Hindawi Publishing Corporation

BioMed Research International

Volume 2014, Article ID 729618, 9 pages
http://dx.doi.org/10.1155/2014/729618

Research Article
Description of a Novel Adhesin of Mycobacterium avium
Subsp. paratuberculosis

Mariana Noelia Viale,1 Gabriela Echeverria-Valencia,1 Pablo Romasanta,2

María Laura Mon,1 Marisa Fernandez,2 Emilio Malchiodi,2 María Isabel Romano,1
Andrea Karina Gioffré,1 and María de la Paz Santangelo1
1
Instituto de Biotecnologı́a, Instituto Nacional de Tecnologı́a Agropecuaria, 1686 Hurlingham, Buenos Aires, Argentina
2
Instituto de Estudios de la Inmunidad Humoral (IDEHU), Facultad de Farmacia y Bioquı́mica, Universidad de Buenos Aires,
1113 Ciudad Autónoma de Buenos Aires, Argentina

Correspondence should be addressed to Marı́a de la Paz Santangelo; [email protected]

Received 7 March 2014; Accepted 29 June 2014; Published 22 July 2014

Academic Editor: Armando Acosta

Copyright © 2014 Mariana Noelia Viale et al. This is an open access article distributed under the Creative Commons Attribution
License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly
cited.

The binding and ingestion of Mycobacterium avium subsp. paratuberculosis (MAP) by host cells are fibronectin (FN) dependent. In
several species of mycobacteria, a specific family of proteins allows the attachment and internalization of these bacteria by epithelial
cells through interaction with FN. Thus, the identification of adhesion molecules is essential to understand the pathogenesis of MAP.
The aim of this study was to identify and characterize FN binding cell wall proteins of MAP. We searched for conserved adhesins
within a large panel of surface immunogenic proteins of MAP and investigated a possible interaction with FN. For this purpose, a
cell wall protein fraction was obtained and resolved by 2D electrophoresis. The immunoreactive spots were identified by MALDI-
TOF MS and a homology search was performed. We selected elongation factor Tu (EF-Tu) as candidate for further studies. We
demonstrated the FN-binding capability of EF-Tu using a ligand blot assay and also confirmed the interaction with FN in a dose-
dependent manner by ELISA. The dissociation constant of EF-Tu was determined by surface plasmon resonance and displayed
values within the 𝜇M range. These data support the hypothesis that this protein could be involved in the interaction of MAP with
epithelial cells through FN binding.

1. Introduction receptors for FN-opsonized mycobacteria in vitro and in vivo

[4]. M cells have the distinctive characteristic of displaying
Paratuberculosis (PTB) is a chronic granulomatous enteri- 𝛽1 integrins on their luminal face at high density; therefore,
tis of domestic and wild ruminants. This disease involves the presence of these integrins on M cells may explain why
extensive mycobacterial shedding, which accounts for the these cells are the entry of the bacteria. The interaction
high contagiousness, and ends with fatal enteritis. Decreases between MAP and FN is explained by the presence of the
in weight, milk production, and fertility produce severe FN-binding proteins called adhesins. Different adhesins from
economic loss [1]. The etiological agent of PTB is Mycobac- several pathogens were identified as virulence factors [5–
terium avium subsp. paratuberculosis (MAP). MAP enters the 9]. Adhesins may also induce strong protective immunity
intestinal tissue through M cells present in the dome epithe- in the host and, thus, remain attractive vaccine targets. For
lium covering the continuous Peyer’s patches in the distal instance a 27-kDa outer membrane protein from Salmonella
ileum [2, 3]. Initially, the pathogen interacts with proteins typhi binds to laminin and induces a strong protective
of the extracellular matrix (ECM), which function as ligands antibody response in animal models and humans [10]. In
for bacterial adhesion. Fibronectin (FN) binding is required addition, the antigen 85 complex (Ag85) was the first family of
for attachment and internalization of MAP by the epithelial mycobacterial proteins to be identified as having FN-binding
cells and 𝛽1 integrins have been identified as the host cell capability. Members of the Ag85 complex were described as
2 BioMed Research International

