Journal of Proteomics 195 (2019) 66–75

Contents lists available at ScienceDirect

Journal of Proteomics
journal homepage: www.elsevier.com/locate/jprot

Lable-free based comparative proteomic analysis of secretory proteins of T

rough Brucella mutants
Peng Lia, Mingxing Tiana, Hai Hua, Yi Yina, Xiang Guana, Chan Dinga, Shaohui Wanga,
Shengqing Yua,b,
⁎

a
Shanghai Veterinary Research Institute, Chinese Academy of Agricultural Sciences (CAAS), Shanghai 200241, PR China
b
Jiangsu Co-innovation Center for Prevention and Control of Important Animal Infectious Diseases and Zoonosis, Yangzhou, PR China

ARTICLE INFO ABSTRACT

Keywords: Brucella rough mutants are reported to induce infected macrophage death, which is type IV secretion system
Brucella (T4SS) dependent. T4SS and its secretory proteins play a major role in host-bacteria interactions, but the crucial
Type IV secretion system secretory proteins to promote macrophage death during Brucella rough mutant infection have not been char-
Proteomics acterized. In this study, we found that T4SS components played no role for macrophage death induced by
Secretory proteins
Brucella rough mutant infection, but some T4SS effectors did. Proteomics of secretory proteins from Brucella
OmpW family protein (BAB1_1579)
rough mutants ΔrfbE and ΔrfbEΔvirB123 was analyzed by liquid chromatography/tandem mass spectrometry and
Protein BAB1_1185
861 unique proteins were identified, among which 37 were differential secretory proteins. Gene ontology and
pathway analysis showed that differential secretory proteins involved in cellular process and metabolic process,
distributed in the cell and membrane, possessed molecular function of catalytic activity and binding, and were
associated with ribosome, NOD-like receptor signaling pathway, two-component system and bacterial secretion
system. Cell death analysis showed that T4SS effector VceC, and two differential secretory proteins OmpW
family protein (BAB1_1579) and protein BAB1_1185 were associated with Brucella cytotoxicity. This study
provides new insights into the molecular mechanisms associated with Brucella cytotoxicity and valuable in-
formation for screening vaccine candidates for Brucella.
Significance: Brucella rough mutants induce infected macrophage death, which is T4SS dependent. In the present
report, a comparative proteomics analysis revealed 37 differential secretory proteins between Brucella rough
mutants ΔrfbE and ΔrfbEΔvirB123. Further study demonstrated OmpW family protein (BAB1_1579) and un-
characterized protein BAB1_1185, two differential secretory proteins, were associated with Brucella cytotoxicity.
This study provides novel information of the secretory proteins from the Brucella rough mutants and their effects
on the Brucella cytotoxicity.

1. Introduction
Brucella, a Gram-negative facultative intracellular pathogen capable
of infecting various cell types, including epithelial cells, placental tro-
phoblasts, dendritic cells and macrophages, cause a worldwide zoonotic
disease called brucellosis in animals and humans [1,2]. A prominent
characteristic of the smooth wild-type Brucella is its ability to maintain
its intracellular growth via using a stealthy strategy to avoid activation
of host cells to produce cellular bactericidal substances [3], and its
virulence also relies on its properties for survival and replication in host
cells [4]. Brucella rough mutants that lack the O-antigen of lipopoly-
saccharide (LPS), however, induce infected macrophage death via ac-
tivating IRE1α pathway of endoplasmic reticulum (ER) stress by
enhanced type IV secretion system (T4SS) secretion, and are taken up in
greater numbers by macrophages than the smooth wild-type strains
[5–7]. However, the crucial components associated with macrophage
death induced by Brucella rough mutant remain to be identified.
Multiple reports have identified Brucella effectors whose transloca-
tion into the host cytosol is directly or indirectly dependent on T4SS, a
multiprotein complex [8–10]. T4SS protein substrates which modulate
varied cellular processes including apoptosis, vesicular traffic and ubi-
quitination were directly translocated into the cytosol of the target host
cell to aid bacterial colonization and survival inside host tissues
[11–13]. Recent studies have identified 15 effectors that are secreted by
Brucella in a T4SS-dependent manner. The initially identified two ef-
fectors VceA and VceC are translocated into macrophages by the

⁎
Corresponding author at: Shanghai Veterinary Research Institute, Chinese Academy of Agricultural Sciences (CAAS), Shanghai 200241, PR China.
E-mail address: yus@shvri.ac.cn (S. Yu).

https://doi.org/10.1016/j.jprot.2019.01.008
Received 9 October 2018; Received in revised form 7 January 2019; Accepted 13 January 2019
Available online 17 January 2019
1874-3919/ © 2019 Elsevier B.V. All rights reserved.
P. Li et al.
Brucella Type IV secretion system and VceC interacted with ER cha-
perone BiP/Grp78 [8,14]. The effector protein RicA interacts with
Rab2, a small GTPase described as recruited to the BCVs, which is es-
sential for Golgi-to-ER vesicle trafficking [9]. A T4SS effector BPE123
targets host cell α-enolase contributing to Brucella abortus intracellular
lifestyle [15]. The effector protein BPE005 induces collagen deposition
and matrix metalloproteinase 9 down-modulation via transforming
growth factor β1 in hepatic stellate cells [16]. Brucella induce an un-
folded protein response via the effector protein TcpB/BtpA that sup-
ports intracellular replication in macrophages [17]. Recent studies
showed that the Brucella effector protein TcpB/BtpA induces degrada-
tion of inflammatory caspases and thereby subverts noncanonical in-
flammasome activation in macrophages to attenuate pyroptosis and
inflammation [18]. BtpB is a Brucella effector that plays a major role in
the modulation of host innate immune response during infection [19].
Brucella modulate secretory trafficking via multiple T4SS effector pro-
teins including BspA, BspB and BspF that likely act coordinately to
promote Brucella pathogenesis [20]. Recently, Miller et al. showed that
B. abortus deliver into macrophages via a type IV-secretion effector
BspB that remodels Golgi-associated membrane traffic to promote
biogenesis of the Brucella replicative vacuole and bacterial proliferation
[21]. Additionally, the effector protein SepA participates in the early
stages of intracellular survival [22]. Taken together, T4SS effectors
were secreted into host cells, which then target various host mechan-
isms and perpetuate the infection. Therefore, identifying the effectors
secreted by Brucella through the T4SS machinery and determining
pathways they target host are critical for understanding their patho-
genesis.
In previous studies, we found that Brucella rough mutant ΔrfbE in-
duced the death of infected macrophages, which is T4SS dependent
[6,23]. In the present study, we found that T4SS components were not
associated with macrophage death caused by infection with Brucella
rough mutant, but some T4SS effectors did. To identify the critical T4SS
effectors, proteomics of secretory proteins from Brucella rough mutants
ΔrfbE and ΔrfbEΔvirB123 was analyzed by liquid chromatography/
tandemmass spectrometry (LC–MS/MS). T4SS effector VceC and two
differential proteins identified in this study, OmpW family protein
(BAB1_1579) and uncharacterized protein BAB1_1185, were associated
with Brucella rough mutant-induced macrophage cytotoxicity.
2. Materials and methods
2.1. Ethics statement
This study was performed in strict accordance with the re-
commendations in the Guide for the Care and Use of Laboratory
Animals of the Institutional Animal Care and Use Committee guidelines
set by Shanghai Veterinary Research Institute, Chinese Academy of
Agricultural Sciences (CAAS). We made all efforts to minimize animal
suffering, and the protocol for animal experiments was approved by the
Committee on the Ethics of Animal Experiments of Shanghai Veterinary
Research Institute, CAAS (shvri-MO-0135).
2.2. Strains, plasmids, macrophages and culture conditions
All strains and plasmids used in this study are listed in Table 1.
Brucella abortus S2308 and its derivatives were grown in tryptic soy
broth (TSB) or on tryptic soy agar (TSA) (Difco, Franklin Lakes, NJ,
USA) plates at 37 °C with 5% CO2. Manipulation of Brucella was per-
formed in a biosafety level 3 laboratory facility at the CAAS. Escherichia
coli strains were cultured at 37 °C in Luria Broth (LB). When appro-
priate, 100 μg/mL kanamycin or 20 μg/mL chloramphenicol (Sig-
ma–Aldrich Inc., St. Louis, MO, USA) was added. Mouse macrophage
RAW264.7 (ATCC, Manassas, VA, USA) was cultured at 37 °C with 5%
CO2, in Dulbecco's modified Eagle's medium (DMEM, Gibco, Grand Is-
land, NY, USA) supplemented with 10% heat-inactivated fetal bovine
67
Journal of Proteomics 195 (2019) 66–75
serum (Gibco).
2.3. Plasmid construction
All primers used in this study are listed in Table S1. Suicide plasmids
were constructed, using an overlap PCR assay, as previously reported
[23]. Briefly, the upstream and downstream fragments of target genes
were amplified by independent PCRs and extracted from agarose gels
that were used as templates for a second round of PCR. The resultant
product that contained joined flanking sequences was purified by gel
extraction and cloned into a pKB plasmid, after being digested with
XbaI, to generate the suicide plasmids.
Overexpression plasmids of pVceC, pVirB10, pOmpW, pBAB1_0970,
pBAB1_1185 and pBAB1_1576 were constructed using conventional
methods. The virB10, ompW, BAB1_0970, BAB1_1185 and BAB1_1576
genes containing the promoter and terminator regions were amplified
by independent PCRs using the primers, respectively, and then cloned
into the plasmid pBCSP31, to generate the plasmids pVceC, pVirB10,
pOmpW, pBAB1_0970, pBAB1_1185 and pBAB1_1576, respectively. All
recombinant plasmids were propagated in E. coli DH5α cells
(InvitrogenCorp., Carlsbad, CA, USA) and then extracted to construct
recombinant Brucella strains as previously described [23].
2.4. Mutant construction
The mutants were constructed by allelic replacement, using a two-
step strategy as previously reported [6,23,24]. The suicide plasmids
(0.5 to 1.0 μg) were transferred to the ΔrfbE strain by electroporation.
The first exchanged recombinants were selected by plating on TSA
containing kanamycin. The second round of exchanged recombinants
was selected by plating on TSA containing 5% sucrose. Analyses of
PCRs were carried out to identify clones. Overexpression plasmids of
pVceC, pVirB10, pOmpW, pBAB1_0970, pBAB1_1185 and pBAB1_1576
(0.5 to 1.0 μg) were transferred to the S2308 strain by electroporation.
The recombinants were then selected by plating on TSA containing
chloramphenicol. The PCR or qPCR analyses were carried out to iden-
tify recombinants.
2.5. Preparation of secretory proteins from Brucella rough mutants ΔrfbE
and ΔrfbEΔvirB123
Brucella rough mutants ΔrfbE and ΔrfbEΔvirB123 were grown for
24 h at 37 °C in 1000 mL TSB to the exponential phase
(OD600 = 1.0–1.2), and then centrifuged at 8000 ×g for 5 min to pre-
cipitate bacteria. The pellets of rough Brucella strains were washed
twice with minimal medium RPMI 1640 and then resuspended in
500 mL same medium for 8 h to induce of the T4SS expression at 37 °C
with shaking at 200 rpm/min [6]. The induced bacterial culture was
centrifuged at 8000 ×g for 5 min at 4 °C, the resulting supernatant
containing the secretory proteins was collected, and filtered through 1 L
Filter System (0.22 μm PES, Corning, USA) to remove the residual
bacteria, and then concentrated using Centrifugal Filters Ultracel-3 K
(Merck Millipore, Ireland) according to the protocol. The resulting
sample was resuspended in 500 μL fresh-cold phosphate buffered saline
(PBS). Protein concentration was determined with Enhanced BCA
Protein Assay Kit (Beyotime, Shanghai, China). Samples of secretory
proteins were stored at −80 °C until used.
2.6. SDS-PAGE analysis
The samples of secretory proteins were separated by sodium do-
decyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) to de-
termine whether samples were suitable for subsequent analysis. Protein
samples were mixed with 5× protein loading buffer (0.1 M Tris buffer,
pH 6.8, 4% SDS, 0.2% β-mercaptoethanol, 40% glycerol, and 0.002%
bromphenol blue) and boiled for 10 min. Samples were subjected to
P. Li et al. Journal of Proteomics 195 (2019) 66–75

