ArticlePDF Available

Mutational Analyses Reveal Overall Topology and Functional Regions of NilB, a Bacterial Outer Membrane Protein Required for Host Association in a Model of Animal-Microbe Mutualism

Journal of Bacteriology

March 2012
194(7):1763-76

DOI:10.1128/JB.06711-11

Source
PubMed

Authors:

Heidi Goodrich-Blair

University of Wisconsin–Madison

The gammaproteobacterium Xenorhabdus nematophila is a mutualistic symbiont that colonizes the intestine of the nematode Steinernema carpocapsae. nilB (nematode intestine localization) is essential for X. nematophila colonization of nematodes and is predicted to encode an integral outer membrane beta-barrel protein, but evidence supporting this prediction has not been reported. The function of NilB is not known, but when expressed with two other factors encoded by nilA and nilC, it confers upon noncognate Xenorhabdus spp. the ability to colonize S. carpocapsae nematodes. We present evidence that NilB is a surface-exposed outer membrane protein whose expression is repressed by NilR and growth in nutrient-rich medium. Bioinformatic analyses reveal that NilB is the only characterized member of a family of proteins distinguished by N-terminal region tetratricopeptide repeats (TPR) and a conserved C-terminal domain of unknown function (DUF560). Members of this family occur in diverse bacteria and are prevalent in the genomes of mucosal pathogens. Insertion and deletion mutational analyses support a beta-barrel structure model with an N-terminal globular domain, 14 transmembrane strands, and seven extracellular surface loops and reveal critical roles for the globular domain and surface loop 6 in nematode colonization. Epifluorescence microscopy of these mutants demonstrates that NilB is necessary at early stages of colonization. These findings are an important step in understanding the function of NilB and, by extension, its homologs in mucosal pathogens.

NilB topology predicted by the PROFtmb (9) hidden Markov model. The model presented is similar to the model predicted by B2TMR-HMM (see Fig. S2 in the supplemental material and reference 36). While the PROFtmb model predicts that the ∼130 N-terminal amino acids are periplasmic, TMBB-PRED predicts that ∼100 of these amino acids form four additional TM domains (5, 6) (drawn in the periplasm in the same manner as TM domains and surface loops; the black box denotes this region). Only mature NilB is pictured (as predicted by SignalP 3.0 [8]), and numbering is according to that of immature NilB. Black arrows, FLAG tag insertions (before indicated amino acid); gray text, deletions (underlining distinguishes adjacent deletions); circles, the predicted TPR domain; SL, surface loop; 18TMSL, surface loop unique to the 18-TM strand model.

…

NilB protein expression is highest in stationary phase, repressed by NilR, and elevated by growth in defined medium. (A) Growth curves of X. nematophila nilR⁺ and nilR mutant strains in defined medium (DM) and LB. White boxes and white circles represent the nilR mutant in LB and defined medium, respectively. Gray boxes and gray circles represent nilR⁺ in LB and defined medium, respectively. The growth curves of both strains are identical in LB. In defined medium, the curves are similar for the two strains, except that the nilR mutant reaches a higher A600 than the nilR⁺ strain in stationary phase. (B) Western blots of NilB protein expression. nilR⁺ and nilR mutant strains were grown in LB or DM as indicated. Crude cell extracts were taken at 6, 12, 24, 36, 54, and 77 h, total protein content was normalized, and samples were loaded onto 12% SDS-PAGE gels for anti-FLAG Western blotting. The arrows indicate NilB, and NF contains extracts from the equivalent X. nematophila strains lacking the FLAG tag. (C) Quantitation of the blots in panel B. White boxes and white circles represent levels of NilB detected in the nilR mutant strain when grown over time in LB and defined medium, respectively. Gray boxes and gray circles represent levels of NilB detected in the nilR⁺ strain when grown over time in LB and defined medium, respectively. Levels of NilB in LB are similar for the two strains; low levels of NilB are detectable at 6, 12, and 24 h (log phase), and maximal expression is at 54 and 77 h (stationary phase). Levels of NilB in the nilR⁺ strain grown in defined medium are similar to levels of NilB in LB, except that the amount of NilB expressed is constant over time. Maximum NilB levels are produced in the nilR mutant strain grown in defined medium; low levels of NilB are detectable at 6, 12, and 24 h (log phase), with an increase in expression at 54 and 77 h (stationary phase). Quantitation was done with ImageQuant IQTL 7.0.

…

NilB is a membrane-localized protein and is surface exposed. (A) X. nematophila cells expressing NilB-FLAG26 in a nilR mutant background (HGB1200) were pelleted, resuspended, and sonicated. Soluble and membrane fractions were separated by ultracentrifugation. The crude extract (lane 1), cleared lysate (lane 2), soluble fraction (lane 3), and membrane fraction (lane 4) were electrophoresed on a 12% SDS-polyacrylamide gel and transferred to a membrane. NilB protein (arrowhead) was detected using anti-FLAG antibody. (B) Whole X. nematophila nilR mutant cells expressing NilB-FLAG26 (HGB1200) were incubated with 38.4 μg/ml proteinase K (PK) (final concentration) for 1 to 30 min at 37°C. At various time points (indicated in minutes above each lane), samples were removed and reactions were stopped with the addition of 5 mM PMSF (final concentration) and kept on ice. SDS-PAGE loading dye was added to all the reaction mixtures, which were heated at 95°C for 10 min and electrophoresed on 12% SDS-PAGE gels. NilB (top panel) (anti-NilB) or NilC (bottom panel) (anti-NilC) was detected in Western blots using anti-FLAG or anti-NilC (15) antibody, respectively. Full-length NilB-FLAG26 protein (black arrowheads, No PK lanes) was present in the no-proteinase K control but not in cells lacking nilB (Δnil) (HGB1251). This species decreases in intensity after the addition of proteinase K, while smaller-molecular-mass proteins (black arrowheads, lanes 1 and 10) appear. M, Thermo Scientific Pierce prestained protein molecular mass marker (sizes in kDa are noted on the left).

…

FLAG tag insertions distinguish whether NilB adopts a predicted 18-TM or 14-TM strand structure. Insertions were made in three N-terminal TM strands (before aa 68, aa 102, and aa 121) predicted only by the 18-TM strand model (D) and in two TM strands predicted by both models (before aa 379 and at aa 421); all mutations were made in nilR mutant cells. The effect of each mutation on nematode colonization (A) and protein production (B) was assessed by nematode sonication and immunoblotting with anti-FLAG, respectively. (A) Insertions before aa 68 or aa 121 reduce nematode colonization below wild-type levels but not to the level observed when nilB is absent (ETn7). Insertions before aa 102, aa 379, and aa 421 cause severe nematode colonization defects. *, P < 0.01, relative to ETn7 values (except with the aa 102 strain, for which P was 0.049). IJ, infective juvenile. (B) When FLAG insertions are in TM strands predicted only by the 18-TM strand model, NilB protein is detectable in whole cells by anti-FLAG Western blotting, but when insertions are in TM strands predicted by both models, NilB is not detectable. (C and D) Schematic of NilB's structure and the relative positions of FLAG tag insertions in the 14-TM (C) and 18-TM (D) strand models. Circles indicate that an insertion mutant colonized better than (white circles) or similarly to (shaded circles) a strain that lacked SR1. NF, Tn7/SR1 without a FLAG26 insertion in NilB; ETn7, Tn7 lacking SR1; FLAG26, Tn7/SR1 with a FLAG26 insertion in NilB (no deletions).

…

Identification of NilB regions necessary for colonization. FLAG tags were inserted into predicted NilB surface loops, the predicted N-terminal periplasmic domain, and the C terminus. The numbers indicate the amino acid locations of the tag; for example, 26 indicates that the FLAG tag is directly before (N-terminal to) amino acid 26 (Fig. 1). Mutants were made in nilR mutant backgrounds. (A) X. nematophila NilB-FLAG mutants were assessed for S. carpocapsae nematode colonization. Asterisks indicate colonization levels significantly different (P < 0.05) from that of X. nematophila expressing NilB without the FLAG tag (NF). (B) NilB protein production by the NilB-FLAG mutants was assayed on whole cells by Western blotting with anti-FLAG antibody. The abundances and SDS-PAGE migrations of the NilB-FLAG proteins (bracket) are varied. (C) Schematic of NilB structure, including predicted locations of FLAG tag insertions in the 14-TM strand model (Fig. 1). Shaded circles represent insertions that abolished or attenuated nematode colonization, while white circles represent insertions that do not significantly affect colonization.

…

Figures - available from: Journal of Bacteriology

This content is subject to copyright. Terms and conditions apply.

A preview of this full-text is provided by American Society for Microbiology.

Learn more

Content available from Journal of Bacteriology

This content is subject to copyright. Terms and conditions apply.

Mutational Analyses Reveal Overall Topology and Functional Regions

of NilB, a Bacterial Outer Membrane Protein Required for Host

Association in a Model of Animal-Microbe Mutualism

Archna Bhasin,

John M. Chaston,

and Heidi Goodrich-Blair

Department of Bacteriology, University of Wisconsin—Madison, Madison, Wisconsin, USA,

and Biology Department, Valdosta State University, Valdosta, Georgia, USA

The gammaproteobacterium Xenorhabdus nematophila is a mutualistic symbiont that colonizes the intestine of the nematode

Steinernema carpocapsae.nilB (nematode intestine localization) is essential for X. nematophila colonization of nematodes and

is predicted to encode an integral outer membrane beta-barrel protein, but evidence supporting this prediction has not been

reported. The function of NilB is not known, but when expressed with two other factors encoded by nilA and nilC, it confers

upon noncognate Xenorhabdus spp. the ability to colonize S. carpocapsae nematodes. We present evidence that NilB is a surface-

exposed outer membrane protein whose expression is repressed by NilR and growth in nutrient-rich medium. Bioinformatic

analyses reveal that NilB is the only characterized member of a family of proteins distinguished by N-terminal region tetratrico-

peptide repeats (TPR) and a conserved C-terminal domain of unknown function (DUF560). Members of this family occur in di-

verse bacteria and are prevalent in the genomes of mucosal pathogens. Insertion and deletion mutational analyses support a be-

ta-barrel structure model with an N-terminal globular domain, 14 transmembrane strands, and seven extracellular surface loops

and reveal critical roles for the globular domain and surface loop 6 in nematode colonization. Epiﬂuorescence microscopy of

these mutants demonstrates that NilB is necessary at early stages of colonization. These ﬁndings are an important step in under-

standing the function of NilB and, by extension, its homologs in mucosal pathogens.

The mutualism between the gammaproteobacterium Xenorh-

abdus nematophila and the nematode Steinernema carpocapsae

is a model to understand the molecular mechanisms underlying

the initiation and maintenance of long-term beneﬁcial animal-

microbe associations (13). The soil-dwelling infective third-stage

juvenile nematode (the infective juvenile) is colonized by a mon-

oculture of X. nematophila bacteria in a structure, known as the

receptacle (previously the vesicle), located at the anterior of the

nematode intestine (10, 21, 42, 58). The infective juvenile infects

the blood of insect hosts, releases the bacteria from the receptacle,

and together with the symbiont rapidly kills the insect host (22, 26,

32). Within the insect cadaver, the nematode reinitiates develop-

ment and reproduces through adult and juvenile stages that lack

the receptacle structure characteristic of the infective juvenile

stage. When nutrients derived from the insect cadaver are exhausted,

progeny juveniles reassociate with the symbiotic bacteria and develop

into the infective juvenile stage, which exits the insect cadaver to

search for a new host (reviewed in reference 50). X. nematophila’s

contributions to mutualism are that it is essential for both insect host

killing and S. carpocapsae nematode growth. In turn, X. nematophila

uses its nematode host as a vector into insects and as a protected

environment between infections (39, 43, 44).

Bacterial colonization of the infective juvenile nematode host

takes place in at least three stages: (i) entry of a few bacterial cells

into the receptacle, (ii) outgrowth of the initiating population to

30 to 200 bacterial cells, entirely ﬁlling the receptacle (37), and (iii)

persistence of the colonizing bacteria until the bacteria are re-

leased into an insect host by a recovering infective juvenile (21,

25). The process by which X. nematophila cells initially gain entry

into the infective juvenile receptacle and the point at which other

species are excluded have not been elucidated. Transposon screens

for colonization-defective X. nematophila mutants revealed genes

involved in both entry and outgrowth (30, 37). Outgrowth mu-

tants, including those with defects in methionine and threonine

biosynthesis, localize to the receptacle but do not replicate to ﬁll it

(21, 38) and ultimately disappear from within the nematode pop-

ulation (38). In contrast, mutants that are defective in early stages

of colonization are never apparent within the receptacle, either

because they fail to enter or because they fail to survive immedi-

ately after entering. Mutations in any of three genes, nilA,-B, and

-C(nematode intestine localization), carried on a 3.5-kb locus

(called SR1), cause entry defects. nilA mutants colonize nema-

todes to approximately 2.5% of the level of the wild type, while

nilB and nilC mutants do not colonize above the level of detection

(0.005 CFU/infective juvenile) (30). Microscopic analysis demon-

strated that the partial colonization phenotype of a nilA mutant

occurs because a few nematodes are partially colonized, while the

rest are devoid of bacteria. This indicates that nilA mutants are

defective primarily in entry but also have outgrowth defects (17).

However, it remains unclear whether nilA mutations cause a de-

fect in adherence to speciﬁc tissues or in survival during coloniza-

tion initiation.

An early observation regarding the nil locus was that it is absent

from other species of Xenorhabdus (17, 30), suggesting that it may

be speciﬁcally required for colonization of S. carpocapsae nema-

todes. Indeed, when nilA,-B, and -Cwere introduced into the

Received 16 December 2011 Accepted 20 January 2012

Published ahead of print 27 January 2012

Address correspondence to Heidi Goodrich-Blair, hgblair@bact.wisc.edu.

A.B. and J.M.C. contributed equally to this work.

Supplemental material for this article may be found at http://jb.asm.org/.

doi:10.1128/JB.06711-11

0021-9193/12/$12.00 Journal of Bacteriology p. 1763–1776 jb.asm.org 1763

chromosome of Xenorhabdus poinarii or Xenorhabdus bovienii,

these strains colonized the receptacles of S. carpocapsae nema-

todes, but the corresponding control strains lacking the nil locus

did not (17). Therefore, the nil locus not only is necessary for S.

carpocapsae nematode colonization by X. nematophila (30) but

also functions as a speciﬁcity determinant.

