and are
activitysuggestedthat
A nuclear PGRP in symbiont-induced development1115
© 2009 Blackwell Publishing Ltd, Cellular Microbiology, 11, 1114–1127

Page 4

integral components of the developmental response to
bacterial cues in this symbiosis.
Results
Characterization of the EsPGRP1 protein and an
anti-EsPGRP1 antibody
Western blot analyses determined that EsPGRP1 is
present in the soluble fraction of lysed cells of whole
newly hatched animals. The EsPGRP1 antibody (Fig. 2A)
reacted with a peptide at ~24 kDa, consistent with the
molecular mass predicted by the derived amino acid
sequence (Fig. 2B). To test the prediction that EsPGRP1
has amidase activity (Goodson et al., 2005), we trans-
fected a FLAG-tagged EsPGRP1 expression plasmid into
Drosophila S2* cells. A protein fraction enriched for
EsPGRP1-FLAG was generated from these transfectants
using anti-FLAG affinity chromatography (Fig. 2C). This
fraction was coincubated with TCT. Within 1 h, 64 ng of
the EsPGRP1-enriched protein fraction degraded over
2.5 mg of TCT (Fig. 2D), in a manner consistent with
PGRP amidase activity (Gelius et al., 2003). In contrast,
24 mg of crude protein extract of S2* cells was found to
have no catalytic activity against TCT. The potential for
direct interaction between EsPGRP1 and PGN was
further supported by the appearance of zones of clearing
on a PGN-gel zymogram at positions corresponding to
EsPGRP1 bands on a companion immunoblot (Fig. S1).
EsPGRP1 localized to epithelial nuclei
To gain insight into the possible functions of EsPGRP1 in
the symbiosis, we performed confocal immunocytochem-
Fig. 1. Events in the early symbiosis of the squid/vibrio association
A. A dorsal view of a hatchling E. scolopes. E. scolopes harvests V. fischeri cells from ambient seawater within hours of hatching and
maintains a population of the symbiont in a specialized ‘light organ’. The light organ appears as a dark region in the centre of the body cavity
(white arrow).
B. A diagram depicting the ventral view of the hatchling light organ; left of the dashed line surface features are shown, and right of the dashed
line interior structures are displayed. Ciliated epithelial fields specific to the surface of the juvenile light organ promote colonization by the
symbiont. These ciliated epithelial fields that undergo morphogenesis include the anterior appendage (aa), the posterior appendage (pa) and
the ciliated ridge (cr).
C. A timeline illustrating relevant characteristics of the early symbiosis. V. fischeri cells aggregate near the ciliated epithelial fields and enter
the light organ around 4–6 h post hatching. Within hours of successful colonization of the deep crypts of the light organ, the symbionts deliver
an irreversible signal that triggers an extensive developmentally programmed tissue destruction (morphogenesis) of the light organ into a
mature form lacking the ciliated epithelial fields. ac, antechamber; d, ducts; dc, deep crypts; EPS, exopolysaccharide; hg, hindgut; p, pores.
1116 J. V. Troll et al.
© 2009 Blackwell Publishing Ltd, Cellular Microbiology, 11, 1114–1127

Page 5

istry to localize the protein in tissues and cells. Reactivity
to the EsPGRP1 antibody occurred in the nuclei of epi-
thelia throughout the body (Fig. 3 and Fig. S2), including
those of the light organ, gills, gut and mantle. Unlike other
epithelia, where variation in staining intensity occurs
among cells, nuclear localization of EsPGRP1 in the
superficial ciliated epithelium of the light organ was
uniform from cell-to-cell in hatchling and in non-symbiotic
animals (n > 100 individuals, from the eggs of 10 different
females) (Fig. 3A–D). EsPGRP1 was detected at low
levels in the connective tissue matrix of the light organ,
and when present, appeared to be entirely non-nuclear
(Fig. 3B and D). The nuclear distribution of EsPGRP1 was
heterogeneous; anti-EsPGRP1 antibodies most intensely
stained subnuclear regions that were weakly labelled for
nucleic acids (Figs 3–6), suggesting that EsPGRP1 does
not directly bind to DNA. Pre-immune serum did not react
detectably with host tissues at comparable gain settings
on the confocal microscope (Fig. S3), and antibodies gen-
erated to the other three EsPGRPs did not show signifi-
cant reactivity in the nuclei (data not shown).
Absence of nuclear EsPGRP1 was correlated
with symbiosis-induced late-stage apoptosis in
the light organ
To understand the function of EsPGRP1 during coloniza-
tion, we first examined its localization at 24 h. All morpho-
genic processes of the superficial light-organ epithelia are
underway at this time. Whereas all epithelial cells of non-
symbiotic animals stained similarly, a subset of the epi-
thelial nuclei in 24-h symbiotic light organs did not stain for
EsPGRP1 (Fig. 4A). The EsPGRP1-negative nuclei con-
tained condensed chromatin, and cells with these pheno-
types occurred at a frequency similar to that observed
previously for apoptotic cells in the 24-h symbiotic light
organ (Foster et al., 1998; Koropatnick et al., 2004).
These data suggested that EsPGRP1-negative nuclei
occur in the cells undergoing apoptosis. This hypothesis
was supported by the finding that all nuclei undergoing
DNA degradation, that is, were TUNEL-positive, a char-
acteristic of the later stages of programmed cell death in
this system (Foster et al., 1998), also lacked EsPGRP1
(Fig. 4B). The reverse did not occur, i.e. only a subset
of EsPGRP1-negative nuclei was TUNEL-positive. These
observations provide evidence that nuclear loss of
EsPGRP1 occurs prior to late-stage apoptosis.
To determine the relationship of EsPGRP1 loss to the
integrity of the nucleus during apoptosis, we examined
EsPGRP1 localization in the context of known nuclear
apoptosis events. We observed an alternate intranuclear
distribution of EsPGRP1, in which the DNA appeared
condensed and EsPGRP1 appeared to fill the remainder
of the nucleoplasm (Fig. 4C). The alternate EsPGRP1/
DNA distribution was observed very rarely, suggesting
that this distribution is highly transient. In addition, we
investigated the temporal/spatial relationship between
EsPGRP1 and nucleoporins, which during apoptosis
undergo spatial redistribution in the nuclear membrane
and are eventually degraded (Kihlmark et al., 2001). In
EsPGRP1/nucleoporin costaining experiments in sym-
Fig. 2. Biochemical characteristics of the EsPGRP1 protein.
A. An alignment of the EsPGRP C-termini (arrow indicates
additional C-terminal sequence for EsPGRP3). Boxed area
highlights antigen sequence used for anti-EsPGRP1 antibody
generation. *, identical residues; :, conserved substitutions;
., semi-conserved substitutions.
B. A representative immunoblot of E. scolopes tissue. The
anti-EsPGRP1 antibody reacts with a major band in the aqueous
soluble fraction at ~24 kDa. C, cytoplasmic (aqueous) fraction;
M, membrane (SDS soluble) fraction.
C. Anti-FLAG immunoblot and companion gel of 24 mg crude
protein extract from untransfected Drosophila S2* cells and 256 ng
of an EsPGRP1-enriched fraction of S2* cells transfected with an
EsPGRP1-FLAG expression construct. The arrow indicates an
anti-FLAG reactive band at the expected molecular weight for
EsPGRP1.
D. An EsPGRP1-enriched EsPGRP1/S2* fraction degraded TCT in
vitro, but a protein extract of untransfected S2* cells had no effect
on TCT levels. Bars indicate mass of TCT (left axis) in each
reaction, and white peaks inside bars are the TCT peaks following
treatment as the absorbance at 215 nm on a reverse-phase HPLC
chromatograph (right axis). UnT/S2*, untransfected S2* cell extract;
EsPGRP1/S2*, EsPGRP1-enriched fraction of S2* cells transfected
with pJT28; mock rxn, TCT incubated in reaction buffer for 3 h;
TCT, 7 ml 1 mM TCT loaded directly onto HPLC column.
A nuclear PGRP in symbiont-induced development1117
© 2009 Blackwell Publishing Ltd, Cellular Microbiology, 11, 1114–1127

