Ethics statement

This study was carried out in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. All animal work was conducted under the animal protocol No. 71557 and was approved by the University of Chicago Institutional Animal Care and Use Committee (IACUC). Cesarean-section was performed under isoflurane anesthesia and all efforts were made to minimize suffering. If a rat pup showed illness during the course of the study, the animal was humanely euthanized and death was not used as an endpoint.

Preparation of bacterial conditioned media

The Lactobacillus plantarum, Lactobacillus acidophilus, and Bifidobacteria infantis used were from ATCC (No. 14917, 53544 and 15697, respectively). Bacteria were first grown and expanded in MRS broth (DeMan Rogosa & Sharpe, Difco) at 37°C and 5% CO2 under anaerobic and non-agitating conditions, centrifuged (20 min, 4,500×g), and resuspended in MRS (L. plantarum) or modified Hank's balanced saline solution (HBSS) (L. acidophilus and B. infantis) supplemented with 0.04 M MgSO4, 0.03 M MnSO4, 1.15 M K2PO4, 0.36 M sodium acetate, 0.88 M ammonium citrate, 10% polysorbate (growth factor for Lactobacillus sp) and 20% dextrose. Bacteria then were propagated overnight at 37°C, 5% CO2 under nonagitating conditions to 2×109 cfu/mL. The culture was again centrifuged and the supernatant was aseptically filtered using 0.22 µm low protein binding cellulose acetate filters (Millipore) and used as conditioned media.

Neonatal rat experimental intestinal injury model

All animal studies were reviewed and approved by the Institutional Animal Care and Use Committee. Animal experiments were conducted following a well described rat pup experimental intestinal injury animal model [36]. In this model, pups were subjected to major risk factors for human NEC (prematurity, formula feeding, bacterial colonization, and hypoxia-ischemia), resulting in changes in the immature rat pups similar to those found in humans as follows: the abdomen was distended, blood was detected in the stool, and the ileum and proximal colon were the most affected parts of the intestine [37].

Neonatal rats from time-dated pregnant Sprague-Dawley dams were delivered by cesarean section at E20 following isoflurane anesthesia. Pups were then stabilized, dried, and maintained in a humidified incubator at 37°C and bowel/bladder function was stimulated by a soft cotton-tip applicator. Pups were fed with Esbilac puppy formula every 3 hours via an orogastric feeding catheter and colonized with 10⁷ CFU of Serratia marcescens, Klebsiella pneumoniae, and Streptococci viridans in 100 µl formula once daily. Pups were also stressed under 5% O₂+95% N₂ for 10 min after feeding three times a day. The feeding volume began at 0.1 ml and was increased incrementally up to 0.25 ml. Naturally born rat pups fed by dams were included as healthy controls. For conditioned medium (CM) groups, CM from Lp, La and Bi were used in different combinations (Lp, La/Bi, or La/Bi/Lp CM) and compared to the experimental control category of vehicle alone. Fifty µl CM was administered once daily 3 hours before administration of Serratia marcescens, Klebsiella pneumoniae, Streptococci viridans. Formula (regular volume minus 50 µl) was fed immediately after CM administration.

Animals were sacrificed when ill or at the end of the experiment on day 5. The health of animals was evaluated by morbidity scale: score 0: active, pink color, no abdominal distension or discoloration, score 1: lethargic, pink color, no abdominal distension or discoloration, score 2: not active, facial cyanosis, slight abdominal distension, score 3: not active, facial and peripheral cyanosis, abdominal distension with or without abdominal discoloration, normal stools, score 4: not active, marked cyanosis, abdominal distension, abdominal discoloration, bloody stools, score 5: death. Animals that reached score 3 were considered ill and sacrificed. The intestines were collected and fixed in 10% buffered formalin overnight for tissue section preparation. H&E-stained intestinal sections were assessed histologically for ileal damage by a pathologist blinded to treatment groups using a published scoring system to evaluate the degree of intestinal injury on a “0–4” scale as follows: 0 no histological damage; 1 (mild), slight submucosal and/or lamina propria separation; 2 (moderate), moderate separation of the submucosa and/or lamina propria and/or edema in the submucosa and muscular layers; 3 (severe), severe separation of the submucosa and/or lamina propria and/or severe edema in the submucosa and muscular layers with regional villous sloughing; and 4 (necrosis), loss of villi and necrosis. Scores ≥2 were defined as NEC-like intestinal injury [36].

