Fucoidan from Sargassum hemiphyllum inhibits infection and inflammation of Helicobacter pylori

Having infected by Helicobacter pylori, the infection often leads to gastritis, gastric ulcer, or even gastric cancer. The disease is typically treated with antibiotics as they used to effectively inhibit or kill H. pylori, thus reducing the incidence of gastric adenoma and cancer to significant extent. H. pylori, however, has developed drug resistance to many clinically used antibiotics over the years, highlighting the crisis of antibiotic failure during the H. pylori treatment. We report here that the fucoidan from Sargassum hemiphyllum can significantly reduce the infection of H. pylori without developing to drug resistance. Fucoidan appears to be a strong anti-inflammation agent as manifested by the RAW264.7 cell model examination. Fucoidan can prohibit H. pylori adhesion to host cells, thereby reducing the infection rate by 60%, especially in post treatment in the AGS cell model assay. Mechanistically, fucoidan intervenes the adhesion of BabA and AlpA of H. pylori significantly lowering the total count of H. pylori and the level of IL-6 and TNF-α in vivo. These results all converge on the same fact that fucoidan is an effective agent in a position to protect the stomach from the H. pylori infection by reducing both the total count and induced inflammation.

It has been well documented that Helicobacter pylori is the causative agent of major gastric associated diseases. The virulence factor CagA of the pathogen has also been characterized abnormally activating the nuclear factor kappa-B (NF-κB) to upregulate the production of the inflammation factor interleukin-8 (IL-8), thus leading to gastritis. If it develops into a chronic condition, gastritis could morph further to gastric ulcer and likely end up gastric cancer 1 .
The H. pylori-related gastric diseases account for approximately two-thirds of all gastric cancer deaths in the Asia-Pacific region in 2012 2 . The most striking fact is that up to 90% of gastric cancer can be attributed to the H. pylori infection worldwise 3,4 . Some countries in East Asia including Taiwan are at the top of the rank 5 . In view of public health and wellbeing, how to prevent the H. pylori infection from occurring and how to effectively treat these diseases become two big concerns. The mainstream treatment of the H. pylori infection remains the use of antibiotics, which usually combines two types of antibiotics in conjunction with a proton pump inhibitor 6 . Because of extensive and/or abusive use of antibiotics, some escaped strains of H. pylori may have hence developed drug resistance, which often results in treatment failure 7 . To find more effective antibiotics or substances that can stop or be conducive to curbing the H. pylori associated chronic gastritis becomes of growing interests in the biomedical community.
Seaweeds are known fraught with various biological activities. Fucoidan, for example, is one of the key polysaccharides in brown seaweeds, which features a high content of fucose and sulfate 8 with strong anti-cancer 9-11 , anti-virus 12 , and anti-pathogen adhesion/infection activities 13 . Recent studies indicated that fucoidan can impede the attachment of H. pylori to the gastric epithelium thereby inhibiting the H. pylori infection 14 . Fucoidan has also been shown capable of suppressing the production of pro-inflammatory cytokines by down-regulating the MAPK and NF-κB-mediated signaling pathways 15 www.nature.com/scientificreports/ In this study, we set out to examine the fucoidan from Sargassum hemiphyllum for its biomedical effects upon the H. pylori infection in vitro and in vivo. First, we chemically analyzed the sulfate and total sugar contents of the fucoidan. Second, the fucoidan was subjected to anti-inflammation and anti-adhesion evaluations against the H. pylori infection using both mouse macrophage (RAW264.7)-cell and gastric adenocarcinoma-cell (AGS) model systems. Finally, the efficacy of the fucoidan was evaluated at the animal level, where its effects on the H. pylori-infected BALB/c mice were gauged in terms of prevention and treatment. Our results demonstrate an overall superb protection effect against the H. pylori infection showing pronounced mitigation of inflammation induced by H. pylori.

