Expression of Fluorescent Proteins in Bifidobacteria for Analysis ofHost-Microbe Interactions

Verena Grimm, Marita Gleinser, Caroline Neu, Daria Zhurina, Christian U. Riedel

Institute of Microbiology and Biotechnology, University of Ulm, Ulm, Germany

Bifidobacteria are an important component of the human gastrointestinal microbiota and are frequently used as probiotics. Thegenetic inaccessibility and lack of molecular tools commonly used in other bacteria have hampered a detailed analysis of the ge-netic determinants of bifidobacteria involved in their adaptation to, colonization of, and interaction with the host. In the presentstudy, a range of molecular tools were developed that will allow the closing of some of the gaps in functional analysis of bifido-bacteria. A number of promoters were tested for transcriptional activity in Bifidobacterium bifidum S17 using pMDY23, a previ-ously published promoter probe vector. The promoter of the gap gene (Pgap) of B. bifidum S17 yielded the highest promoter ac-tivity among the promoters tested. Thus, this promoter and the pMDY23 backbone were used to construct a range of vectors forexpression of different fluorescent proteins (FPs). Successful expression of cyan fluorescent protein (CFP), green fluorescentprotein (GFP), yellow fluorescent protein (YFP), and mCherry could be shown for three strains representing three different Bifi-dobacterium spp. The red fluorescent B. bifidum S17/pVG-mCherry was further used to demonstrate application of fluorescentbifidobacteria for adhesion assays and detection in primary human macrophages cultured in vitro. Furthermore, pMGC-mCherry was cloned by combining a chloramphenicol resistance marker and expression of the FP mCherry under the control ofPgap. The chloramphenicol resistance marker of pMGC-mCherry was successfully used to determine gastrointestinal transit timeof B. bifidum S17. Moreover, B. bifidum S17/pMGC-mCherry could be detected in fecal samples of mice after oraladministration.

Bifidobacteria are Gram-positive bacteria of the class Actino-bacteria, which is highly diverse and comprises many eco-

nomically and clinically relevant microorganisms, such as Myco-bacterium spp., Corynebacterium spp., Streptomyces spp., andothers (1). Natural habitats of the members of the genus Bifido-bacterium include food, sewage, and oral cavities, but the mostimportant ecological niche of bifidobacteria is the intestinal tractof humans and animals (2). Some strains of bifidobacteria havebeen associated with beneficial effects on the health status of thehost and have thus attracted considerable interest by the food anddairy industries. Some of the beneficial effects that have beenclaimed to be related to the presence or administration of bifido-bacteria are cholesterol reduction, improvement of lactose intol-erance, alleviation of constipation, and immunomodulation (2–4). Two of the more promising targets for bifidobacterialtreatments are amelioration of chronic intestinal inflammation(5–10) and protection against infections with enteric pathogens(11–13).

Recently, some of the mechanisms of host colonization of bi-fidobacteria and their effect on the immune system have beenelucidated in more detail. Bifidobacterial genomes contain up toseven gene clusters for pili (14, 15). Some of these genes wereshown to be expressed under in vitro conditions and in the mousegastrointestinal tract (GIT), and electron and atomic force mi-croscopy revealed the presence of pili on the surface of some ofthese strains (14–17). Experiments with recombinant Lactococcuslactis strains expressing two of the pilus clusters of Bifidobacteriumbifidum PRL2010 suggest a role for adhesion to extracellular ma-trix proteins (16). Moreover, the importance of tad pili for colo-nization of the mice was demonstrated for Bifidobacterium breveUCC2003 (17). Similarly, the exopolysaccharide (EPS) producedby B. breve UCC2003 confers resistance against bile and acid andimproves persistence in the mouse GIT (12). Moreover, the pres-

ence of EPS on the surface changes the immunogenicity of B. breveUCC2003. Higher numbers of proinflammatory immune cellswere observed in the spleens of mice colonized with a mutantlacking EPS than in the wild type (WT) (12). Of note, the EPS-producing wild-type B. breve UCC2003 was more efficient in re-ducing the colonization with the murine pathogen Citrobacter ro-dentium (12). Similarly, Bifidobacterium longum subsp. longumJCM 1217T and B. longum subsp. infantis 157F conferred protec-tion against infections in a mouse model of infections with enter-opathogenic Escherichia coli O157 (11), and a fermented milkproduct containing Bifidobacterium animalis subsp. lactis DN-173010 reduced intestinal inflammation by inhibiting colitogenic,Gram-negative bacteria in a model of ulcerative colitis (8). In bothstudies, the effect of bifidobacteria could be at least partially at-tributed to an acidification of the intestinal pH due to acetic andlactic acid produced by bifidobacteria.

Despite the progress in understanding the effect of bifidobac-teria and other probiotics, the European Food Safety Authorityhas so far rejected all of the claims submitted for probiotic bacteriabased on the fact that these claims have not been substantiatedsufficiently by clinical studies in humans (18). This highlights theneed for a better definition of the target groups for probiotic treat-

Received 24 December 2013 Accepted 21 February 2014

Published ahead of print 28 February 2014

Editor: G. T. Macfarlane

Address correspondence to Christian U. Riedel, christian.riedel@uni-ulm.de.

Supplemental material for this article may be found at http://dx.doi.org/10.1128/AEM.04261-13.

Copyright © 2014, American Society for Microbiology. All Rights Reserved.

doi:10.1128/AEM.04261-13

2842 aem.asm.org Applied and Environmental Microbiology p. 2842–2850 May 2014 Volume 80 Number 9

on October 17, 2020 by guest

http://aem.asm

.org/D

ownloaded from

ments as well as for relevant biomarkers, well-performed clinicaltrials, and a more detailed analysis of the underlying immunolog-ical mechanisms and the populations of immune cells that medi-ate the observed effects.

A key problem for analysis of the beneficial effects of bifido-bacteria on the host is their poor genetic accessibility. While someprogress has been made in improving transformation efficiencies,directed gene inactivation, and transposon mutagenesis for singlestrains (19, 20), the vast majority of strains of interest remaindifficult if not impossible to modify genetically. Moreover, theavailable vector systems for bifidobacteria have a rather limitedhost range (i.e., replicons function only in some species of thegenus Bifidobacterium but not in others), and expression systemsare poorly developed (2, 21).

An indispensable prerequisite to study the effect of bifidobac-teria on the immune system and its various cell populations is theability to label bacteria. Labeled bacteria may be tracked either invivo in live animals or used to identify, isolate, and analyze definedpopulations of immune cells that contain bacteria. Two of themethods most frequently used to label bacteria for these purposesare expression of either a luciferase or fluorescent proteins (FPs).However, both methods have been reported only for a single B.breve strain (22–24).

Since the discovery of the green fluorescent protein (GFP) ofthe jellyfish Aequorea victoria in 1962 (25), FPs have been isolatedfrom a variety of (mostly marine) organisms and continuouslymodified and improved for a wide range of applications (26–28).Today, FPs are one of the most versatile and widely used tools forthe investigation of bacterial physiology and interaction with thehost. For example, FPs have been successfully used as a transcrip-tional reporter for expression of genes or to study protein local-ization and dynamics (29–31). Furthermore, FPs are a valuabletool to monitor bacterial biofilms, to identify single, labeled cellsin a complex sample, or to study horizontal gene transfer (32). FPscan also be used to study virtually any aspect of the interaction ofbacteria with the host both in vitro and in vivo. Change in bacterialgene expression can be monitored using FPs as transcriptionalreporters, infected cells can be identified from a pool, and inter-action of bacteria with individual cells of the host or localization toa subcellular compartment can be visualized in vitro in live or fixedcells or tissue samples or even in living animals (33–36).

