Oral Bifidobacterium longum expressing alpha-melanocyte-stimulating hormone to fight experimental colitis

Abstract

The oral delivery of peptides is a highly attractive treatment approach. However, the harsh environment of the gastrointestinal tract limits its application. Here, we utilize Bifidobacterium as a delivery system to orally deliver a potent anti-inflammatory but short duration peptide alpha-melanocyte-stimulating hormone (α-MSH) against experimental colitis. The aim of our study was to facilitate the efficient oral delivery of α-MSH. We designed a vector of pBDMSH and used it to construct a Bifidobacterium longum expressing α-MSH. We then determined the bioactivity of recombinant Bifidobacterium in lipopolysaccharide-induced inflammatory models of HT-29 cells. Finally, we used Bifidobacterium expressing α-MSH against dextran sulfate sodium (DSS)-induced ulcerative colitis mice. Results based on the myeloperoxidase activity, the levels of inflammatory cytokines TNF-α, IL-1β, IL-6, and IL-10 and the histological injury of colon tissue reveal recombinant Bifidobacterium was efficient in attenuating DSS-induced ulcerative colitis, suggesting an alternative way to use Bifidobacterium as a delivery system to deliver α-MSH for DSS-induced ulcerative colitis therapy.

• α-MSH
• Bifidobacterium
• colitis
• peptide delivery

Introduction

For decades, a great deal of attention has been paid to peptide drugs due to their incredible selectivity and their effective and potent function. It is thought that oral delivery of peptides is a highly preferable approach owing to its convenience and patient compliance. However, due to its tremendous degradation in the gastrointestinal tract the extremely low bioavailability of this administration approach poses a significant challenge (Choonara et al., 2014). Despite the considerable advances in technology and the impressive progress that have been made in recent years, true oral peptide formulations still represent a less than desirable treatment option (Chin et al., 2012). Specifically, some peptides have a short duration in the bloodstream making it necessary for continuous administration to sustain the therapeutic effect. Furthermore, some peptides elicit an undesired immune response or lack site-specific targeting, making their applicability more impractical. Therefore, there is a considerable need to develop a new strategy to facilitate the application of these drugs.

Alpha-melanocyte-stimulating hormone (α-MSH) is a neuropeptide containing 13 amino acids that is derived from proopiomelanocortin. Over the past several decades, many studies have demonstrated that α-MSH possesses potent anti-inflammatory and immunomodulating activities. It exerts these activities primarily by mediating various mediators: downregulation of IL-1, IL-4, IL-6, IL-8, TNF-α, IFN-γ, myeloperoxidase (MPO), and so on, as well as upregulation of anti-inflammatory cytokine IL-10. It is now clear that the anti-inflammatory effects of α-MSH have been established in all forms of inflammation diseases, such as irritant dermatitis, sepsis, vasculitis, inflammatory bowel disease (IBD), acute pancreatitis and hepatitis, suggesting that α-MSH is a valuable drug for the treatment of inflammatory disorders (Brzoska et al., 2008,2010; Chin et al., 2012). Unfortunately, given the fact that the half-life of the α-MSH peptide in vivo is only a few minutes (Lipton, 1990), its practical application is limited by the necessity for continuous injection to maintain its therapeutic effect. Thereby, a promising delivery system is required to circumvent this impediment with prolonged biological activity.

Bifidobacteria are anaerobic microorganisms that naturally occur in the intestines. These bacteria constitute a major proportion of the intestinal microbiota and they confer a health benefit on the host by balancing the intestinal microflora, synthesizing nutrients, and improving the immune system (Oatley et al., 2000). In this regard, bifidobacteria are considered to be one of the most important types of probiotic bacteria and they are generally recognized as being safe (Park et al., 2008). Because bifidobacteria are nonpathogenic, they have been widely employed in the fields of food science and medicine in recent years (Reyes Escogido et al., 2007).