Results
Yields and compositions of the fucoidan from Sargassum hemiphyllum. Sargassum hemiphyllum was extracted with hot water, by which the yield of fucoidan was estimated about 13.5%. The GPC analysis showed that the retention time (the estimated molecular weight) of the extracted fucoidan (9.84 min) is similar to that of the fucoidan reference purchased from Sigma (CAS: 9072-19-9) with the molecular weight of 440 kDa 17 (Fig. 1). Fucoidan is sulfated polysaccharides, in which the total sugar and sulfate contents were respectively determined to be 321 mg/g and 12.87%. Its total phenolic and protein contents were 18.59% and 20.60%, respectively. Its monosaccharide components contained 51.2% fucose, 16

Inhibition of H. pylori infection on AGS cells by fucoidan.
We then came to examine the inhibitory activity of the fucoidan upon the H. pylori infection. The infection ratio of H. pylori to cells was set at 100:1 (multiplicity of infection, MOI = 100). Our examination included three testing groups: Post-treatment, Pre-treatment, and Co-treatment. The bacteria-challenge experiments showed that the fucoidan, in general, is effective against the H. pylori infection in a dose-dependent manner in the Post-treatment group. The infection rate was significantly reduced with the concentration of the fucoidan at 250 μg/mL in comparison to that of the untreated group. Given the concentration of the fucoidan at 2000 μg/mL, the infection rate was further decreased to 40% in the Post-treatment group while it had no inhibitory activity in the other two groups (Fig. 3), suggesting that the fucoidan directly interacts with H. pylori.
The activity and mechanism of fucoidan in reducing H. pylori infection. Given that the fucoidan effects on the post-treatment group against the H. pylori infection, we hypothesized that the effectiveness of the fucoidan is attributed to the compromised adhesion of H. pylori to host cells. Therefore, AGS cells infected by H. pylori were collected and subjected to RT-PCR analysis to probe the mechanism behind this phenomenon. The analysis showed that the expression of 16s rRNA was significantly decreased in the cells added with the fucoidan (Fig. 4a). Because AGS cells infected by H. pylori generally trigger the release of pro-inflammatory cytokine Triple independent experiments were performed and values were calculated as mean ± SD for each group. Results were statistically analyzed using the Student's t-test (*p < 0.05, **p < 0.01, ***p < 0.001 compared with the LPS group without fucoidan). www.nature.com/scientificreports/ IL-8, the reduced production of IL-8 in the bacteria-infected AGS cells is thus ascribed to the treatment of the fucoidan (Fig. 4b). The outer membrane proteins (OMPs) AlpA and BabA of H. pylori have been characterized as adhesins or adherence-associated proteins. For this reason, their participation in the infection of H. pylori on AGS cells was analyzed. As expected, the fucoidan tended to inhibit the expression of AlpA and BabA in H. pylori (Fig. 4c,d). Thereby, the fucoidan was concluded an agent that is in a position to reduce the adhesion capability of H. pylori to host cells as well as to inhibit the induction of H. pylori-associated inflammation responses in host cells.
The effect of fucoidan on the total count of H. pylori in BALB/c mice. With these encouraging in vitro results, we took a further step to evaluate the treatment efficacy of the fucoidan on the H. pylori-infected mice. Our initial assay showed that H. pylori massed mainly in the stomach of the infected mice in agreement with empirical results. Concerning the H. pylori count, both the pre-treatment and post-treatment groups (Preand Post-Fu groups) were significantly lower than the Hp + group; the Post-Fu group was relatively better than the Pre-Fu group. The 16s rRNA of H. pylori is generally used as an indicative of its expression; as expected, the Hp + group had the highest expression level of 16s rRNA out of the three groups. The Post-Fu group likewise outperformed the Pre-Fu group in reducing the total count of H. pylori in mice (Fig. 5).