Table 1
Strains and plasmids used in the present study.
Strains and plasmids Characteristics Source or reference

Strains
B. abortus strains S2308 wild-type strain; smooth phenotype ATCC
ΔrfbE rfbE deletion mutant strain; rough phenotype; rfbE encodes ATP-binding protein which flip the O-antigen from the cytoplasmic (Zhang et al., 2013)
face to the periplasmic face of the inner membrane; full deletion.
ΔrfbEΔvirB123 rfbE and virB123 deletion mutant strain; rough phenotype; virB123 encodes VirB1, VirB2, VirB3 which are components of the (Li et al., 2017)
T4SS. VirB1 is a lytic transglycosylase and degrades peptidoglycan to provide space for the T4SS to be assembled; VirB2 and
VirB3 mediates assembly of the pilus; full deletion.
ΔrfbEΔvirB4(domain) rfbE and virB4 (domain) deletion mutant strain; rough phenotype; virB4 (domain), a conserved NTP-binding domain in VirB4 This study
suggests that one or both proteins couple energy by NTP hydrolysis to transport of effectors; full deletion.
ΔrfbEΔvirB11(domain) rfbE and virB11 (domain) deletion mutant strain; rough phenotype; virB11 (domain), a conserved NTP-binding domain in VirB11 This study
suggests that one or both proteins couple energy by NTP hydrolysis to transport of effectors; full deletion.
ΔrfbEΔvirB4 rfbE and virB4 deletion mutant strain; rough phenotype; the ATPases VirB4 is a component of the T4SS and essential for T4SS This study
function by providing energy for T4SS assembly and transport of the effector proteins; full deletion.
ΔrfbEΔvirB7 rfbE and virB7 deletion mutant strain; rough phenotype; VirB7 is a component of the T4SS and has a stabilizing role in the This study
formation of the core complex; full deletion.
ΔrfbEΔvirB8910 rfbE and virB8910 deletion mutant strain; rough phenotype; virB8910 encodes VirB8, VirB9, VirB10 which are components of the This study
T4SS and are the core structure of the T4SS; full deletion.
ΔrfbEΔvirB11 rfbE and virB11 deletion mutant strain; rough phenotype; the ATPases VirB11 is a component of the T4SS and essential for T4SS This study
function by providing energy for T4SS assembly and transport of the effector proteins; full deletion.
ΔrfbEΔompW rfbE and ompW deletion mutant strain; rough phenotype; OmpW is an outer membrane protein and identified by lable free This study
quantitative secretory proteomic analysis of ΔrfbE and ΔrfbEΔvirB123; full deletion.
ΔrfbEΔvirB10 rfbE and virB10 deletion mutant strain; rough phenotype; VirB4 is a component of the T4SS and is the core structure of the T4SS; This study
full deletion.
ΔrfbEΔBAB1_0970 rfbE and BAB1_0970 deletion mutant strain; rough phenotype; BAB1_0970 encodes solute-binding protein/glutamate receptor: This study
Bacterial extracellular solute-binding protein and is identified by lable free quantitative secretory proteomic analysis of ΔrfbE
and ΔrfbEΔvirB123; full deletion.
ΔrfbEΔBAB1_1185 rfbE and BAB1_1185 deletion mutant strain; rough phenotype BAB1_1185 encodes uncharacterized protein and is identified by This study
lable free quantitative secretory proteomic analysis of ΔrfbE and ΔrfbEΔvirB123; full deletion.
ΔrfbEΔBAB1_1576 rfbE and BAB1_1576 deletion mutant strain; rough phenotype; BAB1_1576 encodes uncharacterized protein and is identified by This study
lable free quantitative secretory proteomic analysis of ΔrfbE and ΔrfbEΔvirB123; full deletion.
ΔrfbEΔVceA rfbE and VceA deletion mutant strain; rough phenotype;VceA is the T4SS effector. This study
ΔrfbEΔVceC rfbE and VceC deletion mutant strain; rough phenotype; VceC is the T4SS effector and interacted with ER chaperone BiP/Grp78. This study
ΔrfbEΔBPE123 rfbE and BPE123 deletion mutant strain; rough phenotype; BPE123 is the T4SS effector and targets host cell alpha-enolase (ENO- This study
1) contributing to Brucella intracellular lifestyle; full deletion.
ΔrfbEΔBPE005 rfbE and BPE005 deletion mutant strain; rough phenotype; BPE005 is the T4SS effector and induces collagen deposition and This study
matrix metalloproteinase 9 down-modulation via transforming growth factor β1 in hepatic stellate cells; full deletion.
ΔrfbEΔBPE275 rfbE and BPE275 deletion mutant strain; rough phenotype; BPE275 is the T4SS effector; full deletion. This study
ΔrfbEΔBspA rfbE and BspA deletion mutant strain; rough phenotype; BspA is the T4SS effector and promotes Brucella pathogenesis; full This study
deletion.
ΔrfbEΔBspB rfbE and BspB deletion mutant strain; rough phenotype; BspB is the T4SS effector and remodels Golgi-associated membrane This study
traffic to promote biogenesis of the Brucella replicative vacuole and bacterial proliferation; full deletion.
ΔrfbEΔBspE rfbE and BspE deletion mutant strain; rough phenotype; BspE is the T4SS effector and promotes Brucella pathogenesis; full This study
deletion.
ΔrfbEΔBspF rfbE and BspF deletion mutant strain; rough phenotype;BspF is the T4SS effector and promotes Brucella pathogenesis; full This study
deletion; full deletion.
ΔrfbEΔRicA rfbE and RicA deletion mutant strain; rough phenotype; RicA is the T4SS effector and interacts with Rab2 which is essential for This study
Golgi-to-ER vesicle trafficking; full deletion.
ΔrfbEΔSepA rfbE and SepA deletion mutant strain; rough phenotype; SepA is the T4SS effector and participates in the early stages of This study
intracellular survival; full deletion.
ΔrfbEΔBtpA rfbE and BtpA deletion mutant strain; rough phenotype; BtpA is the T4SS effector and supports intracellular replication in This study
macrophages, induces an unfolded protein response and induces degradation of inflammatory caspases to attenuate pyroptosis
and inflammation; full deletion.
ΔrfbEΔBtpB rfbE and BtpB deletion mutant strain; rough phenotype; BtpB is the T4SS effector with immune modulatory functions; full This study
deletion.
S2308(pVceC) S2308 based VceC overexpression strain; smooth phenotype This study
S2308(pOmoW) S2308 based ompW overexpression strain; smooth phenotype This study
S2308(pVirB10) S2308 based virB10 overexpression strain; smooth phenotype This study
S2308(pBAB1_0979) S2308 based BAB1_0970 overexpression strain; smooth phenotype This study
S2308(pBAB1_1185) S2308 based BAB1_1185 overexpression strain; smooth phenotype This study
S2308(pBAB1_1576) S2308 based BAB1_1576 overexpression strain; smooth phenotype This study
E. coli strain DH 5α F− φ80lacZ∆M15∆(lacZYA-argF)U169 recA1 endA1 hsdR17(rk−,mk+) phoA supE44 thi-1 gyrA96 relA1 λ− Invitrogen