In the present study, we developed a range of vectors for func-tional analysis of various bifidobacteria. We report for the firsttime expression of four fluorescent proteins in different strains ofbifidobacteria and the use of these recombinant strains to imageinteractions with host cells. Furthermore, one of the vectors de-veloped was successfully used to monitor host colonization anddetection of bacteria in fecal samples.

MATERIALS AND METHODSBacterial strains and growth conditions. The bacterial strains and plas-mids used in this study are listed in Table S1 in the supplemental material.Bifidobacterium strains were cultured anaerobically in Lactobacillus MRSmedium (Difco) supplemented with 0.5 g liter�1

L-cysteine (MRSc) at37°C. Anaerobic conditions were achieved by cultivation using Anaero-Gen sachets (Merck) in sealed jars. For cultivation of B. bifidum S17 har-boring plasmids, the respective medium was supplemented with antibi-otics at the following concentrations unless stated otherwise: 1 �g ml�1

ampicillin, 5 �g ml�1 chloramphenicol, 2 �g ml�1 erythromycin, and 100�g ml�1 spectinomycin.

E. coli DH10B was used as the cloning host and was cultivated in 2�

tryptone-yeast extract medium and grown to stationary phase at 37°Cwith agitation. For cultivation of strains harboring plasmids, the respec-tive medium was supplemented with appropriate antibiotics (spectino-mycin and ampicillin, 100 �g ml�1; chloramphenicol, 15 �g ml�1; eryth-romycin, 300 �g ml�1). Sequence integrity of all cloned inserts wasverified by DNA sequencing (Eurofins MWG Operon, Germany). Trans-formation into B. bifidum S17, B. longum subsp. longum E18, and B. breveS27 was performed by electroporation as described previously (19).

Determination of selective antibiotic concentrations. Selective con-centrations for ampicillin, chloramphenicol, erythromycin, and spectino-mycin were determined for each of the Bifidobacterium sp. strains used inthis study by a microtiter plate assay. Fresh overnight cultures were di-luted in fresh MRSc containing different concentrations of antibiotics toan optical density at 600 nm (OD600) of 0.1 and distributed into individualwells of a 96-well tissue culture plate (Sarstedt AG & Co., Newton, NC). Asa positive control, MRSc medium without antibiotics was inoculated tothe same OD600. Plates were incubated under anaerobic conditions for 24h, and growth was determined by measuring the OD600 in a multilabelmicrotiter plate reader (Infinite M200; Tecan Group Ltd.). Experimentswere performed in four technical replicates per strain and condition, andthree independent cultures per strain were tested. Antibiotic concentra-tions 10 times higher than the concentrations allowing minimal growthwere considered selective.

Plasmid stability assay. Stability of plasmids in the absence of antibi-otics was determined as described previously (37). Briefly, plasmid-bear-ing strains were cultured in MRSc without antibiotics and were diluted1:1,000 every 12 h in fresh medium for a total of 100 generations. Plasmidstability was determined every 25 generations by plating on MRSc agarwithout antibiotics. At least 100 colonies were randomly selected andreplica plated on MRSc agar with the appropriate antibiotic at selectiveconcentrations or MRSc agar without antibiotics and incubated at 37°Cfor 48 h under anaerobic conditions. Growth on both plates was verified,and plasmid stability was calculated as the percentage of antibiotic-resis-tant colonies.

�-Glucuronidase assay. The glucuronidase assay was performed asdescribed elsewhere (38) with minor modifications. The assay is based onthe enzymatic conversion of 4-methylumbelliferyl-�-D-glucuronide(MUG) to 4-methylumbelliferone (MU) by the �-glucuronidase GusA.MU is a direct measure of GusA activity, which in turn correlates withpromoter activity. GusA activity was measured in cell extracts of B. bifi-dum S17 harboring pMDY23 derivatives with different promoters clonedupstream of the gusA reporter gene (see Materials and Methods in thesupplemental material). Recombinant strains were inoculated into 100 mlof MRSc at an OD600 of 0.1 from a fresh overnight culture and grownunder anaerobic conditions at 37°C. At the indicated time points, bacteriawere harvested by centrifugation (4°C, 5,000 � g, 10 min). After beingwashed with double-distilled water (ddH2O) and GUS assay buffer (50mM Na2HPO4 [pH 7], 1 mM EDTA, 0.1% Triton X-100, 5 mM dithio-threitol [DTT]), cells were resuspended in 1 ml GUS assay buffer anddisrupted in cryotubes with glass beads (0.1-mm diameter; Carl Roth) at4°C using a RiboLyser (Hybaid) for three cycles of 30 s at full speed, withcooling on ice for 1 min between cycles. Lysates were centrifuged for 15min at 14,000 � g and 4°C to remove beads and bacterial debris. Super-natants were retained as cell extracts, and protein content was measured inthe supernatants using a two-dimensional (2-D) Quant kit (GE Health-care Life Sciences).

GusA activity was measured in cell extracts using 2.5 to 10 �g totalprotein in a final volume of 100 �l in GUS assay buffer. The reaction wasstarted by adding 25 �l of 5 mM MUG. After an incubation time at 37°Cfor 30 min, during which MUG was hydrolyzed quantitatively to MU byGusA, a 10-�l aliquot was removed and mixed with 170 �l stop buffer (0.2M NaCO3, pH 9.5) in a 96-well plate (Sarstedt AG & Co.). MU was quan-tified by fluorescence using an Infinite M200 multimode reader (InfiniteM200; Tecan Group Ltd.) with excitation at 388 nm and measuring fluo-

Expression of Fluorescent Proteins in Bifidobacteria

May 2014 Volume 80 Number 9 aem.asm.org 2843

on October 17, 2020 by guest

http://aem.asm

.org/D

ownloaded from

rescence (emission) at 480 nm. As a negative control, B. bifidum S17harboring the promoterless pMDY23 plasmid was used.

Adhesion assays. Adhesion to Caco-2 cells was determined by theclassical plate counting method described previously (39). Briefly, conflu-ent monolayers of Caco-2 cells in 24-well tissue culture plates were used.At this stage, approximately 1 � 106 cells were counted per well. Anovernight culture of B. bifidum S17/pVG-mCherry was washed threetimes in phosphate-buffered saline (PBS) and resuspended in cell culturemedium (Dulbecco’s modified Eagle’s medium [DMEM] supplementedwith 10% fetal calf serum, 2 mM L-glutamine, 1% [vol/vol] nonessentialamino acids [NEAA]) and adjusted to an OD600 of 2, which equals 2 � 108

CFU ml�1. A total of 500 �l of this suspension was added to a well ofCaco-2 cells. Following an incubation of 1 h to allow adherence, unboundbacteria were removed by three washings in cell culture medium. Fluores-cence of each well was measured in a multilabel microtiter plate reader(Infinite M200; Tecan Group Ltd.) with excitation at 585 nm and emis-sion at 620 nm. Then, the Caco-2 cell monolayer was scraped off thebottom of the well, and the 500 �l of medium containing cellular debriswas quantitatively transferred to a sterile Eppendorf cup. Wells wererinsed with 500 �l of PBS, and the washings were combined with thedebris to give a total volume of 1 ml. The suspension was vigorouslyhomogenized, and 10-fold serial dilutions were plated on MRSc agar sup-plemented with 100 �g ml�1 spectinomycin to determine bifidobacterialcell numbers and calculate adhesion as CFU per well.