Histological changes and eosinophils infiltration in gastric tissues of infected mice. Next, we used the Sydney grading system to evaluate PMN infiltration, chronic inflammation, and hyperplasia in the H. pylori-infected mouse stomach tissue sections as shown in Fig. 6. The H & E staining showed that the muscularis mucosae in the infected group experienced noticeable cell infiltration as manifested by the irregular stomach mucosa layers gathered with a considerable number of immunocytes. In contrast, the pathological damage caused by H. pylori was significantly improved in the fucoidan treatment group based on the histopathological score. As shown in the Post-Fu group, the polymorphonuclear leukocytes (PMNs) activity, chronic inflammation, and hyperplasia scores all were lower than the corresponding ones in the Hp + group (Table 2). Added together, we concluded that the fucoidan can regulate the host immune responses by ameliorating some adverse symptoms upon the H. pylori infection, for example, the eosinophils infiltration.  www.nature.com/scientificreports/ Fucoidan reduced pro-inflammatory cytokines in mouse stomach. One now realizes that the Th2 cells tends to produce cytokines IL-4 and IL-5. To understand whether the fucoidan modulates the immune responses in the H. pylori-infected mice follows the Th2 pathway or not. We measured the expression level of IL-4 and IL-5 ( Fig. 7a,b). We found that the expression level of IL-4 made no significant differences amid the testing groups. The expression level of IL-5, however, was positively correlated with the severity of eosinophils infiltration, which was significantly reduced in the testing groups. IL-6, IL-1β, and TNF-α, on the other hand, are representatives of pro-inflammatory cytokines. The expression levels of IL-6, IL-1β, and TNF-α were subjected to cytokine analysis, of which they were drastically increased in the BALB/c mice infected with H. pylori for two weeks. In contrast, the expression level of IL-6 and TNF-α was significantly decreased in the infected mice treated with the fucoidan; interestingly, only was IL-1β significantly reduced in the Post-Fu group (Fig. 7c-e).

Discussion
It has now been well appreciated that the seaweed Sargassum hemiphyllum contains rich and diverse polysaccharides, not least fucoidan. Fucoidan features copious fucose as well as multiple sulfate modifications on its polysaccharide chain 18 . In this study, we demonstrated that the fucoidan derived from Sargassum hemiphyllum is capable of modulating the level of NO in vitro and in vivo agreeing with previous reports, where fucoidan was praised for its favorable anti-inflammatory activities 19,20 . Several reports independently pointed out that fucoidan can block selectins, thereby modulating inflammatory cells in tissues, particularly the infiltration and inflammation. These two syndromes are commonly manifested, when the immune system is compromised. Sulfated polysaccharides lately re-attract people's attention because some sulfated polysaccharides were reported to interfere the adhesion of H. pylori to macrophages or other host cells [21][22][23][24] , in that H. pylori interacts with host cells likely through the sulfate-polysaccharide-mediated receptors for binding to cell surface. The fucoidan from Cladosiphon okamuranus has been shown able to lower the total count of H. pylori in the human Table 2. Histopathological scores of stomach tissues in H. pylori-infected BALB/c mice according to Sydney grading system. Values were presented as mean ± SD. Values in the same column were statistically analyzed with Student's t-test (*p < 0.05, **p < 0.01 compared with the Hp + group).  www.nature.com/scientificreports/ stomach 13,25 . Given these facts, we were tempted and therefore very keen to explore the protective function and acting mechanism of the fucoidan from Sargassum hemiphyllum upon the H. pylori infection by the antiadhesion assay.

PMNs activity Chronic inflammation Hyperplasia
Helicobacter pylori is known to express adhesin BabA 26 , which commonly takes part in binding with host receptors and thereby enhancing the infection of H. pylori and leading to serious inflammation and eventual tumorigenesis 27 . AlpA is a lipoprotein, which exists in the cell membrane of H. pylori. It is an important adhesin adhering to the human stomach tissues. When overexpressed, AlpA would invoke strong oxidation pressure. Beyond that, AlpA can induce pro-inflammatory cytokines severely damaging the stomach tissues 28 . On the contrary, fucoidan is able to associate with BabA and AlpA thereby preventing H. pylori from adhering to host cells particularly the gastric epithelial cells 14,28 .