Page 6

biotic animals, EsPGRP1 was lost from the nucleus prior
to the redistribution and breakdown of the nucleoporins
(Fig. 4D). Taken together, these data suggest that nuclear
loss of EsPGRP1 rapidly follows chromatin condensation
and precedes the breakdown of the nuclear envelope,
linking nuclear loss of EsPGRP1 to the apoptotic pathway.
Timing of EsPGRP1 loss suggested a role in late-stage
apoptosis and morphogenesis
As described above, all TUNEL-positive nuclei were
EsPGRP1-negative while only a subset of EsPGRP1-
negative nuclei was TUNEL-positive (Fig. 4B), suggesting
that nuclear loss of EsPGRP1 occurs prior to late-stage
apoptosis. To characterize the timing of these processes,
we enumerated EsPGRP1-negative and TUNEL-positive
nuclei in anterior appendages from both symbiotic and
non-symbiotic light organs at the following time points: 8,
12, 13, 18 and 24 h post hatching (Fig. 5A and B). The
loss of nuclear EsPGRP1 was first detected in symbiotic
animals at 12 h, a time point coincident with the delivery
of the irreversible signal for morphogenesis. In addition,
we observed significantly greater numbers of EsPGRP1-
negative nuclei at 13 h, only 1 h after the morphogenesis
signal. An increased level of EsPGRP1-negative nuclei
was present in symbiotic animals at each time point up
to 24 h. In contrast, TUNEL-positive nuclei were first
detected in symbiotic light organs at 13 h, and greater
numbers were not observed until 18 h, followed by
another significant increase at 24 h. Greater numbers of
both EsPGRP1-negative (13–24 h) and TUNEL-positive
(18–24 h) nuclei were found in symbiotic compared with
non-symbiotic light organs, consistent with these events
being components of colonization-induced morphogenic
tissue destruction. Beyond 24 h, cells were rapidly lost
from the anterior appendage due to morphogenesis and
counts from these later time points cannot be compared
with the earlier time points. These data provide evidence
Fig. 3. Localization of EsPGRP1 in host tissue. Confocal
immunocytochemistry of hatchling and 24 h non-symbiotic
E. scolopes localized EsPGRP1 to epithelial nuclei.
A. A low-magnification image of the light organ. Anti-EsPGRP1
antibodies intensely labelled the ciliated epithelial fields on the
lateral surfaces of the light organ.
B–D. Representative higher-magnification images displaying the
anterior appendage and a portion of the more medial ciliated field
(dashed box in A). The nuclei of these epithelial cells (aa and cr)
labelled strongly with the anti-EsPGRP1 antibody (green arrows),
whereas non-epithelial nuclei of the underlying tissues (ctm)
appeared devoid of EsPGRP1 (white arrowheads).
E. Epithelia across the body contain nuclear EsPGRP1. A
representative image taken from gill tissue is shown, note
variations of staining intensity in the nuclei (white arrows). aa,
anterior appendage; cr, ciliated ridge; ctm, connective tissue matrix;
hg, hindgut; p, pores; pa, posterior appendage. EsPGRP1, green;
DNA, red; actin cytoskeleton, blue.
1118J. V. Troll et al.
© 2009 Blackwell Publishing Ltd, Cellular Microbiology, 11, 1114–1127

Page 7

Fig. 4. Loss of EsPGRP1 during progression through apoptosis. Confocal microscopy of symbiotic light organs revealed nuclear loss of
EsPGRP1 and entry into late-stage apoptosis in anterior appendage epithelial nuclei.
A. Micrographs of an optical section of the anterior appendage: individual channels on the left, merged image on the right. Colonization of the
light organ by V. fischeri for 24 h induces a subset of nuclei in the ciliated epithelial fields to lose EsPGRP1 staining (e.g. white arrowheads).
B. High-magnification micrographs of epithelial cells from the anterior appendage. TUNEL staining of cleaved DNA only occurs in nuclei that
have lost EsPGRP1 staining (yellow arrows), but not all EsPGRP1-negative nuclei are TUNEL-positive (white arrowhead). Dashed box in inset
indicates area of the anterior appendage shown in high magnification.
C. Micrograph of an anterior appendage. Dashed box indicates area of high magnification in right panel that displays a transient alteration of
intranuclear (based on size and shape) localization of EsPGRP1 staining after chromatin condensation but prior to the loss of EsPGRP1
signal from the nucleus (yellow arrowhead) as seen in (A). Note also that haemocytes (h), which have migrated to the light organ blood sinus,
fail to stain with EsPGRP1. Colour palette is the same as in (B): EsPGRP1, green; DNA, red; TUNEL, blue.
D. High-magnification micrographs of anterior appendage epithelial cells from non-symbiotic and symbiotic animals. Labelling of the
nucleoporins (red) with the monoclonal Ab 414 indicates that loss of EsPGRP1 from the nucleus (blue) occurs prior to the dissolution of the
nuclear membrane (red arrowheads). Inset in sym panel exhibits an EsPGRP1-negative nucleus with complete breakdown of the nuclear
envelope.
A nuclear PGRP in symbiont-induced development 1119
© 2009 Blackwell Publishing Ltd, Cellular Microbiology, 11, 1114–1127

Page 8

that nuclear loss of EsPGRP1 precedes entry into late-
stage apoptosis.As it is not possible to follow an individual
cell through the entire process, it cannot be ruled out that
a subset of EsPGRP1-negative nuclei never become
TUNEL-positive.
Because the onset of nuclear-EsPGRP1 loss coincided
with the delivery of an irreversible morphogenic signal by
V. fischeri at 12 h (Doino et al., 1995), we hypothesized
that the EsPGRP1 phenotype was a component of the
host response to this morphogenesis signal. To test this
hypothesis, we performed ‘curing’ experiments to deter-
mine whether EsPGRP1 loss is also a component of the
irreversible morphogenic programme. In these experi-
ments, symbiotic juvenile E. scolopes were treated with
the antibiotic chloramphenicol to clear the crypts of sym-
bionts either before or after the 12 h time point and were
subsequently assayed at 24 h (Fig. 5C). Previous experi-
ments showed that only a small proportion of animals
Fig. 5. The timing of EsPGRP1 nuclear loss. EsPGRP1 loss occurs before entry into late-stage apoptosis and is triggered by the irreversible
morphogenesis signal.
A. Representative confocal micrographs from a time-course of 8–24 h symbiotic and non-symbiotic hatchlings. EsPGRP1, green; DNA, red;
TUNEL, blue.
B. Direct-count estimates of EsPGRP1-negative nuclei (black bars) and TUNEL-positive nuclei (grey bars) per anterior appendage (n = 15).
Statistical analysis included ANOVA with Tukey’s pairwise comparisons to identify statistically significant differences between time points (group
error rate P ? 0.05). Symbols represent groupings that are statistically different from other groups. Error bars represent standard error.
C. A qualitative comparison of light organs from a ‘curing experiment.’ Symbiotic hatchlings were treated at 8.5 or 14 h with 20 mg ml-1
chloramphenicol and scored for EsPGRP1-negative and TUNEL-positive nuclei in the anterior appendage at 24 h. The 95% confidence
intervals for EsPGRP1-negative and TUNEL-positive nuclei were determined for 24 h symbiotic animals and the values for each animal from
the two cured and the non-symbiotic groups were compared with the 24 h sym CI. If the number of EsPGRP1-negative or TUNEL-positive
nuclei in a given anterior appendage fell within the 24 h sym CI, then that animal was considered ‘sym-type’ for that category (protein
localization or cell-death).
1120 J. V. Troll et al.
© 2009 Blackwell Publishing Ltd, Cellular Microbiology, 11, 1114–1127