a mycobacterial adhesins firstly in M. tuberculosis [11] and in an ice bath for 15 min with pulses of 1 min on, 1 min off
then in several mycobacterial species [12–15]. The members in a Branson Sonifier S250. Cell wall proteins were obtained
of this complex are found within the outer envelope and as previously described by Hirschfield and collaborators [25].
culture supernatants of mycobacteria and are immunodom- Briefly, the sonic extract was centrifuged at 27,000 ×g for
inant antigens [16, 17]. Furthermore, these proteins possess 20 min, and the resulting cell wall containing pellet was
mycolyltransferase activity and catalyze the synthesis of the subjected to 2% sodium dodecyl sulfate (SDS) extraction
most abundant glycolipid of the mycobacterial cell wall, for 2 h at 50∘ C. The SDS extraction was repeated twice. The
trehalose 6,6-dimycolate (TDM) [18]. Another important protein concentration in the cell wall (CW) fraction was
adhesin described in mycobacteria is the fibronectin attach- evaluated with Kit 2D Quant (GE Healthcare).
ment protein (FAP). FAP is a member of a family of FN-
binding proteins present in several species of mycobacteria 2.3. Two-Dimensional-SDS Polyacrylamide Gel Electrophore-
that mediate the attachment and internalization of these sis (2D-SDS-PAGE). For 2D analysis, CW fractions were
bacteria by epithelial cells in vitro [19–23]. This protein is also first desalted performing a gel filtration step (Sephadex G25
called APA (for alanine-proline-rich antigen) and is encoded column). Proteins were precipitated with cold acetone and
by a gene annotated as MAP1569 in the MAP K-10 strain. resuspended in a reswelling buffer (8 M urea, 2% CHAPS,
Although the MAP-APA is not an immunodominant antigen, 0.5% IPG buffer pH 4–7, 20 mM DTT, and 0.004% bro-
it activates dendritic cells and induces a Th1 polarization [24]. mophenol blue). 2D-SDS-PAGE was performed as described
Furthermore, MAP-infected cattle showed a strong humoral by Xolalpa and collaborators [26]. The gels were transferred
response to recombinant APA assayed by Western blot and onto a nitrocellulose membrane (Hybond-ECL GE Health-
ELISA [7]. In a previous study conducted by our group, care) and the membranes were subjected to Western blot.
APA was detected mainly in the culture supernatant filtrates, The sera from 5 positive animals were pooled and diluted
demonstrating that this protein is predominantly secreted to 1 : 100 to detect immunogenic proteins from the CW
[7]. Other cell wall proteins thus could interact with FN to fraction. The immunoreactive spots were manually excised
facilitate complex formation and, in this way, allow adherence from a replicate Coomassie blue stained gel and sent to
to epithelial cells. the Mass Spectrometry Center for Biological and Chemical
With all this in mind, we hypothesized that molecules Analysis (CEQUIBIEM) at the School of Exact and Natural
with similar structure, even those from nonrelated microor- Sciences, University of Buenos Aires. The mass spectrometry
ganisms, could have conserved adhesin functions. In the platform used is made up of UV-MALDI-TOF/TOF Ultraflex
present study, we searched for conserved adhesins within a II (Bruker Daltonics) and the software Mascot was used
large panel of surface immunogenic proteins of MAP and to identify proteins from peptide sequence databases. The
investigated a possible interaction with FN. By using the protein score was calculated as −10 ∗ Log(𝑃), where 𝑃 is the
ligand blot assay (LBA), we confirmed the binding properties probability in which the observed match was a random event.
of a protein previously described in other bacteria and iden- Protein scores greater than 69 are significant (𝑃 < 0.05).
tified a novel surface component with FN-binding activity in The proteins identified by MALDI-TOF MS were sub-
MAP. The protein-protein interactions revealed by LBA were jected to bioinformatic analysis including similarity searches
confirmed by ELISA binding assays and surface plasmon with proteins with FN-binding domains. Sequence similarity
resonance (SPR) in order to determine the dissociation searches were performed by BlastP (http://blast.ncbi.nlm.nih
constant (KD). .gov/Blast.cgi).

2. Materials and Methods 2.4. Recombinant MAP-EF-Tu: Cloning and Expression Assays.
DNA from MAP was purified by the CTAB method as
2.1. Bacterial Strains and Culture Media. All cloning steps
described previously by van Embden and collaborators [27].
were performed in Escherichia coli DH5𝛼. E. coli BL21(kDE3)
PCR amplification was performed to amplify the complete
was used for recombinant protein production. E. coli was
open reading frame of EF-Tu using the forward primer eftu-
grown in Luria Bertani (LB) broth or on LB agar. When
fw ggatccgcgaaggcgaagttcgag (BamHI site) and the reverse
necessary, ampicillin was added to the medium at a concen-
primer eftu-rev aagcttctacttgatgatcttgac (HindIII site). The
tration of 100 𝜇g/mL. MAP was grown in Middlebrook 7H9
amplified 1,190 bp fragment was cloned into pGEM-T vec-
medium (Difco Laboratories, USA), 0.05% Tween 80, 0.5%
tor (Promega) and directionally subcloned into pRSET-A
glycerol, AD (0.5% bovine serum albumin, 0.2% glucose), and
(Invitrogen). Protein expression was induced with 1 mM
mycobactin (2 𝜇g/mL).
isopropyl 𝛽-D-1-thiogalactopyranoside (IPTG). The protein
was expressed as insoluble inclusion bodies and therefore
2.2. Preparation of Cell Wall Protein Fraction of MAP. MAP purified using a ProBond Ni-NTA Resin column (Invitrogen)
cultures were harvested at midlog phase, centrifuged at under denaturing conditions with 8 M urea. After purifica-
14,000 ×g for 20 min at room temperature and washed twice tion, EF-Tu was refolded by gradually removing the urea
with phosphate buffered saline (PBS). Cell pellets were using a refolding buffer (Tris 50 mM pH 8, Arg-HCl, EDTA,
resuspended in lysis buffer (PBS, 1 mM EDTA) with 1 mM GSH 3 mM, GSSG 0.3 mM, and PMSF 1 mM). Protein quan-
phenylmethanesulfonyl fluoride (PMSF), an inhibitor of ser- tification was performed using BCA protein assay (Pierce)
ine proteases, and then this suspension was probe sonicated following the manufacturer’s recommendations.
BioMed Research International 3