The NEC-like intestinal injury incidence of the animal model performed in our animal facility has been consistent and similar to that of other groups using this model [38], [39]. Based on power and sample size calculation using Stat software to compare two proportions (percentages), 45 animals in each group are needed to document a 30% decrease in the incidence of the disease with an α of 0.05 and a power of 80%. Similar or higher numbers of animals were included in each group to test the efficacy of CM treatment.

Ex vivo intestinal barrier function assay

To assess barrier integrity of different intestinal regions, a previously described ex vivo intestinal loop assay was performed [40], [41]. Explants of ileum (right above and next to cecum) and jejunum/duodenum (at least 5 cm away from ileum) were harvested from experimental rat pups. Explanted intestinal segments were flushed with dialyzed 10 kDa FITC-dextran (1 mg/mL in PBS, Sigma Aldrich), ligated with sutures at one end, filled with the same FITC-dextran solution, and ligated at the other end. Explanted loops were rinsed twice in PBS, transferred into wells containing 0.6 mL PBS, and left at RT. One hundred and forty µl PBS from the wells were sampled at 30, 60, and 90 minutes after explants were placed in PBS and added to 96-well plates in duplicate (60 µl/well) to determine FITC-dextran flux across the loops. Fluorescence intensity (excitation at 490 nm and emission at 525 nm) was measured using a Synergy-2 fluorometer (Bio-Tek Instruments, Winooski, VT). FITC-dextran flux/concentration (µg/ml) was calculated with standard curves and then normalized to intestinal length (cm). Barrier integrity of paired ileum and jejunum/duodenum from same animals (n = 5) at all time points was measured and compared to ileum.

In vivo intestinal barrier function assay

To investigate intestinal lumen-to-blood permeability/barrier function in the NEC-like intestinal injury model, an in vivo permeability assay was performed using FITC-dextran as described previously [42]. Previous publications have used both 4- and 10-kDa FITC-dextran to determine in vivo gut permeability [43]–[46]. Immature intestine is considered leakier than mature intestine [47] and a small molecule tracer does not allow for detection of subtle alterations in barrier function. After optimization, a 10-kDa rather than a 4-kDa FITC-dextran (Sigma Aldrich) was used in our study to better distinguish improved barrier function of leaky immature intestine. At the end of the 5-day intestinal injury model, surviving pups grossly have histologically normal intestine (score “0”). Thus, these pups were used to evaluate barrier function in order to avoid the confounding effect of barrier disruption due to intestinal necrosis associated with disease state.

Briefly, all surviving rat pups at the end of the experiments on day 5 were starved for 4 hours and then received 10 mg/ml 10-kDa FITC-dextran (40 mg/100 g body weight) via an oral gavage tube directly into the stomach. Four hours later, whole blood was collected and centrifuged at 1,500×g for 10 minutes. Twenty five µl of serum was diluted in 100 µl of PBS and the diluted serum was added in a 96-well black wall microplate in duplicate (60 µl per well) for measurement of fluorescence intensity (excitation at 490 nm and emission at 525 nm) using a Synergy-2 fluorometer (Bio-Tek Instruments, Winooski, VT). Standards were included to determine the concentrations of FITC-dextran in the serum. High concentrations of FITC-dextran in the serum indicate greater transmucosal transport of FITC-dextran across the intestinal barrier to blood or poorer barrier function.