Page 9

cured before 12 h progressed through morphogenesis,
whereas 100% of animals cured at 12 h or later under-
went development (Doino et al., 1995). Light organs cured
of V. fischeri after 12 h were similar to fully symbiotic light
organs with respect to nuclear loss of EsPGRP1 and entry
into late-stage apoptosis. In contrast, 100% of light organs
cured before 12 h were protected from cell death and 80%
were protected from loss of EsPGRP1. These data impli-
cate nuclear loss of EsPGRP1 in symbiotic light organs as
a component of the developmental response to the irre-
versible morphogenic signal delivered by V. fischeri.
TCT induced nuclear loss of EsPGRP1 in
light-organ tissues
Because TCT and LPS together irreversibly signal mor-
phogenesis at 12 h (Koropatnick et al., 2004), we per-
formed experiments to determine whether TCT and LPS
are sufficient to trigger EsPGRP1 loss and entry into
late-stageapoptosis.Wetooktwoapproaches:(i)pharma-
cological treatment with TCT and/or LPS (or its active
component lipid A) and (ii) colonization with V. fischeri
mutants defective in TCT release. While treatment with
V. fischeri LPS had no effect on either the nuclear localiza-
tion of EsPGRP1 or late-stage apoptosis, the presence of
10 mM TCT induced a level of nuclear loss of EsPGRP1
closetothatcharacteristicofsymbiotictissues(Fig. 6Aand
B). Unlike regression and haemocyte trafficking (Koropat-
nick et al., 2004), EsPGRP1 loss was not increased by the
addition LPS/lipid A with TCT (Fig. 6B). Because PGN
monomers were not previously reported to induce apopto-
sis in the squid light organ, we postulated that a small
amount of LPS contaminating the filter-sterilized Instant
Ocean(FSIO)couldhavebeensynergizingwiththeTCTto
Fig. 6. TCT induction of EsPGRP1 loss and late-stage apoptosis.
A. Representative confocal micrographs of E. scolopes epithelia. Hatchling E. scolopes were exposed to 10 ng ml-1LPS and/or 10 mM TCT
and assayed for EsPGRP1-negative and TUNEL-positive nuclei at 24 h. TCT induces EsPGRP1-negative (white arrowheads) and
TUNEL-positive (yellow arrows) nuclei. EsPGRP1, green; DNA, red; TUNEL, blue.
B. Direct-count estimates of EsPGRP1-negative and TUNEL-positive nuclei in TCT/LPS-treated light organs (n = 10). Symbols represent
groupings that are statistically different from other groups.
C. Direct-count estimates of EsPGRP1-negative and TUNEL-positive nuclei in light organs colonized by V. fischeri mutants defective
in TCT secretion. Dltg represents mutant strain DMA388, which is defective for three lytic transglycosylases (DltgA, DltgD, ltgY::erm).
Complementation strains carry indicated gene on the plasmid pVSV107 (Dunn et al., 2006). Symbols represent groupings that are statistically
different from non-symbiotic light organs († represents an intermediate level of EsPGRP1 loss that is different from both sym and non-sym).
For both (B) and (C), Statistical analysis included ANOVA with Tukey’s pairwise comparisons to identify statistically significant differences
between time points (group error rate P ? 0.05). Error bars represent standard error.
A nuclear PGRP in symbiont-induced development 1121
© 2009 Blackwell Publishing Ltd, Cellular Microbiology, 11, 1114–1127

Page 10

induce the levels EsPGRP1 loss and TUNEL-positive
nuclei observed in all TCT treatments. Alternatively,
because EsPGRP1 directly interacted with PGN in vitro
(Fig. 2), it remained possible that this phenotype was
distinct and dependent only on the presence of PGN
fragments, such as TCT. To address this matter, we
included a control using a polyclonal anti-LPS antibody to
inhibit any effects from contaminating LPS as previously
described (Koropatnick et al., 2004). The anti-LPS anti-
body had no statistically significant effect on either nuclear
loss of EsPGRP1 or the induction of TUNEL (Fig. 6B),
suggesting that all of the effects could be attributed to the
activity of TCT. A mutant of V. fischeri lacking three lytic-
transglycosylase genes (Dltg), which is unable to release
TCTinvitro(Adinetal.,2009),didnotinducefullEsPGRP1
and TUNEL phenotypes by 24 h when compared with
wild-type colonized hatchlings (Fig. 6C). Some residual
activity was observed in light organs colonized by the Dltg
mutant, which is likely due to PGN release during normal
symbiont cell turnover. Partial complementation of the
triplemutantwitheitheroftwosinglelytic-transglycosylase
genes in trans restored phenotypes comparable to wild-
type colonized animals. The genetic data, coupled with
pharmacological application of TCT, conclusively demon-
strate that PGN monomers are necessary for both normal
localization changes of EsPGRP1 and induction of late-
stage apoptosis in the morphogenesis programme.
The TCT-induced loss of nuclear EsPGRP1 and late-
stage apoptosis was limited to the light-organ epithelia. In
addition to light-organ tissues, we also examined the epi-
thelia and connective tissues of the gut, gills and epider-
mis. In the epithelia of these three tissues, neither TCT
treatment nor symbiosis was observed to have any effect
on either the nuclear localization of EsPGRP1 or the
induction of apoptosis (Fig. 6A and Fig. S2). We observed
an average of four TUNEL-positive epithelial cells per
hindgut of non-symbiotic animals (n = 8). Comparable
levels of TUNEL-positive epithelial nuclei were found in
hindguts from TCT-treated animals (n = 7). Every TUNEL-
positive epithelial nucleus lacked staining for EsPGRP1, a
characteristic that contrasted with nuclei from neighbour-
ing non-apoptotic gut epithelial cells, which strongly
stained for EsPGRP1. These data show that TCT is a
specific morphogen of the light organ and that other
tissues are insensitive to its effects. In addition, it appears
that nuclear loss of EsPGRP1 is a common component of
the apoptotic pathway in E. scolopes epithelial cells
whether they are responding to TCT or some alternate
apoptotic stimulus.
Discussion
In this study, we characterized the cellular dynamics of
EsPGRP1 in symbiont-induced tissue destruction in the
light organ of E. scolopes, a process in which apoptosis is
a hallmark character (Fig. 1C). We provide evidence that:
(i) EsPGRP1 is a nuclear protein of the E. scolopes epi-
thelia; (ii) in the light organ, the nuclear localization of
EsPGRP1 is lost during the course of symbiont-induced
morphogenesis; (iii) the alteration of EsPGRP1 localiza-
tion is morphologically, temporally and spatially correlated
with programmed cell death as a component of light-
organ development; and (iv) nuclear loss of EsPGRP1
and subsequent entry into late-stage apoptosis are
induced by the light-organ morphogen, TCT, implicating
EsPGRP1 dynamics in light-organ development.
The nuclear localization of EsPGRP1 is unprecedented
and, to our knowledge, no other PGRPs have been
reported to localize to the nucleus. While the EsPGRP1
protein contains no obvious nuclear localization signals,
some proteins with molecular weights less than ~50 kDa
can passively diffuse into the nucleus (Talcott and Moore,
1999). Passive diffusion alone would likely result in an
even distribution of protein between the cytoplasm and
the nucleus. However, EsPGRP1 concentrates in the
nucleus relative to the cytoplasm. This uneven distribu-
tion suggests that some mechanism other than passive
diffusion must also contribute to the cellular localization
of EsPGRP1; for example, binding to nuclear ele-
ments may be responsible for sequestering EsPGRP1, or
EsPGRP1 may specifically bind to an unknown protein
with a strong nuclear localization signal resulting in
nuclear import.
Multiple lines of evidence tie the nuclear localization
dynamicsofEsPGRP1to
EsPGRP1 was lost from nuclei with morphologies consis-
tent with apoptosis, and these EsPGRP1-negative nuclei
occurred with a frequency similar to that observed for
apoptotic cells during light-organ morphogenesis (Foster
et al., 1998; 2000; Koropatnick et al., 2004). All TUNEL-
positive nuclei were EsPGRP1-negative and loss of
nuclear EsPGRP1 occurred shortly after chromatin con-
densation, and prior to the re-distribution and breakdown
of nucleoporins (Fig. 4). The timing of EsPGRP1 loss from
the nucleus was consistent with that reported for apopto-
sis in the light organ. Chromatin condensation peaks
between 12 and 14 h (Foster et al., 1998) and the nuclear
loss of EsPGRP1 was first observed to increase signifi-
cantly in this time frame (Fig. 5). Further, we observed the
onset of nuclear loss of EsPGRP1 consistently ahead of
the onset of DNA cleavage (TUNEL), suggesting that
removal of EsPGRP1 is permissive to apoptotic progres-
sion (Fig. 5). The timing of DNA cleavage observed in this
study, using a fluorometric TUNEL assay, is consistent
with the previously published timing of DNA cleavage and
late-stage apoptosis in the light organ based on observa-
tions of cellular morphology and colorimetric TUNEL
assays in histological sections (Montgomery et al., 1994;
theapoptoticprocess.
1122J. V. Troll et al.
© 2009 Blackwell Publishing Ltd, Cellular Microbiology, 11, 1114–1127