In order to allow for quantification of adherent bacteria using fluores-cence, a calibration curve was established in parallel. For this purpose,bacteria were adjusted to an OD600 of 2, 1, 0.5, or 0.25 in cell culturemedium, and 500 �l of these suspensions was added to separate wells ofthe plate containing Caco-2 cells. Following 1 h of incubation, fluores-cence was measured immediately without washing, and CFU were deter-mined as described above. Autofluorescence of Caco-2 cells without bac-teria was measured as a background control and subtracted from all othervalues. For fluorescence microscopy, Caco-2 cells were fixed with 4%(wt/vol) paraformaldehyde (PFA) in PBS.

Generation of primary human macrophages. Primary human mac-rophages were generated essentially as described previously (40). Briefly,lymphocytes were isolated from fresh buffy coats by density gradient cen-trifugation using lymphocyte separation medium (LSM 1077; PAA Lab-oratories). Monocytes were isolated from the lymphocyte pool by nega-tive selection using the monocyte isolation kit II (human; MiltenyiBiotech) according to the manufacturer’s instructions. Following isola-tion, monocytes were cultured in Lumox dishes (Sarstedt) for 7 days withRPMI 1640 medium containing 10% fetal calf serum, 10 mM L-glutamine,1% (vol/vol) penicillin-streptomycin (PAA Laboratories), and 1% (vol/vol) NEAA. For differentiation, 50 ng/ml recombinant human macro-phage colony-stimulating factor (M-CSF; R&D Systems) was added to thedishes. After 3 days, medium was changed and growth factors were freshlyadded. On day 7, cells were detached by being rinsed with PBS (PAALaboratories) and seeded at 1 � 104 cells per well into 24-well plates (BDBiosciences). Prior to experiments, macrophages were grown overnight inantibiotic-free medium containing growth factors.

Macrophages were incubated with B. bifidum S17/pVG-mCherry for 2h at a bacteria/cell ratio of 10:1. Afterwards, macrophages were washed 3times with PBS to remove extracellular bacteria and fixed with 4% (wt/vol) PFA in PBS.

Gastrointestinal transit time of B. bifidum S17 in mice and detectionin fecal slurries. All animal experiments were approved by the ethicalcommittee for animal experimentation of the University of Ulm and theresponsible legal authority at the Regierungspräsidium Tübingen. Micewere bred and kept under specific-pathogen-free conditions at the animalfacility at the University of Ulm and were fed a standard laboratory chowand water ad libitum.

C57BL/6J mice (both sexes, 7 to 12 weeks old; n � 6) were inoculatedwith a single dose of 2 � 109 CFU of B. bifidum S17/pMGC-mCherry peranimal. Fecal pellets were collected at the indicated time points (0 to 24 h

after inoculation), weighed, and homogenized in 1 ml of PBS by vigorousvortexing and used as fecal slurries for fluorescence microscopy. Forquantification of fecal carriage of B. bifidum S17/pMGC-mCherry, serialdilutions were prepared and plated on MRSc agar plates containing 200�g ml�1 mupirocin and 5 �g ml�1 chloramphenicol. Agar plates wereincubated anaerobically for 48 h at 37°C. Bifidobacterial counts were de-termined as CFU g�1 feces.

Fluorescence microscopy. For fluorescence microscopy, overnight cul-tures of fluorescent Bifidobacterium strains were washed three times andresuspended in PBS. One drop of an appropriate dilution in PBS wasplaced on a microscope slide and covered with a coverslip. Macrophagesand Caco-2 cells incubated with fluorescent bifidobacteria were fixed with4% (wt/vol) PFA in PBS and stained for 30 min at room temperature inthe dark with Hoechst (Invitrogen Life Technologies) diluted 1:10,000and phalloidin-Alexa 488 (Invitrogen Life Technologies) diluted 1:500 inPBS.

Fluorescence microscopy was performed with a 100� oil immersionor a 40� objective, as indicated in the figure legends, on a Zeiss AxioObserver.Z1 microscope (Carl Zeiss) equipped with the appropriate filtersets to detect fluorescence of Hoechst (excitation at 365 � 10 nm, emis-sion at 445 � 50 nm), cyan fluorescent protein (CFP) (excitation at 436 �20 nm, emission at 480 � 40 nm), GFP and Alexa 488 (excitation at 470 �40 nm, emission at 525 � 50 nm), yellow fluorescent protein (YFP) (ex-citation at 500 nm � 25 nm, emission at 535 nm � 30 nm), or mCherry(excitation at 545 � 30 nm, emission at 610 � 75 nm). Images wereacquired and analyzed using the software ZEN 2012 (Carl Zeiss).

RESULTSIdentification of a suitable promoter for gene expression in bi-fidobacteria. As a first step towards the generation of plasmids forgene expression in bifidobacteria, five different promoters wereassayed for the transcriptional activity using the previously pub-lished pMDY23 promoter probe vector (41). The range of pro-moters tested included the promoter regions upstream of the gapgene (BBIF_0612) encoding glyceraldehyde-3-phosphate dehy-drogenase (Pgap) and the pk gene (BBIF_0789) for fructose-6-phosphate phosphoketolase (Ppk) of B. bifidum S17. Fructose-6-phosphate phosphoketolase is the key enzyme of the bifidus shuntand was thus hypothesized to be highly expressed in bifidobacteriaduring growth on media containing glucose, such as MRSc. ThePgap promoter was selected based on previous reports about hightranscriptional activity and successful use for protein expressionin bifidobacteria (41–43). Promoter regions were retrieved from thepublically available genome sequence of B. bifidum S17 (GenBankaccession no. CP002220) (44). Furthermore, the synthetic pro-moters Phyper, Phelp, and PCP25 were included, which were previ-ously shown to have high transcriptional activity in a number ofGram-positive bacteria (45–47). These promoters were clonedupstream of the gusA reporter gene in pMDY23 (Fig. 1A), trans-formed into B. bifidum S17, and assayed for transcriptional activ-ity by GusA assay during growth of the recombinant strains (Fig.1B). No reporter activity could be measured for PCP25 and Phelp.Pgap showed consistently measurable transcriptional activity ex-cept for the late stationary growth phase (i.e., after 24 h) andyielded the highest reporter activities of all promoters at any timepoint assayed. GusA reporter activity could also be measured forPpk and Phyper but at somewhat lower levels than Pgap and not asconsistently throughout growth. Thus, Pgap was considered an ap-propriate promoter for gene expression in B. bifidum S17.

Expression of fluorescent proteins in bifidobacteria. In orderto identify a bifidobacterial vector replicon with the widest possi-ble host range, we transformed a number of published E. coli/

Grimm et al.

2844 aem.asm.org Applied and Environmental Microbiology

on October 17, 2020 by guest

http://aem.asm

.org/D

ownloaded from

Bifidobacterium shuttle vectors, including pMDY23, pPKCm1,and pDG7, into various Bifidobacterium strains. Of these plas-mids, pMDY23 had the broadest range of bifidobacterial hosts andwas successfully and reproducibly transformed into Bifidobacte-rium adolescentis NCC251, B. bifidum (strains S17 and S16), B.breve S27, B. longum subsp. longum (strains E18 and NCC2705), B.animalis subsp. lactis NCC362, and B. animalis subsp. animalisMB254, i.e., a total of 8 strains of 5 different species (see Table S2in the supplemental material). The replicon of pMDY23 was thuschosen as the backbone of all further plasmids generated in thisstudy.