In the present study, we used Bifidobacterium longum (B. longum) as a delivery vector that developed a B. longum secreting α-MSH using recombinant DNA technology. We then determined its anti-inflammatory biological activity in lipopolysaccharide (LPS)-induced HT-29 cells and evaluated its effectiveness in the inflammatory models of dextran sulfate sodium (DSS)-induced ulcerative colitis mice. Our ongoing research aimed to facilitate oral administration of the peptide of α-MSH.

Materials and methods

Construction of Bifidobacterium secreting α-MSH

Escherichia coli DH5α was preserved in our laboratory and used for general cloning. It was cultured at 37 °C in Luria-Bertani broth. The B. longum HB25 strains came from Tian et al. (2014) and were grown in a de Man, Rogosa, and Sharpe (MRS) medium supplemented with 0.05% (w/v) L-cysteine at 37 °C under an anaerobic condition. The bifidobacteria amyB signal peptide coding sequence was PCR-amplified from the B. longum and the following PCR primers were used (the portions that are underlined are recognition sites for the restriction enzymes EcoRI and SmaI, respectively): F (GGCGAATTTAAAGGAGCCATGAAACAT), R (GGCCCGGGCTGTTCGAACGTGGTGTTT), 196 bp. The basic plasmid pDG7 (Matteuzzi et al., 1990) and pBV220 (ShineGene Molecular Biotech, Shanghai, China), a synthesized oligonucleotide (Figure 1) using codon optimization for bifidobacteria (Reyes Escogido et al., 2007) containing a furin linker and an α-MSH gene, as well as the amyB signal sequence, were used to construct the pBDMSH secretion vector (Figure 2). The general operation process of restriction enzyme digestion, ligation, and transformation of E. coli were performed as described elsewhere (Sambrook & Russell, 2001). The pBDMSH plasmid was transformed into B. longum by electroporation as previously described (Rossi et al., 1997). The transformants were screened via a solid agarized medium containing chloramphenicol (100 mg/L final concentration) in an anaerobic condition.

Figure 1. Sequence of the synthesized nucleotide fragment and its amino acid sequence. The boxed portion is a furin recognition sequence (furin linker); the sequences coding for α-MSH are shown in the region of the arrow.

Figure 2. Construction of α-MSH expression plasmid pBDMSH. pMB1, plasmid from B. longum represents the replicon for Bifidobacterium; Amp, ampicillin resistance gene; ori, origin of replication; CIts857, a thermo-inducible gene; P_RP_L, promoter of prokaryote; rrnBT1T2, transcriptional terminator; amyB signal, amylase signal peptide.

Expression of the target protein

The B. longum strain transformed with pBDMSH was either inoculated in buffered Dulbecco’s modified Eagle’s medium (DMEM) media containing 3 μg/ml of chloramphenicol at 37 °C or it was cultivated at 37 °C for 8 h first, then induced at 42 °C for 8–24 h. The cells were collected and lysed at 24 h, together with bifidobacterial culture supernatant, then centrifuged at 12 000 rpm for 30 min. The collected supernatant was used to determine the protein expression with Tricine-SDS-PAGE. A portion of bifidobacterial culture supernatant was lysed with 3 μg/L furin (Biocoen, Beijing, China) for 8 h. The levels of α-MSH in the cell culture supernatants were detected in properly diluted samples using an α-MSH enzyme-linked immunosorbent assay (ELISA) kit (IBL-America, Minneapolis, MN) according to the manufacturer’s protocols.

Determination of anti-inflammatory activity

The HT-29 cells were purchased from American Type Culture Collection (Manassas, VA) and cultured in RPMI-1640 media in the presence of 10% fetal bovine serum, 100 IU/ml penicillin, and 100 IU/ml streptomycin in a humidified atmosphere (37 °C, 95% air, 5% CO2). The cells were seeded into 24-well plates at a density of 5 × 10⁵ cells/well. The bifidobacteria culture supernatants (digested with furin when required) were sterilized by filtration through 10 kDa ultrafiltration membranes (Millipore Amicon, Billerica, MA) and adjusted to an anticipated concentration (containing 60 pg/ml α-MSH). After removing the growth medium, the cells were incubated with bifidobacterial culture supernatants diluted 1:1 (v/v) with fresh medium and pretreatment for 1 h, and the cells were then exposed to 50 ng/ml of LPS (E. coli, Sigma). The cell supernatants were collected before they were incubated for 4 h, then an ELISA kit (GenStar, Beijing, China) was used to detect TNF-α in accordance with the manufacturer’s instructions (GenStar, Beijing, China).