Online databases lack nil gene homologs of known function,

and understanding nil gene function therefore requires funda-

mental genetic and biochemical characterization. NilC is a 282-

amino-acid (aa) outer membrane-localized lipoprotein that is

predominantly oriented toward the periplasm (based on its resis-

tance to degradation by proteinase K in whole cells) (15). NilA is

predicted to be a small 90-aa inner membrane protein (30), and

NilB is predicted to encode a 466-aa protein that adopts a beta-

barrel structure in the outer membrane (Fig. 1) (30). The pre-

dicted localization of NilB to the outer membrane suggests that it

may interact directly with host surfaces or molecules, for example,

by binding to nematode surface structures or by transporting a

nutrient provided to the bacteria from the host.

To investigate the topology and function of NilB, we assessed

the colonization phenotypes of nilB mutants created by FLAG tag

insertions and domain deletions. These data reveal certain do-

mains critical for NilB function in colonization. Also, coupled

with in vivo imaging of colonizing bacteria within the nematode

host, our data suggest that NilB functions solely at entry into the

nematode receptacle. Unlike with another speciﬁcity factor, nilA,

we found no indication that nilB is necessary for outgrowth or

persistence within the nematode receptacle.

MATERIALS AND METHODS

Strains, plasmids, media, and growth conditions. X. nematophila strains

(Table 1) were grown aerobically at 30°C on LB agar plates containing

0.1% pyruvate or in the dark in LB broth in the presence of appropriate

antibiotics, unless stated otherwise. Final antibiotic concentrations used

were ampicillin at 100

␮

g/ml, chloramphenicol at 30 (for E. coli)or15(X.

nematophila)

␮

g/ml, erythromycin at 200

␮

g/ml, kanamycin at 50

␮

g/ml,

and streptomycin at 25 (E. coli) or 12.5 (X. nematophila)

␮

g/ml. Deﬁned

medium was made as described previously (41) except that agar and

amino acids were omitted and 0.68 g MgCl

liter

⫺1

and 0.91%

glucose were added after the autoclaving.

An X. nematophila strain (HGB1251) that lacks nilR (encoding a re-

pressor of the nil locus) as well as the locus carrying nilA,nilB, and nilC,

symbiosis region 1 (SR1), was created by conjugating strain HGB1141 (E.

coli S17

␭

pir containing p⌬nilR16::Sm [16]) into the recipient strain

HGB777 (⌬SR1-7::Kan) (15). To facilitate in vivo visualization of strains

FIG 1 NilB topology predicted by the PROFtmb (9) hidden Markov model. The model presented is similar to the model predicted by B2TMR-HMM (see Fig.

S2 in the supplemental material and reference 36). While the PROFtmb model predicts that the ⬃130 N-terminal amino acids are periplasmic, TMBB-PRED

predicts that ⬃100 of these amino acids form four additional TM domains (5, 6) (drawn in the periplasm in the same manner as TM domains and surface loops;

the black box denotes this region). Only mature NilB is pictured (as predicted by SignalP 3.0 [8]), and numbering is according to that of immature NilB. Black

arrows, FLAG tag insertions (before indicated amino acid); gray text, deletions (underlining distinguishes adjacent deletions); circles, the predicted TPR domain;

SL, surface loop; 18TMSL, surface loop unique to the 18-TM strand model.

Bhasin et al.

1764 jb.asm.org Journal of Bacteriology

colonizing nematodes, the green ﬂuorescent protein (GFP) expression

plasmid JMC001 was created. Plasmid pURR25 was digested with SacI

and BglII (Promega) to isolate the 2.5-kb cat-lacZp-GFP fragment.

pECM20 (37) was also digested with SacI and BglII to remove the aphAp-

GFP region. The resulting 5-kb pECM20 fragment was ligated to the

2.5-kb cat-lacZp-GFP fragment using T4 DNA ligase (Promega) and elec-

troporated into E. coli S17-1

␭

pir cells. A colony with green coloration,

visible by the naked eye, was frozen and used as a donor to transfer

pECM20-lacZp-GFP into HGB1251 to create X. nematophila

⌬nilR16::Sm ⌬SR1-7::Km lacZp-GFP (HGB1429) or into HGB777 to cre-

ate X. nematophila ⌬SR1-7::Km lacZp-GFP (HGB1430).

NilB-FLAG insertion mutants were created by PCR ampliﬁcation

(with Platinum PFX DNA polymerase; Invitrogen) of the entire pTn7/SR1

plasmid (15) with primers ﬂanking the mutation site (see Table S1 in the

supplemental material). One primer contained an nilB-complementary se-

quence, the FLAG sequence (nucleotides, GATTACAAGGATGACGACGA

TAAG; amino acids, DYKDDDDK) and the NheI restriction site, while

the other primer contained an nilB-complementary sequence and the

TABLE 1 Xenorhabdus strains used in this study

Strain type Strain Parent strain Relevant characteristic(s)

Source or

reference

Insertionless

HGB007 NA

Sequenced ATCC 19061 X. nematophila wild type ATCC

HGB777 HGB007 ⌬(nilA-nilC)7::Km (⌬SR1) 15

HGB1251 HGB777 ⌬(nilA-nilC)7::Km (⌬SR1) ⌬nilR16::Sm This study

HGB1430 HGB777 ⌬(nilA-nilC)7::Km lacZp-GFP (from pJMC001) This study

HGB1429 HGB1251 ⌬(nilA-nilC)7::Km ⌬nilR16::Sm lacZp-GFP

(from pJMC001)

This study

Non-GFP-expressing nilB FLAG

insertion mutants

HGB778 HGB777 SR1 15

HGB1256 HGB777 SR1/nilB26-FLAG aa026 (nilB-FLAG26) This study

HGB1200 HGB1251 SR1/nilB26-FLAG aa026 (nilB-FLAG26) This study

HGB1805 HGB1251 SR1/nilB27-FLAG aa068 This study

HGB1201 HGB1251 SR1/nilB28-FLAG aa085 This study

HGB1806 HGB1251 SR1/nilB29-FLAG aa102 This study

HGB1807 HGB1251 SR1/nilB30-FLAG aa121 This study

HGB1203 HGB1251 SR1/nilB31-FLAG aa137 This study

HGB1254 HGB1251 SR1/nilB32-FLAG aa186 This study

HGB1204 HGB1251 SR1/nilB33-FLAG aa233 This study

HGB1253 HGB1251 SR1/nilB34-FLAG aa273 This study

HGB1206 HGB1251 SR1/nilB35-FLAG aa322 This study

HGB1260 HGB1251 SR1/nilB36-FLAG aa357 This study

HGB1808 HGB1251 SR1/nilB37-FLAG aa379 This study

HGB1207 HGB1251 SR1/nilB38-FLAG aa399 This study

HGB1809 HGB1251 SR1/nilB39-FLAG aa421 This study

HGB1208 HGB1251 SR1/nilB40-FLAG aa451 This study

HGB1324 HGB1251 SR1/nilB41-FLAG aa466 This study

HGB1255 HGB1251 SR1 This study

GFP-expressing nilB FLAG

insertion mutants

HGB1502 HGB1430 SR1/nilB34-FLAG aa270 This study

HGB1495 HGB1429 Empty This study

HGB1496 HGB1429 SR1 (wild type nilB without FLAG) This study

HGB1484 HGB1429 SR1/nilB26-FLAG aa026 (nilB-FLAG26) This study

HGB1487 HGB1429 SR1/nilB32-FLAG aa186 This study

HGB1489 HGB1429 SR1/nilB34-FLAG aa273 This study

HGB1490 HGB1429 SR1/nilB35-FLAG aa322 This study

HGB1492 HGB1429 SR1/nilB38-FLAG aa399 This study

GFP-expressing nilB-FLAG26

deletion mutants

HGB1460 HGB1429 SR1/nilB42-⌬026–042 This study

HGB1461 HGB1429 SR1/nilB43-⌬043–059 This study

HGB1462 HGB1429 SR1/nilB44-⌬026–059 This study

HGB1463 HGB1429 SR1/nilB45-⌬076–091 This study

HGB1464 HGB1429 SR1/nilB46-⌬092–108 This study

HGB1465 HGB1429 SR1/nilB47-⌬026–108 This study

HGB1510 HGB1429 SR1/nilB48-⌬109–156 This study

HGB1511 HGB1429 SR1/nilB49-⌬026–156 This study

HGB1466 HGB1429 SR1/nilB50-⌬266–279 This study

HGB1467 HGB1429 SR1/nilB51-⌬292–298 This study

HGB1468 HGB1429 SR1/nilB52-⌬314–320 This study

HGB1469 HGB1429 SR1/nilB53-⌬352–357 This study

HGB1470 HGB1429 SR1/nilB54-⌬389-410 This study

All strains except HGB007, HGB777, HGB1251, HGB1430, and HGB1429 had att Tn7insertions.

NA, not applicable.

Topology and Functional Regions of X. nematophila NilB

April 2012 Volume 194 Number 7 jb.asm.org 1765

NheI restriction site. PCR products were cleaned using the QIAquick PCR

puriﬁcation kit (Qiagen), digested with the restriction enzyme NheI at

37°C, ligated with T4 DNA ligase (Promega), and electroporated into E.

coli S17-1

␭

pir cells.

With one exception, nilB in-frame deletion constructs were created by

PCR by amplifying the entire pAB001 plasmid (containing nilB26-

FLAG26), with primers oriented in the direction opposite to that of the

desired deletion site and containing 9 bp of the sequence upstream and 18

bp of the sequence downstream of the desired deletion site (see Table S1 in

the supplemental material), using Takara PrimeStar polymerase per the

manufacturer’s instructions and annealing at 55°C for 5 to 10 s. PCR

products were cleaned using a Zyppy PCR puriﬁcation kit (Zymo Re-

search), digested with DpnI, reverse dialyzed on a 0.025-

␮

m VSWP mem-

brane (Millipore) for 15 min, and electroporated into E. coli S17-1

␭

pir

cells. The exception was that the nilB construct with a deletion of aa 26 to

156 (nilB⌬26 –156 construct) could not be created using this approach, so

the NheI restriction site approach utilized for creating FLAG insertions

was adopted, resulting in insertion of duplicate NheI sites ﬂanking the

FLAG26 insertion (AS-DYKDDDDK-AS). Enzymes used for creating de-

letions were also used to create the nilB⌬26-156 construct and the con-

structs nilB-FLAG aa068 (in which the FLAG tag is inserted before aa 68),

nilB-FLAG aa102, nilB-FLAG aa121, nilB-FLAG aa379, and nilB-FLAG

aa421. The complete SR1 region of each Tn7-nilB mutant construct (plas-

mids are listed in Table 2) was sequenced to ensure that additional muta-

tions did not arise during PCR. Constructs were incorporated into the att

Tn7site of X. nematophila ATCC 19061 ⌬SR1 ⌬nilR (HGB1251), ⌬SR1

(HGB777), ⌬SR1 lacZp-GFP (HGB1430), or ⌬SR1 ⌬nilR lacZp-GFP

(HGB1429) by triparental mating as described previously (7). Veriﬁcation

TABLE 2 Plasmids used in this study

Type of construct Plasmid Relevant properties Reference or source

pEVS107 oriR6K mobRP4; mobilizable suicide miniTn7-Erm delivery

vector; Erm

pUX-BF13 Mobilizable Tn7transposition helper plasmid that expresses Tn7

transposase in trans;Ap

pTn7/SR1 Delivers miniTn7-Erm/SR1; used as a template for creation of

the NilB-FLAG insertion and deletion mutations

pJMC001 Plasmid pECM20 with cat-lacZp-GFP from pURR25 cloned into

the SacI and BglII sites, replacing the original gene encoding

GFP on this plasmid

This study

pECM20 GFP plasmid that contains a 614-bp chromosomal sequence of

X. nematophila ATCC 19061 that facilitates plasmid

integration at a site that does not interfere with nematode

colonization

pURR25 Tn7PA1/03/04gfpmut3; cat-lacZp-GFP source for pJMC001 D. Lies and D. Newmann

via T. Ciche

pTn7/SR1nilB-FLAG insertion

mutant constructs

pAB001 pTn7/SR1/nilB26-FLAG aa026 (nilB-FLAG26) This study

pAB002 pTn7/SR1/nilB28-FLAG aa085 This study

pAB003 pTn7/SR1/nilB31-FLAG aa137 This study

pAB004 pTn7/SR1/nilB32-FLAG aa186 This study

pAB005 pTn7/SR1/nilB33-FLAG aa233 This study

pAB006 pTn7/SR1/nilB34-FLAG aa273 This study

pAB007 pTn7/SR1/nilB35-FLAG aa322 This study

pAB008 pTn7/SR1/nilB36-FLAG aa357 This study

pAB009 pTn7/SR1/nilB38-FLAG aa399 This study

pAB010 pTn7/SR1/nilB40-FLAG aa451 This study

pAB011 pTn7/SR1/nilB41-FLAG aa466 This study

pJMC016 pTn7/SR1/nilB27-FLAG aa068 This study

pJMC017 pTn7/SR1/nilB29-FLAG aa102 This study

pJMC018 pTn7/SR1/nilB30-FLAG aa121 This study

pJMC019 pTn7/SR1/nilB37-FLAG aa379 This study

pJMC020 pTn7/SR1/nilB39-FLAG aa421 This study

pAB001 nilB-FLAG26 deletion

mutant constructs

pJMC002 pTn7/SR1/nilB42-⌬026–042 This study

pJMC003 pTn7/SR1/nilB43-⌬043–059 This study

pJMC004 pTn7/SR1/nilB44-⌬026–059 This study

pJMC005 pTn7/SR1/nilB45-⌬076–091 This study

pJMC006 pTn7/SR1/nilB46-⌬092–108 This study

pJMC007 pTn7/SR1/nilB47-⌬026–108 This study

pJMC014 pTn7/SR1/nilB48-⌬109–156 This study

pJMC015 pTn7/SR1/nilB49-⌬026–156 This study

pJMC008 pTn7/SR1/nilB50-⌬266-279 This study

pJMC009 pTn7/SR1/nilB51-⌬292–298 This study

pJMC010 pTn7/SR1/nilB52-⌬314–320 This study

pJMC011 pTn7/SR1/nilB53-⌬352–357 This study

pJMC012 pTn7/SR1/nilB54-⌬389–410 This study

Bhasin et al.

1766 jb.asm.org Journal of Bacteriology

of insertion of each Tn7/nilB-FLAG or deletion construct at the att Tn7

site was veriﬁed by PCR as described previously (15). Since insertion or

deletion mutations were created in the context of the SR1 genomic region,

the mutant nilB alleles were expressed from the native nilB promoter and

regulatory regions. Constructs were introduced into a strain lacking SR1

so that no SR1 DNA sequence was present in duplicate. Therefore,

HGB777, -1251, -1429, and -1430 complemented with Tn7/SR1 lacking

any modiﬁcations to SR1 are considered wild-type controls in this study.