Based on pMDY23-Pgap, a range of vectors for expression ofdifferent fluorescent proteins was constructed. Four differentplasmids were generated essentially by replacing the gusA reportergene in pMDY23-Pgap with the genes for the fluorescent proteinsCFP, GFP, YFP, and mCherry (Fig. 2A). Plasmids were con-structed as described in the supplemental Materials and Methods.This yielded plasmids pVG-CFP, pVG-GFP, pVG-YFP, and pVG-mCherry, which were successfully transformed into three strainsof bifidobacteria belonging to three different species. Derivativesof B. bifidum S17 containing these different plasmids showed highfluorescence at the expected wavelengths, and virtually all bacteria

FIG 1 (A) Schematic representation of the PCR products of Phyper, Pcp25, Phelp, Ppk, and Pgap (gray bars) cloned into the unique XhoI and BglII restriction sitesof pMDY23. Relevant features of the plasmids are indicated as follows: pMB1 replicon for replication in E. coli by a black bar; repAB for replication inbifidobacteria, the spectinomycin resistance gene (spc), and the gusA reporter gene by gray arrows. (B) Analysis of the transcriptional activity of the clonedpromoters in B. bifidum S17 at the indicated time points during growth in MRSc by measuring glucuronidase activities in crude extracts. Values are means �standard deviations from 3 technical replicates per time point, and data from one representative of three independent cultures per strain are shown.

FIG 2 (A) Schematic representation of the genes encoding different fluorescent proteins cloned into pMDY23-Pgap. Relevant features of the plasmids areindicated as follows: pMB1 replicon for replication in E. coli by a black bar; repAB for replication in bifidobacteria and the spectinomycin resistance gene (spc) bygray arrows; Pgap promoter by a black arrow; genes encoding fluorescent proteins GFP, CFP, YFP, and mCherry by green, blue, yellow, and red arrows,respectively. (B) Fluorescence microscopy of B. bifidum S17 transformed with pVG-CFP, pVG-GFP, pVG-YFP, or pVG-mCherry. (C) Fluorescence microscopyof a mix of the four strains shown in panel B. Images were acquired using a Zeiss Axio Observer.Z1 microscope and a 100� oil immersion objective. The scalebars indicate a size of 10 �m.

Expression of Fluorescent Proteins in Bifidobacteria

May 2014 Volume 80 Number 9 aem.asm.org 2845

on October 17, 2020 by guest

http://aem.asm

.org/D

ownloaded from

were fluorescent (Fig. 2B). Moreover, the spectra of the emittedlight were different enough to distinguish differently labeled B.bifidum S17 strains in a mix (Fig. 2C). Similar results were ob-tained when pVG-CFP, pVG-GFP, pVG-YFP, or pVG-mCherrywere transformed into B. longum subsp. longum E18 (see Fig. S3 inthe supplemental material) or B. breve S27 (see Fig. S4). Collec-tively, these results indicate that these plasmids are useful to labelvarious bifidobacterial strains of different species.

Applications of fluorescent B. bifidum S17. Red fluorescent B.bifidum S17/pVG-mCherry adhering to Caco-2 cells could be vi-sualized by fluorescence microscopy (Fig. 3A). In order to quan-titatively assess adherent bacteria, a calibration curve was estab-lished for B. bifidum S17/pVG-mCherry in the presence of Caco-2cells. This revealed a high degree of correlation of fluorescencewith CFU (Fig. 3B). Using this calibration curve, CFU of B. bifi-dum S17/pVG-mCherry were calculated based on fluorescencemeasurements and compared to actual numbers of bacteria ad-hering to Caco-2 cells as determined by classical plate counting(Fig. 3C). This yielded almost identical results for both methods,indicating that measuring fluorescence gives results that are com-parable to the classical plate counting method.

In a second approach, we sought to check if it is possible toimage fluorescent bifidobacteria inside relevant host cells. For thispurpose, primary human macrophages were generated ex vivo

from monocytes by differentiation for 7 days in the presence ofM-CSF and cocultured with B. bifidum S17/pVG-mCherry. Usingfluorescence microscopy, high numbers of intracellular B. bifidumS17/pVG-mCherry were detected (Fig. 4).

Plasmids for in vivo colonization studies. Initial experimentson gastrointestinal colonization and persistence of bifidobacteriarevealed that mice at the animal facility at the University of Ulmharbor an intrinsic B. animalis strain, which was resistant to spec-tinomycin at 100 �g ml�1, i.e., the concentration generally usedfor selection of pMDY23 derivatives (data not shown). Thisprompted us to generate a range of vectors for in vivo tracingstudies. Based on pMDY23, four plasmids with different antibioticresistance genes and the pBluescript multiple-cloning site (MCS)were generated (see Materials and Methods and Fig. S5A in thesupplemental material). The selective concentrations to the anti-biotics used were tested for a range of bifidobacterial strains fre-quently used in our lab. Using a 96-well microtiter plate assay, theconcentrations that proved to be selective for almost all strainstested were 1, 5, 2, and 100 �g ml�1 for ampicillin, chloramphen-icol, erythromycin, and spectinomycin, respectively (see Table S6in the supplemental material). However, B. animalis subsp. ani-malis MB254 and B. animalis subsp. lactis NCC362 still showedminimal growth at 100 �g ml�1 spectinomycin.

These concentrations were successfully used to select clones ofB. bifidum S17 harboring either pMGA, pMGC, pMGE, or pMGSfollowing electroporation. All plasmids were tested for stability inB. bifidum S17 in the absence of antibiotic pressure. Followingcultivation for 100 generations in the absence of antibiotic, thepercentage of resistant colonies did not drop below 100%, indi-cating high plasmid stability. Transformation of B. bifidum S17with pMGA, pMGS, or pMGE reduced the growth rate and/orfinal OD600 in the presence and/or absence of respective antibiot-ics to various degrees (see Fig. S5B in the supplemental material).No difference was observed for B. bifidum S17/pMGC on MRScwith or without chloramphenicol, and under both conditionsgrowth characteristics of this strain did not differ from those of theWT (see Fig. S5B in the supplemental material).

Thus, B. bifidum S17/pMGC was selected for in vivo studies. To

FIG 3 (A) Fluorescence microscopy of adherent B. bifidum S17/pVG-mCherry on a confluent monolayer of Caco-2 cells. Nuclei of Caco-2 cells werestained with Hoechst (blue). Images were acquired using a Zeiss Axio Ob-server.Z1 microscope and a 40� objective. The scale bar indicates a size of 20�m. (B) Relative fluorescence units (RFU) of B. bifidum S17/pVG-mCherry ona confluent monolayer of Caco-2 cells plotted against CFU per well. The equa-tion of the calculated linear regression and the Pearson’s coefficient of corre-lation (R2) are indicated. (C) CFU per well of adherent B. bifidum S17/pVG-mCherry as quantified directly using classical plate counting (plated) andcalculated from measured RFU using the equation of the linear regressionshown in panel B. Values in panel B and C are means � standard deviationsfrom three independent cultures with both CFU and RFU determined in tech-nical duplicates.

FIG 4 Fluorescence microscopy of B. bifidum S17/pVG-mCherry inside pri-mary human macrophages. Nuclei of macrophages were stained with Hoechst(blue), and the actin cytoskeleton was stained with phalloidin-Alexa 488(green). Images were acquired using a Zeiss Axio Observer.Z1 microscope anda 40� objective. The scale bar indicates a size of 20 �m.

Grimm et al.

2846 aem.asm.org Applied and Environmental Microbiology

on October 17, 2020 by guest

http://aem.asm

.org/D

ownloaded from

combine the advantages of antibiotic selection with expression offluorescent proteins, pMGC-mCherry (see Fig. S7 in the supple-mental material) was generated by cloning the Pgap-mCherry con-struct of pVG-mCherry into the MCS of pMGC as described inMaterials and Methods in the supplemental material. The result-ing plasmid was successfully transformed into B. bifidum S17,yielding B. bifidum S17/pMGC-mCherry, which was then used todetermine gastrointestinal transit time in mice.