Plasmids
pKB Kanr; pUC19 plasmid containing sacB gene
pKBΔvirB4(domain) Kanr; pKB plasmid containing the virB4(domain) gene; used to construct deletion strain This study
pKBΔvirB11(domain) Kanr; pKB plasmid containing the virB11(domain) gene; used to construct deletion strain This study
pKBΔvirB4 Kanr; pKB plasmid containing the virB4 gene; used to construct deletion strain This study
pKBΔvirB7 Kanr; pKB plasmid containing the virB7gene; used to construct deletion strain This study
pKBΔvirB9810 Kanr; pKB plasmid containing the virB8910 gene; used to construct deletion strain This study
pKBΔvirB11 Kanr; pKB plasmid containing the virB11 gene; used to construct deletion strain This study
pKBΔompW Kanr; pKB plasmid containing the ompW gene; used to construct deletion strain This study
pKBΔvirB10 Kanr; pKB plasmid containing the virB10 gene; used to construct deletion strain This study
pKBΔBAB1_0970 Kanr; pKB plasmid containing the BAB1_0970 gene; used to construct deletion strain This study
pKBΔBAB1_1185 Kanr; pKB plasmid containing the BAB1_1185gene; used to construct deletion strain This study
pKBΔBAB1_1576 Kanr; pKB plasmid containing the BAB1_1576 gene; used to construct deletion strain This study
(continued on next page)

68
P. Li et al. Journal of Proteomics 195 (2019) 66–75

Table 1 (continued)

Strains and plasmids Characteristics Source or reference

r
pKBΔVceA Kan ; pKB plasmid containing the VceA gene; used to construct deletion strain This study
pKBΔVceC Kanr; pKB plasmid containing the VceC gene; used to construct deletion strain This study
pKBΔBPE123 Kanr; pKB plasmid containing the BPE123 gene; used to construct deletion strain This study
pKBΔBPE005 Kanr; pKB plasmid containing the BPE005 gene; used to construct deletion strain This study
pKBΔBPE275 Kanr; pKB plasmid containing the BPE275 gene; used to construct deletion strain This study
pKBΔBspA Kanr; pKB plasmid containing the BspA gene; used to construct deletion strain This study
pKBΔBspB Kanr; pKB plasmid containing the BspB gene; used to construct deletion strain This study
pKBΔBspE Kanr; pKB plasmid containing the BspEgene; used to construct deletion strain This study
pKBΔBspF Kanr; pKB plasmid containing the BspFgene; used to construct deletion strain This study
pKBΔRicA Kanr; pKB plasmid containing the RicAgene; used to construct deletion strain This study
pKBΔSepA Kanr; pKB plasmid containing theSepAgene; used to construct deletion strain This study
pKBΔBtpA Kanr; pKB plasmid containing the BtpAgene; used to construct deletion strain This study
pKBΔBtpB Kanr; pKB plasmid containing the BtpBgene; used to construct deletion strain This study
pBCSP31 Cmr; The promoter of bcsp31 gene was inserted into the pBBR1-MCS plasmid (Li et al., 2017)
pVceC Cmr; pBCSP31 plasmid containing the vceC gene; used to construct overexpression strain This study
pOmpW Cmr;pBCSP31 plasmid containing the ompW gene; used to construct overexpression strain This study
pVirB10 Cmr;pBCSP31 plasmid containing the virB10 gene; used to construct overexpression strain This study
pBAB1_0970 Cmr; pBCSP31 plasmid containing the BAB1_0970 gene; used to construct overexpression strain This study
pBAB1_1185 Cmr; pBCSP31 plasmid containing the BAB1_1185 gene; used to construct overexpression strain This study
pBAB1_1576 Cmr; pBCSP31 plasmid containing the BAB1_1576 gene; used to construct overexpression strain This study

SDS-PAGE using 12.5% resolving gels. Gels were stained with coo-
massie brilliant blue (CBB) G250 (Invitrogen, USA). An Odyssey two-
color infrared imaging system (LI-COR Biosciences) was used to vi-
sualise the proteins.
2.7. Protein FASP digestion
The prepared secretory protein sample from section 2.5 (about
300 μg) was solubilized in DTT whose final concentration was up to
100 mM, incubated in boiling water for 5 min, cooled to room tem-
perature, diluted with 200 μL Urea buffer (8 M Urea, 150 mM Tris–HCl
pH 8.0), transferred to Amicon Ultra-15 Centrifugal Filters Ultracel-3 K
(Merck Millipore, Ireland) and then centrifuged at 14,000 ×g for
15 min to discard the filtering medium. The 200 μL Urea buffer was
added and then centrifuged for 15 min under the same conditions to
discard the filtering medium, followed by addition of 100 μL iodoace-
tamide (50 mM iodoacetamide in UA) and incubation for 30 min in the
dark. After 10 min of centrifugation using the above conditions, the
filters were washed three times with 100 μL Urea buffer and centrifuged
for 15 min. The concentrates were diluted three times with 100 μL of
25 mM NH4HCO3 and centrifugation for 15 min at room temperature.
The resulting concentrate was dissolved in 40 μL trypsin buffer [6 μg
trypsin (Promega, Fitchburg, WI, USA) in 40 μL NH4HCO3 buffer], and
then incubated at 37 °C for 16–18 h. The filtrate containing the resulting
peptides was collected by ultrafiltration with Amicon Ultra-15
Centrifugal Filters Ultracel-3 K, (Merck Millipore). Desalting treatment
was processed by C18-SD extraction disk cartridge [Empore SPE
Cartridges C18 (standard density), bed I.D. 7 mm, volume 3 mL
(Sigma)] and the OD280 of the resulting peptides was measured.
2.8. LC–MS/MS
Experiments were performed on a Q Exactive mass spectrometer
that was coupled to HPLC liquid phase system Easy nLC1000 (Thermo
Fisher Scientific，USA). The 2 μg of the peptide mixture was loaded
onto a C18-reversed phase column (RP-C18, 10 cm long, 150 μm inner
diameter, Dionex, USA) packed in-house with RP-C18 5 μm resin in
buffer A (0.1% formic acid in 2% acetonitrile) and separated with a
linear gradient of buffer B (0.1% formic acid in 84% acetonitrile) at a
flow rate of 400 nl/min controlled by intelliflow technology over
120 min. MS data were acquired using a data-dependent top 10 method
dynamically choosing the most abundant precursor ions from the
survey scan (300–1800 m/z) for higher energy collisional dissociation
(HCD) fragmentation. Survey scans were acquired at a resolution of
70,000 at m/z 200, and resolution for HCD spectra was set to 17,500 at
m/z 200. Normalized collision energy was 30 eV and the underfill ratio,
which specifies the minimum percentage of the target value (30 eV)
likely to be reached at maximum fill time (100 ms), was defined as
0.1%. The instrument was run with peptide recognition mode enabled.
2.9. Data analysis and bioinformatics analysis
Protein identifications were performed with Maxquant Software
(version 1.3.0.5). All LC–MS/MS data were searched against the
Uniprot Brucella abortus protein database (97,215 sequences, 01/09/
2017) using the Maxquant software (version 1.3.0.5). For protein
identification, the mass error tolerance for MS scans was first searched
with an error window of 20 ppm and then with a main search error of
6 ppm. Mass tolerance for MS/MS scans was set at 20 ppm for parent ion
and 0.1 Da for fragment ions. Taxonomy was restricted to B. abortus and
tryptic peptides were considered with maximum two missed cleavages.
Carbamidomethylation of cysteine was used as a fixed modification,
and oxidation of methionine was set as a variable protein modification.
Label free quantification (LFQ) was set as TRUE, and LFQ min ratio
count was set as 1. Peptide identifications with false discovery
rates > 5% (q-value > 0.05) were discarded. Perseus software (version
1.3.0.4) was used for identification of deferentially expressed proteins
after Maxquant software based protein identification. Differential se-
cretory proteins were analyzed for significant down- or up-regulation
according to label-free quantification, and the fold (ΔrfbEΔvirB/ΔrfbE)
difference had to be > 1.5 or < 0.65. Gene ontology (GO) and Kyoto
Encyclopedia of Genes and Genomes (KEGG) analysis were performed
to identify possible enrichment of the differential secretory proteins
with particular biological characteristics.
2.10. Cell infection assay
Monolayers of RAW264.7 cells were cultured in 24-well plates and
infected with B. abortus S2308 or its derivatives at a multiplicity of in-
fection (MOI) of 100 colony forming units (CFUs) per cell. To synchronize
the infection, the infected plates were centrifuged at 400 ×g for 5 min, and
cells were then incubated at 37 °C with 5% CO2 for 1 h. The monolayers
were washed twice with PBS (HyClone, GE Lifesciences, Logan, UT, USA)
to remove extracellular nonadherent bacteria, and then incubated with
DMEM containing gentamicin (100 μg/mL) for 1 h to kill extracellular
bacteria. To maintain survival of the infected cells, the monolayers were
incubated with DMEM containing gentamicin (20 μg/mL) and 2% FBS
after being washed thrice with PBS.

