In Vitro Evaluation of the Phytopharmacological Potential of Sargassum incisifolium for the Treatment of Inflammatory Bowel Diseases

Background: Comprised of Crohn’s disease and ulcerative colitis, inflammatory bowel diseases (IBD) are characterized by chronic inflammation of the gastro-intestinal tract, which often results in severe damage to the intestinal mucosa. This study investigated metabolites from the South African endemic alga, Sargassum incisifolium, as potential treatments for IBD. Phytochemical evaluation of S. incisifolium yielded prenylated toluhydroquinones and toluquinones, from which semi-synthetic analogs were derived, and a carotenoid metabolite. The bioactivities of S. incisifolium fractions, natural products, and semi-synthetic derivatives were evaluated using various in vitro assays. Methods: Sargahydroquinoic acid isolated from S. incisifolium was converted to several structural derivatives by semi-synthetic modification. Potential modulation of IBD by S. incisifolium crude fractions, natural compounds, and sargahydroquinoic acid analogs was evaluated through in vitro anti-inflammatory activity, anti-oxidant activity, cytotoxicity against HT-29 and Caco-2 colorectal cancer cells, and PPAR-γ activation. Results: Sargahydroquinoic acid acts on various therapeutic targets relevant to IBD treatment. Conclusions: Conversion of sargahydroquinoic acid to sarganaphthoquinoic acid increases peroxisome proliferator activated receptor gamma (PPAR-γ) activity, compromises anti-oxidant activity, and has no effect on cytotoxicity against the tested cell lines.


Introduction
The incidence of the two major types of inflammatory bowel diseases (IBDs), Crohn's disease (CD) and ulcerative colitis (UC), has become a global health challenge. A systematic review of studies reporting the prevalence and incidence of IBDs, performed by Ng et al., revealed that while the incidence has stabilised in the westernised world, it has steadily been increasing in developing countries over the past decade or two [1]. CD and UC are characterised by chronic inflammation of the intestine with many associated symptoms, complications, and an increased risk for colorectal cancer [2]. Conventional treatment is aimed at reducing intestinal inflammation and modulating the immune system. The most commonly used treatments are aminosalicylate anti-inflammatories (5-ASA,  (2), fucoxanthin (3) and sargaquinal (4), and semi-synthetic derivatives sarganaphthoquinoic acid (5) and sargachromenoic acid (6).

Reagents
Culture mediums were sourced from Sigma Aldrich® (Johannesburg, South Africa) and Hyclone® (Thermo Fisher, Logan, UT, USA) while Fetal Bovine Serum (FBS) was obtained from LONZA® (Basel, Switzerland). Chang Liver cells (HeLa derivative) were purchased from Highveld Biologicals, Johannesburg, South Africa and HT29 and Caco2 colorectal carcinoma cell lines from the American Type Culture Collection (Manassas, VA, USA). The EC50 values of the test compounds were calculated from a minimum 5-point dose-response curve using a GraphPad Prism 4 software package (GraphPad, San Diego, CA, USA). Liquid chromatography utilised HPLC grade solvents supplied by Lichrosolv® (Merck, Germany). NMR experiments were obtained on a Bruker Avance 400 MHz NMR spectrometer (Bruker Corporation, Billerica, MA, USA) using standard pulse sequences. All HPLC solvents were filtered through a 0.45 μm filter before use. Normal phase HPLC was performed using a Spectra-Physics IsoChrom pump (Spectra-Physics, Santa Clara, CA, USA), a Whatman® Partisil 10 (9.5 mm × 500 mm) semi-preparative column (GE healthcare, Chicago, IL, USA) and a Waters 410 differential refractometer (Waters Corporation, Milford, MA, USA) attached to a 100 mV full scale Rikadenki chart recorder (Rikadenki Electronics GmbH, Freiburg im Breisgau, Germany).

Algal Material
The algal specimen of S. incisifolium (collection voucher NDK101124) was collected from Noordhoek, near Port Elizabeth, on the southeast coast of South Africa on 24 November 2010. A specimen ( Figure S1) is kept in the seaweed collection at the School of Pharmacy, University of the Western Cape. The algal specimen was transported to the laboratory on ice where it was immediately frozen and stored until the time of extraction. For purposes of identification and authentication, the algal material was morphologically compared with previous voucher specimens of S. incisifolium. A voucher specimen (NDK06-5) is kept at the Division of Pharmaceutical Chemistry, Rhodes University, Makhanda, South Africa.

Extraction and Isolation of Bioactive Metabolites
The algal extraction procedure was consistent with previously reported methods [16]. The frozen alga (NDK101124) was allowed to defrost under running distilled water. The defrosted alga was then soaked in MeOH for 1 h, after which the MeOH was decanted and the retained algae heated at 40 °C  (4), and semi-synthetic derivatives sarganaphthoquinoic acid (5) and sargachromenoic acid (6).