Page 11

Foster et al., 1998). Thus, the timing described here is
unlikely to be due to differences in sensitivity of the
TUNEL and EsPGRP1 nuclear loss assays. We conclude
that nuclear loss of EsPGRP1 occurs during the apoptotic
process, following chromatin condensation and preced-
ing reorganization of the nuclear envelope and DNA
cleavage.
An association between PGRPs and apoptosis has
begun to gain recognition. Genetic studies of PGRPs
have revealed that they can both activate and be acti-
vated by the transcription factor NF-kB, a known regulator
of apoptosis (Beg and Baltimore, 1996; Michel et al.,
2001; Kaneko et al., 2004; Lang et al., 2008). Drosophila
PGRP-LF mutants are developmentally defective in wing
formation, including the aberrant appearance of apoptotic
cells in the wing imaginal discs (Maillet et al., 2008). The
correlation of EsPGRP1 activity and apoptosis within
E. scolopes epithelial cells further supports the relation-
ship between the PGRP family of innate immunity proteins
and animal development through control and/or inter-
action with apoptosis pathways.
In experiments with bacterial cell envelope components
known to induce morphogenesis in the light organ, TCT
was capable of inducing a full EsPGRP1 response with
little or no contribution from lipid A or LPS. However,
treatment with TCT or TCT + LPS failed to induce the full
apoptotic response that is induced by V. fischeri, that is,
including late-stage apoptosis as visualized by TUNEL
(Fig. 6B). Earlier studies of apoptosis induction by
TCT + LPS used an early-stage cell-death indicator, chro-
matin condensation visualized by acridine orange staining
(Koropatnick et al., 2004). In that study, a dose–response
analysis was performed to determine the concentrations
of TCT + LPS that most closely mimicked the onset of cell
death by the symbiont, which provided the basis for con-
centrations used in the present study. Thus, it is likely that
the failure of the TCT + LPS to induce the full programme
is not due to their concentration. However, subtle differ-
ences may exist in the way these molecules are pre-
sented by intact bacteria compared with pharmacological
treatment, such that purified cell envelope constituents
are fully capable of inducing early-stage, but not late-
stage, apoptosis. Alternatively, the symbionts may be pro-
viding a third morphogenic signal, which complements
TCT + LPS under natural conditions. Candidates for addi-
tionalmorphogenicsignalsincludeothercellsurfacemole-
cules of the symbiont, such as the flagellins and the Syp
exopolysaccharide, both of which are important early in
colonization (Millikan and Ruby, 2004; Yip et al., 2005).
Symbiont luminescence, another possible signal, has
been shown to have specific effects on morphogenesis.
Bioluminescence is induced following colonization of the
deep crypts. V. fischeri luxA mutants, which fail to produce
light, are attenuated in their ability to induce host
haemocyte trafficking (Koropatnick et al., 2007) and cell
swelling (Visick et al., 2000), and are delayed in regres-
sion (T.A. Koropatnick and M.J. McFall-Ngai, unpubl.
data). It is intriguing to speculate that the slowed morpho-
genesis in animals colonized by luxA mutants could be
due to a reduction in the speed of apoptosis due to the
lack of symbiotic bioluminescence. Future studies will
investigate whether dark mutants of V. fischeri are also
deficient in their induction of the nuclear loss of EsPGRP1
and late-stage apoptosis.
Microarray studies of the onset of the symbiosis
revealed that EsPGRP1 mRNA levels are increased at
18 h following inoculation in animals colonized by V. fis-
cheri, that is, concurrent with the loss of EsPGRP1 from
the nuclei of the superficial epithelial field. In addition,
luxA mutants fail to induce the full increase in EsPGRP1
transcript levels characteristic of colonization with wild-
type V. fischeri (Chun et al., 2008). The findings that the
luxA mutants are delayed in regression and show lower
levels of EsPGRP1 mRNA suggest that these symbiont-
triggered increases in EsPGRP1 gene transcription are
important for the induction of full morphogenesis. How this
upregulation interfaces with the eventual loss of the
protein from the nuclei of the cells of the superficial epi-
thelial field remains to be determined.
It is unclear whether the activity of TCT is the result
of a direct interaction with EsPGRP1 or works through
an intermediary receptor/signal transduction pathway.
Genetic evidence from D. melanogaster suggests that
TCT can be transported across host cell membranes
(Kaneko et al., 2006), so it is conceivable that TCT could
gain access to the cytoplasm and the nucleus, via passive
diffusion, of light-organ epithelial cells. We attempted to
determine the location of TCT in host tissue through a
suite of approaches including confocal microscopy of
hatchling E. scolopes treated with FITC-labelled TCT
(data not shown) (Flak and Goldman, 1998) and Nano-
SIMS tracing of stable isotope labelled TCT (Fig. S4)
(Lechene et al., 2006). To our knowledge, the localization
of TCT in host cells has not been accomplished in any
system. In the present study, we were unable to detect
TCT associated with the host tissue. However, apoptotic
cells in tissues that are unresponsive to TCT undergo
nuclear loss of EsPGRP1, suggesting that this process is
a component of apoptosis, independent of TCT induction.
Thus, it seems unlikely that EsPGRP1 is responding
directly to TCT. EsPGRP4 is an obvious candidate for a
TCT-receptor because this protein is expressed in the
light organ, contains a putative transmembrane domain,
and topology prediction programmes suggest that the
PGRP domain would be localized on the extracellular
membrane surface (Goodson et al., 2005). However, the
cytoplasmic domain of this putative integral membrane
protein is small, providing few clues as to possible down-
A nuclear PGRP in symbiont-induced development1123
© 2009 Blackwell Publishing Ltd, Cellular Microbiology, 11, 1114–1127