69
P. Li et al.
2.11. Cell death analysis
Macrophage death was detected using two approaches. In the first
approach, infected cells were stained with annexin V and propidium
iodide (PI) at 8 h, and 12 h post infection (p.i.), using the annexin V-
FITC/PI staining kit (Beyotime, Shanghai, China). In the second ap-
proach, the release of lactate dehydrogenase (LDH) in the supernatant
of Brucella-infected RAW264.7 cells was determined as optical density
at 490 nm (OD490) at 8 h, and 12 h p.i., using a CytoTox 96 non-
radioactive cytotoxicity assay (Promega). Cell death was expressed as a
percentage of maximum LDH release. The percentage was calculated as
follows: (OD490 of infected cells − OD490 of uninfected cells)/(OD490 of
lysed uninfected cells − OD490 of uninfected cells) × 100%.
2.12. RNA extraction and real-time PCR
Total RNA was extracted from bacteria using the TRIzol RNA
Isolation Reagent (Invitrogen) according to the manufacturer's pro-
tocol. Genomic DNA contamination was removed through treatment
with a Turbo DNA-free kit (Ambion). The RNA quantity and quality
were evaluated with calculation of RNA concentration and OD260/
OD280 ratio (1.8–2.0) using the NanoDrop ND-1000 spectrophotometer
(NanoDrop Technologies, Inc.), and the integrity was assessed by cal-
culation of RNA integrity number (RIN) using standard denaturing
agarose gel electrophoresis. RNA (1 μg) was then reverse transcribed
into cDNA, using a PrimeScript RT-PCR kit (Takara) according to the
manufacturer's instructions. A volume of 20 μL RT-PCR mixture was
made comprising 10 μL 2× GoTaq qPCR master mix (Promega), 1 μL
cDNA, 0.5 μL (each) forward and reverse primers (10 μM each), and
8 μL double-distilled water (ddH2O). The mixture was incubated at
95 °C for 2 min, and then subjected to 40 cycles at 95 °C for 15 s, fol-
lowed by 60 °C for 1 min using a MastercyclerepRealplex system
(Eppendorf). All samples were analyzed in triplicate and relative tran-
scription levels of each gene were determined by the 2-ΔΔCt method,
using 16S RNA as an internal control for data normalization.
2.13. Western blot analysis
Two differential secretory proteins, OmpW family protein
(BAB1_1579) and uncharacterized protein BAB1_1185 were used to
confirm the results by Western blot. Aliquots containing 300 μg of se-
cretory proteins of Brucella rough mutant ΔrfbE or ΔrfbEΔvirB123 were
separated by SDS-PAGE and then transferred onto nitrocellulose
membranes (Millipore) using transfer procedure as previously de-
scribed [23]. The membranes were blocked overnight at 4 °C in Tris-
buffered saline containing 5% skim milk, and incubated with specific
mouse anti-BAB1_1185 (ORF) or rabbit anti-OmpW family protein
(BAB1_1579, ORF) antibodies (prepared in our lab, 1:500 dilution).
These specific antibodies were prepared in strict accordance to the
‘Ethics Statement’ in Methods. After washing three times with Tris-
buffered saline and Tween 20, the membrane was incubated with the
respective goat anti-mouse or anti-rabbit IgG conjugated to horseradish
peroxidase (HRP) at 1:1000 dilution for 1 h at room temperature. The
membrane was visualized with ECL substrate solution and the image
was analyzed by Gel-Pro Analyzer 4.0 software.
2.14. Statistical analysis
Statistical analysis was performed using the GraphPad Prism 6.0
software (GraphPadSoftware Inc., La Jolla, CA, USA). All p-values be-
tween identified samples were generated using unpaired two-tailed
Student's t-tests. In the case of group analysis, two-way ANOVA fol-
lowed by Holm–Sidak's multiple tests was used. All experiments were
repeated at least three times and the results were presented as
means ± SD from ≥3 replicates per condition.
70
Journal of Proteomics 195 (2019) 66–75
3. Results
3.1. SDS-PAGE analysis
The samples of secretory proteins were separated by SDS-PAGE
analysis and clearly displayed protein bands were shown (Fig. S1), in-
dicating that secretory proteins were suitable for subsequent analysis
and then subjected to trypsin digestion and LC–MS/MS analysis.
3.2. Proteomic analysis of differential secretory proteins of Brucella rough
mutants ΔrfbE and ΔrfbEΔvirB123
A total of 861 secretory proteins were identified (Table S2), among
which 37 proteins were differentially secreted between Brucella rough
mutants ΔrfbE and ΔrfbEΔvirB123. Of them, 15 were down-regulated in
Brucella rough mutant ΔrfbEΔvirB123. T4SS protein VirB10
(BAB2_0059), OmpW family protein (BAB1_1579), bacterial extra-
cellular solute-binding protein (BAB1_0970), uncharacterized proteins
BAB1_1185 and BAB1_1576 were notable among the down-regulated
secretory proteins in Brucella rough mutant ΔrfbEΔvirB123. The differ-
ential secretory proteins of ΔrfbE and ΔrfbEΔvirB123 were listed in
Table 2.
3.3. GO enrichment and KEGG analysis for the differential secretory
proteins of ΔrfbE and ΔrfbEΔvirB123
Differential secretory proteins between ΔrfbE and ΔrfbEΔvirB123
were annotated by BLAST search in the UniProt database with the GO
vocabulary. Second-level GO terms were applied to classify proteins
into biological process, molecular function and cellular components
according to their functional annotation. The results demonstrated that
the differential secretory proteins were mainly involved in cellular
process and metabolic process, distributed in the cell and membrane,
and possessed molecular function of catalytic activity and binding (Fig.
S2). KEGG pathway analysis showed that the differential secretory
proteins were associated with ribosome, NOD-like receptor signaling
pathway, two-component system and bacterial secretion system (Table
S3).
3.4. T4SS effectors, but not components were crucial for macrophage death
caused by ΔrfbE infection
Our previous studies have shown that Brucella rough mutant ΔrfbE
induces the death of macrophages, which is T4SS dependent [6]. To
determine whether macrophage death caused by ΔrfbE infection is as-
sociated with five differential secretion proteins of VirB10
(BAB2_0059), OmpW family protein (BAB1_1579), bacterial extra-
cellular solute-binding protein (BAB1_0970), uncharacterized proteins
BAB1_1185 and BAB1_1576, we constructed double gene mutants of
ΔrfbEΔvirB10, ΔrfbEΔompW, ΔrfbEΔBAB1_0970, ΔrfbEΔBAB1_1185 and
ΔrfbEΔBAB1_1576. RAW264.7 macrophages were infected with these
mutants, and their LDH releases were determined to assess quantita-
tively the death of macrophages at 8 h and 12 h p.i.. As the results, no
difference was shown for the LDH release from ΔrfbEΔompW,
ΔrfbEΔBAB1_0970, ΔrfbEΔBAB1_1185 and ΔrfbEΔBAB1_1576 infected
cells, compared to the ΔrfbE mutant infected cells (Fig. 1A). However,
the LDH release from the ΔrfbEΔvirB10 mutant-infected macrophages
was reduced to the similar levels of the smooth wild-type strain S2308
infected cells at 8 h, and 12 h p.i. (Fig. 1A).
VirB10 is the T4SS component of Brucella. To determine whether
macrophage death caused by ΔrfbE infection is associated with T4SS
components, we constructed double-gene mutated strains of
ΔrfbEΔvirB123, ΔrfbEΔvirB4, ΔrfbEΔvirB7, ΔrfbEΔvirB8910,
ΔrfbEΔvirB11, ΔrfbEΔvirB4 (domain) and ΔrfbEΔvirB11 (domain). The
LDH release was determined to assess quantitatively the death of
macrophages at 8 h and 12 h p.i., found that LDH release from the
 P. Li et al.