For this purpose, female C57BL/6J mice (n � 6) were inocu-lated with B. bifidum S17/pMGC-mCherry by a single oral dose of2 � 109 CFU per animal. Following inoculation, the numbers ofchloramphenicol-resistant bifidobacteria were enumerated in fe-cal pellets by plate counting on selective agar (Fig. 5A). This re-vealed that fecal shedding of B. bifidum S17/pMGC-mCherrypeaked at around 5 h postinoculation, when approximately 1 �108 CFU g�1 feces was counted. Thereafter, counts continuouslydecreased until 24 h, when levels had dropped below the limit ofdetection, i.e., 1 � 103 CFU g�1 feces, in 4 out of 6 animals. More-over, we were able to detect fluorescent B. bifidum S17/pMGC-mCherry in fecal slurries of samples collected 5 h after inoculation(Fig. 5B).

DISCUSSION

With the present study, we aimed at extending the range of avail-able plasmid-based molecular tools for protein expression in andfunctional analysis of bifidobacteria. By making use of the originalpurpose of pMDY23 as a promoter probe vector, a number ofpromoters were analyzed for transcriptional activity. No GusAreporter activity could be detected for the synthetic promotersPCP25 and Phelp. This is somewhat surprising, since both promot-ers were shown to drive high levels of reporter gene expression inother Gram-positive and/or Gram-negative bacteria (45, 46, 48)and Phelp was successfully used to generate a luminescent B. breve

strain (22). Phyper was successfully used for recombinant proteinexpression in Listeria monocytogenes (47, 49). Similarly, reason-able transcriptional activity was observed for Phyper in B. bifidumS17 but only at early time points during growth. The promotershowing the highest levels of GusA reporter activity during thefirst 12 h of growth was Pgap. This is in line with the observationthat the gap promoter is highly active in B. longum NCC2705 (41).Consequently, the gap promoter of B. longum was successfullyused for recombinant protein expression (42, 43). However, theabsence of reporter activity at later stages of growth, i.e., in sta-tionary phase, suggests that its use is limited to actively growingcultures.

The pMDY23 backbone was chosen for the generation of fur-ther plasmids based on successful transformation into a total of 15strains belonging to 7 different species with 4 subspecies (see TableS2 in the supplemental material) (39, 50, 51). To our knowledge,this represents the widest range of bifidobacterial hosts for any E.coli/Bifidobacterium shuttle vector described so far (2, 21, 52).

Using the Pgap promoter and the backbone of pMDY23, a seriesof plasmids were constructed, which allowed expression of fourdifferent FPs in B. bifidum S17, B. breve S27, and B. longum E18. Tothe best of our knowledge, this represents the first report on thesuccessful tagging of bifidobacterial strains by four different FPs.As observed previously for L. monocytogenes (47), the excitationand emission spectra of the tagged strains were sufficiently differ-ent to distinguish four strains tagged with different constructs byfluorescence microscopy.

Adhesion to the gastrointestinal mucosa is thought to helpprobiotic bacteria to colonize and persist at least transiently in thegastrointestinal tract of the host (53). Thus, adhesion to intestinalepithelial cells or mucus is one of the properties frequently testedfor potential probiotic strains (6, 54–56). Until now, most studieson adhesion of bifidobacteria to cultured epithelial cells have usedmethods involving bacteria labeled by radioisotopes or by classicalplate counting (6, 39, 57–59). The results obtained in this studydemonstrate that the use of fluorescent bifidobacteria in adhesionassays is a promising application yielding results comparable tothose obtained with classical assays, e.g., by plate counting orusing radioactively labeled bacteria. Compared to these classi-cal methods, the use of fluorescent bacteria to determine adhe-sion saves consumables since no further agar plates, materialsfor dilutions series, or expensive and elaborate scintillationequipment are required. Moreover, the results are availableimmediately after the measurement, saving up to 48 h requiredfor growth of colonies.

Fluorescent bifidobacteria were also successfully detected inhuman primary macrophages generated by ex vivo differentiationof monocytes. This demonstrates that detection of bifidobacteriainside relevant host cell populations is possible. Using similar sys-tems, we and others have successfully visualized fluorescentstrains of C. rodentium, Salmonella enterica serovar Typhimu-rium, E. coli, and L. monocytogenes within host tissues and identi-fied specific host cell populations that are associated with or havephagocytosed these labeled strains (60–64).

Cronin and colleagues have created an mCherry-expressing B.breve strain and were able to detect this strain in subcutaneoustumors in athymic mice by fluorescence microscopy followingintravenous application (24). The same strain was tagged using aluciferase and was imaged in the colons and tumors of live mice orin dissected GITs of colonized animals following oral or intrave-

FIG 5 (A) Fecal shedding of B. bifidum S17/pMGC-mCherry during the first24 h after inoculation of C57BL/6J mice (n � 6) with a single oral dose of 2 �109 CFU per animal; (B) detection of B. bifidum S17/pMGC-mCherry by flu-orescence microscopy of fecal slurries prepared from a fresh fecal pellet sam-pled 5 h postinoculation. Images were acquired using a Zeiss Axio Observer.Z1microscope and a 100� oil immersion objective. The scale bars indicate a sizeof 10 �m.

Expression of Fluorescent Proteins in Bifidobacteria

May 2014 Volume 80 Number 9 aem.asm.org 2847

on October 17, 2020 by guest

http://aem.asm

.org/D

ownloaded from

nous inoculation (22–24). Luminescence of this strain correlatedto actual bacterial numbers as determined by plate counting (23,24). Similarly, fluorescence has been used to image bacteria andprotozoan parasites in vivo (65–68). Thus, in theory, imaging offluorescent bifidobacteria in live animals or dissected organsshould be possible. However, bioluminescent reporters are by farmore frequently used for in vivo imaging of bacteria due to limi-tations in tissue penetration of fluorescence (69).

Using the chloramphenicol resistance marker of pMGC-mCherry, we determined the gastrointestinal transit time of B.bifidum S17/pMGC-mCherry in mice after inoculation by a singleoral dose. While these results suggest that this strain is not able tocolonize mice stably in the presence of a normal microbiota, theydemonstrate the general applicability of the plasmid to monitorgastrointestinal transit and colonization of bifidobacteria. More-over, the detection of B. bifidum S17/pMGC-mCherry in fecalslurries following oral administration is a third example for po-tential applications of fluorescent bifidobacteria. One importantpiece of information that can be gathered from these experimentsis that at least a portion of the administered bifidobacteria are ableto transit through the GIT and are shed with feces in a viable state.

Collectively, these results suggest that tagging bifidobacteria byexpression of FPs is a valuable tool to study the interaction ofpotential probiotic strains with the host. Moreover, the successfulexpression of FPs also demonstrates the feasibility of the expres-sion system consisting of the pMDY23 backbone and the Pgap

promoter. Using homologous expression in bifidobacteria willhelp to further understand the role of single genes or proteins inhost colonization and the beneficial effects of bifidobacteria. Like-wise, bifidobacteria can be further optimized by heterologous ex-pression of factors known to confer abilities such as host coloni-zation or adhesion to host structures to other bacteria. Since thepMDY23 backbone was successfully transformed in at least 7 dif-ferent Bifidobacterium spp., these tools will be available for a widerange of bifidobacterial strains.

ACKNOWLEDGMENTS

V.G., D.Z., and C.U.R. are funded by the German Academic ExchangeService/Federal Ministry of Education and Research (grant D/09/04778).M.G. was supported by a Ph.D. fellowship of the Carl Zeiss Foundation.