Immunohistochemistry staining

Intestinal (ileal) segments were formalin fixed and paraffin embedded. Sections were deparaffinized in xylene and hydrated with ethanol. For antigen unmasking, slides were heated in 10 mM sodium citrate buffer (pH 6.0) before treatment with 0.3% hydrogen peroxide. The specimens were blocked with 5% BSA in TBST (Tris-buffered saline with 0.05% vol/vol Tween-20) followed by incubation at 4°C overnight with p-NF-κB, IκBα antibodies (Cell Signaling), or control IgGs isolated before immunization from the same animal species used to generate the p-NF-κB or IκBα antibodies. After washing, the sections were incubated with polymer-HRP secondary antibodies (Dako, Carpinteria, CA). Positive staining and nuclei were visualized with DAB chromogen and hematoxylin counterstaining, respectively. For quantitation of NF-κB activity, nuclear p-NF-κB positive versus total IEC from three fields of each slide (n = 3) were counted and presented as mean of percentages of p-NF-κB-positive over total IEC ± SE. For quantitation of IκBα staining, H&E intensity was scored on a “0” to “4” scale with “0” being the lowest and “4” being the highest intensity among all samples examined [48]. Results are presented as mean of scores ± SE.

Immunoblot Analysis

Immunoblotting was conducted as described previously [49]. For tissue lysate preparation, 2-cm distal intestine (ileum) was homogenized with a hand-held homogenizer (Pellet Pestle, Kimble/Kontes, Vineland, NJ) in lysis buffer (50 mM Tris⋅HCl pH 7.4, 150 mM NaCl, 1 mM EDTA, 1% SDS, 50 mM DTT, 50 µg/ml aprotinin, 50 µg/ml leupeptin, and 5 mM PMSF). Lysates were sonicated and then centrifuged at 10,000×g for 10 minutes at 4°C. The protein concentration of supernatants was determined using the BCA protein assay kit (Pierce). Lysates were resolved by SDS-PAGE and transferred to PVDF membranes. Membranes were blocked in 5% milk PBS-T [0.1% Tween 20 (v/v) in PBS] for 1 h at RT and then probed with primary and appropriate secondary antibodies in 5% milk PBS-T. Immunoreactive bands were visualized by chemiluminescence reaction using ECL reagents (Amersham Biosciences) followed by exposure of the membranes to autoradiography film (Midsci, St. Louis, MO). Protein levels were quantified by densitometry using ImageJ software. The IκBα and GAPDH antibodies were purchased from Transduction Laboratories, Inc. (San Diego). The occludin antibody was obtained from Invitrogen (Grand Island, NY).

Expression and purification of mUIM-2 for proteasome purification

pET15b-mUIM-2 plasmid, kindly provided by Dr. Lenore K. Beitel [50], was transformed into Escherichia coli BL21 (DE3) competent cells by heat shock. The bacteria were then grown in Luria-Bertani (LB) broth containing 100 ug/ml ampicillin to an optical density (O.D.) of 0.6 (A600 nm). The culture was then induced with 0.25 mM IPTG for 16 hours and harvested by centrifugation at 3000×g for 15 minutes at 4°C. Culture pellets were resuspended in buffer A (50 mM HEPES, pH 7.5, 300 mM NaCl, 25 mM imidazole (Fisher Scientific, Ottawa, ON), lysed by sonication, and centrifuged at 30,000×g for 30 minutes at 4°C. The resulting supernatant was loaded to a Ni-NTA column, and the His(6)-tagged mUIM-2 was eluted with a gradient from 25 mM to 500 mM imidazole. The fractions eluted were then analyzed with 12% SDS-PAGE gels followed by Coomassie Blue G-250 (Fisher Scientific) staining. Fractions containing pure mUIM-2 were pooled and subsequently concentrated with Millipore Amicon Ultra centrifugal devices (10,000 MWCO).

Expression and Purification of GST-UBL for proteasome purification

The pGEX-2T-UBL plasmid, kindly provided by Juli Feigon [51], was expanded and expressed as in the mUIM-2 purification except that the bacteria were induced with IPTG for 4 hours instead of 16 hours. After 4 hours at 37°C, the bacteria were harvested by centrifugation at 3000×g at 4°C for 15 minutes. The cell pellets were then resuspended in purification buffer (50 M HEPES, pH 7.5, 5 mM MgCl₂, 10% glycerol (Fisher Scientific)) and lysed by sonication. The lysed cells were then centrifuged at 30,000×g at 4°C for 30 minutes and the supernatant was loaded onto Glutathione-Agarose beads, with gentle agitation at 4°C for 2 hours. The agarose beads were centrifuged at 500×g for 2 minutes, supernatant was removed, and beads were washed with 10 column volumes of purification buffer to obtain GST-UBL.