Table 2
The differential secretory proteins of ΔrfbE and ΔrfbEΔvirB123.
Protein name Gene accession PepCounta Unique Sequence MWc (KDa) LFQd intensity of ΔrfbEΔvirB (a) LFQ intensity of ΔrfbE (b) a/b
PepCountbc coverage (%)
Mean SD Mean SD Ratio P value

Type IV secretion system protein VirB10 BAB2_0059 8 8 32.2 41.484 0 Ne 84,182,333 12,190,633 0 N
OmpW family BAB1_1579 6 6 55.8 27.593 0 N 13,254,300 2,116,861 0 N
Solute-binding protein/glutamate receptor: Bacterial extracellular BAB1_0970 10 10 47.5 32.703 0 N 7,074,550 633,412 0 N
solute-binding protein
Uncharacterized protein BAB1_1185 9 9 69.7 18.292 0 N 32,249,000 3,029,813 0 N
Uncharacterized protein BAB1_1576 3 3 6.8 75.291 0 N 31,431,000 2,212,856 0 N
Protein ApaG BAB1_0334 2 2 29.9 106,053,000 8,451,453 0 N +∞ N
Histidine ammonia-lyase BAB2_0305 17 3 43.7 53.785 37,827,333 1,394,378 0 N +∞ N
UDP-N-acetylmuramoylalanine-D-glutamate ligase BAB1_1452 4 4 12 48.693 11,360,300 1,406,274 0 N +∞ N
Outer membrane efflux protein BAB1_0963 3 3 5.9 48.51 11,169,100 1,783,662 0 N +∞ N
Uncharacterized protein BAB1_1295 2 2 15.7 18.047 45,516,500 1,416,345 0 N +∞ N
Glutamate synthase amidotransferase BAB2_0053 5 5 36.7 45.897 58,176,500 1,870,029 0 N +∞ N
Glutathione S-transferase BAB1_1513 4 4 16.7 27.236 19,311,500 3,385,674 0 N +∞ N
IS66 family element, orf2, putative BAB2_0683 3 3 10.9 13.913 11,293,100 2,732,070 0 N +∞ N
Acyl-CoA dehydrogenase BAB1_1109 6 6 18 43.231 227,071,000 32,073,030 0 N +∞ N
ChaperoninclpA/B:AAA ATPase BAB1_1573 4 4 16.4 36.647 56,269,333 1,788,249 0 N +∞ N
Bacterial luciferase BAB1_2052 3 3 3.8 37.655 7,495,033 477,974 0 N +∞ N
dCTPdeaminase BAB1_0336 2 2 10.2 41.997 13,648,550 1,515,777 0 N +∞ N
Transporter, putative BAB1_0306 4 4 13.6 40.868 12,500,000 1,634,067 0 N +∞ N

71
Uncharacterized protein BAB2_0010 2 2 13.4 22.664 21,022,000 2,640,971 0 N +∞ N
AMP nucleosidase BAB1_0647 10 10 25.2 55.509 40,566,000 3,954,972 7,565,500 1,607,400 5.36 0.015
Glutaredoxin:Thioredoxin type domain BAB1_1879 3 3 27.2 10.167 161,174,333 33,451,384 36,033,667 1,116,633 4.47 0.021
HpcH/HpaIaldolase BAB1_0864 7 3 38.5 27.908 136,030,000 20,497,618 23,309,000 4,330,296 4.09 0.002
Thioredoxin BAB1_2107 3 3 38.3 11.421 271,853,333 65,330,058 125,686,666 20,342,903 2.16 0.020
Glucokinase BAB2_1010 8 2 40.8 37.606 346,100,000 100,343,077 169,953,333 28,821,563 2.04 0.043
Response regulator receiver protein CpdR BAB2_0042 5 5 58.3 13.586 88,878,666.67 2,095,398 54,470,500 3,078,720 1.63 0.005
Homoserine dehydrogenase BAB1_1293 5 5 11.2 46.81 146,930,000 12,705,144 92,663,667 3,554,330.828 1.58 0.048
50S ribosomal protein BAB1_1265 2 2 18.2 7.036 7,053,533,333 324,677,456 4,468,233,333 190,096,608 1.57 0.018
DNA gyrase, subunit B BAB1_1675 5 5 40.1 17.357 525,040,000 28,977,506 901,820,000 26,114,388 0.58 0.022
Uncharacterized protein BAB1_1917 9 9 37.8 36.790 130,583,333 27,563,111 198,556,667 12,155,608 0.65 0.021
ABC transporter, periplasmic binding protein BAB2_0919 17 17 53.8 36.849 1,387,300,000 136,753,086 2,253,633,333 3,502,287,851 0.61 0.025
Transcriptional regulatory protein BAB1_1227 6 6 35.1 21.92 833,710,000 174,995,675 1,404,666,667 2,172,935,858 0.59 0.007
Uncharacterized protein BAB1_0863 8 4 72 27.884 490,260,000 35,638,489 1,063,866,667 1,650,953,720 0.54 0.048
Uncharacterized protein BAB1_0691 5 5 47.2 14.256 120,189,333 20,934,340 226,720,000 21,489,162.38 0.53 0.022
30S ribosomal protein S3 BAB1_1249 12 12 47 26.603 119,818,000 52,638,549 242,483,333 55,251,069 0.49 0.049
Uncharacterized protein BAB2_0314 4 4 20.2 31.889 85,886,000 16,277,166 179,173,333 35,972,045 0.48 0.014
30S ribosomal protein S7 BAB1_1259 5 5 47.4 17.625 120,006,000 25,170,853 369,945,000 59,435,000 0.32 0.013
Type IV secretion system protein VirB9 BAB2_0060 3 3 17.2 32.163 37,235,333 5,824,328 362,463,333 193,556,345 0.10 0.043

P-values < .05.

a
PepCount indicated the total number of detected peptides for the individual protein.
b
Unique PepCount referred to the number of different peptides assigned to the proteins.
c
MW (KDa) was the molecular weight of each identified protein.
d
LFQ, label-free quantification.
e
N, Non.
Journal of Proteomics 195 (2019) 66–75
P. Li et al. Journal of Proteomics 195 (2019) 66–75

Fig. 1. Determination of the LDH release. RAW264.7 cells cultured in a 24-well plate were infected with mutants at a MOI of 100, and cell death was determined at
8, and 12 h p.i.. Uninfected cells were used as negative controls (Mock). (A) LDH released from ΔrfbEΔompW, ΔrfbEΔvirB10, ΔrfbEΔBAB1_970, ΔrfbEΔBAB1_1185 or
ΔrfbEΔBAB1_1576 infected cells. ns, no significant, ****p < .0001. (B) LDH released from ΔrfbEΔvirB4(domain), ΔrfbEΔvirB11(domain), ΔrfbEΔvirB123,
ΔrfbEΔvirB4, ΔrfbEΔvirB7, ΔrfbEΔvirB8910, or ΔrfbEΔvirB11 infected cells. ****p < .0001. (C) LDH released from ΔrfbEΔVceC, ΔrfbEΔBPE005, ΔrfbEΔBPE275,
ΔrfbEΔBspB, ΔrfbEΔBspE, ΔrfbEΔBspF, ΔrfbEΔRicA or ΔrfbEΔSepA infected cells. ****p < .0001. (D) LDH released from ΔrfbEΔBtpA, ΔrfbEΔBtpB, ΔrfbEΔVceA,
ΔrfbEΔBPE123 or ΔrfbEΔBspA infected cells. ns, no significant. LDH released from S2308 and ΔrfbE was used as controls.