2.5. Western Blot. Proteins were fractionated on 12% SDS- conditions using a nonlinear BIA evaluation program. The
PAGE, according to Laemmli procedure [28], and then nonspecific binding control consisted of passing the analytes
stained with 0.25% Coomassie Brilliant Blue R250 (Sigma) on a free surface that had been previously activated and
or transferred onto nitrocellulose membranes (Hybond-ECL blocked.
GE Healthcare). EF-Tu was assayed by Western blotting using
1 : 3,000 dilution anti-His (GE Healthcare) as primary anti- 2.9. Humoral Response Evaluation by Line Print Immunoassay.
body and an alkaline phosphatase-conjugated anti-mouse The protein EF-Tu and a set of antigens, purified protein
IgG (Sigma) as secondary antibody (1 : 3,000 dilution). A derivative of M. avium (PPDa), purified protein derivative
colorimetric detection was performed using BCIP/NBT (5- of M. bovis (PPDb), and paratuberculosis protoplasmic anti-
bromo-4-chloro-3-indolylphosphate/nitroblue tetrazolium) gen (PPA3), were evaluated with different sera. Sera were
Color Development (Promega), according to the manufac- obtained from 10 healthy animals, from 8 animals with bovine
turer’s instructions. tuberculosis (TBB experimentally infected animals, positive
for delayed-type hypersensitivity—DTH—with PPDb and
2.6. Ligand Blot Assay (LBA). Five 𝜇g of the purified recom- with lesions at the end of the experience), and from 25
binant proteins was electrophoresed in 12% SDS-PAGE gels PTB naturally infected animals, positive for DTH with
and transferred onto nitrocellulose membranes (Hybond- PPDa and fecal culture positive. A total of 20 𝜇L of each
ECL, GE Healthcare). Ag85 [15] and AhpC were used as of the antigens was applied onto a nitrocellulose mem-
positive and negative controls, respectively. The membranes brane (Amersham Hybond TM-ECL) using a semiautomatic
were blocked with 5% BSA in PBS buffer for 1 h at room aerosolizer (Camag Scientific Inc., Wilmington, Delaware) at
temperature and incubated with 20 𝜇g/mL of FN for 24 hs a concentration of 100 𝜇g/mL. The membranes were blocked
at 4∘ C. The membranes were then washed three times with and placed in a “mini blotter” (Isogen BioSolutions). This
PBS and incubated with anti-FN in PBS-BSA 5% (1 : 100) for procedure allowed simultaneous analysis of the 45 sera, which
2 h at room temperature, followed by a final incubation with were evaluated at dilutions of 1 : 100. After 1 h incubation,
anti-Mouse IgG alkaline phosphatase antibody (1 : 30,000) sera were aspirated and the membranes were washed. The
for 1 h at room temperature. The membranes were washed
membranes were then incubated with protein G conjugated
and the colorimetric detection of the bound bait protein was
to peroxidase (1 : 1,500), washed, and finally developed with
performed.
chemiluminescence substrate (Pierce ECL Western blotting
substrate, Thermo Scientific) according to the manufacturer’s
2.7. Dose-Response Curves. The 96-well plates (Polysorp directions.
Nunc) were coated with 1 𝜇g of EF-Tu in 200 𝜇L carbonate
buffer pH 9.5 at 4∘ C and incubated overnight. AhpC was
included as a negative control. Plates were then blocked 3. Results
and increasing concentrations of FN (0, 1, 10, 20, 50, and 3.1. Analysis of MAP-Cell Wall Proteins. As a first screening
100 𝜇g/mL) were added in a final volume of 200 𝜇L. Protein for surface-exposed immunogenic proteins, the MAP-cell
binding was assessed with hyperimmune anti-FN serum at wall (CW) protein fraction was obtained and resolved by 2D-
the dilution of 1 : 100 followed by incubation with HRP- SDS-PAGE and transferred onto a nitrocellulose membrane.
conjugated anti-mouse IgG (Sigma) (1 : 500). Both incu-
Then, the membranes were analyzed by Western blot using
bations were performed at 37∘ C for 1 h. The wells were
a pool of sera from MAP-infected animals. The images were
washed three times, and the colorimetric detection was
digitalized and visually analyzed (Figure 1). A total of 41 spots
performed using 280 𝜇L of developing solution 2,2󸀠 -azino-
bis3-ethylbenzothiazoline-6-sulphonic acid (ABTS) at a con- corresponding to proteins that were recognized by positive
centration of 10 𝜇g/mL (Sigma) and 12 𝜇L of H2 O2 30% sera were excised from a replicate Coomassie blue stained gel
in 10 mL of 0.1 M citrate phosphate buffer at pH 5. The and subsequently identified by MALDI-TOF MS. A total of 18
absorbance at 405 nm was determined in a microplate reader proteins were identified by this method (Table 1). We focused
Multiskan Spectrum (Thermo Scientific). our interest in proteins that were previously characterized
as adhesins in pathogenic microorganisms. Among these
proteins, we selected elongation factor Tu (EF-Tu) because
2.8. Surface Plasmon Resonance (SPR). Protein-protein inter-
the homologous of Mycoplasma pneumoniae (identity 65%)
actions were assessed by SPR [29], using a BIAcoreT100
system (GE Healthcare). Briefly, FN was covalently immobi- and Acinetobacter baumannii (identity 72%) functions as a
lized on the BIAcore carboxymethylated dextran matrix −5 FN-binding protein that facilitates the interactions between
sensor chip (GE Healthcare) according to the manufacturer’s bacteria and extracellular matrix [30–32].
instructions. Protein solution of EF-Tu (0, 0.312, 0.625, 1.25, The eftu gene from MAP was evaluated for ortholo-
2.5, 5.0, and 10.0 𝜇M) in 10 mM HEPES, 150 mM NaCl, 3 mM gous in M. avium strain 104 (GenBank accession number
EDTA, 0.005% surfactant P20, pH 7.4 was injected over NC 008595) and M. tuberculosis (Mtb) H37Rv (GenBank
immobilized FN at a flow rate of 30 𝜇L/min for 1 min at accession number NC 000962). In the MAP genome (MAP
25∘ C. The dissociation was carried out with PBS-Tween 0.05% strain K-10 GenBank accession number NC 002944), this
or HBS-P20. Surfaces were regenerated by applying pulses gene was annotated as MAP 4143 and shares 93% of
of 10 mM HCl. The KD was determined under equilibrium nucleotide identity with that of Mtb and 100% with that of
4 BioMed Research International

170
130
100
70

55

40

35

25

(a) (b)

Figure 1: Analysis of MAP-cell wall proteins by 2D-SDS-PAGE. The cell wall protein fraction (CW) of MAP was resolved by 2D-SDS PAGE
per duplicate and the resulting gels were (a) stained with Coomassie blue or (b) transferred onto a nitrocellulose membrane and subjected to
Western blot. Sera from 5 positive animals were pooled and diluted to 1 : 100 to detect immunogenic proteins from the CW fraction. Molecular
weight standards are shown on the left.