REFERENCES1. Ventura M, Canchaya C, Tauch A, Chandra G, Fitzgerald GF, Chater

KF, van Sinderen D. 2007. Genomics of Actinobacteria: tracing the evo-lutionary history of an ancient phylum. Microbiol. Mol. Biol. Rev. 71:495–548. http://dx.doi.org/10.1128/MMBR.00005-07.

2. Lee J-H, O’Sullivan DJ. 2010. Genomic insights into bifidobacteria. Mi-crobiol. Mol. Biol. Rev. 74:378 – 416. http://dx.doi.org/10.1128/MMBR.00004-10.

3. Leahy SC, Higgins DG, Fitzgerald GF, van Sinderen D. 2005. Gettingbetter with bifidobacteria. J. Appl. Microbiol. 98:1303–1315. http://dx.doi.org/10.1111/j.1365-2672.2005.02600.x.

4. Picard C, Fioramonti J, Francois A, Robinson T, Neant F, MatuchanskyC. 2005. Review article: bifidobacteria as probiotic agents—physiologicaleffects and clinical benefits. Aliment. Pharmacol. Ther. 22:495–512. http://dx.doi.org/10.1111/j.1365-2036.2005.02615.x.

5. Riedel C-U, Foata F, Philippe D, Adolfsson O, Eikmanns B-J, Blum S.2006. Anti-inflammatory effects of bifidobacteria by inhibition of LPS-induced NF-kappaB activation. World J. Gastroenterol. 12:3729 –3735.http://www.wjgnet.com/1007-9327/full/v12/i23/3729.htm.

6. Preising J, Philippe D, Gleinser M, Wei H, Blum S, Eikmanns BJ, NiessJ-H, Riedel CU. 2010. Selection of bifidobacteria based on adhesion andanti-inflammatory capacity in vitro for amelioration of murine colitis.

Appl. Environ. Microbiol. 76:3048 –3051. http://dx.doi.org/10.1128/AEM.03127-09.

7. Philippe D, Heupel E, Blum-Sperisen S, Riedel CU. 2011. Treatmentwith Bifidobacterium bifidum 17 partially protects mice from Th1-driveninflammation in a chemically induced model of colitis. Int. J. Food Mi-crobiol. 149:45– 49. http://dx.doi.org/10.1016/j.ijfoodmicro.2010.12.020.

8. Veiga P, Gallini CA, Beal C, Michaud M, Delaney ML, DuBois A,Khlebnikov A, van Hylckama Vlieg JET, Punit S, Glickman JN, Onder-donk A, Glimcher LH, Garrett WS. 2010. Bifidobacterium animalissubsp. lactis fermented milk product reduces inflammation by altering aniche for colitogenic microbes. Proc. Natl. Acad. Sci. U. S. A. 107:18132–18137. http://dx.doi.org/10.1073/pnas.1011737107.

9. Di Giacinto C, Marinaro M, Sanchez M, Strober W, Boirivant M. 2005.Probiotics ameliorate recurrent Th1-mediated murine colitis by inducingIL-10 and IL-10-dependent TGF-beta-bearing regulatory cells. J. Immu-nol. 174:3237–3246.

10. Jeon SG, Kayama H, Ueda Y, Takahashi T, Asahara T, Tsuji H, TsujiNM, Kiyono H, Ma JS, Kusu T, Okumura R, Hara H, Yoshida H,Yamamoto M, Nomoto K, Takeda K. 2012. Probiotic Bifidobacteriumbreve induces IL-10-producing Tr1 cells in the colon. PLoS Pathog.8:e1002714. http://dx.doi.org/10.1371/journal.ppat.1002714.

11. Fukuda S, Toh H, Hase K, Oshima K, Nakanishi Y, Yoshimura K, TobeT, Clarke JM, Topping DL, Suzuki T, Taylor TD, Itoh K, Kikuchi J,Morita H, Hattori M, Ohno H. 2011. Bifidobacteria can protect fromenteropathogenic infection through production of acetate. Nature 469:543–547. http://dx.doi.org/10.1038/nature09646.

12. Fanning S, Hall LJ, Cronin M, Zomer A, Macsharry J, Goulding D,O’Connell Motherway M, Shanahan F, Nally K, Dougan G, van Sin-deren D. 2012. Bifidobacterial surface-exopolysaccharide facilitates com-mensal-host interaction through immune modulation and pathogen pro-tection. Proc. Natl. Acad. Sci. U. S. A. 109:2108 –2113. http://dx.doi.org/10.1073/pnas.1115621109.

13. Asahara T, Shimizu K, Nomoto K, Hamabata T, Ozawa A, Takeda Y.2004. Probiotic bifidobacteria protect mice from lethal infection withShiga toxin-producing Escherichia coli O157:H7. Infect. Immun. 72:2240 –2247. http://dx.doi.org/10.1128/IAI.72.4.2240-2247.2004.

14. Westermann C, Zhurina DS, Baur A, Shang W, Yuan J, Riedel CU.2012. Exploring the genome sequence of Bifidobacterium bifidum S17 forpotential players in host-microbe interactions. Symbiosis 58:191–200.http://dx.doi.org/10.1007/s13199-012-0205-z.

15. Foroni E, Serafini F, Amidani D, Turroni F, He F, Bottacini F,O’Connell Motherway M, Viappiani A, Zhang Z, Rivetti C, van Sin-deren D, Ventura M. 2011. Genetic analysis and morphological identifi-cation of pilus-like structures in members of the genus Bifidobacterium.Microb. Cell Fact. 10(Suppl 1):S16. http://dx.doi.org/10.1186/1475-2859-10-S1-S16.

16. Turroni F, Serafini F, Foroni E, Duranti S, O’Connell Motherway M,Taverniti V, Mangifesta M, Milani C, Viappiani A, Roversi T, Sánchez B,Santoni A, Gioiosa L, Ferrarini A, Delledonne M, Margolles A, Piazza L,Palanza P, Bolchi A, Guglielmetti S, van Sinderen D, Ventura M. 2013. Roleof sortase-dependent pili of Bifidobacterium bifidum PRL2010 in modulat-ing bacterium-host interactions. Proc. Natl. Acad. Sci. U. S. A. 110:11151–11156. http://dx.doi.org/10.1073/pnas.1303897110.

17. O’Connell Motherway M, Zomer A, Leahy SC, Reunanen J, Bottacini F,Claesson MJ, O’Brien F, Flynn K, Casey PG, Munoz JAM, Kearney B,Houston AM, O’Mahony C, Higgins DG, Shanahan F, Palva A, de VosWM, Fitzgerald GF, Ventura M, O’Toole PW, van Sinderen D. 2011.Functional genome analysis of Bifidobacterium breve UCC2003 revealstype IVb tight adherence (Tad) pili as an essential and conserved host-colonization factor. Proc. Natl. Acad. Sci. U. S. A. 108:11217–11222. http://dx.doi.org/10.1073/pnas.1105380108.

18. Anonymous. 2013. Culture shock. Nature 493:133–134. http://dx.doi.org/10.1038/493133b.

19. Brancaccio VF, Zhurina DS, Riedel CU. 2013. Tough nuts to crack:site-directed mutagenesis of bifidobacteria remains a challenge. Bioengi-neered 4:197–202. http://dx.doi.org/10.4161/bioe.23381.

20. Ruiz L, Motherway MO, Lanigan N, van Sinderen D. 2013. Transposonmutagenesis in Bifidobacterium breve: construction and characterizationof a Tn5 transposon mutant library for Bifidobacterium breve UCC2003.PLoS One 8:e64699. http://dx.doi.org/10.1371/journal.pone.0064699.

21. Sun Z, Baur A, Zhurina D, Yuan J, Riedel CU. 2012. Accessing theinaccessible: molecular tools for bifidobacteria. Appl. Environ. Microbiol.78:5035–5042. http://dx.doi.org/10.1128/AEM.00551-12.