macrophages infected with ΔrfbEΔvirB123, ΔrfbEΔvirB4, ΔrfbEΔvirB7,
ΔrfbEΔvirB8910, ΔrfbEΔvirB11, ΔrfbEΔvirB4 (domain) or ΔrfbEΔvirB11
(domain) was reduced, compared to the ΔrfbE mutant infected cells, but
similar to those from the smooth wild-type strain S2308 infected cells at
8 h and 12 h p.i. (Fig. 1B).
To determine which T4SS effector is associated with macrophage
death caused by infection with the ΔrfbE mutant, we constructed
ΔrfbEΔVceC, ΔrfbEΔBPE005, ΔrfbEΔBPE275, ΔrfbEΔBspB, ΔrfbEΔBspE,
ΔrfbEΔBspF, ΔrfbEΔRicA, ΔrfbEΔSepA, ΔrfbEΔBtpA, ΔrfbEΔBtpB,
ΔrfbEΔVceA, ΔrfbEΔBPE123 and ΔrfbEΔBspA to infect RAW264.7 mac-
rophages. The LDH release from the cells infected with ΔrfbEΔBPE005,
ΔrfbEΔBPE275, ΔrfbEΔBspB, ΔrfbEΔBspE, ΔrfbEΔBspF, ΔrfbEΔBtpA,
ΔrfbEΔBtpB, ΔrfbEΔVceA, ΔrfbEΔBPE123 or ΔrfbEΔBspA showed no
difference with those from the ΔrfbE mutant infected cells (Fig. 1C and
D). However, LDH release from ΔrfbEΔVceC, ΔrfbEΔRicA or ΔrfbEΔSepA
infected cells was increased, compared to the ΔrfbE mutant infected
cells (Fig. 1C).
Taken together, T4SS components were not crucial for macrophage
death caused by infection with the ΔrfbE mutant, but some T4SS ef-
fectors did.
3.5. OmpW family protein (BAB1_1579) and uncharacterized protein
BAB1_1185 were associated with macrophage death caused by ΔrfbE
infection
To identify the crucial differential secretory proteins associated with
macrophage death caused by infection with the ΔrfbE mutant, vceC,
virB10, ompW, BAB1_0970, BAB1_1576 or BAB1_1576 gene over-
expression strains were constructed from the smooth wild-type strain
S2308, respectively, thereby generating overexpression strains S2308
(pVceC), S2308(pVirB10), S2308(pOmpW), S2308(pBAB1_0970),
S2308(pBAB1_1576) and S2308(pBAB1_1185). The overexpression of
the genes was detected by qPCR, indicating that the expression of these
genes were significantly increased in the smooth wild-type strain S2308
(Fig. 2A). The LDH release assay was then performed at 8 h and 12 h p.i.
to assess quantitatively the death of macrophages infected with over-
expression strains S2308(pVceC), S2308(pVirB10), S2308(pOmpW),
S2308(pBAB1_0970), S2308(pBAB1_1576) and S2308(pBAB1_1185).
The results showed that the S2308(pVceC), S2308(pOmpW),
S2308(pBAB1_1185)-infected macrophages significantly increased LDH
release, compared to the smooth wild-type strain S2308 infected cells
(Fig. 2B), indicating that overexpression of vceC, ompW and BAB1_1576
in the smooth wild-type strain S2308 increased cytotoxicity within
macrophages. Furthermore, the death of Brucella-infected macrophages
was analyzed using annexin V-FITC and PI staining, which was used to
detect translocation of phosphatidylserine from the inner cell mem-
brane to the outer cell membrane during the early stages of apoptosis.
The PI stains the DNA of necrotic cells and/or cells at the late stage of
apoptosis [23]. The results showed that S2308(pVceC),
S2308(pOmpW), S2308(pBAB1_1185) strains significantly increased
macrophage death at 8 h and 12 h p.i., compared to the smooth wild-
type strain S2308 infected cells (Fig. 2C). However, the macrophage
death induced by infection of S2308(pVceC), S2308(pOmpW), or
S2308(pBAB1_1185) is less than that induced by the rough mutant
strain ΔrfbE infection (Fig. 2C).
Taken together, we demonstrated that T4SS effectors VceC, secre-
tory proteins OmpW family protein (BAB1_1579) and uncharacterized
protein BAB1_1185 were associated with macrophage death caused by
infection with the ΔrfbE mutant.
3.6. Validation of differential secretory proteins from ΔrfbE and
ΔrfbEΔvirB123 at protein level
Western blot analysis showed that OmpW family protein

72
P. Li et al. Journal of Proteomics 195 (2019) 66–75

Fig. 2. BAB1_1579 and BAB1_1185 overexpression enhanced cytotoxicity in Brucella S2308 infected RAW264.7 cells. (A) qPCR analysis of the overexpression
strains. The expression of gene X (vceC, ompW, virB10, BAB1_970, BAB1_1185 and BAB1_1576) in the strains S2308 and S2308 (pBCSP31-X) was detected.
****p < .0001. (B) LDH detection. RAW264.7 cells were infected with S2308(pVceC), S2308(pVirB10), S2308(pOmpW), S2308(pBAB1_0970), S2308(pBAB1_1576)
or S2308(pBAB1_1185) at an MOI of 100. The supernatants were collected at 8 and 12 h p.i., and LDH release was detected using the CytoTox 96 nonradioactive
cytotoxicity assay. The supernatants of uninfected RAW264.7 cells were used as negative controls (medium). ns, no significant, ****p < .0001. (C) Phase-contrast
microscopy and Annexin V-FITC/PI staining. RAW264.7 cells were infected with S2308(pVceC), S2308(pVirB10), S2308(pOmpW), S2308(pBAB1_0970),
S2308(pBAB1_1576) and S2308(pBAB1_1185) at an MOI of 100. The cells were stained with FITC-annexin (green) and PI (red) at 8 and 12 h p.i., and observed using
fluorescence microscopy at a magnification of ×100. Uninfected RAW264.7 cells were used as negative controls (Mock). (For interpretation of the references to color
in this figure legend, the reader is referred to the web version of this article.)

(BAB1_1579) and uncharacterized protein BAB1_1185 were detected in

the secretory proteins of ΔrfbE mutant, but not in those of the
ΔrfbEΔvirB123 mutant (Fig. 3). These data further confirmed the
changes in the secretory proteins detected by the Lable-free based
comparative proteomic analysis.

4. Discussion

Brucella are facultative intracellular bacteria which target to mac-

rophages. Wild-type Brucella strain can survive and multiply in mac-
rophages [4,25], and rough Brucella strain can induce the death of in-
fected macrophages, which is T4SS dependent [5,26]. In comparison to
the Brucella smooth wild-type strain, VjbR upregulation in the Brucella
rough mutant increases transcription of the virB operon, resulting in
overexpression of the T4SS gene, accompanied by the over-secretion of
effectors, thereby causing the death of infected macrophages via
strongly activating IRE1α pathway of ER stress [6]. However, the T4SS
Fig. 3. Validation of differential secretory proteins from ΔrfbE and
effectors responsible for the cytotoxicity have not been identified. The ΔrfbEΔvirB123 at the protein level. The protein levels of OmpW family
present study describes the use of a Lable-free based comparative pro- protein (BAB1_1579) and uncharacterized protein BAB1_1185 were identified
teomics method to identify differential secretory proteins between by Western blot. SDS-PAGE of secretory proteins from ΔrfbE and ΔrfbEΔvirB123
Brucella rough mutants ΔrfbE and ΔrfbEΔvirB123. A total of 861 secre- served as an internal control for protein normalization. Lane M, protein ladders.
tory proteins were identified, among which 37 were differential se- Lane 1, the sample from ΔrfbE. Lane 2, the sample from ΔrfbEΔvirB123. The
cretory proteins between ΔrfbE and ΔrfbEΔvirB123. GO enrichment arrows indicate the protein bands.
analysis demonstrated that the differential secretory proteins were
mainly involved in cellular process and metabolic process, distributed
in the cell and membrane, possessed molecular function of catalytic