Fractionation, NilB detection by Western blotting, and glycosyla-

tion. Two Western blotting methods were used in this study. In method 1,

used for all experiments except where stated otherwise, strains were grown

as described above prior to harvesting of cell pellets. Pellets were washed

and resuspended in cold 1⫻phosphate-buffered saline (PBS) and soni-

cated on ice using a Sonic Dismembrator with 5 bursts of 3 s each or until

the cells lysed. Lysed cells were spun at 13,000 rpm in a microcentrifuge to

remove cell debris. The total protein concentration of the cleared lysates

was determined by the Bradford assay (Bio-Rad), and equal amounts of

protein were heated for 10 min at 95°C in 1⫻SDS-PAGE loading dye

prior to being loaded onto 12% SDS-PAGE gels. In other cases, the cleared

lysate was spun at 434,513 ⫻gin a Beckman TLA 100.2 rotor for 30 min

at 4°C for separation of the membrane and soluble fractions. The mem-

brane fraction was solubilized in 1⫻PBS plus 0.5% Sarkosyl by stirring for

30 min. Fractions were heated for 10 min at 95°C in 1⫻SDS-PAGE load-

ing dye prior to being loaded onto 12% SDS-PAGE gels. After transfer to

polyvinylidene diﬂuoride (PVDF) membranes (Bio-Rad), membranes

were blocked for 1 h with 5% dry milk in 1⫻Tris-buffered saline with

Tween 20 (TBS-T) at room temperature, washed, and incubated with

1:1,000-diluted anti-FLAG antibody (Cell Signaling) in 1⫻TBS-T plus

5% bovine serum albumin (BSA) overnight at 4°C. The next day, mem-

branes were washed, incubated with 1:10,000-diluted horseradish perox-

idase (HRP)-linked anti-rabbit IgG (Cell Signaling) for1hatroom tem-

perature, and washed again prior to detection using the Amersham ECL

Plus Western blotting detection system (GE Healthcare) and the Storm/

Typhoon scanner or the LAS4010 biomolecular imager (ImageQuant)

according to the manufacturers’ instructions.

Western blotting method 2 was used for the experiments from which

results are shown in Fig. 4 and for which data are not shown. In this

method, strains were subcultured from LB at a 1:100 dilution into deﬁned

medium for ⬃24 h prior to harvesting of cell pellets. Pellets were washed

in an equal volume of PBS, concentrated 4-fold in PBS, and normalized to

their optical density at 600 nm (OD

600

) prior to being mixed with 2⫻

SDS-PAGE loading dye, heated to 95°C for 10 min, and run at 100 V on a

7.5% SDS-PAGE gel. Gels were transferred to a PVDF membrane (Bio-

Rad) by a Trans-Blot SD semidry electrophoretic transfer cell (Bio-Rad)

set to transfer at 22 V for at least 15 min. Membranes were blocked over-

night in PBS plus 5% dry milk prior to being sequentially washed and

incubated with a 1:10,000 or 1:5,000 dilution of anti-FLAG antibody

(Sigma F7425) and HRP-conjugated anti-rabbit secondary antibody

(Pierce catalog number 32260) according to the manufacturers’ instruc-

tions. Blots were detected using the ECL Plus Western blotting substrate

(Pierce catalog number 32209) and XAR ﬁlm (Kodak XAR catalog num-

ber 165-1454). Adobe Photoshop CS3 was used to alter the intensity and

contrast of scanned images of ﬁlms, with all samples within an individual

panel (including the controls) manipulated in the same way.

The glycosylation state of NilB was determined using the GelCode

glycoprotein staining kit (Pierce). X. nematophila crude protein was ex-

tracted from all nilB-FLAG insertion mutants in the ⌬SR1 ⌬nilR back-

ground and electrophoresed as described above, except that transfer was

onto a nitrocellulose membrane (Bio-Rad) rather than a PVDF mem-

brane. Positive (horseradish peroxidase) and negative (soybean trypsin

inhibitor) control proteins were also loaded on the protein gel. Upon

transfer, the membrane was washed with acetic acid, oxidizing solution,

staining reagent, and reducing agent according to the manufacturer’s pro-

tocol.

Proteinase K assay. Strain HGB1200 (⌬SR1 ⌬nilR Tn7/SR1nilB-

FLAG26) was grown at 30°C overnight in LB broth. Cells were pelleted

and then washed and resuspended in 50 mM Tris-HCl, pH 8, 1 mM

CaCl

. The resuspended whole cells were split into two aliquots. One

aliquot was treated with 0.1% Triton X-100 (ﬁnal concentration) to break

open the cells. Both aliquots were treated with 38.4

␮

g/ml proteinase K

(ﬁnal concentration; Fermentas) for 0 to 30 min at 37°C. At various time

points, samples were removed and reactions were stopped with the addi-

tion of 5 mM PMSF (Sigma catalog number P-7626) and kept on ice.

SDS-PAGE loading dye was added to all the reaction mixtures, which were

heated at 95°C for 10 min, sonicated for8stoshear the DNA, and elec-

trophoresed on 12% SDS-PAGE gels. NilB and NilC (as a control) were

detected in Western blots using anti-FLAG and anti-NilC antibody (15),

respectively.

Growth curves and time courses of NilB protein levels. Strains

HGB1200 (⌬SR1 ⌬nilR Tn7/SR1 nilB-FLAG26) and HGB1256 (⌬SR1

Tn7/SR1 nilB-FLAG26) were streaked from ⫺80°C onto LB pyruvate

plates with antibiotics. Colonies from these plates were inoculated into

liquid LB with antibiotics and grown at 30°C overnight in a shaking water

bath. Overnight cultures were subcultured 1:100 into 1 ml LB or deﬁned

medium (no antibiotics) in the wells of a 24-well plate. Absorbance at 600

nm was measured at 1-h intervals using the SpectraMax M5

plate reader

(Molecular Devices). At 6, 12, 24, 36, 54, and 77 h, 0.5 ml of each strain in

either LB or deﬁned medium was removed, microcentrifuged to remove

the medium, washed with 1⫻cold PBS, resuspended in 1⫻PBS, and

sonicated at approximately 14 W for 10 s. Sample protein concentrations

were assessed by the Bradford assay (Bio-Rad), and all samples were nor-

malized to each other. Samples at later time points were diluted in more

PBS than earlier time points (PBS resuspension volumes ranged from 37.5

to 150

␮

l). After resuspension, 4⫻SDS-PAGE loading dye was added and

the samples were stored at ⫺20°C until Western blotting was performed.

Quantitation of blots was performed using the ImageQuant TL program

(GE Healthcare).

Nematode-bacterium cocultivations. The effect of FLAG insertions

in NilB on nematode colonization was tested by growing each mutant on

lipid agar (54) and inoculating greater than 200 sterile nematode eggs (55)

or surface-sterilized axenic infective juveniles onto each plate. The latter

were prepared by inoculating sterile nematode eggs onto a GFP-express-

ing strain of X. nematophila (HGB1430 ⌬SR1::Km lacZp-GFP) defective

in nematode colonization. Infective juveniles were harvested from White

traps and stored at 25°C until they were surface sterilized for 3 min in 0.5%

NaOCl (30), and then they were inoculated onto bacterial lawns. Infective

juveniles that developed on the FLAG-tagged mutant bacteria were har-

vested in White traps (56), and bacterial colonization of the nematode

host was assessed either by dilution plating surface-sterilized nematode

sonicates to calculate average numbers of CFU per infective juvenile (15)

or by analyzing nematode carriage of GFP-labeled bacteria by epiﬂuores-

cence microscopy as described previously (37). For microscopy, nema-

todes were analyzed at a 100⫻magniﬁcation and scored as either colo-

nized (bacteria were localized to the receptacle) or uncolonized (no

bacteria were observed within the receptacle). If greater than 30 in 500

nematodes in the population were colonized, nematodes were scored in-

dividually. If fewer than 30 nematodes in 500 were colonized, samples of

nematode populations were scanned for colonized nematodes in 24-well

plates. The number of colonized nematodes in each well was determined,

and the entire nematode population in each well was enumerated by di-

lution counting. Organisms in additional wells were counted until either

30 colonized nematodes or all nematode progeny from the inoculated

plate were counted (whichever occurred ﬁrst). Data are reported as the

percentages of total nematodes that carried GFP-labeled bacteria.

For data presented in Fig. 5, sterile nematode eggs were added to

each lawn of bacteria and nematode sonications were performed on

samples from at least ﬁve different experiments performed at different

times. For Fig. 4, surface-sterilized axenic infective juveniles were

added to each lawn of bacteria. Nematode sonications were performed

Topology and Functional Regions of X. nematophila NilB

April 2012 Volume 194 Number 7 jb.asm.org 1767

on samples from two experiments performed at different times. For

the ﬁrst experiment, two otherwise isogenic exconjugants were inoc-

ulated, and three replicate experiments were performed for each (six

replicate experiments, total) (data not shown). Colonization results

for each exconjugant pair were the same, so for the second experiment

(data shown), the results of two replicate samples for each of three

isogenic cultures of a single representative exconjugant were assessed.

The results of replicate samples were averaged to give biological-rep-

licate (i.e., isogenic strain) values, and averages and standard devia-

tions among biological-replicate values for the second experiment are

presented in Fig. 4. A linear mixed-effects model was applied to each

experiment individually, and similar statistical trends were observed

between the two experiments. For all other colonization tests, surface-

sterilized axenic infective juveniles were used to inoculate bacterial

lawns. Assays were performed as two experiments at different times

with two replicate samples for each of three biological replicates of

each strain. For graphical presentation, the results of replicate samples

were averaged to give biological-replicate values, and averages and

standard deviations among biological-replicate values for a single rep-

resentative experiment are presented. A linear mixed-effects model or

a generalized linear mixed-effects model was applied to raw values or

data after combination of the results of both experiments.

Statistical analysis. Statistical differences in Fig. 5 were assessed in

Microsoft Excel using a two-tailed heteroscedastic Student ttest. Other-

wise, differences between sonicated samples were performed using a lin-

ear mixed-effects model, and differences between samples analyzed by

microscopy were assessed using a generalized linear mixed-effects model.

Statistical tests were performed in R (48). All collected data points were

included in the analyses unless otherwise indicated, and the Pvalue cutoffs

are at least ⬍0.05.

RESULTS

nilB represents a family of proteins found within symbionts of

animals. In 2002, it was reported that, while X. nematophila nilB is

absent from other Xenorhabdus species tested to date, it has ⬃18

to 25% full-length sequence similarity (BLASTp) to proteins of

unknown function encoded by Neisseria meningitidis,Pasteurella

multocida,Haemophilus inﬂuenzae, and Moraxella catarrhalis (2,

30). A more recent BLASTX analysis probing the NCBI database

(December 2011) revealed additional proteins with 18 to 26%

sequence identity to NilB (cutoff E value of 10

⫺5

) (see Table S2 in

the supplemental material). All BLASTX hits were encoded by

harmful or beneﬁcial animal symbionts, including many Neisseria

and Actinobacillus spp., and isolates revealed by the Human Mi-

crobiome Project. Two species of Psychrobacter, normally found

in cold-adapted environments, appear on this list, and both are

also host associated. Psychrobacter sp. strain PRwf-1 is found on

ﬁsh skin and gills and was isolated in Puerto Rico (see http://ge-

nome.jgi-psf.org/psy_p/psy_p.home.html); Psychrobacter sp.

strain 1510 was isolated from blood as part of the Human Micro-

biome Project (NCBI accession number AFHU00000000). In the

enterobacteriaceae, which are well represented with sequenced

isolates, NilB homologs have been identiﬁed in only three species:

X. nematophila,Salmonella enterica subsp. arizonae, a reptile

pathogen, and Edwardsiella tarda, a vertebrate symbiont and ﬁsh

pathogen.

NilB has not been found in any Xenorhabdus species tested to

date other than X. nematophila (17, 30). We also did not observe

any Xenorhabdus proteins in our nilB BLASTX analysis. However,

we did identify a protein in the X. nematophila genome (14) with

sequence similarity to NilB homologs (this protein was identiﬁed

by reciprocally comparing low-scoring NilB homologs against the

X. nematophila genome [cutoff E value of 10

⫺5

]). This X. nema-

tophila protein (XNC1_0074, YP_003710424) is found in the

same genomic context in X. nematophila,X. bovienii,Photorhab-

dus luminescens, and Photorhabdus asymbiotica (close relatives of

Xenorhabdus spp.) (see Fig. S1 in the supplemental material) (12,

26). We distinguish NilB homologs from NilB-like proteins based

on their identiﬁcation by BLASTX: NilB homologs are classiﬁed as

those retrieved by nilB BLASTX (E value of ⬍10

⫺5

; 109 hits);

NilB-like proteins are those identiﬁed using XNC1_0074 BLASTX

(E value of ⬍10

⫺5

; 326 hits). There is substantial overlap between

these protein groups, since 80 proteins are identiﬁed by both nilB

and XNC1_0074 BLASTX queries.

NilB is predicted to be an outer membrane protein with 7 or

9 surface-exposed loops and two conserved domains. Prediction

programs indicate that NilB homologs and NilB-like proteins lo-

calize to the outer membrane and adopt beta-barrel structures (5,

6, 9) (Fig. 1 and data not shown). The N-terminal 25 amino acids

of X. nematophila NilB are predicted to encode a sec secretion

signal sequence that is cleaved prior to outer membrane localiza-

tion (8, 30). Of the online programs available for predicting the

structure of beta-barrels, hidden Markov models (HMM) have

the highest accuracy (4). Two HMM programs (PROFtmb [9] and

B2TMR-HMM [36]) each predict that NilB adopts an outer mem-

brane structure with 14 transmembrane (TM) strands, 7 surface

loops (SLs), and a 138- or 140-aa (postcleavage) N-terminal

periplasmic domain, respectively (Fig. 1). These structures differ

from the model predicted by the HMM program PRED-TMBB

(6), which predicts that the 350 C-terminal amino acids of NilB

comprise 18 TM strands and 9 SLs and that the N terminus is a

38-aa periplasmic domain (Fig. 1). All three programs predict

similar locations for the seven extracellular SLs (Fig. 1), but the

PROFtmb and B2TMR-HMM models predict that the two

additional SLs in the PRED-TMBB model make up a longer N-

terminal periplasmic domain.

The family of NilB/NilB-like proteins shares two conserved

domains, neither of which is unique to this family. At least one

copy of a 34-aa tetratricopeptide repeat (TPR)-like domain, pre-

dicted to fold into two packed

␣

-helices (1, 3, 11), is found within

the N-terminal one-third of the proteins. TPR motifs are widely

distributed and occur within proteins involved in diverse func-

tions, including transcription, cell cycle control, and protein

transport complexes (19). The TPR domain is thought to mediate

protein-protein interactions (11). In NilB, a TPR domain was not

identiﬁed by a TPR prediction program (TPRpred; http://toolkit

.tuebingen.mpg.de/tprpred) using standard maximum cutoffs

(1e⫺4), but a single TPR domain (aa 112 to 145) was identiﬁed by

lowering the stringency of the cutoff value (Pvalue 1.2e⫺03).