Page 12

stream signalling pathways that could potentially result in
activation of apoptosis and morphogenesis. Recently,
an intranuclear bacterial pathogen of bivalve mollusks,
Candidatus Endonucleobacter bathymodioli, was char-
acterized (Zielinski et al., 2009), suggesting a selection
pressure that could explain the nuclear localization of
EsPGRP1. It is easy to envision that rapidly inducing
apoptosis upon intranuclear invasion by a pathogen
would limit the infectious potential of such a pathogen by
inhibiting its access to the energy-rich molecules of the
nucleus. In such a scenario, it would seem likely that this
immune associated apoptosis was co-opted for develop-
mental regulation during the establishment of the squid/
vibrio mutualism.
Whereas this is the first report of a nuclear PGRP
associated with apoptosis, we expect that this pheno-
menon is not unique. An interesting avenue for further
research will be the determination of the mechanistic link
to apoptosis. We predict that future studies of intracellular
localization of PGRPs among the marine invertebrates
will reveal other nuclear members of the PGRP protein
family. Further, while the study of PGRPs has primarily
focused on their roles in ameliorating pathogenesis, we
demonstrate a role for these proteins in mutualistic asso-
ciations of animals with their bacterial partners.
Experimental procedures
General methods
Unless otherwise noted, all chemicals were purchased from
Sigma-Aldrich (St. Louis, MO).
Adult E. scolopes were caught in shallow coastal waters of
Oahu and maintained and bred in the lab as previously described
(Montgomery and McFall-Ngai, 1993). Juvenile squid used in
experiments were collected within 10 min of hatching, washed
three times in FSIO, and placed in scintillation vials containing
either FSIO (non-symbiotic) or FSIO + 5 ¥ 103cfu ml-1of V. fis-
cheri ES114 grown in LBS or SWT broth at 28°C (symbiotic). The
Dltg mutant (ES114-derivative strain DMA388: DltgA, DltgD,
ltgY::erm) and its derivatives were used at 2 ¥ 103cfu ml-1(Adin
et al., 2009). In these experiments, ES114 was used at identical
concentrations to V. fischeri mutants. All animal experiments
were carried out on a 12 h light/dark cycle and colonization by
V. fischeri was monitored by following luminescence using a
TD-20/20 luminometer (Turner Designs, Sunnyvale, CA) (Ruby
and Asato, 1993).
In experiments in which animals were cured of their symbionts,
symbiotic juveniles were exposed to 20 mg ml-1of chlorampheni-
col in FSIO beginning at 8.5 or 14 h post colonization (Doino
et al., 1995) and were scored at 24 h post colonization. Non-
symbiotic juveniles exposed to chloramphenicol were also analy-
sed as a control for any effects of the antibiotic on EsPGRP1. For
experiments involving bacterial cell envelope components, juve-
nile E. scolopes were exposed to 10 mM TCT and/or 10 ng ml-1of
either V. fischeri LPS or lipid A in FSIO for 24 h. Cell envelope
componentswerepurified from
described (Cookson et al., 1989; Apicella et al., 1994).
V. fischerias previously
Antibody generation and western blots
Rabbit polyclonal antibodies were raised against an EsPGRP1
synthetic peptide conjugated to ovalbumin (Harlan Biosciences,
Indianapolis, IN). The synthetic peptide of 13 amino acids,
WSHYGLRNHNKSA, was selected from the C-terminal region of
EsPGRP1 because it is predicted to be highly antigenic (Laser-
gene Protean, DNA Star, Madison, WI) and because it contains
little similarity to the other EsPGRPs (Fig. 2A and Fig. S1) or
other sequences in the light-organ EST database. Upon western
blot analysis to test the specificity of the antisera for the
EsPGRP1 antigen, squid tissues were homogenized in 20 mM
sodium phosphate pH 7.4, 100 mM sodium chloride. The
aqueous soluble or cytosolic fraction was obtained as the super-
natant of a centrifugation of the sample at 14 000 g for 15 min.
The pellet was then washed three times with the homogenization
buffer and the membranes were extracted with homogenization
buffer containing 1% SDS. Protein concentrations were deter-
mined using the Bradford assay. The proteins were then sepa-
rated on 12.5% SDS-PAGE gel and blotted onto a PVDF
membrane (Bio-Rad, Hercules, CA). The blot was blocked
overnight in 20 mM Tris, pH 7.4, 300 mM sodium chloride,
0.5% bovine serum albumin (BSA) and 1% goat serum
(Jackson ImmunoResearch Laboratories, West Grove, PA). Anti-
EsPGRP1 antibody was diluted 1:100 in the block solution and
incubated with the blot overnight at room temperature. The blot
was then incubated with the secondary antibody, a goat anti-
rabbit horseradish peroxidase-conjugate (Jackson Immuno-
Research Laboratories, West Grove, PA) at a 1:3000 dilution.
Reactivity was visualized using Pierce ECL Chemiluminescent
western blot kit (Pierce, Rockford, IL).
TCT degradation assays
The EsPGRP1 coding sequence was PCR amplified with a
C-terminal FLAG-tag using the following DNA primers: JT73,
ACAGTGGATCCATGCACCATCACCATCACCATATGCACGCG
TTCGGAG; and JT74, ACTGTGCGGCCGCTTATTTATCGTCAT
CGTCTTTGTAGTCACCAATAATGGCACTTT. The PCR product
was subsequently cloned into the pPAC-PL plasmid using
the BamHI and NotI restriction enzymes. Plasmid pJT28 was
constructed by subcloning the FLAG-tagged EsPGRP1 coding
sequence into the pRmHa-3 expression plasmid downstream of a
Cu2+-inducible metallothionien promoter by EcoRI and SalI
restriction of the PCR product of the following primers: JT136,
TACTTGAATTCATGCACCATCACCATCACCATATGCACGCGT
TCG; and JT134, ACTCTGTCGACTTATTTATCGTCATCGTCT
TTGTAGTC. Drosophila S2* cells were stably transfected with
pJT28 as previously described (Silverman et al., 2000).
EsPGRP1 protein was induced by addition of 500 mM CuSO4into
the media for 8–12 h. Cells were then harvested, washed with
PBS and frozen at -80°C.
The Cu2+-induced pJT28/S2* cells were resuspended in lysis
buffer (20 mM Tris, pH 7.4, protease inhibitor cocktail III and
1 mM PMSF) and incubated on ice for 10 min. Sodium chloride
was then added to a final concentration of 100 mM and the
insoluble material was separated by centrifugation at 20 000 g for
10 min at 4°C. The supernatant was passed over an anti-FLAG
antibody affinity column three times and the column was subse-
quently washed with 500 column volumes of TBS (20 mM Tris,
pH 7.4, 150 mM Sodium chloride). Bound proteins were eluted
1124J. V. Troll et al.
© 2009 Blackwell Publishing Ltd, Cellular Microbiology, 11, 1114–1127