Proteasome purification from rat livers

Proteasomes from experimental rat livers were purified according to T.C. Scanlon et al and H.C. Besche et al with some modifications [50], [52]. The ubiquitin-like (UBL) domain of the proteasome associated protein hHR23B has high affinity for the 26S proteasome and GST-UBL served as the affinity matrix. After binding to the UBL domain, proteasomes were eluted with an excess of a recombinant His-tagged ubiquitin-interacting motif (UIM). The UIM has a higher affinity for UBL than the proteasomes and competes with the 26S proteasome for UBL binding, allowing proteasomes to be eluted from the GST-UBL beads. The His tag then allows subsequent removal of any excess UIM from the purified proteasome preparation through use of Ni-NTA beads as described below.

In brief, 300 mg to 1 g of rat livers were homogenized in purification buffer (50 mM HEPES, pH 7.5, 5 mM MgCl2, 10% glycerol) supplemented with 2 mM ATP and protease inhibitors (Roche, Laval, QC) and centrifuged at 21,100×g for 1 hour at 4°C. The supernatants were then added to the affinity matrix (GST-UBL beads), and the suspension was incubated at 4°C for 4 hours with rotation. The beads were then washed with 10 column volumes of purification buffer containing 2 mM ATP. Proteasomes bound to the GST-UBL matrix were eluted by addition of 20-fold molar excess of purified mUIM-2 and subsequent overnight incubation with continuous rotation at 4°C. Five hundred microlitres of Ni-NTA agarose beads were added to the suspension and incubated at 4°C for 2 hours to remove excess mUIM-2. The supernatants containing the purified proteasomes were then collected by centrifuging at 500×g for 3 minutes and subjected to a BCA assay (Pierce, Nepean, ON) to determine protein concentration.

26S Proteasome activity assay with purified proteasome

Chymotrypsin-like activity of the 26S proteasome was measured using a Varian Cary Eclipse fluorescence spectrophotometer by a fluorogenic substrate assay. In brief, 25 µg of affinity-purified proteasome was incubated in assay buffer (25 mM HEPES/pH 7.6, 10 mM EDTA, 0.03% SDS, 2 mM ATP) supplemented with 65 µM of suc-leu-leu-val-tyr-AMC (SLLVY-AMC) as the substrate. Proteasome chymotrypsin-like activity was determined by measuring the fluorogenic signal generated by cleavage and release of the fluorogenic compound AMC (7-amino-4-methylcoumarin) over time. Fluorescence generated from free AMC was monitored every minute for 30 minutes over the linear range of the reaction (Excitation: 380 nm, emission: 460 nm), and proteasome activity was determined by calculating the rate (relative fluorescence units/minute or RFU/minute) of free AMC released over time using Varian Cary Eclipse kinetics application software. Experiments were performed with 96-well half area black polystyrene plates (Corning Incorporated, Corning, NY) in quadruplicates. Epoxomicin (Enzo Life Sciences Inc, Farmingdale, NY), a naturally occurring selective proteasome inhibitor [53] was added to separate wells to 5 µM. Epoxomicin inhibition of proteasome activity served as a control to confirm specificity of the assay.

Analysis of TNF-α in the intestine

Local production of TNF-α, an inflammatory cytokine elevated in human and animal NEC, was analyzed from whole intestinal tissues. Intestinal tissues in RIPA buffer were homogenized by sonication and cleared by centrifugation. Protein concentration of the resulting supernatant/intestinal lysates was determined by BCA protein assay (Pierce, Nepean, ON). Fifty µg of lysate was used to determine intestinal TNF-α levels using the xMAP technology, a flow-assisted and bead-based Cytokine Assay (Bio-Rad) and analyzed with five-parameter curve fitting generated standard curve using the Bioplex Manager 6.0 software (Bio-Rad). The cytokine concentration in lysates is expressed as picograms cytokine per mg of intestinal lysate (pg/mg).

Immunofluorescence Staining of ZO-1

After deparaffinization and rehydration, intestinal sections were blocked in 5% BSA for 1 h at RT and then incubated with ZO-1 and occludin antibodies (Invitrogen, Grand Island, NY.) at 1200 dilution, followed with an Alexa Fluor 488-conjugated second antibody (Molecular Probes, Eugene, OR). Nuclei were labeled with DAPI (100 ng/ml). Images were acquired by confocal microscopy using Leica SP2.