Several NilB homologs tested (e.g., those of Aggregatibacter aphro-

philus,Actinobacillus minor, and Haemophilus parasuis) had mul-

tiple consecutive TPR domains, while some NilB homologs had

only a single copy (e.g., that of Mannheimia haemolytica). The X.

nematophila NilB-like protein XNC1_0074 contains four consec-

utive predicted TPR domains (data not shown). The second con-

served domain, DUF560/pfam04575 (35), spans the C-terminal

two-thirds of NilB homologs, essentially constituting the barrel

domain predicted by the 14-TM strand model. This domain of

unknown function is present in over 900 proteins, including all

NilB homologs evaluated. In some cases, the TPR and DUF560

domains overlap. No structure or function has been assigned to

the DUF560 domain.

Bhasin et al.

1768 jb.asm.org Journal of Bacteriology

NilB protein levels are elevated by nilR mutation and nutri-

ent-limiting conditions. The current working model for Xenorh-

abdus-Steinernema symbiosis is that nutrient limitation in the in-

sect cadaver is the signal for Xenorhabdus to colonize the

nematode intestine. This idea, combined with previous data dem-

onstrating that nilC transcription is upregulated during stationary

phase, led us to investigate whether NilB production is also growth

phase dependent (15). To monitor NilB protein levels, we at-

tempted to generate polyclonal antibodies against NilB but were

unsuccessful. Instead, we engineered a FLAG (DYKDDDDK)-

tagged version of NilB, in which the tag amino acids were inserted

immediately preceding amino acid 26 (predicted to be the ﬁrst

amino acid in the mature protein after signal sequence cleavage).

FLAG-tagged nilB was provided in trans at the att Tn7site of an X.

nematophila strain lacking wild-type nilB. Since under laboratory

growth conditions NilR suppresses transcription of the nil genes

(16), we also tested NilB production in the presence and absence

of nilR. Neither the nilR mutation (16) nor the FLAG insertion in

NilB (see below; also, data not shown) negatively affects nematode

colonization (i.e., NilB function). To determine the effect of the

nilR mutation and growth phase on NilB protein levels, we mon-

itored growth and NilB-FLAG26 levels of nilR

⫹

(HGB1256) and

nilR mutant (HGB1200) strains in LB broth and deﬁned medium.

nilR

⫹

and nilR mutant strains have similar growth curves in LB. In

deﬁned medium, both strains grow similarly until a plateau is

reached at ⬃25h(A

600

⬇0.22), after which nilR mutant cells

continue to grow at a higher rate and ultimately reach a higher

600

(⬃0.7) than do nilR

⫹

cells (A

600

,⬃0.5) (Fig. 2A). The differ-

ence in the maximum levels of absorbance of the two strains was

not due to differences in viability (both strains yielded 1 ⫻10

CFU/ml at 77 h), differences in absorbance of the culture super-

natant, or gross differences in cell size (monitored by microscopy)

(data not shown).

At various times during growth in deﬁned medium and LB,

crude cell extracts were taken and evaluated for NilB protein pro-

duction by Western blotting (Fig. 2C). As expected based on pre-

vious analyses of nilB-lacZ transcriptional fusions (16), quantiﬁ-

cation of the Western blot revealed that NilB protein levels were

higher in the nilR mutant strain than in the nilR

⫹

strain (Fig. 2B).

Maximum levels of NilB production are found in the nilR mutant

strain grown in deﬁned medium, with low-but-detectable

amounts of NilB produced at 6, 12, and 24 h (log phase), an inter-

mediate amount at 36 h, and maximal amounts at 54 and 77 h

(stationary phase) of growth in deﬁned medium. In contrast, the

nilR

⫹

strain exhibited low and constant levels of NilB at all time

points of growth in deﬁned medium (Fig. 2B and C). In LB me-

dium, NilB levels in both nilR

⫹

and nilR mutant strains increased

in stationary phase but never reached the levels observed in the

nilR mutant strain during growth in deﬁned medium (Fig. 2B and

C). In some samples, another FLAG-reactive band was apparent.

This protein is not a variant of NilB, since it is present in the

no-FLAG (NF) control (Fig. 2C). The data shown in Fig. 2 indicate

FIG 2 NilB protein expression is highest in stationary phase, repressed by

NilR, and elevated by growth in deﬁned medium. (A) Growth curves of X.

nematophila nilR

⫹

and nilR mutant strains in deﬁned medium (DM) and LB.

White boxes and white circles represent the nilR mutant in LB and deﬁned

medium, respectively. Gray boxes and gray circles represent nilR

⫹

in LB and

deﬁned medium, respectively. The growth curves of both strains are identical

in LB. In deﬁned medium, the curves are similar for the two strains, except that

the nilR mutant reaches a higher A

600

than the nilR

⫹

strain in stationary phase.

(B) Western blots of NilB protein expression. nilR

⫹

and nilR mutant strains

were grown in LB or DM as indicated. Crude cell extracts were taken at 6,

12, 24, 36, 54, and 77 h, total protein content was normalized, and samples

were loaded onto 12% SDS-PAGE gels for anti-FLAG Western blotting.

The arrows indicate NilB, and NF contains extracts from the equivalent X.

nematophila strains lacking the FLAG tag. (C) Quantitation of the blots in

panel B. White boxes and white circles represent levels of NilB detected in

the nilR mutant strain when grown over time in LB and deﬁned medium,

respectively. Gray boxes and gray circles represent levels of NilB detected in

the nilR

⫹

strain when grown over time in LB and deﬁned medium, respec-

tively. Levels of NilB in LB are similar for the two strains; low levels of

NilB are detectable at 6, 12, and 24 h (log phase), and maximal expression is at

54 and 77 h (stationary phase). Levels of NilB in the nilR

⫹

strain grown in

deﬁned medium are similar to levels of NilB in LB, except that the amount of

NilB expressed is constant over time. Maximum NilB levels are produced in

the nilR mutant strain grown in deﬁned medium; low levels of NilB are detect-

able at 6, 12, and 24 h (log phase), with an increase in expression at 54 and 77

h (stationary phase). Quantitation was done with ImageQuant IQTL 7.0.

Topology and Functional Regions of X. nematophila NilB

April 2012 Volume 194 Number 7 jb.asm.org 1769

that under laboratory growth conditions, NilB levels are highest

under stationary-phase and nutrient-limited conditions. Further-

more, NilR negatively impacts NilB levels in all growth phases,

perhaps indicating that the signal for NilR derepression is not

present in laboratory growth medium. Also, nilB is repressed by

other factors (e.g., Lrp [18]) that may contribute to the effects that

we observed here.

NilB is a surface-exposed outer membrane protein. To con-

ﬁrm predictions that NilB is localized to the outer membrane, we

monitored NilB-FLAG levels in X. nematophila cellular fractions.

NilB-FLAG detection by Western blotting was facilitated by pre-

paring extracts from the X. nematophila nilB-FLAG26 strain lack-

ing nilR, which represses nilB transcription (16) and protein levels

(Fig. 2). Using anti-FLAG antibody, NilB-FLAG26 (⬃53 kDa) was

detected in whole cells, crude sonicates, and membrane fractions

isolated from X. nematophila but not in soluble, cleared X. nema-

tophila lysates (Fig. 3A), supporting predictions that NilB is local-

ized to the outer membrane and therefore likely adopts a beta-

barrel structure (33). To determine if any portion of NilB is

surface exposed, we treated whole cells expressing NilB-FLAG26

with proteinase K and then monitored NilB levels over time using

anti-FLAG immunoblotting. Levels of full-length NilB-FLAG in

whole cells decreased dramatically by 1 min after exposure to pro-

teinase K and were not detectable after 4 min of incubation, indi-

cating that at least some regions of NilB are surface exposed and

susceptible to cleavage (Fig. 3B). NilC, which is not sensitive to

proteinase K digestion (15), remained detectable over the same

time period (Fig. 3B). Both NilB-FLAG and NilC were sensitive to

proteinase K digestion in lysed cell extracts (data not shown).

After ⬃1 min of incubation with proteinase K, concomitant

with the reduction in full-length NilB-FLAG, another band that

migrated slightly faster by SDS-PAGE than full-length mature

NilB-FLAG appeared (Fig. 3B), and it had an approximate molec-

ular mass of ⬃48 kDa. Given that the FLAG tag is encoded at the

amino-terminus of the protein, the appearance of these bands

is consistent with proteinase K cleavage somewhere in the C-ter-

minal region of the protein. The estimated molecular masses of

NilB if it is cleaved in the middle of one of the three SLs at the

C-terminal end of the protein are ⬃40 (SL5), ⬃46 (SL6), and ⬃51

(SL7) kDa. Based on this, initial proteinase K cleavage may occur

within or near SL6 (aa 387 to 410), the longest predicted surface-

exposed loop, which may therefore be particularly susceptible to

cleavage. The intensity of the band decreases progressively over

time of exposure to proteinase K, while a third band of ⬃40 kDa

appears by 10 min of incubation with proteinase K, possibly rep-

resenting another cleavage product (perhaps within SL5).

FLAG tag insertions support a model of NilB with an

⬃140-aa periplasmic domain and 7 surface-exposed loops. To

probe NilB structure and differentiate among bioinformatic to-

pology predictions (Fig. 1), we exploited the fact that insertions

within TM strands should destabilize the protein and prevent nor-

mal protein localization, while insertions within extracellular or

periplasmic domains are more likely to be tolerated (28, 49).

Based on HMM models predicted by PRED-TMBB, PROFtmb,

and B2TMR-HMM, we created ﬁve unique FLAG tag insertion

nilB mutants in addition to a nilB-FLAG26 strain to distinguish

between the HMM models predicting either 14 (PROFtmb or

B2TMR-HMM) or 18 (PRED-TMBB) TM strands (with 7 or 9

SLs, respectively) (Fig. 1).

We inserted FLAG tags into three regions predicted to be trans-

membrane strands in the 18-strand model but periplasmic strands

in the 14-strand model (before aa 68, aa 102, or aa 121). As a

control, we also inserted FLAG tags into two domains consistently

predicted by all HMM programs to be TMs (before aa 379 and aa

421, ﬂanking SL6). As predicted, X. nematophila expressing NilB

FLAG insertions before aa 379 and aa 421 are colonization defec-

tive (Fig. 4A) and fail to yield detectable protein (Fig. 4B), indicat-

ing that these insertions destabilize the NilB protein. In contrast,

X. nematophila carrying NilB-FLAG variants before aa 68 or aa

121 colonize at levels signiﬁcantly higher than in a strain lacking

NilB entirely (Fig. 4A). The difference in colonization phenotype

of a strain carrying NilB-FLAG aa102 and a strain lacking nilB

entirely was barely signiﬁcant (P⫽0.049) (Fig. 4A) and may in-

dicate that FLAG insertion at aa 102 disrupts NilB function. Re-

gardless, NilB-FLAG protein is detectable in strains expressing

NilB-FLAG insertions at aa 68, aa 102, and aa 121 (Fig. 4B), and

this is consistent with a periplasmic localization of these domains

as predicted in the 14-TM model (Fig. 1 and 4C) but not the

18-TM model (Fig. 4D).

Colonization function of nilB insertion mutants. We next

used FLAG tag insertions to assess which regions of NilB are nec-

essary for colonization, as well as for further veriﬁcation of the

FIG 3 NilB is a membrane-localized protein and is surface exposed. (A) X.

nematophila cells expressing NilB-FLAG26 in a nilR mutant background

(HGB1200) were pelleted, resuspended, and sonicated. Soluble and mem-

brane fractions were separated by ultracentrifugation. The crude extract (lane

1), cleared lysate (lane 2), soluble fraction (lane 3), and membrane fraction

(lane 4) were electrophoresed on a 12% SDS-polyacrylamide gel and trans-

ferred to a membrane. NilB protein (arrowhead) was detected using anti-

FLAG antibody. (B) Whole X. nematophila nilR mutant cells expressing NilB-

FLAG26 (HGB1200) were incubated with 38.4

␮

g/ml proteinase K (PK) (ﬁnal

concentration) for 1 to 30 min at 37°C. At various time points (indicated in

minutes above each lane), samples were removed and reactions were stopped

with the addition of 5 mM PMSF (ﬁnal concentration) and kept on ice. SDS-

PAGE loading dye was added to all the reaction mixtures, which were heated at

95°C for 10 min and electrophoresed on 12% SDS-PAGE gels. NilB (top panel)

(anti-NilB) or NilC (bottom panel) (anti-NilC) was detected in Western blots

using anti-FLAG or anti-NilC (15) antibody, respectively. Full-length NilB-

FLAG26 protein (black arrowheads, No PK lanes) was present in the no-pro-

teinase K control but not in cells lacking nilB (⌬nil) (HGB1251). This species

decreases in intensity after the addition of proteinase K, while smaller-molec-

ular-mass proteins (black arrowheads, lanes 1 and 10) appear. M, Thermo

Scientiﬁc Pierce prestained protein molecular mass marker (sizes in kDa are

noted on the left).

Bhasin et al.

1770 jb.asm.org Journal of Bacteriology

14-TM strand model shown in Fig. 1, relative to other possible

topologies predicted by non-HMM algorithms (see Fig. S2 in the

supplemental material). To this end, we created 10 insertions at

regions of NilB predicted to be periplasmic or surface exposed

(Fig. 5C) in the 14-TM model (before aa 85, 137, 186, 233, 273,

322, 357, 399, 451, and 466), and at least one of which is within a

predicted TM domain in all of the other models that we identiﬁed

by various prediction programs (Fig. S2).

Nematode colonization phenotypes varied among the inser-

tion mutants. The insertion at aa 84 in the predicted periplasmic

domain completely abolishes colonization, measured as the aver-

age number of CFU/infective juvenile determined by nematode

sonication and dilution plating (Fig. 5A). Colonization by strains

carrying insertions at aa 322 (SL4), aa 399 (SL6), and aa 451 (SL7)

was detectable but signiﬁcantly attenuated relative to that by a

non-FLAG-tagged wild-type strain, while strains carrying inser-

tions at aa 137, 186, 233, 273, 357, and 466 colonized at levels not

signiﬁcantly different from that of the wild-type nilB control (Fig.