Grimm et al.

2848 aem.asm.org Applied and Environmental Microbiology

on October 17, 2020 by guest

http://aem.asm

.org/D

ownloaded from

22. Cronin M, Sleator RD, Hill C, Fitzgerald GF, van Sinderen D. 2008.Development of a luciferase-based reporter system to monitor Bifidobac-terium breve UCC2003 persistence in mice. BMC Microbiol. 8:161. http://dx.doi.org/10.1186/1471-2180-8-161.

23. Cronin M, Morrissey D, Rajendran S, El Mashad SM, van Sinderen D,O’Sullivan GC, Tangney M. 2010. Orally administered bifidobacteria asvehicles for delivery of agents to systemic tumors. Mol. Ther. J. Am. Soc.Gene Ther. 18:1397–1407. http://dx.doi.org/10.1038/mt.2010.59.

24. Cronin M, Akin AR, Collins SA, Meganck J, Kim J-B, Baban CK, JoyceSA, van Dam GM, Zhang N, van Sinderen D, O’Sullivan GC, KasaharaN, Gahan CG, Francis KP, Tangney M. 2012. High resolution in vivobioluminescent imaging for the study of bacterial tumour targeting. PLoSOne 7:e30940. http://dx.doi.org/10.1371/journal.pone.0030940.

25. Shimomura O, Johnson FH, Saiga Y. 1962. Extraction, purification andproperties of aequorin, a bioluminescent protein from the luminous hy-dromedusan, Aequorea. J. Cell. Comp. Physiol. 59:223–239. http://dx.doi.org/10.1002/jcp.1030590302.

26. Zhang J, Campbell RE, Ting AY, Tsien RY. 2002. Creating new fluores-cent probes for cell biology. Nat. Rev. Mol. Cell Biol. 3:906 –918. http://dx.doi.org/10.1038/nrm976.

27. Shaner NC, Steinbach PA, Tsien RY. 2005. A guide to choosing fluorescentproteins. Nat. Methods 2:905–909. http://dx.doi.org/10.1038/nmeth819.

28. Giepmans BNG, Adams SR, Ellisman MH, Tsien RY. 2006. The fluo-rescent toolbox for assessing protein location and function. Science 312:217–224. http://dx.doi.org/10.1126/science.1124618.

29. Lissemore JL, Jankowski JT, Thomas CB, Mascotti DP, de Haseth PL.2000. Green fluorescent protein as a quantitative reporter of relative pro-moter activity in E. coli. Biotechniques 28:82– 89.

30. Southward CM, Surette MG. 2002. The dynamic microbe: green fluores-cent protein brings bacteria to light. Mol. Microbiol. 45:1191–1196. http://dx.doi.org/10.1046/j.1365-2958.2002.03089.x.

31. Bongaerts RJM, Hautefort I, Sidebotham JM, Hinton JCD. 2002. Greenfluorescent protein as a marker for conditional gene expression in bacte-rial cells. Methods Enzymol. 358:43– 66. http://dx.doi.org/10.1016/S0076-6879(02)58080-0.

32. Larrainzar E, O’Gara F, Morrissey JP. 2005. Applications of autofluo-rescent proteins for in situ studies in microbial ecology. Annu. Rev. Mi-crobiol. 59:257–277. http://dx.doi.org/10.1146/annurev.micro.59.030804.121350.

33. Coombes JL, Robey EA. 2010. Dynamic imaging of host-pathogen inter-actions in vivo. Nat. Rev. Immunol. 10:353–364. http://dx.doi.org/10.1038/nri2746.

34. Konjufca V, Miller MJ. 2009. Two-photon microscopy of host-pathogeninteractions: acquiring a dynamic picture of infection in vivo. Cell. Micro-biol. 11:551–559. http://dx.doi.org/10.1111/j.1462-5822.2009.01289.x.

35. Campisi L, Brau F, Glaichenhaus N. 2008. Imaging host-pathogen in-teractions. Immunol. Rev. 221:188 –199. http://dx.doi.org/10.1111/j.1600-065X.2008.00586.x.

36. Bumann D, Valdivia RH. 2007. Identification of host-induced pathogengenes by differential fluorescence induction reporter systems. Nat. Protoc.2:770 –777. http://dx.doi.org/10.1038/nprot.2007.78.

37. Cronin M, Knobel M, O’Connell-Motherway M, Fitzgerald GF, vanSinderen D. 2007. Molecular dissection of a bifidobacterial replicon.Appl. Environ. Microbiol. 73:7858 –7866. http://dx.doi.org/10.1128/AEM.01630-07.

38. Wickes BL, Edman JC. 1995. The Cryptococcus neoformans GAL7 geneand its use as an inducible promoter. Mol. Microbiol. 16:1099 –1109. http://dx.doi.org/10.1111/j.1365-2958.1995.tb02335.x.

39. Gleinser M, Grimm V, Zhurina D, Yuan J, Riedel CU. 2012. Improvedadhesive properties of recombinant bifidobacteria expressing the Bifido-bacterium bifidum-specific lipoprotein BopA. Microb. Cell Fact. 11:80.http://dx.doi.org/10.1186/1475-2859-11-80.

40. Neu C, Sedlag A, Bayer C, Förster S, Crauwels P, Niess J-H, vanZandbergen G, Frascaroli G, Riedel CU. 2013. CD14-dependent mono-cyte isolation enhances phagocytosis of Listeria monocytogenes by proin-flammatory, GM-CSF-derived macrophages. PLoS One 8:e66898. http://dx.doi.org/10.1371/journal.pone.0066898.

41. Klijn A, Moine D, Delley M, Mercenier A, Arigoni F, Pridmore RD.2006. Construction of a reporter vector for the analysis of Bifidobacteriumlongum promoters. Appl. Environ. Microbiol. 72:7401–7405. http://dx.doi.org/10.1128/AEM.01611-06.

42. Khokhlova EV, Efimov BA, Kafarskaia LI, Shkoporov AN. 2010. Het-erologous expression of secreted biologically active human interleukin-10

in Bifidobacterium breve. Arch. Microbiol. 192:769 –774. http://dx.doi.org/10.1007/s00203-010-0606-4.

43. Shkoporov AN, Efimov BA, Khokhlova EV, Kafarskaia LI, SmeianovVV. 2008. Production of human basic fibroblast growth factor (FGF-2) inBifidobacterium breve using a series of novel expression/secretion vectors.Biotechnol. Lett. 30:1983–1988. http://dx.doi.org/10.1007/s10529-008-9772-8.

44. Zhurina D, Zomer A, Gleinser M, Brancaccio VF, Auchter M, Waid-mann MS, Westermann C, van Sinderen D, Riedel CU. 2011. Completegenome sequence of Bifidobacterium bifidum S17. J. Bacteriol. 193:301–302. http://dx.doi.org/10.1128/JB.01180-10.

45. Riedel CU, Monk IR, Casey PG, Morrissey D, O’Sullivan GC, TangneyM, Hill C, Gahan CGM. 2007. Improved luciferase tagging system forListeria monocytogenes allows real-time monitoring in vivo and in vitro.Appl. Environ. Microbiol. 73:3091–3094. http://dx.doi.org/10.1128/AEM.02940-06.

46. Jensen PR, Hammer K. 1998. The sequence of spacers between the con-sensus sequences modulates the strength of prokaryotic promoters. Appl.Environ. Microbiol. 64:82– 87.

47. Balestrino D, Hamon MA, Dortet L, Nahori M-A, Pizarro-Cerda J,Alignani D, Dussurget O, Cossart P, Toledo-Arana A. 2010. Single-celltechniques using chromosomally tagged fluorescent bacteria to study Lis-teria monocytogenes infection processes. Appl. Environ. Microbiol. 76:3625–3636. http://dx.doi.org/10.1128/AEM.02612-09.