Analysis of mRNA expression by quantitative real-time RT-PCR

RNA was isolated from frozen intestine using RNeasy Plus Mini Kit (Qiagen, Valencia, CA) according to the manufacturer's protocols. One µg total RNA was reverse transcribed in 20 µl using RNA to cDNA EcoDry Premix (Double primed) (Clontech, Mountain View, CA). TaqMan real-time PCR (qPCR) was performed in triplicate using 0.2 µl cDNA and primers and probes from Integrated DNA Technologies, Inc (Coralville, Iowa) on a 7900 Detector (Applied Biosystems, Foster City, CA). The primers and probe used to detect expression of rat ZO-1 (NM_001106266) are 5′-GAAAGGTAAGGGACTGGAGATG-3′, 5′-CATATCCTCCTTACTCACCACAA-3′, and 5′-/56-FAM/AGACAGCCT/ZEN/CTCAACAGAAAGCAGAAG/3IABkFQ/-3′. The primers and probe used to detect expression of rat GAPDH (NM_017008) are 5′-GTAACCAGGCGTCCGATAC-3′, 5′-GTTCTAGAGACAGCCGCATC-3′, and 5′-/56-FAM/CGTTCACAC/ZEN/CGACCTTCACCATCTT/3IABkFQ/-3′. The threshold cycle (Ct) value for each well was obtained using the instrument's default setting. Gene expression was normalized and calculated using the ΔΔCt method: 2^-ΔΔCt = 2^{–[Ct (ZO-1) –Ct (GAPDH)]} experiment/2^{–[Ct (ZO-1) –Ct (GAPDH)]} dam-fed and presented as fold changes to the dam-fed healthy control.

Statistical analysis

All data are presented as means ± SE. Statistical analysis was performed using the χ2 test for NEC-like intestinal injury incidence, Student's t test for paired data, or one-way analysis of variance (ANOVA) with a Bonferroni correction for multiple comparisons with GraphPad InState software. Differences were considered to be significant with p values <0.05.

Combined conditioned media (CM) from Lactobacillus plantarum (Lp), Lactobacillus acidophilus (La), and Bifidobacterium infantis (Bi) synergistically reduced intestinal injury incidence

An experimental rodent model with intestinal injury similar to NEC was used to examine the effect of oral administration of CM from Lp, La and Bi. Non-stressed and naturally born dam-fed pups served as healthy controls. Vehicle treatment alone served as an experimental control. Pups undergoing animal studies were fed with formula every 3 hours via an orogastric feeding tube. Due to this nature of the animal model, only certain numbers/categories of animals were included in one experiment. The vehicle-treated control group was always included in experiments, thus has a greater number of animals compared to other groups. The degree of intestinal injury was scored on a “0–4” scale and score ≥2 is defined as diseased injury. All dam-fed healthy controls had score “0”. In the triple combined La/Bi/Lp CM, the histological scoring of intestinal damage shows no score 4 (the most severe injury) and statistical analysis confirmed that this combined CM significantly reduced disease severity when compared with vehicle-treated group (Fig. 1). Oral administration of Lp CM and La/Bi CM reduced intestinal injury incidence (score ≥2) from 43% to 32% (n = 50, p = 0.1780) and to 28% (n = 70, p = 0.0433), respectively (Table 1). Combined CM from all three organisms further reduced NEC-like intestinal injury incidence from 43% to 20% (n = 41, p = 0.0064).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Figure 1. Histological scores of intestinal injury.