Colitis induction and administration of B. longum

The entire experiment was approved by the Animal Care and Use Committee of Jinan University. The male SPF BALB/c mice, aged 6–12 weeks and weighing 18 ± 2 g, were obtained from the Third Military Medical University. They were housed in comfortable cages at Nanyang Qi Wei Microecology Genes Science and Technology Development Co., Ltd (China) and allowed to acclimatize to standard lighting and temperature conditions with food and water freely available before the experiment was performed. The mice were randomly divided into four groups (n = 10 mice per group) as follows: control (C) group, DSS group, B. longum empty cloning vector (EV) group, and B. longum expressing α-MSH (pBDMSH) group. For induction of colitis, the DSS group, EV group, and pBDMSH group received 3.0% DSS via drinking water casually for 7 d; the control group was only given tap water. After DSS treatment for 7 d, the EV group and the pBDMSH group were orally administered the B. longum empty cloning vector and the engineered B. longum (B. longum expressing α-MSH) (1 × 10¹⁰ CFU/mouse/d) daily, respectively. The DSS group and the control group were orally administered PBS daily. In the indicated time after administration, the mice were euthanized and their colon tissues were harvested and snap frozen for tissue analysis.

Determination of MPO activity

On d 2 and d 9 after administration of B. longum, the tissue samples from the distal colon were obtained and weighed, then homogenized in 0.5% (w/v) hexadecyltrimethylammonium bromide in 50 mM potassium phosphate buffer (pH 6.0). The homogenates were then centrifuged for 15 min at 15 000 rpm at 4 °C. After they were resuspended for 20 min, insoluble material was removed and the supernatants were collected. Next, 50 μl of the supernatants sample was added to 200 μl of o-Dianisidine solution (30 μl of 20 mg/ml o-Dianisidine dihydrochloride and 30 μl of 20 mM hydrogen peroxide in 50 mM potassium phosphate buffer, pH 6.0). After another 20 min, the reaction was terminated with sodium azide, and the change of absorbance at 460 nm was read using a microplate reader. One unit of MPO activity was defined as the degrading of 1 mmol of peroxide per minute at 25 °C.

Measurement of the cytokines and α-MSH

Colon specimens weighing 0.1 g were homogenized in ice-cooled PBS at pH 7.2 and then centrifuged at 12 000 rpm for 10 min at 4 °C. The supernatants were collected and the cytokine levels of TNF-α, IL-1β, IL-6, and IL-10 were determined using ELISA kits in accordance with the manufacturer’s instructions (GenStar, Beijing, China). Similar to the cytokines, the mucous membranes from the colonic mucosa were harvested after removing feces and then centrifuged at 5000 rpm for 10 min. The resulting supernatants (digested with furin when needed) were used for α-MSH measurement by ELISA as described above. This α-MSH was determined on d 1, 3, 5, 7, and 9 after administration of B. longum.

Histological evaluation

On d 9, the mice were sacrificed to collect segments of the colon tissue. These colon samples were cleaned with physiological saline, fixed in 4% buffered formaldehyde (pH 7.4), sliced into 5 μm-thick sections, and stained with hematoxylin and eosin, in turn. The morphological changes of the colonic membranes were examined with a microscope.

Statistical analysis

All data in the figures and text are expressed as arithmetic means ± standard deviation (SD). The data are representative of three or more independent experiments. The data were processed using GraphPad Prism 5.0. Statistical analysis for significant differences was performed using one-way ANOVA, where appropriate. A p < 0.05 was considered statistically significant.