5A). To determine if the colonization defects of the insertion mu-

tants are due to a lack of NilB protein production, Western blot-

ting with anti-FLAG antibody was performed on whole-cell ex-

tracts from all mutants. NilB protein was detected at various

amounts and levels of mobility in X. nematophila strains express-

ing each nilB variant. The amount of NilB produced during labo-

ratory growth did not correlate with nematode colonization, since

FIG 4 FLAG tag insertions distinguish whether NilB adopts a predicted

18-TM or 14-TM strand structure. Insertions were made in three N-terminal

TM strands (before aa 68, aa 102, and aa 121) predicted only by the 18-TM

strand model (D) and in two TM strands predicted by both models (before aa

379 and at aa 421); all mutations were made in nilR mutant cells. The effect of

each mutation on nematode colonization (A) and protein production (B) was

assessed by nematode sonication and immunoblotting with anti-FLAG, re-

spectively. (A) Insertions before aa 68 or aa 121 reduce nematode colonization

below wild-type levels but not to the level observed when nilB is absent (ETn7).

Insertions before aa 102, aa 379, and aa 421 cause severe nematode coloniza-

tion defects. *, P⬍0.01, relative to ETn7values (except with the aa 102 strain,

for which Pwas 0.049). IJ, infective juvenile. (B) When FLAG insertions are in

TM strands predicted only by the 18-TM strand model, NilB protein is detect-

able in whole cells by anti-FLAG Western blotting, but when insertions are in

TM strands predicted by both models, NilB is not detectable. (C and D) Sche-

matic of NilB’s structure and the relative positions of FLAG tag insertions in

the 14-TM (C) and 18-TM (D) strand models. Circles indicate that an inser-

tion mutant colonized better than (white circles) or similarly to (shaded cir-

cles) a strain that lacked SR1. NF, Tn7/SR1 without a FLAG26 insertion in

NilB; ETn7,Tn7lacking SR1; FLAG26, Tn7/SR1 with a FLAG26 insertion in

NilB (no deletions).

FIG 5 Identiﬁcation of NilB regions necessary for colonization. FLAG tags

were inserted into predicted NilB surface loops, the predicted N-terminal

periplasmic domain, and the C terminus. The numbers indicate the amino

acid locations of the tag; for example, 26 indicates that the FLAG tag is directly

before (N-terminal to) amino acid 26 (Fig. 1). Mutants were made in nilR

mutant backgrounds. (A) X. nematophila NilB-FLAG mutants were assessed

for S. carpocapsae nematode colonization. Asterisks indicate colonization lev-

els signiﬁcantly different (P⬍0.05) from that of X. nematophila expressing

NilB without the FLAG tag (NF). (B) NilB protein production by the NilB-

FLAG mutants was assayed on whole cells by Western blotting with anti-FLAG

antibody. The abundances and SDS-PAGE migrations of the NilB-FLAG pro-

teins (bracket) are varied. (C) Schematic of NilB structure, including predicted

locations of FLAG tag insertions in the 14-TM strand model (Fig. 1). Shaded

circles represent insertions that abolished or attenuated nematode coloniza-

tion, while white circles represent insertions that do not signiﬁcantly affect

colonization.

Topology and Functional Regions of X. nematophila NilB

April 2012 Volume 194 Number 7 jb.asm.org 1771

two strains with barely detectable levels of NilB (those with FLAG

inserted before aa 233 and aa 466) displayed colonization proﬁ-

ciency (Fig. 5B). The migration distance of each mutant protein

also did not correlate with colonization proﬁciency or deﬁciency.

For example, NilB proteins with insertions at aa 137 and 186 mi-

grated differently than NilB protein with an insertion at aa 26, but

both strains carrying these nilB variants colonized the nematode

receptacle at levels not different from that of the wild type. The

cause of variability in NilB protein mobility is unknown, but a

similar phenomenon has been reported previously in some inser-

tion (including FLAG tag) mutagenesis studies (28, 49) but not

others (40). In studies where migration variability was observed,

differences were attributed to heat-modiﬁable stability (27). NilB-

FLAG mutants were analyzed for heat modiﬁability by either boil-

ing crude extracts for 15 min at 97°C or not boiling them; samples

were then electrophoresed by SDS-PAGE in a 4°C cold room and

Western blotted. No heat-dependent changes in mobility were

observed (data not shown). Alternatively, insertion mutants may

display differential detergent binding as a reﬂection of variable

protein-lipid or intraprotein interactions (46). When we tested if

the protein migration differences were due to variable NilB glyco-

sylation, results showed that in glycoprotein stains of whole-cell X.

nematophila extracts, no X. nematophila proteins were found to be

glycosylated (data not shown). Despite mobility differences, the

presence of NilB protein in the colonization-defective strains with

a FLAG insertion before aa 85, aa 322, aa 399, and aa 451 supports

the conclusion that these regions are critical for the function of

NilB.

nilB mutants are defective in the entry stage of nematode

colonization. To gain further insight into the effect of nilB-FLAG

mutations on nematode colonization, we assessed nematode col-

onization phenotypes in vivo using GFP-expressing nilB-FLAG

mutants and epiﬂuorescence microscopy. We expected that par-

tial-colonization mutants would fall into one of two classes: out-

growth mutants, with nearly all nematodes carrying one or a few

bacteria in each receptacle, or entry or persistence mutants, with a

mixture of nematodes with fully colonized or completely empty

receptacles. We chose FLAG tag insertion constructs that cause

normal to high (with FLAG before aa 26 and aa 273), low (with

FLAG before aa 322 and aa 399), and intermediate (with FLAG

before aa 186) levels of colonization and incorporated them into

the att Tn7site of a GFP-expressing ⌬SR1 strain (HGB1430). The

percentage of nematodes within each population where the GFP-

labeled bacteria were observed in the receptacle was determined.

An oligo-colonization phenotype, in which a receptacle contained

one or a few bacterial cells (29, 38), was not observed for any of the

5 NilB FLAG tag mutants tested in this experiment (data not

shown). Further, nematode receptacles qualitatively appeared to

be full of bacteria or were empty (e.g., Fig. 6A), suggesting that

these nilB mutations affect entry, not outgrowth. This idea was

further addressed by quantifying the average bacterial load per

nematode through surface sterilization and sonication of 10,000

nematodes. Sonication (average bacterial load) and microscopy

(colonization frequency) data are presented normalized against a

strain carrying wild-type nilB without a FLAG insertion (Table 3).

Fold differences reveal good correlation between the outputs of

the two tests, suggesting that bacteria that gained access to the

receptacle fully colonized it and that the colonization defect of a

nilB mutant occurs prior to its gaining access to the receptacle.

This ﬁnding was not necessarily expected, since a similar compar-

ison of sonication and microscopy data for a nilA mutant revealed

a⬎25-fold difference between nematode colonization frequency

(65% of nematodes) and bacterial load (2.5% of wild-type cells)

(17). Thus, the partial colonization defects observed for nilA and

nilB mutants are distinct. These ﬁndings raise the possibility that

the nil genes, which are together sufﬁcient for nematode coloni-

zation, may act at distinct stages of nematode colonization.

Colonization function of nilB deletion mutants. NilB sur-

face-exposed loops and portions of the predicted N-terminal

periplasmic domain (Fig. 6D) were deleted to further assess the

importance of these regions in nematode colonization. Since mi-

croscopic analysis gave greater sensitivity and smaller errors be-

tween replicates than sonication assays, and results were generally

similar to those derived from sonication assays (Table 3), deletion

constructs were introduced into a GFP-expressing X. nematophila

background and colonization was assessed by epiﬂuorescence mi-

croscopy. In each of these deletion mutants, nematode receptacles

appeared to be empty or fully colonized, consistent with our pre-

vious observations. Deletions in SL4 (of aa 314 to 320), SL5 (aa

352 to 357), and SL6 (aa 389 to 410) had signiﬁcant negative yet

variable effects on nematode colonization (Fig. 6B), and deleting

SL6 caused the most severe colonization deﬁciency (equivalent to

a strain lacking nilB entirely). The ﬁnding that the SL3 (aa 266 to

279) deletion strain colonized greater than 50% of the nematodes

in the population emphasizes that some extracellular domains

play a more signiﬁcant role in colonization than others.

Some regions of the predicted N-terminal globular domain (aa

26 to 156) are also important for nematode colonization (Fig. 6B).

Strains carrying deletions of aa 92 to 108 or aa 109 to 156 (the latter

encompassing the predicted TPR domain) reduce colonization to

levels equivalent to that of a strain lacking nilB, whereas deletions

of aa 76 to 91, 26 to 108, and 26 to 156 have a milder, yet signiﬁ-

cant, colonization defect. In contrast, deleting aa 26 to 42 and aa

26 to 59 does not reduce colonization levels, and deleting aa 43 to

59 has only a mildly deleterious effect on nematode colonization

frequency.

Although NilB protein levels and mobilities on SDS-PAGE gels

vary for the different deletion mutants, the migration differences

correlate with the expected size change due to deletion of residues

(Fig. 6C). NilB is detectable for every deletion mutant except that

with a deletion of aa 292 to 298, consistent with the prediction that

these amino acids form a TM domain. As seen with the FLAG

insertion mutants, colonization levels do not correlate with NilB

protein levels. For example, NilB is relatively abundant in the col-

onization-defective ⌬SL6 strain (deletion of aa 389 to 410), in the

colonization-defective strain with a deletion of aa 76 to 91, and in

the colonization-proﬁcient strain with a deletion of aa 26 to 42 but

not in the colonization-proﬁcient strain with a deletion of aa 26 to

59. Therefore, low protein levels likely do not fully explain the

colonization defects of deletion mutants.

DISCUSSION

We have shown that NilB is a surface-exposed protein that is

poorly expressed under laboratory growth conditions. When nilB

expression is derepressed by deletion of nilR, NilB protein levels

are highest under nutrient-limiting conditions (e.g., in deﬁned

medium and/or stationary phase) (Fig. 2), indicating that NilB

expression responds to nutrient conditions independently of

NilR. Because of this, we predict that, during nematode coloniza-

tion, NilB expression is triggered by multiple signals, both NilR

Bhasin et al.

1772 jb.asm.org Journal of Bacteriology

FIG 6 In-frame amino acid deletions identify structurally and functionally required domains of NilB. (A) Partial nematode colonization mutants fully

colonize few nematodes instead of partially colonizing many nematodes in the population. Representative images of nematodes grown on colonization

mutants are shown. Images from left to right are as follows: a 40⫻magniﬁcation of 2 colonized (HGB1502) nematodes (the green dot is an epiﬂuorescent

overlay) and 1 uncolonized nematode (the black arrow points to the location of an empty receptacle), a zoomed image of a colonized nematode

(HGB1502) with an epiﬂuorescent overlay on a DIC image, and a zoomed image of an uncolonized nematode (HGB1429) with an epiﬂuorescent overlay

on a DIC image, with the empty nematode receptacle highlighted (white outlining). Scale bars from left to right: 50

␮

m, 5

␮

m, and 5

␮

m. (B) Small regions

in the N-terminal periplasmic domain were individually sequentially deleted (⌬026–042, ⌬043– 059, ⌬076 – 091, ⌬092–108, ⌬109 –156), along with

larger regions of this domain (⌬026 –059, ⌬026 –108, ⌬026–156). Amino acids were also deleted from SL3, SL4, SL5, or SL6, as well as a region within the

predicted TM domain between SL3 and SL4 (as a negative control). Each of these deletions was created in the plasmid pTn7/SR1/nilB-FLAG26, and Tn7

integration at the att Tn7site of GFP-expressing X. nematophila ATCC 19061 ⌬SR1 ⌬nilR was conﬁrmed. (A) Nematode colonization of each deletion

mutant assessed by microscopy. Pwas ⬍0.05 for differences from values for the negative control (eTn7) (#) or positive control (NF) (*). Strain eTn7lacks

nilB, while NF contains nilB but no FLAG tag. (C) NilB-FLAG proteins of various abundances and mobilities are produced in whole cells by all deletion

mutants except that lacking aa 292 to 298. ⫹indicates detectable NilB, and ⫺indicates no detectable NilB protein. NF contains nilB but no FLAG tag.

FLAG26 contains the FLAG tag before aa 26 of nilB. (D) Schematics of NilB structures and deletions. The bars represent the N-terminal periplasmic

domain from aa 26 to 164. The length of the shaded area represents the size of the deletion. The color of the shading represents the effect of the deletion

on nematode colonization; black indicates no effect relative to that of the wild type, gray indicates an effect intermediate between those of the wild type

and the negative control (no NilB), and the cross-hatched shading indicates no signiﬁcant difference from the effect of the negative control.

Topology and Functional Regions of X. nematophila NilB

April 2012 Volume 194 Number 7 jb.asm.org 1773

dependent and NilR independent. Since under laboratory growth

conditions NilR-dependent repression occurred regardless of

growth phase or medium, we suggest that the signal necessary for

NilR derepression is absent under these conditions and may be

speciﬁc to the nematode host. Analysis of the colonization pheno-

types of nilB mutant alleles revealed that NilB functions at the

stage of entry into the nematode receptacle, rather than during

later stages of growth and persistence, indicating that the signals

necessary for NilB expression likely are present at the very early

stages of colonization.

Our analysis of nilB mutants supports the model that NilB

adopts a 14-TM beta-barrel structure, based on the premise that

insertions within, or deletions of, TM domains prevent proper

localization and lead to protein degradation, while insertions or

deletions in other regions (e.g., those that are surface exposed) are

generally tolerated (23, 49). Insertions or deletions that prevented

detection of NilB protein were in TM strands (aa 292 to 298, 379,

and 421) predicted by the 14-TM model. In contrast, NilB protein

was detectable in mutants with insertions or deletions in regions

predicted to be TM in an 18-TM model but periplasmic in a

14-TM model (aa 68, 92 to 108, 102, 109 to 156, and 121). We also

show that NilB is membrane localized and has surface-exposed

regions, based on its sensitivity to proteinase K digestion in whole

cells (Fig. 3). The appearance of truncated, large-molecular-

weight FLAG-reactive proteins after proteinase K digestion is con-

sistent with cleavage of the C-terminal but not the N-terminal

(antibody-reactive) domain (Fig. 3), supporting the bioinformati-

cally predicted location of the N terminus in the periplasm. Taken

together, these data support a protein structure prediction model

where NilB has 7 SLs (SL1 to -7) and an ⬃140-aa N-terminal

periplasmic domain. Below we discuss how these results provide

insight into possible mechanisms of NilB function in coloniza-

tion.