An animal model with intestinal injury similar to NEC was performed as described in “Materials and Methods”. Intestine from healthy dam-fed controls and experimental pups treated with vehicle or different CM were H&E stained and intestinal injury was evaluated on a “0–4” scale by a pathologist blinded to experiments. The healthy dam-fed pups show no intestinal injury and the three combined CM significantly decreased severity of the intestinal injury. “*” and “**” depict p<0.05 and 0.001, respectively, by one-way ANOVA with Bonferroni correction.

https://doi.org/10.1371/journal.pone.0065108.g001

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Combined conditioned media of probiotics lowered intestinal injury incidence.

https://doi.org/10.1371/journal.pone.0065108.t001

Conditioned media from Lp, La, and Bi decreased NF-κB activation and maintained IκBα expression

As the intestinal inflammation in NEC has been shown to correlate with activation of NF-κB [54], IHC staining of the active/phosphorylated form of NF-κB p65 (p-NF-κB) was used to determine if CM protection was associated with altered NF-κB signaling. As in Figure 2A and B, increased numbers of p-NF-κB-positive intestinal epithelial cells (IEC) are readily observed in the intestines of vehicle-treated non-diseased experimental pups (42.2%) compared to healthy dam-fed control pups (18.2%). Elevated p-NF-κB-positive cell numbers are further increased in vehicle-treated diseased (score ≥2) experimental model pups (55.6%). These results are consistent with previous findings demonstrating importance of NF-κB activation during the development of the disease [54]. In both diseased and non-diseased groups, Lp, La/Bi, and La/Bi/Lp CM significantly decreased p-NF-κB-positive cell numbers (Fig. 2A and B). Whether CM suppression of NF-κB activity was associated with enhanced expression of the NF-κB inhibitor IκBα was determined by IHC. IκBα was readily detected in the intestines of healthy dam-fed pups, but was diminished in vehicle-treated experimental model intestines with or without disease (Fig. 2C). La/Bi CM protected IκBα in non-diseased experimental model pups and all the CM protected IκBα expression in diseased experimental model pups, suggesting that the anti-inflammatory property of Lp, La, and Bi CM can be mediated through protected IκBα expression (Fig. 2C and D). IκBα levels in the intestine evaluated by immunoblotting and densitometry confirm H&E results that CM protected IκBα expression in the intestinal injury model (Fig. 2E and F).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Figure 2. CM from L. plantarum, L. acidophilus, and B. infantis suppressed NF-κB signaling and preserved IκBα expression.

Intestines from healthy dam-fed controls and experimental intestinal injury pups treated with vehicle or different CM were IHC stained for the active/phosphorylated form of NF-κB p65 (p-NF-κB) and IκBα or lysed for immunoblotting for IκBα. A. Representative IHC staining (at 400×) shows that CM decreased IEC NF-κB activity/nuclear p-NF-κB (arrows) in experimental pups with or without intestinal injury (n = 3 per group). B. NF-κB activity from A. was quantified by percentage of nuclear p-NF-κB-positive IEC in three fields of each intestinal section (n = 3). C. Representative IHC staining shows La/Bi CM and all CM protection of IEC IκBα expression in experimental pups with or without intestinal injury, respectively, (n = 3 per group). D. IHC intensity of IκBα from C. was scored on a 0–4 scale to relatively quantify levels of IκBα. E. Immunoblotting shows CM protection of IEC IκBα expression in experimental pups. Intestines of vehicle- and CM-treated experimental pups with disease and healthy dam-fed pups were lysed and subjected to IκBα immunoblotting. IκBα protein levels were quantified by densitometry using ImageJ software and normalized to GAPDH and presented in F. Results are presented as mean of percentages, scores, or relative density ± SE. “*” depicts p<0.05 compared to the vehicle group by one-way ANOVA with Bonferroni correction.

https://doi.org/10.1371/journal.pone.0065108.g002

Conditioned media from Lp, La, and Bi decreased proteasome activity

IκBα is degraded by the chymotrypsin-like (CTL) activity of the proteasome, allowing release and activation of NF-κB [55]; hence, the effect of CM on chymotrypsin-like proteasome activity was investigated. Colitis has been demonstrated to globally affect proteasome activity in non-gastrointestinal tissues [56]. Among organs tested for proteasome activity in neonatal rats, liver tissues provided the optimal yield and activity (data not shown). Therefore, the effect of CM on proteasome activity was assessed using liver homogenates. In liver homogenates from the experimental model, the vehicle-treated group had the highest proteasome activity. Gavage with Lp and La/Bi CM significantly reduced liver proteasome activity in the experimental model (Fig. 3). Epoxomicin, a naturally occurring proteasome inhibitor was used in parallel and its inhibition of proteasome served as a positive control to confirm specificity of the assay.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Figure 3. CM from L. plantarum (Lp CM) and L. acidophilus and B. infantis (La/Bi CM) inhibited proteasome activity.