Page 13

from the column with 100 mM FLAG-peptide. Fractions containing
high levels of the EsPGRP1 protein were determined by anti-
FLAG immunoblots. These fractions were combined and dialysed
against storage buffer (20 mM Tris, pH 7.4, 300 mM Sodium
chloride, 1 mM DTT and 50% glycerol) and stored at -20°C.
Anti-FLAG immunoblots were performed as above except using
4% non-fat dry milk/TBS as the blocking solution and a 1:1000
dilution of rabbit polyclonal anti-FLAG antibodies for 1 h at room
temperature.
To perform the TCT-degradation assay, ~370 ng of the FLAG-
purified EsPGRP1 protein fraction was diluted into 500 ml of
reaction buffer (50 mM ammonium acetate, pH 7, 550 mM
sodium chloride, 4 mM ZnSO4and 1% BSA) and concentrated to
9.2 ng ml-1on a 3-kDa-molecular-weight cut-off microcon column
(Millipore, Billerica, MA). Sixty-four nanograms of the EsPGRP1
protein fraction was mixed 1:1 with 1 mM TCT. The reaction was
incubated at room temperature for 1 h, diluted with 86 ml HPLC-
grade water and loaded onto a Betasil C18 reverse phase
column (Thermo Hypersil-Keystone, Bellefonte, PA). The TCT
was mobilized using a 0–15% acetonitrile gradient and detected
by absorbance at 215 nm. The crude S2* cell protein extract was
generated as above, excluding the anti-FLAG affinity fraction-
ation step. To control for S2* cell-derived amidase activity, which
could be contaminating the EsPGRP1-enriched protein fraction,
24 mg of the crude S2* cell protein extract was incubated with
TCT for 2 h.
Immunocytochemical detection of EsPGRP1 in
E. scolopes tissues
Unless otherwise noted, all fluorophores were obtained from
Molecular Probes (Invitrogen, Carlsbad, CA). To prepare juvenile
squid for immunocytochemistry, they were anaesthetized in
FSIO + 2% ethanol and fixed with 4% paraformaldehyde in
mPBS (50 mM sodium phosphate, pH 7.4, 0.45 M sodium chlo-
ride) for 18 h at 4°C. Post-fix samples were kept at 4°C through-
out the remainder of the protocol. Fixed squid were washed four
times for 30 min in mPBS. The light organs were removed from
each juvenile, permeabolized with 1% Triton-X-100 in mPBS
(mPBST) for 2 days and then blocked overnight (mPBST, 0.5%
BSA, 1% goat serum). Light organs were then exposed to the
EsPGRP1 antisera diluted 1:1000 in the block solution for 7 days.
Light organs were washed four times for 1 h each wash with
mPBST and incubated overnight again in blocking solution.
Samples were then exposed overnight to fluorescent goat anti-
rabbit antibody conjugates (FITC or TRITC, Jackson ImmunoRe-
search Laboratories; AlexaFluor 633) diluted 1:25 in blocking
solution overnight. To counterstain the actin cytoskeleton, light
organs were incubated with 165 nM Alexa Fluor 633-phalloidin in
mPBST overnight. To counterstain nuclear DNA, light organs
were incubated with 100 mg ml-1RNaseA in 2¥ SSC (30 mM
sodium citrate, pH 7.2, 300 mM sodium chloride) for 30 min at
37°C and then stained with either 1.5 mM propidium iodide for
5 min or 2 mM TOTO-3 for 30 min at room temperature.Apoptotic
chromosomal cleavage was detected using the DeadEnd Fluori-
metric TUNEL Assay kit (Promega, Madison, WI) according to
manufacturer’s instructions. The nuclear pore complex was
visualized by immunocytochemistry as above, except the anti-
nucleoporin mAb 414 (Covance Research Products, Denver, PA)
was co-diluted 1:1000 with the anti-EsPGRP1 Ab. Light organs
were mounted in Vectashield anti-bleaching medium (Vector
Laboratories, Burlingame, CA) and viewed on an LSM510
laser scanning confocal microscope (Carl Zeiss Microimaging,
Thornwood, NY).
Acknowledgements
We are grateful to Michael Apicella for kindly providing the V. fis-
cheri LPS and lipid A used in this study. We thank N. Bekiares,
C. Brennan, J. Dillard, H. Goodrich-Blair, L. Knoll, M. Mandel, B.
Rader and E. Ruby for helpful comments on the manuscript. This
work was funded by NIH RO1-AI50661 to M.M.N., NSF IOS
0817232 to M.M.N. and E.G. Ruby, NIH RR R01-12294 to E.G.
Ruby, and by NIH NRSA AI55397 to J.V.T. through the Microbial
Pathogenesis and Host Responses Training Program.
References
Adin, D.M., Engle, J.T., Goldman, W.E., McFall-Ngai, M.J.,
and Stabb, E.V. (2009) Mutations in ampG and lytic trans-
glycosylase genes affect the net release of peptidoglycan
monomers from Vibrio fischeri. J Bacteriol 191: 2012–2022.
Anselme, C., Perez-Brocal, V., Vallier, A., Vincent-Monegat,
C., Charif, D., Latorre, A., et al. (2008) Identification of the
weevil immune genes and their expression in the bacteri-
ome tissue. BMC Biol 6: 43.
Anselme, C., Vallier, A., Balmand, S., Fauvarque, M.O., and
Heddi, A. (2006) Host PGRP gene expression and bacte-
rial release in endosymbiosis of the weevil Sitophilus
zeamais. Appl Environ Microbiol 72: 6766–6772.
Apicella, M.A., Griffiss, J.M., and Schneider, H. (1994) Isola-
tion and characterization of lipopolysaccharides, lipooli-
gosaccharides, and lipid A. Meth Enzymol 235: 242–252.
Beg, A.A., and Baltimore, D. (1996) An essential role for
NF-kBin preventingTNF-alpha-induced
Science 274: 782–784.
Bischoff, V., Vignal, C., Duvic, B., Boneca, I.G., Hoffmann,
J.A., and Royet, J. (2006) Downregulation of the Droso-
phila immune response by peptidoglycan-recognition pro-
teins SC1 and SC2. PLoS Pathog 2: e14.
Chang, C.I., Chelliah, Y., Borek, D., Mengin-Lecreulx, D., and
Deisenhofer, J. (2006) Structure of tracheal cytotoxin in
complex with a heterodimeric pattern-recognition receptor.
Science 311: 1761–1764.
Chun, C.K., Troll, J.V., Koroleva, I., Brown, B., Manzella, L.,
Snir, E., et al. (2008) Effects of colonization, luminescence,
and autoinducer on host transcription during development
of the squid–vibrio association. Proc Natl Acad Sci USA
105: 11323–11328.
Cloud, K.A., and Dillard, J.P. (2002) A lytic transglycosylase
of Neisseria gonorrhoeae is involved in peptidoglycan-
derived cytotoxin production. Infect Immun 70: 2752–2757.
Cloud-Hansen, K.A., Peterson, S.B., Stabb, E.V., Goldman,
W.E., McFall-Ngai, M.J., and Handelsman, J. (2006)
Breaching the great wall: peptidoglycan and microbial
interactions. Nat Rev Microbiol 4: 710–716.
Cookson, B.T., Cho, H.L., Herwaldt, L.A., and Goldman, W.E.
(1989) Biological activities and chemical composition of
purified tracheal cytotoxin of Bordetella pertussis. Infect
Immun 57: 2223–2229.
Coteur, G., Mellroth, P., De Lefortery, C., Gillan, D., Dubois,
P., Communi, D., and Steiner, H. (2007) Peptidoglycan
celldeath.
A nuclear PGRP in symbiont-induced development1125
© 2009 Blackwell Publishing Ltd, Cellular Microbiology, 11, 1114–1127