We identiﬁed several connections between NilB function

and nutrient availability/transport through our studies on NilB

expression, genomic context, and predicted topology. As

discussed above, NilB expression is elevated by nutrient limi-

tation during laboratory growth. Further, pathogens such as

Mannheimia haemolytica and Actinobacillus pleuropneumo-

niae, which derive iron from host-speciﬁc transferrin, lactofer-

rin, or hemoglobin molecules (51), encode NilB homologs near

iron binding (YP_002173032) or host transferrin utilization

(ZP_04464085) genes, respectively. This genomic context sug-

gests that NilB homologs, and perhaps NilB, may also function

in nutrient acquisition. Finally, our data suggest that NilB is a

14-TM beta-barrel protein with a large periplasmic N-terminal

domain. In many beta-barrel proteins, large periplasmic do-

mains function as plugs that help control beta-barrel pore per-

meability. For example, such topology is typical of small-mol-

ecule transporters (47, 53, 57) that depend on the inner

membrane protein TonB to provide the energy for nutrient

transport (e.g., iron, vitamin B

) (31). TonB-dependent trans-

porters include transferrin receptors (31), such as those en-

coded near nilB homologs. At ﬁrst glance, NilB seems unlikely

to be a TonB-dependent protein since it lacks a canonical TonB

box, since most TonB-dependent transporters are 22-TM do-

main proteins, and since an X. nematophila tonB mutant does

not have a colonization defect (38). However, TonB box iden-

tities are varied among TonB-dependent transporters (34), and

the X. nematophila genome encodes two additional TonB-like

genes (13), one of which is oriented in tandem with the NilB-

like protein XNC1_0074 (see Fig. S1 in the supplemental ma-

terial). Taken together, the links between NilB and TonB de-

tailed above are consistent with a role for NilB in nutrient

transport. However, our data do not exclude the possibility of

an adhesin function for NilB, and adhesion and nutrient acqui-

sition models are not mutually exclusive.

Our work has revealed several critical regions of NilB necessary

for its function in nematode colonization, including SL6 and the

N-terminal TPR-like domain. Deletion of SL6 reduced nematode

colonization to levels observed when nilB was absent, but this

reduction in nematode colonization is not a result of NilB insta-

bility since NilB is detected in Western blots of both whole-cell

lysates (Fig. 5B and 6C) and membrane preparations (data not

shown). SL6 (28 residues) is the longest SL in NilB (9 residues

longer than the next longest, SL1), and in a full-length alignment

of all NilB homologs, NilB has more residues at SL6 than any other

NilB homolog (data not shown). The longer length of SL6 in NilB

than in NilB homologs may indicate the importance of SL6 resi-

dues in the speciﬁcity of NilB in S. carpocapsae association. Vari-

ous functions could be envisioned for this region, including di-

rectly binding to host surfaces or molecules. For example, in beta-

TABLE 3 Fold change in two independent experiments between bacterial load and nematode colonization frequency of nilB-FLAG insertion

mutants

FLAG

insertion

Expt 1 Expt 2

Relative bacterial

load (%)

Relative

frequency (%)

Fold

difference

Relative bacterial

load (%)

Relative

frequency (%)

Fold

difference

aa 26 55.882 89.490 1.60 119.156 102.007 1.17

aa 186 52.574 52.352 1.00 61.039 49.763 1.23

aa 273 106.985 96.734 1.11 99.675 91.455 1.09

aa 322 32.279 25.762 1.25 60.390 38.542 1.57

aa 399 1.397 0.960 1.46 4.789 4.721 1.01

eTn70.042 0.015 2.86 0.009 0.009 1.00

None 100.000 100.000 1.00 100.000 100.000 1.00

Bacterial load was assessed by sonication of 10,000 nematodes and dilution plating to determine bacterial counts. Data are presented as percentages of the value for cells expressing

wild-type nilB with no FLAG insertion.

Nematode colonization frequency was assessed by epiﬂuorescence microscopy of nematodes grown on GFP-expressing bacteria. Data are presented as percentages of the value for

cells expressing wild-type nilB with no FLAG insertion.

Fold difference represents the ratio of relative colonization frequency to relative bacterial load. A ratio of 1 indicates that the colonization defect of the strain is explained by a

reduced frequency of colonization (e.g., initiation). A ratio of greater than 1 indicates that a defect in outgrowth contributes to observed defects in bacterial load.

Bhasin et al.

1774 jb.asm.org Journal of Bacteriology

barrel transporters, surface loops act by trapping and directing

transport of speciﬁc molecules through the pore and/or occluding

foreign molecules during transport as part of a molecular airlock

system (see, e.g., references 20 and 45).

Deletion of the TPR-like domain (⌬109 –156) caused a coloni-

zation defect that could be ameliorated by concomitant deletion

of aa 26 to 108 (in Fig. 6B, compare infective juvenile colonization

of the ⌬109 –156 mutant [⬃0.008% of nematodes] versus that of

the ⌬26 –156 mutant [0.8%]). Such a result could be explained if

the NilB periplasmic domain acts as a plug whose “open” and

“closed” states, or some other pore modiﬁcation, are regulated

within the TPR region. Deletion of the domain necessary to tran-

sition to the “open” state would effectively block the pore and

prevent transport, whereas deletion of the entire plug might result

in a constitutively leaky pore that could function, albeit less efﬁ-

ciently than wild-type NilB, in colonization. Alternatively, the

TPR-like domain may interact with other regions of the protein

(e.g., a pore) to impact the structure and function of surface-

exposed loops that may be necessary for adhesion, while deletion

of additional residues of the N-terminal domain may allow NilB to

adopt a more wild-type structure. Regardless, our data suggest

that the TPR-like domain is critical for NilB function, possibly due

to its role in mediating protein-protein interactions. In this re-

gard, it will be of interest to determine if NilB interacts with other

proteins and if such interactions depend on the TPR-like domain.

NilC, an outer membrane, periplasmically oriented lipoprotein

(15), is an obvious candidate for a NilB-interacting protein since it

is encoded on the same genetic locus as NilB (30), is repressed by

NilR (16), and is a species speciﬁcity factor for S. carpocapsae nem-

atodes (17).

Although, based on our data, we are unable to deﬁnitively con-

clude that NilB functions as a transporter or an adhesin, either

model warrants some speculation regarding the identity of its sub-

strate and whether this molecule is speciﬁc to S. carpocapsae. NilB

is a speciﬁcity determinant, and S. carpocapsae nematodes are col-

onized only by Xenorhabdus species that express nilB (17). Since

no defects in nilB mutant growth, secondary metabolism, or other

aspects of physiology have been observed in nilR

⫹

cells during

laboratory culture (30), including testing with the full panel of

Biolog metabolic phenotype arrays (PM1 to PM10; Biolog) (data

not shown), NilB may be involved in the metabolism of a nutrient

or molecule that is absent under laboratory conditions. Alterna-

tively, NilR repression of NilB under these conditions may pre-

clude identiﬁcation of NilB-dependent phenotypes. Therefore,

further investigations of NilB-dependent laboratory growth phe-

notypes in a nilR mutant background are warranted. However,

intriguing hypotheses are that NilB is necessary to transport a

host-speciﬁc nutrient or adhere to a host-speciﬁc molecule and

that NilR derepression occurs only in the presence of this mole-

cule to prevent inappropriate expression of NilB (e.g., in the insect

host, where it may be immunogenic).

We previously suggested (13, 17) that species speciﬁcity in Xe-

norhabdus does not follow the paradigm of host range speciﬁcity

in the well-studied leguminous plant-Rhizobium symbioses. In

these symbioses, each rhizobium species produces a common sig-

naling molecule, the nod factor, and speciﬁcity is determined by

modiﬁcations of the core nod factor molecule (24). In contrast,

the host range speciﬁcity determinant NilB is lacking from other

Xenorhabdus species, suggesting that it is a novel, derived trait that

determines speciﬁcity in this bacterium. However, this idea is

challenged by the results presented here. Structural predictions of

NilB-like proteins, including one in X. bovienii, reveal similar to-

pologies despite relatively low amino acid sequence similarity

(⬃18 to 25%), raising the possibility that NilB-like proteins with

structural (and perhaps functional) similarity are produced by

other Xenorhabdus species, but they have insufﬁciently low coding

sequence identity to be identiﬁed by PCR (30), Southern hybrid-

ization (17; J. Chaston and H. Goodrich-Blair, unpublished data),

or genome sequencing (14). Therefore, NilB-like proteins may

play similar roles among all Xenorhabdus species in mediating host

interactions and speciﬁcity, with NilB representing a highly di-

verged, horizontally acquired (17) member of this family. Subtle

differences in surface-exposed loops (e.g., SL6) may determine

speciﬁcity, much like nod factor modiﬁcations. Future studies

that determine the mechanism of action of NilB and species spec-

iﬁcity factors in other Xenorhabdus spp. will help test if the para-

digm of speciﬁcity determined by “variations on a common

theme” pertains to Xenorhabdus-nematode interactions.

ACKNOWLEDGMENTS

This work was supported by grants awarded to H.G.-B. from the National

Science Foundation (NSF) (IOS-0950873) and the National Institutes of

Health (NIH) (GM059776). J.M.C. was supported by NIH National Re-

search Service Award T32 (grant AI55397, Microbes in Health and Dis-

ease) and an NSF Graduate Research Fellowship. A.B. was supported by

NIH grant F32 GM072342.

We are grateful to Tom Silhavy for helpful discussions and to anony-

mous reviewers, whose comments signiﬁcantly improved the manuscript.

We are also grateful to Aaron Andersen for technical support and Xinxin

Yu and Yang Zhao in the CALS statistical consulting service (UW—Mad-

ison) for assistance with choosing and applying statistical models.

REFERENCES

1. Abe Y, et al. 2000. Structural basis of presequence recognition by the

mitochondrial protein import receptor Tom20. Cell 100:551–560.

2. Altschul SF, et al. 1997. Gapped BLAST and PSI-BLAST: a new genera-

tion of protein database search programs. Nucleic Acids Res. 25:3389–

3402.

3. Andrade MA, Perez-Iratxeta C, Ponting CP. 2001. Protein repeats:

structures, functions, and evolution. J. Struct. Biol. 134:117–131.

4. Bagos PG, Liakopoulos TD, Hamodrakas SJ. 2005. Evaluation of meth-

ods for predicting the topology of beta-barrel outer membrane proteins

and a consensus prediction method. BMC Bioinformatics 6:7.

5. Bagos PG, Liakopoulos TD, Spyropoulos IC, Hamodrakas SJ. 2004. A

hidden Markov model method, capable of predicting and discriminating

beta-barrel outer membrane proteins. BMC Bioinformatics 5:29.

6. Bagos PG, Liakopoulos TD, Spyropoulos IC, Hamodrakas SJ. 2004.

PRED-TMBB: a web server for predicting the topology of beta-barrel

outer membrane proteins. Nucleic Acids Res. 32:W400–W404.

7. Bao Y, Lies DP, Fu H, Roberts GP. 1991. An improved Tn7-based system

for the single-copy insertion of cloned genes into chromosomes of Gram-

negative bacteria. Gene 109:167–168.

8. Bendtsen JD, Nielsen H, von Heijne G, Brunak S. 2004. Improved

prediction of signal peptides: SignalP 3.0. J. Mol. Biol. 340:783–795.

9. Bigelow HR, Petrey DS, Liu J, Przybylski D, Rost B. 2004. Predicting

transmembrane beta-barrels in proteomes. Nucleic Acids Res. 32:2566–

2577.

10. Bird AF, Akhurst RJ. 1983. The nature of the intestinal vesicle in nema-

todes of the family Steinernematidae. Int. J. Parasitol. 13:599– 606.

11. Blatch GL, Lassle M. 1999. The tetratricopeptide repeat: a structural

motif mediating protein-protein interactions. Bioessays 21:932–939.

12. Boemare NE, Akhurst RJ, Mourant RG. 1993. DNA relatedness between

Xenorhabdus spp. (Enterobacteriaceae), symbiotic bacteria of ento-

mopathogenic nematodes, and a proposal to transfer Xenorhabdus lumi-

nescens to a new genus, Photorhabdus, gen. nov. Int. J. Syst. Bacteriol.

43:249–255.

13. Chaston J, Goodrich-Blair H. 2010. Common trends in mutualism re-

Topology and Functional Regions of X. nematophila NilB

April 2012 Volume 194 Number 7 jb.asm.org 1775

vealed by model associations between invertebrates and bacteria. FEMS

Microbiol. Rev. 34:41–58.

14. Chaston JM, et al. 2011. The entomopathogenic bacterial endosymbionts

Xenorhabdus and Photorhabdus: convergent lifestyles from divergent ge-

nomes. PLoS One 6:e27909.

15. Cowles CE, Goodrich-Blair H. 2004. Characterization of a lipoprotein,

NilC, required by Xenorhabdus nematophila for mutualism with its nem-

atode host. Mol. Microbiol. 54:464– 477.

16. Cowles CE, Goodrich-Blair H. 2006. nilR is necessary for co-ordinate

repression of Xenorhabdus nematophila mutualism genes. Mol. Microbiol.

62:760–771.

17. Cowles CE, Goodrich-Blair H. 2008. The Xenorhabdus nematophila nil-

ABC genes confer the ability of Xenorhabdus spp. to colonize Steinernema

carpocapsae nematodes. J. Bacteriol. 190:4121–4128.

18. Cowles KN, Cowles CE, Richards GR, Martens EC, Goodrich-Blair H.

2007. The global regulator Lrp contributes to mutualism, pathogenesis

and phenotypic variation in the bacterium Xenorhabdus nematophila.

Cell. Microbiol. 9:1311–1323.

19. D’Andrea LD, Regan L. 2003. TPR proteins: the versatile helix. Trends

Biochem. Sci. 28:655– 662.

20. Ferguson AD, et al. 2002. Structural basis of gating by the outer mem-

brane transporter FecA. Science 295:1715–1719.

21. Flores-Lara Y, Renneckar D, Forst S, Goodrich-Blair H, Stock P. 2007.

Inﬂuence of nematode age and culture conditions on morphological and

physiological parameters in the bacterial vesicle of Steinernema carpocap-

sae (Nematoda: Steinernematidae). J. Invertebr. Pathol. 95:110–118.

22. Forst S, Dowds B, Boemare N, Stackebrandt E. 1997. Xenorhabdus and

Photorhabdus spp.: bugs that kill bugs. Annu. Rev. Microbiol. 51:47–72.

23. Freudl R. 1989. Insertion of peptides into cell-surface-exposed areas of

the Escherichia coli OmpA protein does not interfere with export and

membrane assembly. Gene 82:229–236.

24. Garg N, Geetanjali. 2007. Symbiotic nitrogen ﬁxation in legume nodules:

process and signaling. A review. Agron. Sustain. Dev. 27:59– 68.

25. Goetsch M, Owen H, Goldman B, Forst S. 2006. Analysis of the PixA

inclusion body protein of Xenorhabdus nematophila. J. Bacteriol. 188:

2706–2710.

26. Goodrich-Blair H, Clarke DJ. 2007. Mutualism and pathogenesis in

Xenorhabdus and Photorhabdus: two roads to the same destination. Mol.

Microbiol. 64:260–268.