Results

Construction of the vector

The pBV220 was used as the expression vector in Bifidobacterium because it has been shown to work well in this system (Fu et al., 2005). However, to ensure the consistency of the plasmid, we linked a replication of Bifidobacterium-pMB1 (Matteuzzi et al., 1990) into pBV220. To allow for secretion of the peptide, we introduced an amyB signal peptide (Park et al., 2005). In order to obtain an intact α-MSH peptide in vivo, we introduced a furin-cleavable sequence between the amyB signal peptide and α-MSH; this would allow us to obtain an integral α-MSH after furin cleavage. Thus, the amyB signal sequence, a synthesized oligonucleotide containing a furin linker and an α-MSH gene as well as pMB1, was ligated into pBV220 to yield pBDMSH (Figure 1).

α-MSH expression

After the protein expression was induced at 42 °C, it was analyzed using Tricine-SDS-PAGE. As shown in Figure 3, a prominent band of about 4.35 kDa in the supernatant of the bacteria was transformed with pBDMSH. This band matched the calculated molecular weight of the α-MSH precursor protein and it was not present in the B. longum empty cloning vector, allowing us to conclude that this band corresponded to the α-MSH precursor protein. We then measured the expression of α-MSH in the DMEM culture supernatants. ELISA showed that the concentration of α-MSH was 206 ± 9.5 pg/ml in 24 h at 37 °C, but when it was induced at 42 °C after 8 h the concentration was 336 ± 13.7 pg/ml, indicating that the pBDMSH plasmid did well in this system.

Figure 3. Tricine-SDS-PAGE of the protein expressions in B. longum. In lane 1 and lane 2, the strains are transformations of pBDMSH; lane 3 is the empty vector strain. The arrow indicates the dark bands corresponding to the expressed α-MSH precursor protein on Tricine-SDS-PAGE.

Determining the anti-inflammatory activity of α-MSH expressed by B. longum

To determine the biological activity of α-MSH expressed by the B. longum, we tested its anti-inflammation effect with an LPS-induced inflammatory model of HT-29 cells. We pretreated the HT-29 cells with B. longum culture supernatants. The cells were then stimulated with LPS and the expression of TNF-α was measured by ELISA. Here, we selected the supernatants of the B. longum empty cloning vector (EV) as a negative control. α-MSH is a strong anti-inflammation peptide that is effective at doses ranging from10⁻¹⁷ to 10⁻¹² M (Brzoska et al., 2008). Data showed that the supernatants of recombinant B. longum (pBDMSH, lysed by furin) containing 30 pg/ml α-MSH were capable of significantly blocking the expression of TNF-α (p < 0.01) in comparison to the EV group, thereby indicating a biological α-MSH (Figure 4). Interestingly, we also observed that recombinant B. longum supernatants with no furin digest could significantly reduce the TNF-α expression (p < 0.05). This might be due to the fact that the active site of α-MSH is positioned in the C-terminal peptide fragments (Brzoska et al., 2008); hence, the α-MSH precursor might exhibit an integrant anti-inflammation effect. However, the true reason for this requires further investigation.

Figure 4. Inhibition of TNF-α production by Bifidobacterium supernatants containing α-MSH in the LPS-stimulated HT-29 cells. The HT-29 cells were pretreated for 1 h with supernatants of the B. longum empty cloning vector (EV) or the transformant (pBDMSH) bifidobacterial culture. TNF-α was then determined after LPS stimulation for 4 h. **p < 0.001 versus LPS and EV groups, *p < 0.05 versus LPS and EV groups.

Expression of α-MSH in the colon

To profile the α-MSH expression in the colon, the concentration of α-MSH in the colonic mucus was determined at d 1, 3, 5, 7, and 9 after oral administration of B. longum. As shown in Figure 5, α-MSH was constantly increased on d 1–5 after administrating B. longum expressing α-MSH, while it was found to enter an equilibrium phase on d 5–9 (20.5 ± 0.8 pg/ml on d 9). Yet, we observed that part of the α-MSH precursor did not mature, and this might be interpreted as being the result of the furin of the mucosa hitting its limit in that concentration of α-MSH.