Page 14

recognition proteins with amidase activity in early deuteros-
tomes (Echinodermata). Dev Comp Immunol 31: 790–804.
Doino, J.A., and McFall-Ngai, M.J. (1995) A transient expo-
sure to symbiosis-competent bacteria induces light organ
morphogenesis in the host squid. Biol Bull 189: 347–355.
Dunn, A.K., Millikan, D.S., Adin, D.M., Bose, J.L., and Stabb,
E.V. (2006) New rfp- and pES213-derived tools for analyz-
ing symbiotic Vibrio fischeri reveal patterns of infection and
lux expression in situ. Appl Environ Microbiol 72: 802–810.
Dziarski, R., and Gupta, D. (2006a) Mammalian PGRPs:
novel antibacterial proteins. Cell Microbiol 8: 1059–1069.
Dziarski, R., and Gupta, D. (2006b) The peptidoglycan rec-
ognition proteins (PGRPs). Genome Biol 7: 232.
Dziarski, R., Platt, K.A., Gelius, E., Steiner, H., and Gupta, D.
(2003) Defect in neutrophil killing and increased suscepti-
bility to infection with nonpathogenic gram-positive bacteria
in peptidoglycan recognition protein-S (PGRP-S) -deficient
mice. Blood 102: 689–697.
Filipe, S.R., Tomasz, A., and Ligoxygakis, P. (2005) Require-
ments of peptidoglycan structure that allow detection by
the Drosophila Toll pathway. EMBO Rep 6: 327–333.
Flak, T.A., and Goldman, W.E. (1998) Muramyl peptide
probes derived from tracheal cytotoxin of Bordetella per-
tussis. Anal Biochem 264: 41–46.
Foster, J.S., Apicella, M.A., and McFall-Ngai, M.J. (2000)
Vibrio fischeri lipopolysaccharide induces developmental
apoptosis, but not complete morphogenesis, of the
Euprymna scolopes symbiotic light organ. Dev Biol 226:
242–254.
Foster, J.S., and McFall-Ngai, M.J. (1998) Induction of apo-
ptosis by cooperative bacteria in the morphogenesis of
host epithelial tissues. Dev Genes Evol 208: 295–303.
Gelius, E., Persson, C., Karlsson, J., and Steiner, H. (2003)
A mammalian peptidoglycan recognition protein with
N-acetylmuramoyl-L-alanine amidase activity. Biochem
Biophys Res Commun 306: 988–994.
Goodson, M.S., Kojadinovic, M., Troll, J.V., Scheetz, T.E.,
Casavant, T.L., Soares, M.B., and McFall-Ngai, M.J.
(2005) Identifying components of the NF-kB pathway in the
beneficial Euprymna scolopes-Vibrio fischeri light organ
symbiosis. Appl Environ Microbiol 71: 6934–6946.
Gupta, D. (2008) Peptidoglycan
maintaining immune homeostasis and normal develop-
ment. Cell Host Microbe 3: 273–274.
Kaneko, T., Goldman, W.E., Mellroth, P., Steiner, H., Fukase,
K., Kusumoto, S., et al. (2004) Monomeric and polymeric
gram-negative peptidoglycan but not purified LPS stimu-
late the Drosophila IMD pathway. Immunity 20: 637–649.
Kaneko, T., Yano, T., Aggarwal, K., Lim, J.H., Ueda, K.,
Oshima, Y., et al. (2006) PGRP-LC and PGRP-LE have
essential yet distinct functions in the drosophila immune
response to monomeric DAP-type peptidoglycan. Nat
Immunol 7: 715–723.
Kihlmark, M., Imreh, G., and Hallberg, E. (2001) Sequential
degradation of proteins from the nuclear envelope during
apoptosis. J Cell Sci 114: 3643–3653.
Koropatnick, T.A., Engle, J.T., Apicella, M.A., Stabb, E.V.,
Goldman, W.E., and McFall-Ngai, M.J. (2004) Microbial
factor-mediated development in a host-bacterial mutual-
ism. Science 306: 1186–1188.
Koropatnick, T.A., Kimbell, J.R., and McFall-Ngai, M.J.
recognition proteins-
(2007) Responses of host hemocytes during the initiation
of the squid-vibrio symbiosis. Biol Bull 212: 29–39.
Lang, M.F., Schneider, A., Kruger, C., Schmid, R., Dziarski,
R., and Schwaninger, M. (2008) Peptidoglycan recognition
protein-S (PGRP-S) is upregulated by NF-kB. Neurosci
Lett 430: 138–141.
Lechene, C., Hillion, F., McMahon, G., Benson, D., Kleinfeld,
A.M., Kampf, J.P., et al. (2006) High-resolution quantitative
imaging of mammalian and bacterial cells using stable
isotope mass spectrometry. J Biol 5: 20.
Lim, J.H., Kim, M.S., Kim, H.E., Yano, T., Oshima, Y., Aggar-
wal, K., et al. (2006) Structural basis for preferential rec-
ognition of diaminopimelic acid-type peptidoglycan by a
subset of peptidoglycan recognition proteins. J Biol Chem
281: 8286–8295.
Lu, X., Wang, M., Qi, J., Wang, H., Li, X., Gupta, D., and
Dziarski, R. (2006) Peptidoglycan recognition proteins are
a new class of human bactericidal proteins. J Biol Chem
281: 5895–5907.
Maillet, F., Bischoff, V., Vignal, C., Hoffmann, J., and Royet,
J. (2008) The Drosophila peptidoglycan recognition protein
PGRP-LF blocks PGRP-LC and IMD/JNK pathway activa-
tion. Cell Host Microbe 3: 293–303.
McFall-Ngai, M.J., and Ruby, E.G. (1991) Symbiont recogni-
tion and subsequent morphogenesis as early events in an
animal-bacterial mutualism. Science 254: 1491–1494.
Mellroth, P.,andSteiner,
N-acetylmuramoyl-L-alanine amidase with antibacterial
activity. Biochem Biophys Res Commun 350: 994–999.
Michel, T., Reichhart, J.M., Hoffmann, J.A., and Royet, J.
(2001) Drosophila Toll is activated by Gram-positive
bacteria through a circulating peptidoglycan recognition
protein. Nature 414: 756–759.
Millikan, D.S., and Ruby, E.G. (2004) Vibrio fischeri flagellin
A is essential for normal motility and for symbiotic com-
petence during initial squid light organ colonization.
J Bacteriol 186: 4315–4325.
Montgomery, M.K., and McFall-Ngai, M. (1993) Embryonic-
development of the light organ of the sepiolid squid
Euprymna-scolopes berry. Biol Bull 184: 296–308.
Montgomery, M.K., and McFall-Ngai, M. (1994) Bacterial
symbionts induce host organ morphogenesis during early
postembryonic development of the squid Euprymna scol-
opes. Development 120: 1719–1729.
Nyholm, S.V., and McFall-Ngai, M.J. (2004) The winnowing:
establishing the squid-vibrio symbiosis. Nat Rev Microbiol
2: 632–642.
Park, J.W., Kim, C.H., Kim, J.H., Je, B.R., Roh, K.B., Kim,
S.J., et al. (2007) Clustering of peptidoglycan recognition
protein-SA is required for sensing lysine-type peptidogly-
can in insects. Proc Natl Acad Sci USA 104: 6602–6607.
Ramet, M., Manfruelli, P., Pearson, A., Mathey-Prevot, B.,
and Ezekowitz, R.A. (2002) Functional genomic analysis of
phagocytosis and identification of a Drosophila receptor for
E. coli. Nature 416: 644–648.
Rosenthal, R.S., Nogami, W., Cookson, B.T., Goldman,
W.E., and Folkening, W.J. (1987) Major fragment of
soluble peptidoglycan released from growing Bordetella
pertussis is tracheal cytotoxin. Infect Immun 55: 2117–
2120.
Royet, J., and Dziarski, R. (2007) Peptidoglycan recognition
H.(2006) PGRP-SB1:an
1126 J. V. Troll et al.
© 2009 Blackwell Publishing Ltd, Cellular Microbiology, 11, 1114–1127