70

55

40

35

(a) (b)

Figure 2: Purified EF-Tu protein using a ProBond Ni-NTA Resin column (Invitrogen). (a) Detection of the protein by Western blot using an
anti-Histidine Antibody (Promega). (b) Coomassie blue stained gel showing the purified protein.

M. avium. The identity at the protein level is 100% between was assayed as a positive control. No signal was observed for
MAP and M. avium and 97% with its orthologous in Mtb. the negative control, AhpC (Figure 3).
We searched for two FN-binding regions (FBR) in the
MAP-EF-Tu protein sequence, which has been previously 3.3. Dose-Response Curves. We further confirmed FN-EF-Tu
identified in the carboxyl terminus of Mycoplasma pneumo- interactions using an ELISA assay with some modifications.
niae EF-Tu [32, 33]. The BlastP analysis is shown in Table 2. Plates were coated with 1 𝜇g of EF-Tu or AhpC used as a
The comparison of both regions yielded an identity of 73% for
negative control and then incubated with increasing concen-
FBR1 and 69% for FBR2.
trations of FN. After incubation with the hyperimmune anti-
FN serum followed by HRP-conjugated anti-mouse IgG and
3.2. Recombinant Expression and Ligand Blot Assay (LBA). the corresponding substrate, the absorbance was determined
The MAP-EF-Tu protein was heterologously expressed in E.
in a microplate reader at 492 nm. We observed a dose-
coli as a recombinant His-tagged protein and its product
dependent interaction confirming the binding of EF-Tu with
was purified under denaturing conditions. Purified EF-Tu
protein is showed in Figure 2. The purified protein was FN (Figure 4).
used to perform LBA using FN as “bait.” Interactions were
detected with anti-FN polyclonal antiserum and an alkaline 3.4. Surface Plasmon Resonance (SPR). To finally determine
phosphatase-conjugated antibody with further colorimetric the EF-Tu dissociation constant (KD), protein-protein inter-
detection of the bound bait. A strong positive signal was actions were also assessed through SPR with a BIAcoreT100
observed for EF-Tu and for Ag85 recombinant protein, which system (GE Healthcare). FN was covalently immobilized
BioMed Research International 5

Table 1: Immunogenic CW proteins of MAP identified by MALDI-TOF MS. Proteins were excised from Coomassie blue stained gels and
subsequently identified by MALDI-TOF at the Mass Spectrometry Center for Biological and Chemical Analysis (CEQUIBIEM) at the School
of Exact and Natural Sciences, University of Buenos Aires. ND: no data.

Spot number MAP Locus Mtb Locus Gene Protein function/family ORF size (bp) Previously described as
envelope protein
Glyceraldehyde-3-
1 MAP1164 Rv1436 gapdh 1,020 He and de Buck, 2010
phosphate
(MAP) [34]
dehydrogenase
2 MAP1205 Rv1479 moxR 1,143 Mawuenyega et al., 2005
ATPase
(Mtb) [35]
3 MAP1325 Rv1630 rpsA S1 30S ribosomal protein S1 1,443 Gu et al., 2003 (Mtb) [36]
4 MAP1367 Rv1658 argG Argininosuccinate synthase 1,197 ND
5 MAP1889c Rv2145c wag31 783 He and de Buck, 2010
DivIVA family protein
(MAP) [34]
6 MAP1962 Rv2220 glnA1 Glutamine synthetase A1 1,437 Gu et al., 2003 (MAP) [36]
7 MAP1998 Rv2245 kasA 1,251 He and de Buck, 2010
3-Oxoacyl synthetase
(MAP) [34]
8 MAP1999 Rv2246 kasB 1 1,323 Mawuenyega et al., 2005
3-Oxoacyl sintase 2 B
(Mtb) [35]
9 MAP2453c Rv1308 atpA 1,665 He and de Buck, 2010
ATP synthase subunit alpha
(MAP) [34]
10 MAP2855c Rv2744c 35kd ag Phage shock protein A 828 ND
5 MAP3152c Rv3075c hpcH/hpaI Aldolase/citrate lyase 921 Gu et al., 2003 (MAP) [36]
11 MAP3404 Rv3285 accA3 Carbamoyl-phosphate 1,824 Mawuenyega et al., 2005
synthase subunit A (Mtb) [35]
12 MAP3567 Rv0148 Hypothetical protein Short-chain 864 He and de Buck, 2010
dehydrogenases/reductases (MAP) [34]
13 MAP3651c Rv0215c fadE3 2 Acyl-CoA dehydrogenase 1,218 He and de Buck, 2010
FadE3 (MAP) [34]
7 MAP3692c Rv0242c fabG4 1,365 He and de Buck, 2010
3-Ketoacyl reductase
(MAP) [34]
14 MAP3853 Rv0384c clpB ATP dependent protease 2,547 He and de Buck, 2010
ClpB (MAP) [34]
3 MAP3936 Rv0440 groEL2 1,626 He and de Buck, 2010
GroEL chaperonin
(MAP) [34]
15 MAP4143 Rv0685 eftu 1,191 He and de Buck, 2010
Elongation factor Tu
(MAP) [34]