73
P. Li et al.
activity and binding. Some proteins, such as structural proteins, which
are theoretically located within the cell, were detected among the
identified secretory proteins. The presence of these proteins among the
differential secretory proteins might be due to incomplete LPS, extra-
cellular vesicles or a novel secretion mechanism. The KEGG pathway
analysis showed that the differential secretory proteins were associated
with the ribosome (50S ribosomal protein, 30S ribosomal protein S3
and 30S ribosomal protein S7), NOD-like receptor signaling pathway
(Thioredoxin), two-component system (Outer membrane efflux protein)
and bacterial secretion system (Outer membrane efflux protein). In-
flammation and IL-6 production triggered by infection with B. abortus,
which induces ER stress by injecting the T4SS effector protein VceC into
host cells, is TRAF2, NOD1/2 and RIP2-dependent [27]. Two-compo-
nent systems represent a predominant signal transduction mechanism,
allowing an appropriate response to be mounted when a stimulus is
sensed [28]. B. abortus ribosomes incorporated in incomplete adjuvant
induced high titres of agglutinins, complement fixing antibodies and
precipitins for B. abortus antigens [29]. Rough Brucella strain has lost
their ability to survive and multiply in both macrophages and epithelial
cells as well as in the mouse virulence model. Thus, the different pro-
teins may account for this appearance.
This study aimed to identify the effector proteins associated with
macrophage death caused by infection with the ΔrfbE mutant. We found
that Brucella rough mutant ΔrfbE secreted T4SS protein VirB10
(BAB2_0059), OmpW family protein (BAB1_1579), bacterial extra-
cellular solute-binding protein (BAB1_0970), uncharacterized proteins
BAB1_1185 and BAB1_1576, but deletion of virB123 in the ΔrfbE
abolished the secretion of these proteins. Cell death analysis demon-
strated that deletion of virB10 in the ΔrfbE mutant significantly reduced
its cytotoxicity in macrophages. However, deletion of ompW,
BAB1_0970, BAB1_1185 or BAB1_1576 in the ΔrfbE mutant did not re-
duce cytotoxicity within macrophages. Remarkably, VirB10 is the T4SS
component which serves as the outer membrane channel with a unique
hydrophobic double helical trans-membrane region and crosses both
membranes of Gram-negative bacteria [30,31]. The absence of B. canis
virB10 and virB11 genes from the virB operon is associated with a re-
duction of virulence in mice, which may be considered as candidates for
studies to be conducted in dogs against canine brucellosis [32]. Thus,
we predict macrophage death caused by ΔrfbE infection might be as-
sociated with T4SS components.
In some other bacteria, pathogenesis has been associated with the
integrity of a T4SS which contain 11 genes highly similar to the 11 virB
genes of Agrobacterium tumefaciens and an extra ORF12 that shares
homology with an adhesin of Pseudomonas fluorescens [33]. The lack of
functional T4SS for example deletion of virB2, virB3, virB4, virB5, virB6,
virB8, virB9, virB10, or virB11 significantly attenuated B. abortus in
macrophage in vitro and in the spleen in mouse model [34,35]. Brucella
rough strain can induce the death of infected macrophages, which is
T4SS dependent [6]. In this study, we investigated whether macro-
phage death caused by infection with the ΔrfbE mutant is associated
with T4SS components. Our data showed that deletion of either T4SS
component in the ΔrfbE mutant significantly reduced its cytotoxicity in
macrophages, indicating that integrity of T4SS in the ΔrfbE mutant is
important for inducing the death of infected macrophages. ATPases
VirB4 and VirB11 are essential for T4SS function by providing energy
for T4SS assembly and transport of the effector proteins [36,37]. A
conserved NTP-binding domain in VirB4 or VirB11 suggests that one or
both proteins couple energy by NTP hydrolysis to transport of effectors
[36]. Furthermore, the NTP-binding region mutated VirB4 and VirB11
strains were constructed in the ΔrfbE mutant and the cytotoxicities in
macrophages were determined. Both mutants with abolished T4SS se-
cretory function, but maintained the integrity, significantly reduced its
cytotoxicity in macrophages, demonstrating that the crucial factor for
macrophage death caused by ΔrfbE infection was not T4SS components,
but T4SS effectors. Recent studies have identified 15 effectors that are
secreted by Brucella in a T4SS-dependent manner. VceC, RicA and SepA
74
Journal of Proteomics 195 (2019) 66–75
contributed to Brucella intracellular survival [8,9,22]. In this study, cell
death analysis demonstrated that deletion of vecC, ricA or sepA in the
ΔrfbE mutant increased their cytotoxicities within macrophages. One
possible explanation is that deletion of vecC, ricA or sepA in the ΔrfbE
mutant decreased Brucella intracellular survival. However, deletion of
other T4SS effectors in the ΔrfbE mutant did not reduce cytotoxicity
within macrophages, suggesting that macrophage death caused by
ΔrfbE infection was associated with several T4SS effectors, but not just
one. Taken together, T4SS components were not crucial to macrophage
death caused by ΔrfbE infection, but some T4SS effectors did.
The cytotoxicity of Brucella for macrophages is probably mediated
by increased secretion of effector proteins that results from over-
expression of virB or an increase in the number of bacterial cells [38].
To determine the differential secretory proteins between ΔrfbE and
ΔrfbEΔvirB123 were crucial for macrophage death caused by ΔrfbE in-
fection, vceC, virB10, ompW, BAB1_0970, BAB1_1576 or BAB1_1576
gene was overexpressed in the smooth wild-type strain S2308, respec-
tively. VceC, a Brucella T4SS effector, is involved in the induction of
inflammatory responses by binding chaperone BiP, to trigger ER stress
[14]. In this study, cell death analysis showed that overexpression of
vceC in the smooth wild-type strain S2308 increased cytotoxicity within
macrophages, compared to the smooth wild-type strain S2308 infected
cells. Outer membrane proteins (OMPs) of Gram-negative bacteria have
diverse functions, such as iron uptake, antimicrobial peptide resistance,
serum resistance, multi-drug resistance and bile resistance, and are
directly involved in the interaction with various environments en-
countered by pathogenic organisms [39]. Overexpression of ompW
(BAB1_1579) or BAB1_1185 in the smooth wild-type strain S2308 also
increased cytotoxicity within macrophages, compared to the smooth
wild-type strain S2308 infected cells. Notablely, all three over-
expression strains of S2308(pVceC), S2308(pOmpW), and
S2308(pBAB1_1185) with vceC, ompW (BAB1_1579) or BAB1_1185
overexpressed in S2308 induced less cytotoxicity within macrophages
than the rough mutant ΔrfbE did, indicating multiple effectors may
work together to cause the cell death. Cytotoxicity in macrophages that
have been infected by Brucella rough mutants is reportedly T4SS de-
pendent [6,23,26]. Thus, T4SS effector VceC, secretory proteins OmpW
family protein (BAB1_1579) and uncharacterized protein BAB1_1185
were the crucial secretory proteins which were associated with mac-
rophage death caused by infection with Brucella rough mutant ΔrfbE.
However, whether OmpW family protein (BAB1_1579) and un-
characterized protein BAB1_1185 were secreted by T4SS in Brucella
remain to be identified.
Taken together, we analyzed the secretory proteins from Brucella
rough mutants ΔrfbE and ΔrfbEΔvirB123 using LC–MS/MS proteomics
and 861 unique proteins were identified in this study. Two secretory
proteins which were associated with macrophage death caused by ΔrfbE
infection were identified. Our findings provide important data to im-
prove our understanding of macrophage death caused by ΔrfbE infec-
tion and to screen vaccine candidates for Brucella.
Supplementary data to this article can be found online at https://
doi.org/10.1016/j.jprot.2019.01.008.
Conflict of interest statement
The authors declare no competing financial interests.
Authors and contributors
SY, MT, and CD conceived and designed the experiments; PL per-
formed the experiments and analyzed the data; MT, HH, YY, XG helped
to perform some experiments; PL wrote the paper, SY revised the
manuscript and coordinated the research. All authors have read and
approved the manuscript.
P. Li et al.
Acknowledgements
We thank the funds from the Scientific and Technical Innovation
Project of the Chinese Academy of Agricultural Sciences (SHVRI-ASTIP-
2014-8) and the National Natural Science Foundation of China
(31602070).
References
[1] J.P. Gorvel, Brucella: a Mr "Hide" converted into Dr Jekyll, Microbes Infect. 10