CM inhibited proteasome activity in the experimental intestinal injury model. At the end of each experiment, liver was harvested, lysed, and proteasome was purified and subjected to a proteasome chymotrypsin-like activity assay as described in “Materials and Methods”. Five µM of the proteasome specific inhibitor epoxomicin (Epox) was added in separate wells as a control to confirm specificity of the assay. Proteasome chymotrypsin-like activity was determined by the rate of generation of fluorogenic product over time. Five µM of Epox was added in separate wells as a control to confirm specificity of the assay. Data (n = 4) are presented as mean ± SE. “*” and “**” depict p<0.05 and 0.001 compared to vehicle groups, respectively, by one-way ANOVA with Bonferroni correction.

https://doi.org/10.1371/journal.pone.0065108.g003

Conditioned media from Lp, La, and Bi decreased local production of TNF-α

To investigate the effect of CM on inflammatory cytokine production, the pro-inflammatory cytokine TNF-α, an NF-κB target elevated in clinical and animal experimental NEC, in intestinal homogenates was next measured. As shown in Fig. 4, healthy dam-fed control pups had the lowest intestinal TNF-α levels. Increased TNF-α was detected in vehicle-treated diseased experimental model pups. Lp, La/Bi, and La/Bi/Lp CM significantly decreased TNF-α levels in pups exposed to the experimental intestinal injury model.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Figure 4. CM from L. plantarum, L. acidophilus, and B. infantis decreased the pro-inflammatory cytokine TNF-α in the intestine.

Intestine was collected from vehicle- or CM-treated experimental pups with intestinal injury and from healthy dam-fed controls and lysed in RIPA buffer. Fifty µgs of lysates were used to determine intestinal TNF-α levels using the xMAP technology. Data (n≥5) are presented as mean ± SE. “*” depicts p<0.05 compared to the vehicle group by one-way ANOVA with Bonferroni correction.

https://doi.org/10.1371/journal.pone.0065108.g004

Combined conditioned media from Lp, La, and Bi protected intestinal barrier function and maintained the tight junction (TJ) protein ZO-1 at TJs

NEC is associated with intestinal barrier disruption [8], [37], [57]. To determine if CM had a cytoprotective effect in addition to an anti-inflammatory effect, intestinal barrier function was assessed by transmucosal transport of fluorescent FITC-dextran (10 kDa) to blood in surviving pups at the end of the animal experiments [49]. Surviving pups were used as these pups all have an intestinal injury histologic score “0”, thus alterations in barrier function seen are not confounded by alterations induced by tissue necrosis associated with disease state. The healthy dam-fed control pups had the lowest level of FITC-dextran in the blood (intact barrier function) whereas the vehicle-treated experimental model pups had the highest levels of FITC-dextran in the blood (poorer barrier function) (Fig. 5A). La/Bi and La/Bi/Lp CM significantly decreased blood FITC-dextran levels, indicating protection of barrier function by these CM in the experimental intestinal injury model. In contrast to combined CM, Lp CM failed to protect barrier function.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Figure 5. CM from L. plantarum, L. acidophilus, and B. infantis protected intestinal barrier integrity.