Table 2: BlastP comparison between the two fibronectin binding confirmed the FN/EF-Tu binding with a moderate affinity
regions (FBRs) of EF-Tu identified by Balasubramanian and collab- through conformational sites.
orators [37] in Mycoplasma pneumoniae (MP) and the homologous
regions in MAP.
3.5. Humoral Response Evaluation by Line Print Immunoassay.
Identity FN-binding proteins could play a role in adhesion to the host
MP amino MAP amino
between MP and immunomodulation. With this in mind, we analyzed
acid region acid region
and MAP (%)
whether EF-Tu is able to stimulate antibody production.
FBR1 192–219 193–220 73 Using a line print assay, we evaluated a broader set of sera
FBR2 340–358 342–360 69 including TBB-infected animals and negative controls. The
sera were obtained from 25 MAP-naturally infected cattle,
8 M. bovis-experimentally infected cattle, and 10 healthy
on the BIAcore CM-5 sensor chip. Solutions of EF-Tu at bovines (negative controls). EF-Tu was recognized by 64%
different concentrations were injected over immobilized FN. of the MAP positive sera. However, this protein was also
The obtained KD was within the 𝜇M order, (3.1 ± 0.9) 10−6 M recognized by sera of healthy and TBB animals, suggest-
(Figure 5). This value is similar to the values of other adhesins ing the presence of antigenic epitopes conserved among
previously described [38]. The control using denaturized EF- mycobacteria species, including environmental mycobacteria
Tu was not able to bind FN. Therefore, all these experiments that sensitizes healthy cattle (Table 3).
6 BioMed Research International

MW EF-Tu Ag85 AhpC 30

100 25
70
20 KD = (3.1 ± 0.9) 10−6 M

Response (RU)
55
15

40 10

5
35 0
0 1e − 6 2e − 6 3e − 6 4e − 6 5e − 6 6e − 6 7e − 6 8e − 6
25 Concentration (M)

Figure 5: KD determination by surface plasmon resonance. Protein

solutions of EF-Tu at different concentrations were injected over
Figure 3: Analysis of FN-binding capability of EF-Tu through a immobilized FN. The experiment confirmed FN-EFTu binding, with
LBA. The blot with the recombinant proteins was incubated with a KD of (3.1 ± 0.9) 10−6 M.
20 𝜇g/mL FN. Colorimetric detection of the bound bait protein was
performed. We observed positive signal indicating the FN-binding
capability of EF-Tu and the positive control Ag85 (red circles). AhpC Table 3: Reactivity of bovine sera to the protein EF-Tu by line print
was used as a negative control. immunoassay. 20 𝜇L of antigens was applied onto a nitrocellulose
membrane and simultaneously confronted to sera from 10 healthy
animals, 8 animals with bovine tuberculosis (TBB), and 25 animals
with paratuberculosis (PTB). Sixty four percent of MAP positive sera
0.7
recognized EF-Tu.
∗∗
0.6 Number of sera with positive recognition
Healthy PTB infected TBB infected
0.5 Antigen
(𝑛 = 10) (𝑛 = 25) (𝑛 = 8)
OD (492 nm)

0.4 ∗∗ PPDA 4 16 3
∗
∗ PPA-3 1 18 2
0.3 PPDB 6 1 4
∗ EF-Tu 8 16 7
0.2

0.1
In this study, we have identified immunogenic CW
0 proteins of MAP by 2D and MALDI-TOF MS analysis. From
0 1 10 20 50 100
the identified proteins, we selected one candidate based on
FN (𝜇g/mL)
the similarity with other proteins having the ability to bind
Figure 4: Dose-response curves assayed by ELISA. Plates were ECM molecules in other pathogenic bacteria: elongation
coated with EF-Tu (black bars) or AhpC (white bars) used as factor Tu (EF-Tu).
negative control and incubated with different concentrations of FN. We first screened its FN-binding capability through a
The absorbance, measured at 492 nm, showed a dose-dependent ligand blot assay using FN and anti-FN antibodies. This
interaction confirming the binding of EF-Tu with FN. Significantly screening confirmed that, in these conditions, EF-Tu binds
different from values of the control protein AhpC ∗∗ 𝑃 < 0.01, FN (Figure 3). Through ELISA assays, using increasing con-
∗
𝑃 < 0.05. centrations of FN, we confirmed the interaction in a dose-
dependent manner (Figure 4). We finally determine the
dissociation constant KD by SPR analysis, which yielded a KD
4. Discussion within the order of 𝜇M (Figure 5) consistent with previous
reports [38].
MAP invades the intestinal tissue primarily through M cells. Several Mycobacterium adhesins capable to bind ECM
This could occur through FN-dependent mechanisms that proteins have been identified in mycobacteria, such as antigen
involve the binding of FN to proteins of the MAP CW 85 complex [11–15], APA [19–23], and GlnA1 [26]. Several
and to integrin receptors present on the luminal surface of reports have demonstrated that these proteins are involved
M cells. 𝛽1 integrins have been identified as the host cell in bacterial dissemination. EF-Tu is a protein responsible
receptors. In addition, the attachment and internalization of for critical steps in protein synthesis [39]. Moreover, this
MAP by epithelial cells in vitro and in vivo depend on APA- protein is a cytoplasmic protein with unusual CW location
FN interactions [4]. However, other MAP proteins could be among microorganisms. For instance, when E. coli is starved
involved in FN binding and they could be important for host- for carbohydrates, nitrogen, and phosphate, this protein
pathogen interactions. becomes methylated and associates to the membrane [40]. In
BioMed Research International 7