27. Hancock RE, Carey AM. 1979. Outer membrane of Pseudomonas aerugi-

nosa: heat-2-mercaptoethanol-modiﬁable proteins. J. Bacteriol. 140:902–

910.

28. Hay ID, Rehman ZU, Rehm BH. 2010. Membrane topology of outer

membrane protein AlgE, which is required for alginate production in

Pseudomonas aeruginosa. Appl. Environ. Microbiol. 76:1806–1812.

29. Herbert Tran EE, Andersen AW, Goodrich-Blair H. 2009. CpxRA in-

ﬂuences Xenorhabdus nematophila colonization initiation and outgrowth

in Steinernema carpocapsae nematodes through regulation of the nil locus.

Appl. Environ. Microbiol. 75:4007–4014.

30. Heungens K, Cowles CE, Goodrich-Blair H. 2002. Identiﬁcation of

Xenorhabdus nematophila genes required for mutualistic colonization of

Steinernema carpocapsae nematodes. Mol. Microbiol. 45:1337–1353.

31. Jarosik GP, Maciver I, Hansen EJ. 1995. Utilization of transferrin-bound

iron by Haemophilus inﬂuenzae requires an intact tonB gene. Infect.

Immun. 63:710–713.

32. Kaya HK, Gaugler R. 1993. Entomopathogenic nematodes. Annu. Rev.

Entomol. 38:181–206.

33. Knowles TJ, Scott-Tucker A, Overduin M, Henderson IR. 2009. Mem-

brane protein architects: the role of the BAM complex in outer membrane

protein assembly. Nat. Rev. Microbiol. 7:206–214.

34. Krewulak KD, Vogel HJ. 2011. TonB or not TonB: is that the question?

Biochem. Cell Biol. 89:87–97.

35. Marchler-Bauer A, et al. 2007. CDD: a conserved domain database for

interactive domain family analysis. Nucleic Acids Res. 35:D237–D240.

36. Martelli PL, Fariselli P, Krogh A, Casadio R. 2002. A sequence-proﬁle-

based HMM for predicting and discriminating beta barrel membrane pro-

teins. Bioinformatics 18(Suppl 1):S46 –S53.

37. Martens EC, Heungens K, Goodrich-Blair H. 2003. Early colonization

events in the mutualistic association between Steinernema carpocapsae

nematodes and Xenorhabdus nematophila bacteria. J. Bacteriol. 185:3147–

3154.

38. Martens EC, Russell FM, Goodrich-Blair H. 2005. Analysis of Xenorh-

abdus nematophila metabolic mutants yields insight into stages of Steiner-

nema carpocapsae nematode intestinal colonization. Mol. Microbiol. 51:

28– 45.

39. Morgan JAW, Kuntzelmann V, Tavernor S, Ousley MA, Winstanley C.

1997. Survival of Xenorhabdus nematophilus and Photorhabdus lumine-

scens in water and soil. J. Appl. Microbiol. 83:665–670.

40. Nguyen KA, et al. 2009. Veriﬁcation of a topology model of PorT as an

integral outer-membrane protein in Porphyromonas gingivalis. Microbi-

ology 155:328–337.

41. Orchard SS, Goodrich-Blair H. 2004. Identiﬁcation and functional char-

acterization of a Xenorhabdus nematophila oligopeptide permease. Appl.

Environ. Microbiol. 70:5621–5627.

42. Poinar GO. 1966. The presence of Achromobacter nematophilus in the

infective stage of a Neoaplectana sp. (Steinernematidae: Nematoda).

Nematologica 12:105–108.

43. Poinar GO, Jr. 1979. Nematodes for biological control of insects. CRC

Press, Boca Raton, FL.

44. Poinar GO, Jr, Thomas GM. 1967. The nature of Achromobacter nema-

tophilus as an insect pathogen. J. Invertebr. Pathol. 9:510–514.

45. Postle K, Kadner RJ. 2003. Touch and go: tying TonB to transport. Mol.

Microbiol. 49:869– 882.

46. Rath A, Glibowicka M, Nadeau VG, Chen G, Deber CM. 2009. Deter-

gent binding explains anomalous SDS-PAGE migration of membrane

proteins. Proc. Natl. Acad. Sci. U. S. A. 106:1760–1765.

47. Ratledge C, Dover LG. 2000. Iron metabolism in pathogenic bacteria.

Annu. Rev. Microbiol. 54:881–941.

48. R Development Core Team. 2006. R: a language and environment for

statistical computing. R Foundation for Statistical Computing, Vienna,

Austria.

49. Rehm BH, Hancock RE. 1996. Membrane topology of the outer mem-

brane protein OprH from Pseudomonas aeruginosa: PCR-mediated site-

directed insertion and deletion mutagenesis. J. Bacteriol. 178:3346–3349.

50. Richards GR, Goodrich-Blair H. 2009. Masters of conquest and pillage:

Xenorhabdus nematophila global regulators control transitions from viru-

lence to nutrient acquisition. Cell. Microbiol. 11:1025–1033.

51. Schryvers AB, Gonzalez GC. 1990. Receptors for transferrin in patho-

genic bacteria are speciﬁc for the host’s protein. Can. J. Microbiol. 36:145–

147.

52. Stabb EV, Ruby EG. 2002. RP4-based plasmids for conjugation between

Escherichia coli and members of the Vibrionaceae. Methods Enzymol. 358:

413–426.

53. Usher KC, O

¨zkan E, Gardner KH, Deisenhofer J. 2001. The plug domain

of FepA, a TonB-dependent transport protein from Escherichia coli, binds

its siderophore in the absence of the transmembrane barrel domain. Proc.

Natl. Acad. Sci. U. S. A. 98:10676–10681.

54. Vivas EI, Goodrich-Blair H. 2001. Xenorhabdus nematophilus as a model

for host-bacterium interactions: rpoS is necessary for mutualism with

nematodes. J. Bacteriol. 183:4687–4693.

55. Volgyi A, Fodor A, Szentirmai A, Forst S. 1998. Phase variation in

Xenorhabdus nematophilus. Appl. Environ. Microbiol. 64:1188–1193.

56. White GF. 1927. A method for obtaining infective nematode larvae from

cultures. Science 66:302–303.

57. Wiener MC. 2005. TonB-dependent outer membrane transport: going

for Baroque? Curr. Opin. Struct. Biol. 15:394– 400.

58. Wouts WM. 1980. Biology, life cycle, and redescription of Neoaplectana

bibionis Bovien, 1937 Nematoda: Steinernematidae. J. Nematol. 12:62–72.

Bhasin et al.

1776 jb.asm.org Journal of Bacteriology

Content uploaded by Heidi Goodrich-Blair

Content may be subject to copyright.

Bacterial hemophilin homologs and their specific type eleven secretor proteins have conserved roles in heme capture and are diversifying as a family

Article

Full-text available

Mar 2024
J BACTERIOL

Cellular life relies on enzymes that require metals, which must be acquired from extracellular sources. Bacteria utilize surface and secreted proteins to acquire such valuable nutrients from their environment. These include the cargo proteins of the type eleven secretion system (T11SS), which have been connected to host specificity, metal homeostasis, and nutritional immunity evasion. This Sec-dependent, Gram-negative secretion system is encoded by organisms throughout the phylum Proteobacteria, including human pathogens Neisseria meningitidis, Proteus mirabilis, Acinetobacter baumannii, and Haemophilus influenzae. Experimentally verified T11SS-dependent cargo include transferrin-binding protein B (TbpB), the hemophilin homologs heme receptor protein C (HrpC), hemophilin A (HphA), the immune evasion protein factor-H binding protein (fHbp), and the host symbiosis factor nematode intestinal localization protein C (NilC). Here, we examined the specificity of T11SS systems for their cognate cargo proteins using taxonomically distributed homolog pairs of T11SS and hemophilin cargo and explored the ligand binding ability of those hemophilin cargo homologs. In vivo expression in Escherichia coli of hemophilin homologs revealed that each is secreted in a specific manner by its cognate T11SS protein. Sequence analysis and structural modeling suggest that all hemophilin homologs share an N-terminal ligand-binding domain with the same topology as the ligand-binding domains of the Haemophilus haemolyticus heme binding protein (Hpl) and HphA. We term this signature feature of this group of proteins the hemophilin ligand-binding domain. Network analysis of hemophilin homologs revealed five subclusters and representatives from four of these showed variable heme-binding activities, which, combined with sequence-structure variation, suggests that hemophilins are diversifying in function. IMPORTANCE The secreted protein hemophilin and its homologs contribute to the survival of several bacterial symbionts within their respective host environments. Here, we compared taxonomically diverse hemophilin homologs and their paired Type 11 secretion systems (T11SS) to determine if heme binding and T11SS secretion are conserved characteristics of this family. We establish the existence of divergent hemophilin sub-families and describe structural features that contribute to distinct ligand-binding behaviors. Furthermore, we demonstrate that T11SS are specific for their cognate hemophilin family cargo proteins. Our work establishes that hemophilin homolog-T11SS pairs are diverging from each other, potentially evolving into novel ligand acquisition systems that provide competitive benefits in host niches.

A Surface Exposed, Two-Domain Lipoprotein Cargo of a Type XI Secretion System Promotes Colonization of Host Intestinal Epithelia Expressing Glycans

Article

Full-text available

Apr 2022

The only known required component of the newly described Type XI secretion system (TXISS) is an outer membrane protein (OMP) of the DUF560 family. TXISSOMPs are broadly distributed across proteobacteria, but properties of the cargo proteins they secrete are largely unexplored. We report biophysical, histochemical, and phenotypic evidence that Xenorhabdus nematophila NilC is surface exposed. Biophysical data and structure predictions indicate that NilC is a two-domain protein with a C-terminal, 8-stranded β-barrel. This structure has been noted as a common feature of TXISS effectors and may be important for interactions with the TXISSOMP. The NilC N-terminal domain is more enigmatic, but our results indicate it is ordered and forms a β-sheet structure, and bioinformatics suggest structural similarities to carbohydrate-binding proteins. X. nematophila NilC and its presumptive TXISSOMP partner NilB are required for colonizing the anterior intestine of Steinernema carpocapsae nematodes: the receptacle of free-living, infective juveniles and the anterior intestinal cecum (AIC) in juveniles and adults. We show that, in adult nematodes, the AIC expresses a Wheat Germ Agglutinin (WGA)-reactive material, indicating the presence of N-acetylglucosamine or N-acetylneuraminic acid sugars on the AIC surface. A role for this material in colonization is supported by the fact that exogenous addition of WGA can inhibit AIC colonization by X. nematophila. Conversely, the addition of exogenous purified NilC increases the frequency with which X. nematophila is observed at the AIC, demonstrating that abundant extracellular NilC can enhance colonization. NilC may facilitate X. nematophila adherence to the nematode intestinal surface by binding to host glycans, it might support X. nematophila nutrition by cleaving sugars from the host surface, or it might help protect X. nematophila from nematode host immunity. Proteomic and metabolomic analyses of wild type X. nematophila compared to those lacking nilB and nilC revealed differences in cell wall and secreted polysaccharide metabolic pathways. Additionally, purified NilC is capable of binding peptidoglycan, suggesting that periplasmic NilC may interact with the bacterial cell wall. Overall, these findings support a model that NilB-regulated surface exposure of NilC mediates interactions between X. nematophila and host surface glycans during colonization. This is a previously unknown function for a TXISS.

A Widespread Bacterial Secretion System with Diverse Substrates

Article

Full-text available

Aug 2021

The microbial constituency of a host-associated microbiome emerges from a complex physical and chemical interplay of microbial colonization factors, host surface conditions, and host immunological responses. To fill unique niches within a host, bacteria encode surface and secreted proteins that enable interactions with and responses to the host and cooccurring microbes.

Beyond Slam: the DUF560 family outer membrane protein superfamily SPAM has distinct network subclusters that suggest a role in non-lipidated substrate transport and bacterial environmental adaptation

Preprint

Full-text available

Jan 2020

In host-associated bacteria, surface and secreted proteins mediate acquisition of nutrients, interactions with host cells, and specificity of tissue-localization. In Gram-negative bacteria, the mechanism by which many proteins cross or become tethered to the outer membrane remains unclear. The d omain of u nknown function (DUF)560 occurs in outer membrane proteins throughout Proteobacteria and has been implicated in host-bacteria interactions and lipoprotein surface exposure. We used sequence similarity networking to reveal three subfamilies of DUF560 homologs. One subfamily includes those DUF560 proteins experimentally characterized to date: NilB, a host-range determinant of the nematode-mutualist Xenorhabdus nematophila , and the s urface lipoprotein a ssembly m odulators Slam1 and Slam2, which facilitate msurface exposure of lipoproteins in Neisseria meningitidis (1, 2). We show that DUF560 proteins from a second subfamily facilitate secretion of soluble, non-lipidated proteins across the outer membrane. Using in silico analysis, we demonstrate that DUF560 gene complement correlates with bacterial environment at a macro level and host association at a species level. The DUF560 protein superfamily represents a newly characterized Gram-negative secretion system capable of lipoprotein surface exposure and soluble protein secretion with conserved roles in facilitating symbiosis. In light of these data, we propose that it be titled the type eleven s ecretion s ystem (TXISS). Importance The microbial constituents of a host associated microbiome are decided by a complex interplay of microbial colonization factors, host surface conditions, and host immunological responses. Filling such niches requires bacteria to encode an arsenal of surface and secreted proteins to effectively interact with the host and co-occurring microbes. Bioinformatic predictions of the localization and function of putative bacterial colonization factors are essential for assessing the potential of bacteria to engage in pathogenic, mutualistic, or commensal activities. This study uses publicly available genome sequence data, alongside experimental results from representative gene products from Xenorhabdus nematophila , to demonstrate a role for DUF560 family proteins in the secretion of bacterial effectors of host interactions. Our research delineates a broadly distributed family of proteins and enables more accurate predictions of the localization of colonization factors throughout Proteobacteria.

An Entomopathogenic Nematode, Oscheius tipulae, Exhibiting a Wide Pathogenic Spectrum and Its Application to Control Dipteran Insect Pests

Preprint

Full-text available

Apr 2024

An entomopathogenic nematode, Oscheius tipulae, was isolated from a soil sample. The identification of this species was supported by morphological and molecular markers. The nematode isolate exhibited pathogenicity against different target insects including lepidopteran, coleopteran, and dipteran insects. The virulence of this nematode was similar to that of a well-known entomopathogenic nematode, Steinernema carpocapsae, against the same insect targets. A comparative metagenomics analysis of these two nematode species predicted the existence of a combined total of 272 bacterial species in their intestines, of which 51 bacterial species were shared between the two nematode species. In particular, the common gut bacteria included several entomopathogenic bacteria including Xenorhabdus nematophila, which is known as a symbiotic bacterium to S. carpocapsae. The nematode virulence of O. tipulae to insects was enhanced by an addition of dexamethasone but suppressed by an addition of arachidonic acid, suggesting that the immune defenses of the target insects against the nematode infection is mediated by eicosanoids, which would be manipulated by the symbiotic bacteria of the nematode. Unlike S. carpocapsae, O. tipulae showed high virulence against dipteran insects including fruit flies, onion flies, and mosquitoes. O. tipulae showed particularly high control efficacies against the onion maggot, Delia platura, infesting the Welsh onion in the rhizosphere in both pot and field assays.