48. Riedel CU, Casey PG, Mulcahy H, O’Gara F, Gahan CGM, Hill C. 2007.Construction of p16Slux, a novel vector for improved bioluminescentlabeling of Gram-negative bacteria. Appl. Environ. Microbiol. 73:7092–7095. http://dx.doi.org/10.1128/AEM.01394-07.

49. Shen A, Higgins DE. 2005. The 5= untranslated region-mediated en-hancement of intracellular listeriolysin O production is required for Lis-teria monocytogenes pathogenicity. Mol. Microbiol. 57:1460 –1473. http://dx.doi.org/10.1111/j.1365-2958.2005.04780.x.

50. Berger B, Moine D, Mansourian R, Arigoni F. 2010. HspR mutations arenaturally selected in Bifidobacterium longum when successive heat shocktreatments are applied. J. Bacteriol. 192:256 –263. http://dx.doi.org/10.1128/JB.01147-09.

51. Ruiz L, Alvarez-Martín P, Mayo B, de Los Reyes-Gavilán CG, Guei-monde M, Margolles A. 2012. Controlled gene expression in bifidobac-teria by use of a bile-responsive element. Appl. Environ. Microbiol. 78:581–585. http://dx.doi.org/10.1128/AEM.06611-11.

52. Rossi M, Brigidi P, Matteuzzi D. 1998. Improved cloning vectors forBifidobacterium spp. Lett. Appl. Microbiol. 26:101–104. http://dx.doi.org/10.1046/j.1472-765X.1998.00285.x.

53. González-Rodríguez I, Ruiz L, Gueimonde M, Margolles A, Sánchez B.2013. Factors involved in the colonization and survival of bifidobacteria inthe gastrointestinal tract. FEMS Microbiol. Lett. 340:1–10. http://dx.doi.org/10.1111/1574-6968.12056.

54. Laparra JM, Sanz Y. 2009. Comparison of in vitro models to study bac-terial adhesion to the intestinal epithelium. Lett. Appl. Microbiol. 49:695–701. http://dx.doi.org/10.1111/j.1472-765X.2009.02729.x.

55. Tuomola EM, Salminen SJ. 1998. Adhesion of some probiotic and dairyLactobacillus strains to Caco-2 cell cultures. Int. J. Food Microbiol. 41:45–51. http://dx.doi.org/10.1016/S0168-1605(98)00033-6.

56. Vélez MP, De Keersmaecker SCJ, Vanderleyden J. 2007. Adherence factorsof Lactobacillus in the human gastrointestinal tract. FEMS Microbiol. Lett.276:140 –148. http://dx.doi.org/10.1111/j.1574-6968.2007.00908.x.

57. Bernet MF, Brassart D, Neeser JR, Servin AL. 1993. Adhesion of humanbifidobacterial strains to cultured human intestinal epithelial cells andinhibition of enteropathogen-cell interactions. Appl. Environ. Microbiol.59:4121– 4128.

58. Riedel CU, Foata F, Goldstein DR, Blum S, Eikmanns BJ. 2006. Inter-action of bifidobacteria with Caco-2 cells-adhesion and impact on expres-sion profiles. Int. J. Food Microbiol. 110:62– 68. http://dx.doi.org/10.1016/j.ijfoodmicro.2006.01.040.

59. Vesterlund S, Paltta J, Karp M, Ouwehand AC. 2005. Measurement ofbacterial adhesion-in vitro evaluation of different methods. J. Microbiol.Methods 60:225–233. http://dx.doi.org/10.1016/j.mimet.2004.09.013.

60. Niess JH, Brand S, Gu X, Landsman L, Jung S, McCormick BA, Vyas JM,Boes M, Ploegh HL, Fox JG, Littman DR, Reinecker H-C. 2005. CX3CR1-mediated dendritic cell access to the intestinal lumen and bacterial clearance.Science 307:254–258. http://dx.doi.org/10.1126/science.1102901.

61. Manta C, Heupel E, Radulovic K, Rossini V, Garbi N, Riedel CU, NiessJH. 2013. CX(3)CR1(�) macrophages support IL-22 production by in-

Expression of Fluorescent Proteins in Bifidobacteria

May 2014 Volume 80 Number 9 aem.asm.org 2849

on October 17, 2020 by guest

http://aem.asm

.org/D

ownloaded from

nate lymphoid cells during infection with Citrobacter rodentium. MucosalImmunol. 6:177–188. http://dx.doi.org/10.1038/mi.2012.61.

62. Rossini V, Zhurina D, Radulovic K, Manta C, Walther P, Riedel CU,Niess JH. 16 October 2013. CX3CR1(�) cells facilitate the activation ofCD4 T cells in the colonic lamina propria during antigen-driven colitis.Mucosal Immunol. http://dx.doi.org/10.1038/mi.2013.70.

63. Nikitas G, Deschamps C, Disson O, Niault T, Cossart P, Lecuit M. 2011.Transcytosis of Listeria monocytogenes across the intestinal barrier uponspecific targeting of goblet cell accessible E-cadherin. J. Exp. Med. 208:2263–2277. http://dx.doi.org/10.1084/jem.20110560.

64. Rescigno M, Urbano M, Valzasina B, Francolini M, Rotta G, Bonasio R,Granucci F, Kraehenbuhl JP, Ricciardi-Castagnoli P. 2001. Dendritic cellsexpress tight junction proteins and penetrate gut epithelial monolayers tosample bacteria. Nat. Immunol. 2:361–367. http://dx.doi.org/10.1038/86373.

65. Zhao M, Yang M, Baranov E, Wang X, Penman S, Moossa AR, Hoff-man RM. 2001. Spatial-temporal imaging of bacterial infection and anti-biotic response in intact animals. Proc. Natl. Acad. Sci. U. S. A. 98:9814 –9818. http://dx.doi.org/10.1073/pnas.161275798.

66. Zhao M, Geller J, Ma H, Yang M, Penman S, Hoffman RM. 2007.Monotherapy with a tumor-targeting mutant of Salmonella Typhimu-rium cures orthotopic metastatic mouse models of human prostate can-cer. Proc. Natl. Acad. Sci. U. S. A. 104:10170 –10174. http://dx.doi.org/10.1073/pnas.0703867104.

67. Zhao M, Yang M, Li X-M, Jiang P, Baranov E, Li S, Xu M, PenmanS, Hoffman RM. 2005. Tumor-targeting bacterial therapy with aminoacid auxotrophs of GFP-expressing Salmonella Typhimurium. Proc.Natl. Acad. Sci. U. S. A. 102:755–760. http://dx.doi.org/10.1073/pnas.0408422102.

68. Mehta SR, Huang R, Yang M, Zhang X-Q, Kolli B, Chang K-P, Hoff-man RM, Goto Y, Badaro R, Schooley RT. 2008. Real-time in vivo greenfluorescent protein imaging of a murine leishmaniasis model as a new toolfor Leishmania vaccine and drug discovery. Clin. Vaccine Immunol. 15:1764 –1770. http://dx.doi.org/10.1128/CVI.00270-08.

69. Andreu N, Zelmer A, Wiles S. 2011. Noninvasive biophotonic imagingfor studies of infectious disease. FEMS Microbiol. Rev. 35:360 –394. http://dx.doi.org/10.1111/j.1574-6976.2010.00252.x.

Grimm et al.

2850 aem.asm.org Applied and Environmental Microbiology

on October 17, 2020 by guest

http://aem.asm

.org/D

ownloaded from