(2008) 1010–1013.
[2] M.L. Boschiroli, V. Foulongne, D. O'Callaghan, Brucellosis: a worldwide zoonosis,
Curr. Opin. Microbiol. 4 (2001) 58–64.
[3] E. Barquero-Calvo, E. Chaves-Olarte, D.S. Weiss, C. Guzman-Verri, C. Chacon-Diaz,
A. Rucavado, I. Moriyon, E. Moreno, Brucella abortus uses a stealthy strategy to
avoid activation of the innate immune system during the onset of infection, PLoS
One 2 (2007) e631.
[4] J.P. Gorvel, E. Moreno, Brucella intracellular life: from invasion to intracellular
replication, Vet. Microbiol. 90 (2002) 281–297.
[5] J. Pei, T.A. Ficht, Brucella abortus rough mutants are cytopathic for macrophages in
culture, Infect. Immun. 72 (2003) 440–450.
[6] P. Li, M. Tian, Y. Bao, H. Hu, J. Liu, Y. Yin, C. Ding, S. Wang, S. Yu, Brucella Rough
Mutant Induce Macrophage Death via Activating IRE1alpha Pathway of
Endoplasmic Reticulum stress by Enhanced T4SS Secretion, Front. Cell. Infect.
Microbiol. 7 (2017) 422.
[7] D.N. Bronner, B.H. Abuaita, X. Chen, K.A. Fitzgerald, G. Nunez, Y. He, X.M. Yin,
M.X. O'Riordan, Endoplasmic Reticulum stress Activates the Inflammasome via
NLRP3- and Caspase-2-Driven Mitochondrial damage, Immunity 43 (2015)
451–462.
[8] M.F. de Jong, Y.H. Sun, A.B. den Hartigh, J.M. van Dijl, R.M. Tsolis, Identification
of VceA and VceC, two members of the VjbR regulon that are translocated into
macrophages by the Brucella type IV secretion system, Mol. Microbiol. 70 (2008)
1378–1396.
[9] M. de Barsy, A. Jamet, D. Filopon, C. Nicolas, G. Laloux, J.F. Rual, A. Muller,
J.C. Twizere, B. Nkengfac, J. Vandenhaute, D.E. Hill, S.P. Salcedo, J.P. Gorvel,
J.J. Letesson, X. De Bolle, Identification of a Brucella spp. secreted effector specifi-
cally interacting with human small GTPase Rab2, Cell. Microbiol. 13 (2011)
1044–1058.
[10] M.I. Marchesini, C.K. Herrmann, S.P. Salcedo, J.P. Gorvel, D.J. Comerci, In search
of Brucella abortus type IV secretion substrates: screening and identification of four
proteins translocated into host cells through VirB system, Cell. Microbiol. 13 (2011)
1261–1274.
[11] P.J. Christie, K. Atmakuri, V. Krishnamoorthy, S. Jakubowski, E. Cascales,
Biogenesis, architecture, and function of bacterial type IV secretion systems, Annu.
Rev. Microbiol. 59 (2005) 451–485.
[12] S. Backert, T.F. Meyer, Type IV secretion systems and their effectors in bacterial
pathogenesis, Curr. Opin. Microbiol. 9 (2006) 207–217.
[13] I.S. Franco, H.A. Shuman, X. Charpentier, The perplexing functions and surprising
origins of Legionella pneumophila type IV secretion effectors, Cell. Microbiol. 11
(2009) 1435–1443.
[14] M.F. de Jong, T. Starr, M.G. Winter, A.B. den Hartigh, R. Child, L.A. Knodler,
J.M. van Dijl, J. Celli, R.M. Tsolis, Sensing of bacterial type IV secretion via the
unfolded protein response, MBio 4 (2013) (e00418-12).
[15] M.I. Marchesini, S.M. Morrone Seijo, F.F. Guaimas, D.J. Comerci, A T4SS Effector
Targets Host Cell Alpha-Enolase contributing to Brucella abortus Intracellular
Lifestyle, Front. Cell. Infect. Microbiol. 6 (2016) 153.
[16] P.C. Arriola Benitez, D. Rey Serantes, C.K. Herrmann, A.I. Pesce Viglietti,
S. Vanzulli, G.H. Giambartolomei, D.J. Comerci, M.V. Delpino, The Effector Protein
BPE005 from Brucella abortus Induces Collagen Deposition and Matrix
Metalloproteinase 9 Downmodulation via Transforming Growth factor beta1 in
Hepatic Stellate Cells, Infect. Immun. 84 (2015) 598–606.
[17] J.A. Smith, M. Khan, D.D. Magnani, J.S. Harms, M. Durward, G.K. Radhakrishnan,
Y.P. Liu, G.A. Splitter, Brucella induces an unfolded protein response via TcpB that
supports intracellular replication in macrophages, PLoS Pathog. 9 (2013)
e1003785.
[18] P. Jakka, S. Namani, S. Murugan, N. Rai, G. Radhakrishnan, The Brucella effector
protein TcpB induces degradation of inflammatory caspases and thereby subverts
noncanonical inflammasome activation in macrophages, J. Biol. Chem. 292 (2017)
75
Journal of Proteomics 195 (2019) 66–75
20613–20627.
[19] S.P. Salcedo, M.I. Marchesini, C. Degos, M. Terwagne, K. Von Bargen, H. Lepidi,
C.K. Herrmann, T.L. Santos Lacerda, P.R. Imbert, P. Pierre, L. Alexopoulou,
J.J. Letesson, D.J. Comerci, J.P. Gorvel, BtpB, a novel Brucella TIR-containing ef-
fector protein with immune modulatory functions, Front. Cell. Infect. Microbiol. 3
(2013) 28.
[20] S. Myeni, R. Child, T.W. Ng, J.J. Kupko 3rd, T.D. Wehrly, S.F. Porcella,
L.A. Knodler, J. Celli, Brucella modulates secretory trafficking via multiple type IV
secretion effector proteins, PLoS Pathog. 9 (2013) e1003556.
[21] C.N. Miller, E.P. Smith, J.A. Cundiff, L.A. Knodler, J. Bailey Blackburn, V. Lupashin,
J. Celli, A Brucella Type IV Effector Targets the COG Tethering complex to Remodel
Host Secretory Traffic and Promote Intracellular Replication, Cell Host Microbe 22
(2017) (317-329 e7).
[22] P.H. Dohmer, E. Valguarnera, C. Czibener, J.E. Ugalde, Identification of a type IV
secretion substrate of Brucella abortus that participates in the early stages of in-
tracellular survival, Cell. Microbiol. 16 (2014) 396–410.
[23] M. Tian, J. Qu, X. Han, C. Ding, S. Wang, D. Peng, S. Yu, Mechanism of Asp24
upregulation in Brucella abortus rough mutant with a disrupted O-antigen export
system and effect of Asp24 in bacterial intracellular survival, Infect. Immun. 82
(2014) 2840–2850.
[24] M.M. Kahl-McDonagh, T.A. Ficht, Evaluation of protection afforded by Brucella
abortus and Brucella melitensis unmarked deletion mutants exhibiting different rates
of clearance in BALB/c mice, Infect. Immun. 74 (2006) 4048–4057.
[25] J. Celli, Surviving inside a macrophage: the many ways of Brucella, Res. Microbiol.
157 (2006) 93–98.
[26] J. Pei, Q. Wu, M. Kahl-McDonagh, T.A. Ficht, Cytotoxicity in macrophages infected
with rough Brucella mutants is type IV secretion system dependent, Infect. Immun.
76 (2008) 30–37.
[27] A.M. Keestra-Gounder, M.X. Byndloss, N. Seyffert, B.M. Young, A. Chavez-Arroyo,
A.Y. Tsai, S.A. Cevallos, M.G. Winter, O.H. Pham, C.R. Tiffany, M.F. de Jong,
T. Kerrinnes, R. Ravindran, P.A. Luciw, S.J. McSorley, A.J. Baumler, R.M. Tsolis,
NOD1 and NOD2 signalling links ER stress with inflammation, Nature 532 (2016)
394–397.
[28] A. Mirabella, R.M. Yanez Villanueva, R.M. Delrue, S. Uzureau, M.S. Zygmunt,
A. Cloeckaert, X. De Bolle, J.J. Letesson, The two-component system PrlS/PrlR of
Brucella melitensis is required for persistence in mice and appears to respond to ionic
strength, Microbiology 158 (2012) 2642–2651.
[29] M.J. Corbel, The immunogenic activity of ribosomal fractions derived from Brucella
abortus, J Hyg (Lond). 76 (1976) 65–74.
[30] V. Chandran, R. Fronzes, S. Duquerroy, N. Cronin, J. Navaza, G. Waksman,
Structure of the outer membrane complex of a type IV secretion system, Nature 462
(2009) 1011–1015.
[31] M.F. de Jong, R.M. Tsolis, Brucellosis and type IV secretion, Future Microbiol 7
(2012) 47–58.
[32] E. Palomares-Resendiz, B. Arellano-Reynoso, R. Hernandez-Castro, V. Tenorio-
Gutierrez, E. Salas-Tellez, F. Suarez-Guemes, E. Diaz-Aparicio, Immunogenic re-
sponse of Brucella canis virB10 and virB11 mutants in a murine model, Front. Cell.
Infect. Microbiol. 2 (2012) 35.
[33] D. O'Callaghan, C. Cazevieille, A. Allardet-Servent, M.L. Boschiroli, G. Bourg,
V. Foulongne, P. Frutos, Y. Kulakov, M. Ramuz, A homologue of the Agrobacterium
tumefaciens VirB and Bordetella pertussis Ptl type IV secretion systems is essential for
intracellular survival of Brucella suis, Mol. Microbiol. 33 (1999) 1210–1220.
[34] A.B. den Hartigh, H.G. Rolan, M.F. de Jong, R.M. Tsolis, VirB3 to VirB6 and VirB8
to VirB11, but not VirB7, are essential for mediating persistence of Brucella in the
reticuloendothelial system, J. Bacteriol. 190 (2008) 4427–4436.
[35] Y.H. Sun, H.G. Rolan, A.B. den Hartigh, D. Sondervan, R.M. Tsolis, Brucella abortus
virB12 is expressed during infection but is not an essential component of the type IV
secretion system, Infect. Immun. 73 (2005) 6048–6054.
[36] M. Watarai, S. Makino, T. Shirahata, An essential virulence protein of Brucella
abortus, VirB4, requires an intact nucleoside-triphosphate-binding domain,
Microbiology 148 (2002) 1439–1446.
[37] C.E. Alvarez-Martinez, P.J. Christie, Biological diversity of prokaryotic type IV se-
cretion systems, Microbiol. Mol. Biol. Rev. 73 (2009) 775–808.
[38] Z. Zhong, Y. Wang, F. Qiao, Z. Wang, X. Du, J. Xu, J. Zhao, Q. Qu, S. Dong, Y. Sun,
L. Huang, K. Huang, Z. Chen, Cytotoxicity of Brucella smooth strains for macro-
phages is mediated by increased secretion of the type IV secretion system,
Microbiology 155 (2009) 3392–3402.
[39] J. Lin, S. Huang, Q. Zhang, Outer membrane proteins: key players for bacterial
adaptation in host niches, Microbes Infect. 4 (2002) 325–331.