addition, this protein was detected as a major CW protein the negative siRNA transfection). This finding suggests that
of Mycobacterium leprae [41]. On the other hand, EF-Tu FN is not the only bacterial adhesion ligand contributing to
has been identified with periplasmic location in Neisseria the MAP ability to bind to host cells. Additionally, the MAP
gonorrhoeae [42] and in E. coli [43]. In Lactobacillus johnsonii Ag85 interaction to FN only partially accounted for MAP’s
and Listeria monocytogenes, EF-Tu is also associated to the ability to bind host cells. Other surface antigens of MAP K-10
membrane; in these bacteria this protein mediates binding to probably participate in additional ECM interactions.
mucin [44] and fibrinogen [37], respectively. Recently, it has In conclusion, we have identified a novel FN-binding
been reported that EF-Tu binds factor H and plasminogen MAP protein, EF-Tu, that could be implicated in the entrance
in Pseudomonas aeruginosa [45]. Thus, EF-Tu joins the of MAP into the epithelial cells, which is the first step of
group of housekeeping enzymes, which includes enolase mycobacteria infection necessary for the progression of PTB.
[46], glyceraldehyde-3-phosphate dehydrogenase [47], and
pyruvate dehydrogenase [30], that exhibit unexpected bio-
logical functions in addition to their well-defined enzymatic Conflict of Interests
activities. Despite the classification of EF-Tu as a cytosolic The authors declare that they have no conflict of interests
protein, Balasubramanian and collaborators [32] have shown regarding the publication of this paper.
that EF-Tu can relocate to the mycoplasma membrane sur-
face with an exposed carboxyl terminus that facilitates FN
binding. On the other hand, other reports document the Authors’ Contribution
presence of cytosolic proteins, including EF-Tu, on surfaces of
different bacteria. Important questions regarding how these Andrea Karina Gioffré and Marı́a de la Paz Santangelo are
proteins translocate to the surface remain unanswered, espe- contributed equally to this work.
cially since conventional secretion or anchoring signals are
absent. The cytoplasmic enzymefructose-1, 6-bisphosphate Acknowledgments
aldolase (FBA) has also been found on the surface of several
pathogenic bacteria. FBA is a glycolytic enzymethat, despite The authors are grateful to Dr. Julia Sabio y Garcı́a for
lacking secretion signals, translocates across the different critical reading of this paper. This work was funded by
compartments of the bacterial cell to access the surface, where ANCyPT Grant PICT1170 and ANCyPT Grant FONARSEC
it binds host molecules and exhibits nonglycolytic functions FS Agrobiotecnologı́a 2010. M. N. Viale and M. P. Santangelo
[40, 41]. are CONICET fellows.
The interaction of EF-Tu with the ECM could be a
key mechanism during host-pathogen interactions, However, References
the overall contribution of this adhesin to the host cell
binding remains unclear. Although there is a dose-dependent [1] C. Cocito, P. Gilot, M. Coene, M. de Kesel, P. Poupart, and
specific binding of EF-Tu to immobilized FN, the affinity P. Vannuffel, “Paratuberculosis,” Clinical Microbiology Reviews,
of EF-Tu to FN is intermediate. In the present study, EF-Tu vol. 7, no. 3, pp. 328–345, 1994.
displayed a KD in the 𝜇M range, similar to other proteins [2] E. Momotani, D. L. Whipple, A. B. Thiermann, and N. F.
proposed as adhesins. For instance, a surface-exposed protein Cheville, “Role of M cells and macrophages in the entrance
of leptospira, Lsa20, binds to plasminogen and laminin of Mycobacterium paratuberculosis into domes of ileal Peyer’s
with a KD in the 𝜇M range, postulated as a protein with patches in calves.,” Veterinary Pathology, vol. 25, no. 2, pp. 131–
137, 1988.
adherence function and proteolytic activity [38]. Moreover,
these adhesins can also bind other proteins of the ECM, such [3] Ó. G. Sigurđardóttir, C. M. Press, and Ø. Evensen, “Uptake
as elastin, laminin, and plasminogen, which contributes to of Mycobacterium avium subsp. paratuberculosis through the
distal small intestinal mucosa in goats: an ultrastructural study,”
adhesion, to invasion, and to the virulence of the bacteria
Veterinary Pathology, vol. 38, pp. 184–189, 2001.
modulating the immune response [15, 24, 45].
[4] T. E. Secott, T. L. Lin, and C. C. Wu, “Mycobacterium avium
In our study, recombinant EF-Tu was recognized by 64%
subsp. paratuberculosis fibronectin attachment protein facili-
of sera from MAP-infected cattle, suggesting a putative role in
tates M-cell targeting and invasion through a fibronectin bridge
the host immune response. However, sera from noninfected with host integrins,” Infection and Immunity, vol. 72, no. 7, pp.
animals also recognize EF-Tu. This result could be due to the 3724–3732, 2004.
presence of EF-Tu in environmental mycobacteria, such as [5] D. Joh, E. R. Wann, B. Kreikemeyer, P. Speziale, and M.
M. avium, which shares 100% identity with the MAP protein. Höök, “Role of fibronectin-binding MSCRAMMs in bacterial
Moreover, the suggested role of EF-Tu in the virulence of adherence and entry into mammalian cells,” Matrix Biology, vol.
pathogenic bacteria could be related to the host-pathogen 18, no. 3, pp. 211–223, 1999.
interactions. [6] C. Abou-Zeid, T. Garbe, R. Lathigra et al., “Genetic and
Kuo and collaborators [15] demonstrated that reducing immunological analysis of Mycobacterium tuberculosis fibro-
the expression of FN on Caco-2 cells that were transfected nectin-binding proteins,” Infection and Immunity, vol. 59, no. 8,
with FN-siRNA impaired the ability of Ag85 or Ag85- pp. 2712–2718, 1991.
expressing MAP K-10 to bind to the Caco-2 cells. How- [7] A. Gioffré, G. Echeverrı́a-Valencia, A. Arese et al., “Charac-
ever, MAP K-10 strain binding to FN siRNA-transfected terization of the Apa antigen from M. avium subsp. paratu-
Caco-2 cells showed only a 9.7% reduction (compared with berculosis: a conserved Mycobacterium antigen that elicits a
8 BioMed Research International