Hosts promptly respond to the early-onset infection of H. pylori by secreting a considerable amount of IL-8. IL-8 is an important immunological effector, which is deemed as an index of recurrent tendency of cancer because IL-8 stimulates angiogenesis, inflammation, and proliferation/translocation of epidermal cells 29,30 . At a high level of IL-8, the risk of contracting cancer is substantially increased. The onset of tissue lesions is often resulted from the recalcitrant and hard-to-remove H. pylori infection, which is further compounded by the persistent secretion of IL-8 14,31 . In this study, we identified that fucoidan can significantly reduce both the count of H. pylori and the level of IL-8 in the Post-Fu group.
Generally speaking, the higher the fucoidan level the better the efficiency in view of inhibiting infection and inflammation from the H. pylori infection 28 . To extend this in vitro phenomenon to in vivo efficacy, we arranged two groups pre-treatment and post-treatment groups by omitting the co-treatment group as the samples need to be mixed with H. pylori so that the condition will not stand on the same basis as do the other two groups. As expected, fucoidan exerts its effects reducing the count of H. pylori significantly in the stomach of the mice in the group of Post-Fu. Post-Fu relatively outweighed Pre-Fu, while the latter was still better than the control (infected group).
IL-6, IL-1β, and TNF-α are three typical pro-inflammatory cytokines 33,34 . They all are upregulated when infected with H. pylori. Among them, IL-6 activates the STAT3 pathway, of which the activation is correlated with the H. pylori-associated gastritis and gastric cancer 32,35,36 . Similarly, the H. pylori infection also increases the expression of IL-1β, which results in the decreased secretion of gastric acid. This phenomenon agrees with the report that given the increased production of IL-1β mouse would develop gastric cancer 31,37 . Moreover, a high gene expression level of the H. pylori 16s rRNA in feces is positively correlated with the TNF-α activity. A high level of TNF-α normally is indicative of chronic inflammation 37 . Although fucoidan did not reduce the level of IL-1β, it, however, significantly decreased the level of IL-6 and TNF-α in mice. Our results showed that the fucoidan improved the stomach inflammation-associated cytokine via the reduction of the H. pylori colonization in the stomach. Importantly, the fucoidan did possess the potential to lower the risk of the H. pylori-associated gastric ulcer or stomach cancer in vivo.
Based on our in vitro and in vivo results, fucoidan was fairly stable to maintain its structural integrity and sulfate residues in the stomach 38 to reduce the risk of the H. pylori infection as well as the expression of inflammatory effectors. The post-treatment regimen was more efficient than the pre-treatment in terms of effectiveness in vivo. By the same token, Back et al. 39 also demonstrated that the patients with the H. pylori infection receiving fucoidan supplement for four weeks showed improved the H. pylori infection. Mechanistically, fucoidan could intervene the adhesion process of H. pylori to AGS cells and inhibit the production of NO in Raw 264.7 cells. Despite such a superb biomedical prospect, fucoidan per se is currently not enough to clear H. pylori to complete extent in mice (Fig. 8). It, nevertheless, is more than capable of acting as an auxiliary substance, such as a medical supplement or a health care product, complementary to formal medical therapeutics.

Materials and methods
Preparation of the fucoidan. Sargassum hemiphyllum was collected from Badouzi Chaojing Park (25°08′31.5″ N 121°48′08.2″ E) in Keelung, Taiwan. The collection procedures for seaweed Sargassum hemiphyllum were reviewed and approved by the Department of Economic Affairs of Keelung City Government (Approval Number: KLCG-1050018424). We followed the guideline in collection of the specified plant material. Having been washed with clean freshwater (to remove impurities and salt), Sargassum hemiphyllum was dried using a cold-air dryer at 40 °C for 48 h. A ratio of 1:40 for the powder of Sargassum hemiphyllum (g) to ddH 2 O (mL) was prepared and then heated up to 100 °C for one hour. After cooling down, 10% cellulose APF and 2% proteinase NY 100 (Amano Enzyme Inc., Aichi, Japan) were simultaneously added in and reacted for another three hours in a shaking incubator. The solution was heated up again to inactivate proteinase NY 100. The solution was filtered out to remove impurities before centrifugation (5000×g, 4 °C, 10 min) to collect the supernatant. Fucoidan was obtained after freeze-drying.