A. Combined La/Bi and La/Bi/Lp CM protected intestinal barrier function. Experimental pups were fed with formula plus vehicle or CM from La, Bi, and Lp and stressed to induce intestinal injury. In vivo barrier function was determined in all surviving experimental as well as healthy dam-fed pups at the end of the animal experiment on day 5. Data from several animal experiments (n≥14) are presented as mean ± SE. Higher levels of the tracer FITC-dextran in blood indicate poorer intestinal barrier function. B. An image of sutured intestinal explant loop for ex vivo loop/barrier function assay. The loop was filled with trypan blue for demonstration. C. Ileum exhibited poor barrier function compared to jejunum/duodenum in three-day old experimental pups. Explants of ileum and jejunum/duodenum were harvested from euthanized rat pups, flushed, filled with dialyzed 10 kDa FITC-dextran (1 mg/mL in PBS), and ligated to form closed loops. Explanted loops were placed in PBS and fluorescence in PBS was measured at 30, 60, and 90 minutes. Barrier integrity of paired ileum and jejunum/duodenum from experimental pups (n = 5) at all time points (n = 15) was measured and compared to ileum. D. Combined La/Bi and La/Bi/Lp CM protected the localization and expression of the TJ protein ZO-1 in the intestine in experimental intestinal injury model. Intestines from animals used in A. were H&E and IF stained for ZO-1 and nuclei and then subjected to confocal microscopy. Representative immunofluorescence staining shows preserved expression and localization of ZO-1 at the TJs of healthy, La/Bi and Lp/La/Bi CM-treated experimental intestines but not in the vehicle- or Lp CM-treated ones. H&E staining demonstrates intestinal injury score “0” (disease free) for all sections. E. Combined La/Bi and La/Bi/Lp CM protected intestinal ZO-1 mRNA expression in experimental pups. Total RNA was isolated from intestine and reversed transcribed to obtain cDNA, which was then subjected to TaqMan real-time PCR to determine expression of ZO-1 and GAPDH mRNA in triplicate (n = 3). Normalized ZO-1 mRNA levels are presented in fold to dam-fed controls. F. CM from La, Bi, or La did not protect expression of the TJ protein occludin in the intestine in experimental intestinal injury. Intestines used as in D. were IF stained for occludin and nuclei and then subjected to confocal microscopy. Representative immunofluorescence staining shows increased occludin expression in the intestinal injury model compared to the healthy dam-fed control and all CM failed to normalize occludin levels to that in healthy controls. G. CM did not normalize occludin expression in experimental pups to that of healthy controls. Intestines from vehicle- and CM-treated experimental pups surviving to the end of the experiments (disease free) and healthy dam-fed pups were lysed and subjected to immunoblotting for occludin and GAPDH. Protein levels were quantified by densitometry using ImageJ software and GAPDH normalized occludin levels are presented as mean ± SE in H. “*” and “**” depict p<0.05 and 0.001, respectively, by t-test in C and compared to the vehicle group by one-way ANOVA with Bonferroni correction in A, E, and H.

https://doi.org/10.1371/journal.pone.0065108.g005

The ileum of the intestine is the classically described affected area in NEC. An ex vivo loop assay was conducted to measure and compare the barrier integrity of paired ileum and non-ileal regions in the intestine, jejunum and duodenum, from experimental rat pups. Our result shows that ileum is leakier than jejunum/duodenum in the pups (Fig. 5B and C).

Intestinal barrier function can be regulated by both TJ structure and proper TJ protein expression. Others and we have previously shown that occludin is up-regulated while expression and localization of ZO-1 are lost during the development of NEC [20], [49], [58]. Since La/Bi and La/Bi/Lp CM protected intestinal barrier function in the experimental model (Fig. 5A), immunofluorescence staining for ZO-1 and occludin in the ileum was evaluated in intestines from the same animals used for the in vivo barrier function assay. Immunofluorescence staining revealed normal TJ structure and ZO-1 protein expression predominantly at the TJs in healthy dam-fed controls. In contrast, ZO-1 staining at TJs was lost in ileal villi of vehicle-treated experimental pups (Fig. 5D). This alteration in ZO-1 by experimental stress was prevented by feeding pups formula supplemented with La/Bi and La/Bi/Lp CM but not by Lp CM. Quantitative real-time RT-PCR shows protected ZO-1 mRNA expression in La/Bi and La/Bi/Lp CM-treated groups (Fig. 5E), supporting the staining results that show combined CM protection of ZO-1 expression in the experimental model. H&E staining confirmed no histological damage in all the pups used for this experiment (Fig. 5D, top panels), suggesting that the loss of ZO-1 at the TJ was not due to intestinal necrosis. Immunoflourescence staining and immunoblotting for occludin show increased occludin in the intestine of vehicle-treated pups exposed to the experimental model compared to healthy controls (Fig. 5F, G, and H). All CM failed to normalize occludin levels to that in healthy controls, demonstrating that combined CM protection of barrier function is not mediated through regulation of occludin.