strong humoral response in cattle,” Veterinary Immunology and [22] T. E. Secott, T. L. Lin, and C. C. Wu, “Fibronectin attachment
Immunopathology, vol. 132, no. 2–4, pp. 199–208, 2009. protein homologue mediates fibronectin binding by Mycobac-
[8] S. J. Shin, C. F. Chang, S. P. McDonough, B. Thompson, H. terium avium subsp. paratuberculosis,” Infection and Immunity,
S. Yoo, and Y. Chang, “In vitro cellular immune responses to vol. 69, no. 4, pp. 2075–2082, 2001.
recombinant antigens of Mycobacterium avium subsp. paratu- [23] T. E. Secott, T. L. Lin, and C. C. Wu, “Fibronectin attachment
berculosis,” Infection and Immunity, vol. 73, no. 8, pp. 5074–5085, protein is necessary for efficient attachment and invasion of
2005. epithelial cells by Mycobacterium avium subsp. paratuberculo-
sis,” Infection and Immunity, vol. 70, no. 5, pp. 2670–2675, 2002.
[9] J. Mullerad, I. Michal, Y. Fishman, A. Hovav, R. G. Barletta, and
H. Bercovier, “The immunogenicity of Mycobacterium paratu- [24] S. L. Jun, J. S. Sung, M. T. Collins et al., “Mycobacterium avium
berculosis 85B antigen,” Medical Microbiology and Immunology, subsp. paratuberculosis fibronectin attachment protein activates
vol. 190, no. 4, pp. 179–187, 2002. dendritic cells and induces a Th1 polarization,” Infection and
Immunity, vol. 77, no. 7, pp. 2979–2988, 2009.
[10] S. Ghosh, K. Chakraborty, T. Nagaraja et al., “An adhesion
[25] G. R. Hirschfield, M. McNeil, and P. J. Brennan, “Peptidoglycan-
protein of Salmonella enterica serovar Typhi is required for
associated polypeptides of Mycobacterium tuberculosis,” Journal
pathogenesis and potential target for vaccine development,”
of Bacteriology, vol. 172, no. 2, pp. 1005–1013, 1990.
Proceedings of the National Academy of Sciences of the United
States of America, vol. 108, no. 8, pp. 3348–3353, 2011. [26] W. Xolalpa, A. J. Vallecillo, M. Lara et al., “Identification of novel
bacterial plasminogen-binding proteins in the human pathogen
[11] C. Abou-Zeid, T. L. Ratliff, H. G. Wiker, M. Harboe, J. Benned- Mycobacterium tuberculosis,” Proteomics, vol. 7, no. 18, pp. 3332–
sen, and G. A. W. Rook, “Characterization of fibronectin- 3341, 2007.
binding antigens released by Mycobacterium tuberculosis and
[27] J. D. A. van Embden, D. van Soolingen, P. M. Small, and P.
Mycobacterium bovis BCG,” Infection and Immunity, vol. 56, no.
W. M. Hermans, “Genetic markers for the epidemiology of
12, pp. 3046–3051, 1988.
tuberculosis,” Research in Microbiology, vol. 143, no. 4, pp. 385–
[12] J. E. R. Thole, R. Schöningh, A. A. M. Janson et al., “Molecular 391, 1992.
and immunological analysis of a fibronectin-binding protein [28] U. K. Laemmli, “Cleavage of structural proteins during the
antigen secreted by Mycobacterium leprae,” Molecular Microbi- assembly of the head of bacteriophage T4,” Nature, vol. 227, no.
ology, vol. 6, no. 2, pp. 153–163, 1992. 5259, pp. 680–685, 1970.
[13] H. Kitaura, N. Ohara, M. Naito, K. Kobayashi, and T. [29] R. Karlsson, A. Michaelsson, and L. Mattsson, “Kinetic anal-
Yamada, “Fibronectin-binding proteins secreted by Mycobac- ysis of monoclonal antibody-antigen interactions with a new
terium avium,” APMIS, vol. 108, no. 9, pp. 558–564, 2000. biosensor based analytical system,” Journal of Immunological
[14] V. Dheenadhayalan, K. Shin, C. Chang et al., “Cloning and Methods, vol. 145, no. 1-2, pp. 229–240, 1991.
characterization of the genes coding for antigen 85A, 85B [30] S. F. Dallo, T. R. Kannan, M. W. Blaylock, and J. B. Base-
and 85C of Mycobacterium avium subsp. paratuberculosis,” man, “Elongation factor Tu and E1 𝛽 subunit of pyruvate
Mitochondrial DNA, vol. 13, no. 5, pp. 287–294, 2002. dehydrogenase complex act as fibronectin binding proteins in
[15] C. J. Kuo, H. Bell, C. L. Hsieh, C. P. Ptak, and Y. F. Chang, “Novel Mycoplasma pneumoniae,” Molecular Microbiology, vol. 46, no.
mycobacteria antigen 85 complex binding motif on fibronectin,” 4, pp. 1041–1051, 2002.
The Journal of Biological Chemistry, vol. 287, no. 3, pp. 1892– [31] S. F. Dallo, B. Zhang, J. Denno et al., “Association of Acinetobac-
1902, 2012. ter baumannii EF-Tu with cell surface, outer membrane vesicles,
[16] S. L. Baldwin, C. D. D’Souza, I. M. Orme et al., “Immunogenic- and fibronectin,” The Scientific World Journal, vol. 2012, Article
ity and protective efficacy of DNA vaccines encoding secreted ID 128705, 10 pages, 2012.
and non-secreted forms of Mycobacterium tuberculosis Ag85A,” [32] S. Balasubramanian, T. R. Kannan, and J. B. Baseman, “The
Tubercle and Lung Disease, vol. 79, no. 4, pp. 251–259, 1999. surface-exposed carboxyl region of Mycoplasma pneumoniae
elongation factor Tu interacts with fibronectin,” Infection and
[17] A. T. Kamath, C. G. Feng, M. Macdonald, H. Briscoe, and
Immunity, vol. 76, no. 7, pp. 3116–3123, 2008.
W. J. Britton, “Differential protective efficacy of DNA vaccines
expressing secreted proteins of Mycobacterium tuberculosis,” [33] S. Balasubramanian, T. R. Kannan, P. J. Hart, and J. B. Baseman,
Infection and Immunity, vol. 67, no. 4, pp. 1702–1707, 1999. “Amino acid changes in elongation factor Tu of Mycoplasma
pneumoniae and Mycoplasma genitalium influence fibronectin
[18] J. T. Belisle, V. D. Vissa, T. Sievert, K. Takayama, P. J. Brennan, binding,” Infection and Immunity, vol. 77, no. 9, pp. 3533–3541,
and G. S. Besra, “Role of the major antigen of Mycobacterium 2009.
tuberculosis in cell wall biogenesis,” Science, vol. 276, no. 5317,
[34] Z. He and J. de Buck, “Localization of proteins in the cell wall of
pp. 1420–1422, 1997.
Mycobacterium avium subsp. paratuberculosis K10 by proteomic
[19] T. L. Ratliff, R. McCarthy, W. B. Telle, and E. J. Brown, “Purifi- analysis,” Proteome Science, vol. 8, article 21, 2010.
cation of a mycobacterial adhesin for fibronectin,” Infection and [35] K. G. Mawuenyega, C. V. Forst, K. M. Dobos et al., “Mycobac-
Immunity, vol. 61, no. 5, pp. 1889–1894, 1993. terium tuberculosis functional network analysis by global sub-
[20] J. S. Schorey, Q. Li, D. W. McCourt et al., “A Mycobacterium cellular protein profiling,” Molecular Biology of the Cell, vol. 16,
leprae gene encoding a fibronectin binding protein is used for no. 1, pp. 396–404, 2005.
efficient invasion of epithelial cells and Schwann cells,” Infection [36] S. Gu, J. Chen, K. M. Dobos, E. M. Bradbury, J. T. Belisle, and
and Immunity, vol. 63, no. 7, pp. 2652–2657, 1995. X. Chen, “Comprehensive proteomic profiling of the membrane
[21] J. S. Schorey, M. A. Holsti, T. L. Ratliff, P. M. Allen, and E. J. constituents of a Mycobacterium tuberculosis strain,” Molecular
Brown, “Characterization of the fibronectin-attachment protein and Cellular Proteomics, vol. 2, no. 12, pp. 1284–1296, 2003.
of Mycobacterium avium reveals a fibronectin-binding motif [37] J. Schaumburg, O. Diekmann, P. Hagendorff et al., “The cell wall
conserved among mycobacteria,” Molecular Microbiology, vol. subproteome of Listeria monocytogenes,” Proteomics, vol. 4, no.
21, no. 2, pp. 321–329, 1996. 10, pp. 2991–3006, 2004.
 BioMed Research International 9