Analytical method of polysaccharide. Molecular weight was measured by gel permeation chromatography (GPC) using the OHpak SB-804 HQ column (Shodex, Tokyo, Japan) 40 . The fucoidan was analyzed by the phenol-sulfuric acid method 41 to estimate the total sugar content. The sulfate content was determined by the turbidimetric method 42 ; the protein and total phenolic contents were measured by the Bradford assay 43  The adhesion test of H. pylori. First, AGS cells (1 × 10 5 cells/well) were cultured in 96-well plates overnight, the medium was removed next day, and then the cells were washed twice with PBS. After that, three protocols were followed: (1) Pre-treatment: a given concentration of fucoidan was added in AGS cells for 2 h, and washed three times with PBS. The H. pylori suspension was then added in to infect the AGS cells for 2 h. (2) Co-treatment: a given concentration of fucoidan and the H. pylori suspension were cultured at the same time for 2 h; the mixed medium was added in the AGS cells and then cultured for 2 h.  were purchased from National Laboratory Ani-mal Center (Taipei, Taiwan). All animal experimental protocols were reviewed and approved by the Institutional Animal Care and Use Committee (IACUC) of National Taiwan Ocean University (NTOU), and the study conformed to the guidelines of the protocol IACUC-101006 approved by the IACUC ethics committee of NTOU. This study was conducted in accordance with the ARRIVE guidelines. Twenty-four mice were divided into four groups, which were respectively designated as Hp − , Hp + , pre-treatment group (Pre-Fu), and post-treatment group (Post-Fu), six mice for each group. Before proceeding with the treatment, mice were acclimated to the new diet and environment for one week. Mice were challenged by H. pylori (1 × 10 9 CFU/mL) on Day 0 and 2. In Pre-Fu group, before mice were challenged by H. pylori, mice were treated 800 mg/kg/day fucoidan for one week; the protocol was continued till sacrifice. On the other hand, after mice were challenged by H. pylori for one week, mice were treated 800 mg/kg/day fucoidan in Post-Fu group. Hp − and Hp + groups were fed with deionized water. At the end of the experiment, the mice were sacrificed and the stomachs were removed under an aseptic environment. 50 μL gastric homogenate was serially diluted and spread on a selective medium (egg yolk emulsion agar, Creative Microbiologicals, Taiwan), and incubated at 37 °C for 3-5 days in a microaerobic environment. When H. pylori colonies grew, the number of colonies (CFU/ mL) in red was counted.

Detection of H. pylori and cytokine RNA expression in AGS cells and mouse gastric tissues.
The H. pylori infected cells were added with fucoidan and incubated for 6 h. The cells were washed with PBS, and the cell pellets were collected to extract RNA. For the in vivo experiment, the stomach tissue was trimmed between 25 to 30 mg to extract the total RNA with the RNeasy mini kit (QIAGEN, Germantown, USA). The RNA primers for 16s rRNA, BabA, AlpA, IL-8, IL-4, IL-5, IL-6, IL-1β, and TNF-α were prepared, of which the sequences were listed in Table 3. RT-PCR was set at 95 °C for 30 s, followed by 40 cycles of 95 °C for 15 s, 55 °C for 15 s, 72 °C for 45 s. The gene expression level for each sample was normalized using GAPDH.
Histopathological analysis. After the mice were sacrificed, the stomach tissue was sampled from each mouse. The samples were fixed in 10% formalin for 24 h at room temperature. The fixed stomach tissue was trimmed into an appropriate size; and paraffin sections were cut at 4 μm. The sections were stained with hematoxylin-eosin (H&E) and Giemsa for observing the presence of inflammatory responses in the muscularis mucosae and eosinophil granulocytes in the specimen. Microscopic observations were carried out at 400 × mag-Abs 560nm sample Abs 560nm blank × 100% Table 3. Primer name and sequences for RT-PCR analysis.