Figure 5. Expression of α-MSH in the mucous membranes of colonic mucosa d 1, 3, 5, 7, and 9 after administration of Bifidobacterium. ***p < 0.001 versus EV and EV + furin groups.

Administration of B. longum expressing α-MSH ameliorates DSS-induced colitis

We then paid attention to how this genetically modified B. longum worked in DSS-induced colitis. As is known, colitis is distinctly characterized by various factors, such as increased MPO activity, unbalanced inflammatory cytokines and colon histological damage. We analyzed these to evaluate the effects of recombinant B. longum on colitis treatment. The mice were orally administered B. longum daily after receiving 3.0% DSS for 7 d; then MPO, TNF-α, IL-1β, IL-6, and IL-10 were detected and histological evaluation was performed.

A previous study found that MPO is expressed by neutrophils and closely related to inflammatory cell infiltration (Fichna et al., 2012). In IBD, MPO is regarded as being a typical feature of inflammatory status and tissue injury and elevated MPO activity follows an aggravated state of that illness (Gambero et al., 2007). In our experiments, after oral administration for 9 d, the MPO activity of the colon tissues was found to significantly decline (p < 0.05) in the pBDMSH group as compared to the DSS and EV groups (Figure 6), suggesting a reduced neutrophil infiltration in the colon tissues.

Figure 6. Administration of Bifidobacterium expressing α-MSH for 9 d decreases the MPO activity in DSS-induced colitis mice, *p < 0.05.

We next evaluated the expression of pro-inflammatory cytokines TNF-α, IL-1β, and IL-6, and anti-inflammatory cytokine IL-10 in DSS-induced colitis. It has been universally acknowledged that these cytokines serve a critical function in the modulation of intestinal immunity and inflammation. Considerable research has reported that large amounts of TNF-α, IL-1β, and IL-6 are found in the colonic mucosa of ulcerative colitis patients (Sanchez-Munoz et al., 2008), and these cytokines are able to exacerbate colitis by means of their pro-inflammatory effects. It has been reported that blockage of TNF-α and IL-6 exerts a therapeutic effect in IBD (Van Kemseke et al., 2000; Neuman, 2007). In our study, the data showed a significant decrease in the cytokines TNF-α, IL-1β, and IL-6 in the pBDMSH group (p < 0.05, Figure 7) in comparison to the DSS and EV groups after administration for 9 d. In contrast, IL-10, one of the anti-inflammatory cytokines that have been shown to have potential as a therapeutic drug for colitis (Trifunović et al., 2015), was dramatically increased (p < 0.01, Figure 7) in the inflammatory tissue in the pBDMSH group.

Figure 7. Effect of administrating B. longum expressing α-MSH on the expression of TNF-α, IL-1β and IL-6, IL-10 in colon tissue on d 2 and d 9, *p < 0.05 versus the DSS and EV groups, **p < 0.01 versus the DSS and EV groups.

Improvement of histological features

To further characterize the inflammatory process of the mice that were administered B. longum, we observed the degree of histological damage. As shown in Figure 8, a high-grade infiltration of inflammatory cells and submucosal edema were observed in the DSS group and the EV group, suggesting a typical inflammatory feature. Compared to the DSS group, the pBDMSH group showed attenuation in infiltration of lymphocytes, representing a moderate ulcer in the colon biopsies.

Figure 8. Findings from the histological analysis of the colon tissue samples on d 9 (100×). Control group: healthy colonic histology. DSS and EV groups: the normal colonic mucosa was eroded showing severe mucosal infiltration. pBDMSH group: a slight inflammatory infiltrate.