OMPdb: A Global Hub of Beta-Barrel Outer Membrane Proteins

Article

Full-text available

Apr 2021

OMPdb ( www.ompdb.org ) was introduced as a database for β-barrel outer membrane proteins from Gram-negative bacteria in 2011 and then included 69,354 entries classified into 85 families. The database has been updated continuously using a collection of characteristic profile Hidden Markov Models able to discriminate between the different families of prokaryotic transmembrane β-barrels. The number of families has increased ultimately to a total of 129 families in the current, second major version of OMPdb. New additions have been made in parallel with efforts to update existing families and add novel families. Here, we present the upgrade of OMPdb, which from now on aims to become a global repository for all transmembrane β-barrel proteins, both eukaryotic and bacterial.

The insect pathogenic bacterium Xenorhabdus innexi has attenuated virulence in multiple insect model hosts yet encodes a potent mosquitocidal toxin

Article

Full-text available

Dec 2017
BMC GENOMICS

Background: Xenorhabdus innexi is a bacterial symbiont of Steinernema scapterisci nematodes, which is a cricket specialist parasite and together the nematode and bacteria infect and kill crickets. Curiously, X. innexi expresses a potent extracellular mosquitocidal toxin activity in culture supernatants. We sequenced a draft genome of X. innexi and compared it to the genomes of related pathogens to elucidate the nature of specialization. Results: Using green fluorescent protein-expressing X. innexi we confirm previous reports using culture-dependent techniques that X. innexi colonizes its nematode host at low levels (~3–8 cells per nematode), relative to other Xenorhabdus-Steinernema associations. We found that compared to the well-characterized entomopathogenic nematode symbiont X. nematophila, X. innexi fails to suppress the insect phenoloxidase immune pathway and is attenuated for virulence and reproduction in the Lepidoptera Galleria mellonella and Manduca sexta, as well as the dipteran Drosophila melanogaster. To assess if, compared to other Xenorhabdus spp., X. innexi has a reduced capacity to synthesize virulence determinants, we obtained and analyzed a draft genome sequence. We found no evidence for several hallmarks of Xenorhabdus spp. toxicity, including Tc and Mcf toxins. Similar to other Xenorhabdus genomes, we found numerous loci predicted to encode non-ribosomal peptide/polyketide synthetases. Anti-SMASH predictions of these loci revealed one, related to the fcl locus that encodes fabclavines and zmn locus that encodes zeamines, as a likely candidate to encode the X. innexi mosquitocidal toxin biosynthetic machinery, which we designated Xlt. In support of this hypothesis, two mutants each with an insertion in an Xlt biosynthesis gene cluster lacked the mosquitocidal compound based on HPLC/MS analysis and neither produced toxin to the levels of the wild type parent. Conclusions: The X. innexi genome will be a valuable resource in identifying loci encoding new metabolites of interest, but also in future comparative studies of nematode-bacterial symbiosis and niche partitioning among bacterial pathogens.

Identification of a Large Family of Slam-Dependent Surface Lipoproteins in Gram-Negative Bacteria

Article

Full-text available

May 2017

The surfaces of many Gram-negative bacteria are decorated with soluble proteins anchored to the outer membrane via an acylated N-terminus; these proteins are referred to as surface lipoproteins or SLPs. In Neisseria meningitidis, SLPs such as transferrin-binding protein B (TbpB) and factor-H binding protein (fHbp) are essential for host colonization and infection because of their essential roles in iron acquisition and immune evasion, respectively. Recently, we identified a family of outer membrane proteins called Slam (Surface lipoprotein assembly modulator) that are essential for surface display of neisserial SLPs. In the present study, we performed a bioinformatics analysis to identify 832 Slam related sequences in 638 Gram-negative bacterial species. The list included several known human pathogens, many of which were not previously reported to possess SLPs. Hypothesizing that genes encoding SLP substrates of Slams may be present in the same gene cluster as the Slam genes, we manually curated neighboring genes for 353 putative Slam homologs. From our analysis, we found that 185 (~52%) of the 353 putative Slam homologs are located adjacent to genes that encode a protein with an N-terminal lipobox motif. This list included genes encoding previously reported SLPs in Haemophilus influenzae and Moraxella catarrhalis, for which we were able to show that the neighboring Slams are necessary and sufficient to display these lipoproteins on the surface of Escherichia coli. To further verify the authenticity of the list of predicted SLPs, we tested the surface display of one such Slam-adjacent protein from Pasteurella multocida, a zoonotic pathogen. A robust Slam-dependent display of the P. multocida protein was observed in the E. coli translocation assay indicating that the protein is a Slam-dependent SLP. Based on multiple sequence alignments and domain annotations, we found that an eight-stranded beta-barrel domain is common to all the predicted Slam-dependent SLPs. These findings suggest that SLPs with a TbpB-like fold are found widely in Proteobacteria where they exist with their interaction partner Slam. In the future, SLPs found in pathogenic bacteria can be investigated for their role in virulence and may also serve as candidates for vaccine development.

Nematodes as Models for Symbiosis

Chapter

May 2022

Translocation of lipoproteins to the surface of gram negative bacteria

Article

Mar 2018

The surface of many Gram-negative bacteria contains lipidated protein molecules referred to as surface lipoproteins or SLPs. SLPs play critical roles in host immune evasion, nutrient acquisition and regulation of bacterial stress response, and have been extensively studied as vaccine antigens. The aim of this review is to summarize the recent studies that have investigated the biosynthetic and translocation pathways used by different bacterial species to deliver SLPs to the surface. We will specifically focus on Slam, a novel outer membrane protein first discovered in pathogenic Neisseria sp., that is involved in translocation of SLPs across the outer membrane.

Gapped BLAST and PSI-BLAST: A new generation of protein database search programs

Article

Full-text available

Sep 1997

The tetratricopeptide repeat: a structural motif mediating protein‐protein interactions

Article

Full-text available

Nov 1999

Petromagnetic Properties In The Naica Mining District, Chihuahua, Mexico: Searching For Source of Mineralization

Article

Full-text available

Jan 2003
EARTH PLANETS SPACE

Ore mineral and host lithologies have been sampled with 89 oriented samples from 14 sites in the Naica District, northern Mexico. Magnetic parameters permit to charac- terise samples: saturation magnetization, density, low- high-temperature magnetic sus- ceptibility, remanence intensity, Koenigsberger ratio, Curie temperature and hystere- sis parameters. Rock magnetic properties are controlled by variations in titanomag- netite content and hydrothermal alteration. Post-mineralization hydrothermal alter- ation seems the major event that affected the minerals and magnetic properties. Curie temperatures are characteristic of titanomagnetites or titanomaghemites. Hysteresis parameters indicate that most samples have pseudo-single domain (PSD) magnetic grains. Alternating filed (AF) demagnetization and isothermal remanence (IRM) ac- quisition both indicate that natural and laboratory remanences are carried by MD-PSD spinels in the host rocks. The trend of NRM intensity vs susceptibility suggests that the carrier of remanent and induced magnetization is the same in all cases (spinels). The Koenigsberger ratio range from 0.05 to 34.04, indicating the presence of MD and PSD magnetic grains. Constraints on the geometry of the intrusive source body devel- oped in the model of the magnetic anomaly are obtained by quantifying the relative contributions of induced and remanent magnetization components.

Gapped BLAST and PSI-BLAST: A new generation of protein database search programs

Article

Full-text available

Sep 1997

Entomopathogenic nematodes. Annu Rev Entomol

Article

Full-text available

Jan 1993
ANNU REV ENTOMOL

The monogeneric nematode families Steinernematidae (Steinernema) and Heterorhabditidae (Heterorhabditis), mutually associated with the pathogenic bacteria Xenorhabdus, are similar in their actions. The free-living, non-feeding infective juveniles possess attributes of both insect parasitoids or predators (they have chemoreceptors and are motile) and microbial pathogens (virulence, high reproductive capacity, numerical but no functional response to host population changes). Their potential as agents of biological control are accentuated by their dispersability using spray equipment, their compatibility with many pesticides, and their amenability to genetic selection. Major sections of this review are on: taxonomy; biology of the nematode-bacterium complex; host range (insects are killed so rapidly that highly adapted host-parasite relationships do not form); behaviour; ecology (including dispersal and host finding, survival, interspecific competition, and recycling and epizootiology); genetics; commercialisation; and efficacy. -P.J.Jarvis

The Presence of Achromobacter Nematophilus in the Infective Stage of a Neoaplectana Sp. (Steinernematidae: Nematoda)

Article

Full-text available

Jan 1966

George Poinar

Most infective juveniles of DD-136 (Neoaplectana, Steinernematidae) were found to contain cells of Achromobacter nematophilus Poinar & Thomas in the ventricular portion of their intestinal lumen. In two instances, anterior intestinal cells of the infective juveniles were found to contain bacterial cells, presumably those of A. nematophilus. When the infective stage penetrated into the body cavity of a suitable host, the bacteria were released through the anus and multiplied rapidly in the host's body, resulting in a fatal septicemia.

Entomopathogenic nematodes for biological control of the leafminer,Liriomyza trifolii(Dipt.: Agromyzidae)

Article

Full-text available

Jan 1993

In the laboratory, mortality rates of the agromyzid leafminer larvae, Liriomyza trifolii (Burgess), ranged from 48 to 98% by 20 strains and/or species of steinernematid and heterorhabditid nematodes. In the greenhouse, abamectin provided superior control of larval leafminers, killing 100% of them as compared with Steinernema carpocapsae (Weiser) All strain (24 to 43% leafminer mortality) or S. carpocapsae Liriomyza-selected strain (8 to 44% leafminer mortality); the maximum relative humidity (r.h.) ranged between 81 and 91% and the minimum r.h. between 50 and 70%. In the foghouse under high r.h., the commercially available All strain and the Hawaiian isolate of S. feltiae (Filipjev) MG-14 strain caused 69 and 67% mean mortality, respectively. There was a significant correlation (P<0.01) between nematode mortality of leafminers and r.h., including the mean, standard deviation, and minimum r.h. during the 48 h after treatment. Average r.h. >92% with a standard deviation of <9% r.h. and a minimum of 72% r.h. provided S. carpocapsae All strain mortality rates of leafminers >65%. The major constraint against the use of nematodes against leafminers in the foliar environment is low r.h. The use of nematodes against L. trifolii can be successful if the r.h. remains high and if nematodes enter leafmines before desiccation, and the nematodes should be integrated with chemical insecticides such as abamectin to manage pesticide resistance in L. trifolii.

The nature of Achromobacter nematophilus as an insect pathogen

Article

Dec 1967

The bacterium Achromobacter nematophilus, which has been isolated from two populations of nematodes belonging to the genus Neoaplectana, was found to be extremely lethal to larvae of Galleria mellonella when injected into the hemocoel. Although producing no obvious effects if fed per os, when one to three bacterial cells were introduced into the body cavity, the insects usually died. In nature, this bacterium occurs in the body of insects attacked by certain neoaplectanid nematodes and, between periods of insect infection, inside the gut of infective stage nematodes. The cells are very fastidious and appear to have a very transient existence in soil or water.

Article

Apr 1993

The levels of DNA relatedness for a broad sample ofXenorhabdus strains isolated from different species of entomopathogenic nematodes (Steinernematidae and Heterorhabditidae) and from different geographical sources were estimated by the hydroxyapatite method. The level of DNA-DNA relatedness for the two phases of each isolate tested was not significantly different from lo%, demonstrating unequivocally that the phase variation demonstrated by all Xenorhabdus spp. is not due to contamination. The isolates of the described Xenorhabdus species coalesced into different DNA relatedness groups, confirming that Xenorhubdus nemuto- philus, Xenorhabdus bovienii, Xenorhubdus poinarii, and Xenorhabdus beddingii, defined on the basis of phenotypic differences, are valid species. The symbiont of Steinernemu intermedia also coalesced with the X. bovienii isolates. This was the only symbiont of seven recently described and unamed Steinernema spp. (including Steinernemu ritteri, Steinemem rara, and Steinernemu anomali) that formed a group with any of the previously described Xenorhabdus species; new species descriptions are required to accommodate the other taxa, but too few isolates were available to allow satisfactory descriptions of them. The DNA relatedness data also showed that the bacteria currently classified as Xenorhabdus luminescens are significantly different from all other Xenorhubdus strains. These data strongly support indications from previous studies of phenotypic characteristics, cellular fatty acids, and DNA relatedness that X. luminescens should be classified as a separate genus, A new genus, Photorhabdus, with an amended description of the type species, Photorhabdus luminescens, is proposed.

Survival of Xenorhabdus nematophilus and Photorhabdus luminescens in water and soil

Article

Dec 1997
J APPL MICROBIOL

Strains of Xenorhabdus nematophilus and Photorhabdus luminescenswere genetically marked with kanamycin resistance and the xylE gene to aid theirdetection in water and soil. Following release in river water, cells declined to undetectable levelsin 6 d. In sterile river water, this decline was enhanced with cells detectable for only 2 d. In sterileMilli-Q purified water, the decline was slower than in either sterile or non-sterile river water.Survival in soil was also restricted with cells only detectable for 7 d. These experiments indicatedthat both X. nematophilus and P. luminescens have limited survival orcompetitive abilities in these environments. The faster decline of populations in sterile river waterwas unexpected, and the possible formation of specialized survival stages was investigated. Insterile water, a non-culturable but viable population of cells was detected, indicating that cellsmay survive longer than anticipated in the environment and remain undetectable using standardmicrobiological methods. The implications of this work to the use of these strains in biologicalcontrol and the release of genetically-modified micro-organisms is discussed.

Mutational Analyses Reveal Overall Topology and Functional Regions of NilB, a Bacterial Outer Membrane Protein Required for Host Association in a Model of Animal-Microbe Mutualism

Abstract and Figures

Recommended publications

Tracking the unfolding pathway of a multirepeat protein via tryptophan scanning - Evidence of locali...

The hsp90-based Chaperone System: Involvement in Signal Transduction from a Variety of Hormone and G...

A tetratricopeptide repeat mutation in yeast transcription factor IIIC131 (TFIIIC131) facilitates re...

The p53 cofactor Strap exhibits an unexpected TPR motif and oligonucleotide-binding (OB)-fold struct...