Page 15

proteins: pleiotropic sensors and effectors of antimicrobial
defences. Nat Rev Microbiol 5: 264–277.
Ruby, E.G., and Asato, L.M. (1993) Growth and flagellation of
Vibrio fischeri during initiation of the sepiolid squid light
organ symbiosis. Arch Microbiol 159: 160–167.
Silverman, N., Zhou, R., Stoven, S., Pandey, N., Hultmark,
D., and Maniatis, T. (2000) A Drosophila IkappaB kinase
complex required for Relish cleavage and antibacterial
immunity. Genes Dev 14: 2461–2471.
Steiner, H. (2004) Peptidoglycan recognition proteins: on and
off switches for innate immunity. Immunol Rev 198: 83–96.
Swaminathan, C.P., Brown, P.H., Roychowdhury, A., Wang,
Q., Guan, R., Silverman, N., et al. (2006) Dual strategies
for peptidoglycan discrimination by peptidoglycan recogni-
tion proteins (PGRPs). Proc Natl Acad Sci USA 103: 684–
689.
Takehana, A., Katsuyama, T., Yano, T., Oshima, Y., Takada,
H., Aigaki, T., and Kurata, S. (2002) Overexpression of
a pattern-recognition receptor, peptidoglycan-recognition
protein-LE, activatesimd/relish-mediated
defense and the prophenoloxidase cascade in Drosophila
larvae. Proc Natl Acad Sci USA 99: 13705–13710.
Talcott, B., and Moore, M.S. (1999) Getting across the
nuclear pore complex. Trends Cell Biol 9: 312–318.
Visick, K.L., Foster, J., Doino, J., McFall-Ngai, M., and Ruby,
E.G. (2000) Vibrio fischeri lux genes play an important role
in colonization and development of the host light organ.
J Bacteriol 182: 4578–4586.
Visick, K.L., and Ruby, E.G. (2006) Vibrio fischeri and its
host: it takes two to tango. Curr Opin Microbiol 9: 632–638.
Werner, T., Liu, G., Kang, D., Ekengren, S., Steiner, H., and
Hultmark, D. (2000) A family of peptidoglycan recognition
proteins in the fruit fly Drosophila melanogaster. Proc Natl
Acad Sci USA 97: 13772–13777.
Yip, E.S., Grublesky, B.T., Hussa, E.A., and Visick, K.L.
(2005) A novel, conserved cluster of genes promotes sym-
biotic colonization and sigma-dependent biofilm formation
by Vibrio fischeri. Mol Microbiol 57: 1485–1498.
Yoshida, H., Kinoshita, K., and Ashida, M. (1996) Purification
of a peptidoglycan recognition protein from hemolymph of
the silkworm, Bombyx mori. J Biol Chem 271: 13854–
13860.
Zaidman-Remy, A., Herve, M., Poidevin, M., Pili-Floury, S.,
Kim, M.S., Blanot, D., et al. (2006) The Drosophila amidase
PGRP-LB modulates the immune response to bacterial
infection. Immunity 24: 463–473.
Zhang, S.M., Zeng, Y., and Loker, E.S. (2007) Characteriza-
tion of immune genes from the schistosome host snail
Biomphalaria glabrata that encode peptidoglycan recogni-
tion proteins and gram-negative bacteria binding protein.
Immunogenetics 59: 883–898.
antibacterial
Zielinski, F.U., Pernthaler, A., Duperron, S., Raggi, L., Giere,
O., Borowski, C., and Dubilier, N. (2009) Widespread
occurrence of an intranuclear bacterial parasite in vent
and seep bathymodiolin mussels. Environ Microbiology
in press. Doi: 10.1111/j.1462-2920.2008.01847.x
Supporting information
Additional Supporting Information may be found in the online
version of this article:
Fig. S1. Antibody generation and PGN-gel zymography.
A. Clustal W alignment of all four EsPGRPs. Boxed areas indi-
cate sequences used for synthetic peptide antigens to raise
specific anti-EsPGRP antibodies.
B. A representative non-reducing zymogram of E. scolopes
protein extracts and accompanying immunoblot. Non-reducing
SDS-PAGE gels of the aqueous protein fraction show correspon-
dence between proteins that react with anti-EsPGRP1 antibody
and those that degrade PGN (dark bands on zymogram). 1, MW
standards; 2,Coomassie-stained
EsPGRP1 immunoblot; 4, empty zymogram lane; 5, zymogram of
aqueous protein fraction.
Fig. S2. Localization of EsPGRP1 in symbiotic or TCT-treated
E. scolopes epithelia. Representative confocal micrographs of
gill, gut and head tissues from animals colonized by V. fischeri or
treated with 10 mM TCT + 10 ng ml-1LPS. Epithelial nuclei con-
tained EsPGRP1 (yellow arrowheads) whereas non-epithelial
nuclei from the gut or head connective tissues or haemocytes in
the gills did not stain for EsPGRP1 (red arrows). EsPGRP1,
green; DNA, red; TUNEL, blue.
Fig. S3. Non-reactivity of pre-immune serum with E. scolopes
tissues. Representative confocal micrographs of the E. scolopes
light organ exposed to a 1:100 dilution of serum obtained prior to
exposure to the EsPGRP1 antigen. No significant cross-reactivity
was observed. Pre-immune serum, green; actin cytoskeleton,
blue.
Fig. S4. NanoSIMS tracing of
Representative NanoSIMS images of
the anterior appendage or (B) the duct. (C) Representative
NanoSIMS image of the distribution of
distribution of
terrestrial ratio of12C14N to12C15N is ~270 (colour indicated by
arrow on colour scale). No observed deviations from this value
were statistically significant. Apparent patterns, such as that in
the duct lumen, reflect increased variation due to low signal. c,
cytoplasm; d, duct lumen; mv, microvilli; n, nucleus.
tissueextract; 3,anti-
15N-TCT in light-organ sections.
12C14N distribution in (A)
12C14N divided by the
12C15N in the area shown in (B). The standard
Please note: Wiley-Blackwell are not responsible for the content
or functionality of any supporting materials supplied by the
authors. Any queries (other than missing material) should be
directed to the corresponding author for the article.
A nuclear PGRP in symbiont-induced development 1127
© 2009 Blackwell Publishing Ltd, Cellular Microbiology, 11, 1114–1127