[38] R. S. Mendes, M. von Atzingen, Z. M. de Morais et al., “The

novel leptospiral surface adhesin Lsa20 binds laminin and
human plasminogen and is probably expressed during infec-
tion,” Infection and Immunity, vol. 79, no. 11, pp. 4657–4667, 2011.
[39] G. R. Jacobson and J. P. Rosenbusch, “Abundance and mem-
brane association of elongation factor Tu in E. coli,” Nature, vol.
261, no. 5555, pp. 23–26, 1976.
[40] C. C. Young and R. W. Bernlohr, “Elongation factor Tu is
methylated in response to nutrient deprivation in Escherichia
coli,” Journal of Bacteriology, vol. 173, no. 10, pp. 3096–3100, 1991.
[41] M. A. M. Marques, S. Chitale, P. J. Brennan, and M. C. V.
Pessolani, “Mapping and identification of the major cell wall-
associated components of Mycobacterium leprae,” Infection and
Immunity, vol. 66, no. 6, pp. 2625–2631, 1998.
[42] S. F. Porcella, R. J. Belland, and R. C. Judd, “Identification of
an EF-Tu protein that is periplasm-associated and processed in
Neisseria gonorrhoeae,” Microbiology, vol. 142, no. 9, pp. 2481–
2489, 1996.
[43] B. D. Beck, P. G. Arscott, and A. Jacobson, “Novel properties
of bacterial elongation factor Tu,” Proceedings of the National
Academy of Sciences of the United States of America, vol. 75, no.
3, pp. 1250–1254, 1978.
[44] D. Granato, G. E. Bergonzelli, R. D. Pridmore, L. Marvin, M.
Rouvet, and I. E. Corthésy-Theulaz, “Cell surface-associated
elongation factor tu mediates the attachment of lactobacillus
johnsonii NCC533 (La1) to human intestinal cells and mucins,”
Infection and Immunity, vol. 72, no. 4, pp. 2160–2169, 2004.
[45] A. Kunert, J. Losse, C. Gruszin et al., “Immune evasion of the
human pathogen Pseudomonas aeruginosa: elongation factor
Tuf is a factor H and plasminogen binding protein,” Journal of
Immunology, vol. 179, no. 5, pp. 2979–2988, 2007.
[46] S. Bergmann, M. Rohde, G. S. Chhatwal, and S. Hammer-
schmidt, “𝛼-Enolase of Streptococcus pneumoniae is a plas-
min(ogen)-binding protein displayed on the bacterial cell sur-
face,” Molecular Microbiology, vol. 40, no. 6, pp. 1273–1287, 2001.
[47] V. Pancholi and V. A. Fischetti, “A major surface protein on
group a streptococci is a glyceraldehyde-3-phosphate-dehydro-
genase with multiple binding activity,” Journal of Experimental
Medicine, vol. 176, no. 2, pp. 415–426, 1992.