Discussion

Making full use of peptides in the medicine domain remains a difficult challenge. In the present work, we established a Bifidobacterium vector system to deliver α-MSH. In this system, we introduced a functional signal peptide amyB. Considering that the residual amino acids of a signal peptide might damage the activity of α-MSH, we inserted an efficient furin linker between α-MSH and the amyB signal peptide; thus, the secreted α-MSH precursor containing α-MSH and signal peptide residual amino acids would develop an intact α-MSH in the presence of furin which is a fundamental proteinase that is widely distributed in all kinds of cells and tissues of humans, including extracellular and intracellular fluids (Ayoubi et al., 1994; Henrich et al., 2003). The α-MSH precursor immediately converts to α-MSH as soon as it interacts with furin in vivo. The test for anti-inflammatory activity determined that the α-MSH precursor exerted an appreciable anti-inflammatory effect. This might be interpreted as the furin transforming the α-MSH precursor into a mature form of α-MSH.

Colitis, known as IBD, is a sophisticated inflammatory disease that occurs in the colon. α-MSH is a typical peptide derived from human beings that exerts a potent anti-inflammatory effect. In particular, this peptide is well established in IBD treatment since it is impacted by a variety of mediators, such as the regulation of the production of pro-inflammatory cytokines, reactive oxygen metabolites, and nuclear factor κB as well as the inhibition of apoptosis (Singh & Mukhopadhyay, 2014), and a previous study supported this view (Rajora et al., 1997). However, it was widely acknowledged that peptides never transport to the colon when they are taken orally since these drugs are easily destroyed by the proteases of the stomach and small intestine (Park et al., 2011). Therefore, a different strategy must be taken to achieve α-MSH applicability.

Bifidobacterium is a major type of anaerobic bacteria that lives in the colon and is friendly to humans. In this study, we designed Bifidobacterium longum as a delivery system that engineered B. longum secreting an α-MSH precursor. When administered orally, this Bifidobacterium α-MSH precursor would be directly delivered to colon, the local site of ulcerative colitis. Subsequently, an α-MSH precursor was matured by the furin located in the colonic mucous (Xu et al., 2000; Gendron et al., 2006). Our results showed that α-MSH was detected by ELISA in the mucous membranes of colonic mucosa thereby supporting the potential positive effect of this delivery system. This delivery system might present two difficulties. On the one hand, Bifidobacterium directly transports α-MSH to the colon, thereby avoiding the unfavorable degradation of peptides in the stomach and small intestine since the colon is a practical position for drug delivery (Rubinstein, 2005), consequently, improving the bioavailability of the peptides; moreover, Bifidobacterium growth in the colon is capable of adhering to the mucus membranes, thereby increasing the residence time of therapeutic peptide contributing to an increased in its bioavailability (Muheem et al., 2014). On the other hand, recombinant B. longum was able to sustain the expression of α-MSH in the colon, which would support “continuous administration”, thereby presenting one way to reverse the short biological half-life of α-MSH. Importantly, considering that Bifidobacterium live in the intestinalcanal, where ulcerative colitis occurs, via the help of Bifidobacterium, α-MSH would be immediately transported to the inflammatory site, thereby providing a valuable site-specific treatment for ulcerative colitis.

In our study, after oral administration of recombinant bacteria, α-MSH worked in the sick mice and they got some relief from their inflammatory symptoms. These results are based on the data that showed a significant decrease in MPO activity, are markable reduction in the pro-inflammatory cytokines TNF-α, IL-1β, and IL-6, a noticeable increase in anti-inflammatory cytokine IL-10 and attenuated tissue injury in the colon of the colitis mice. These findings collectively suggest that orally administered B. longum expressing α-MSH was effective in controlling ulcerative colitis induced by DSS. This proposed deliver system uses Bifidobacterium as a carrier. It is easy to cultivate and can be delivered in an easy-to-use form that does not require sophisticated manufacturing facilities, leading to low production costs. In conclusion, Bifidobacterium as a delivery system to deliver α-MSH for DSS-induced ulcerative colitis therapy is an efficient and alternative strategy.

Declaration of interest

The authors report no conflicts of interest. The authors alone are responsible for the content and writing of this article. This work was sponsored by the Innovation Fund for Technology-based SMEs (No. 12C26214104321), and the National Spark Program of China (NSPC; No. 2011GA750017).