Reagents
Culture mediums were sourced from Sigma Aldrich ® (Johannesburg, South Africa) and Hyclone ® (Thermo Fisher, Logan, UT, USA) while Fetal Bovine Serum (FBS) was obtained from LONZA ® (Basel, Switzerland). Chang Liver cells (HeLa derivative) were purchased from Highveld Biologicals, Johannesburg, South Africa and HT29 and Caco2 colorectal carcinoma cell lines from the American Type Culture Collection (Manassas, VA, USA). The EC 50 values of the test compounds were calculated from a minimum 5-point dose-response curve using a GraphPad Prism 4 software package (GraphPad, San Diego, CA, USA). Liquid chromatography utilised HPLC grade solvents supplied by Lichrosolv ® (Merck, Germany). NMR experiments were obtained on a Bruker Avance 400 MHz NMR spectrometer (Bruker Corporation, Billerica, MA, USA) using standard pulse sequences. All HPLC solvents were filtered through a 0.45 µm filter before use. Normal phase HPLC was performed using a Spectra-Physics IsoChrom pump (Spectra-Physics, Santa Clara, CA, USA), a Whatman ® Partisil 10 (9.5 mm × 500 mm) semi-preparative column (GE healthcare, Chicago, IL, USA) and a Waters 410 differential refractometer (Waters Corporation, Milford, MA, USA) attached to a 100 mV full scale Rikadenki chart recorder (Rikadenki Electronics GmbH, Freiburg im Breisgau, Germany).

Algal Material
The algal specimen of S. incisifolium (collection voucher NDK101124) was collected from Noordhoek, near Port Elizabeth, on the southeast coast of South Africa on 24 November 2010. A specimen ( Figure S1) is kept in the seaweed collection at the School of Pharmacy, University of the Western Cape. The algal specimen was transported to the laboratory on ice where it was immediately frozen and stored until the time of extraction. For purposes of identification and authentication, the algal material was morphologically compared with previous voucher specimens of S. incisifolium. A voucher specimen (NDK06-5) is kept at the Division of Pharmaceutical Chemistry, Rhodes University, Makhanda, South Africa.

Anti-Inflammatory Assay
The murine peritoneal macrophage cells (RAW267.4) were cultured in DMEM containing 10 % FCS. Cells were seeded into 96 well plates at a density of 8 × 10 4 cells/well and allowed to attach overnight. The cells were then treated with 1 µg/mL of bacterial lipopolysaccharide (LPS) (SIGMA ® ) and two concentrations of the test sample (12.5 and 25 µg/mL) for 18 h. To measure nitrate levels, 50 µL of the spent culture medium was removed and added to an equal volume of Griess reagent (SIGMA ® ). The absorbance was measured at 540 nm using a microplate reader and the nitrate concentrations were calculated by comparison with the absorbance to sodium nitrate standard solutions. Aminogaunidine (Sigma ® ) was used as positive control to demonstrate the inhibition of nitrate production. Cell viability was simultaneously measured using the standard MTT assay.

2,2-diphenyl-1-picrylhydrazyl (DPPH) Radical Scavenging Assay
Test samples were diluted in EtOH/H 2 O (1:1) from 10 mg/100 µL stocks prepared in DMSO. A total of 5 µL of each sample was placed into each well of a 96-well plate, followed by the addition of 120 µL of Tris-HCl buffer (50 mM, pH7.4) and 120 µL of freshly prepared DPPH solution (0.1 mM in EtOH). The plate was incubated for 20 min at room temperature, with the absorbance read at 513 nm. The percentage of DPPH radical scavenging was calculated as ((A − B/A) × 100) where A represents the absorbance in the absence of test samples and B represents the absorbance in the presence of test samples. Ascorbic acid was used as a positive control (EC 50 = 24.07 µg/mL).
HeLa derivative cells were seeded into 96-well culture plates (TTP) at 10,000 cells/well in EMEM supplemented with 10% fetal bovine serum (FBS) and left for 24 h. Algal extracts and compounds were added and the cells incubated for a further 48 h after which the medium was replaced with 200 µL of MTT (Sigma ® ) (0.5 mg/mL in EMEM). After a further 2 h of incubation at 37 • C, the MTT was removed and the purple formazan product dissolved in 200 µL of DMSO.
Absorbance was measured at 560 nm using a multiwell scanning spectrophotometer (Multiscan MS, Labsystems). All incubation steps were carried out in a 37 • C humidified incubator with 5% CO 2 . IC 50 and EC 50 values were calculated from a minimum 5-point dose-response curves using the GraphPad Prism 4 software package.

3T3-L1 Preadipocyte Differentiation Assay
Prior to the induction of differentiation, 3T3-L1 cells were routinely maintained in DMEM containing newborn calf serum. Cells were seeded at a density of 3000 cells/well into 96-well plates and allowed to reach 100% confluence. Two days post-confluence, the cells were treated for a further two days with DMEM medium, now supplemented with FBS (to induce mitotic clonal expansion) and the indicated concentrations of test compounds or the control substances rosiglitazone and troglitazone (1 µM, final concentration). Cells were then cultured for an additional 7 days in normal culture medium (DMEM, 10% FBS with inducers) and the medium replaced every two to three days. Triglyceride accumulation, a marker for adipocyte differentiation, was measured by Oil red-O staining. The Oil Red-O stained lipids were extracted in isopropanol and measured at 510 nm. The sample results were then compared to controls using a two-tailed Student's t-test assuming equal variances.

Anti-Inflammatory Potential of S. incisifolium
S. incisifolium fractions were evaluated for anti-inflammatory activity. Fractions Fr C, Fr D (SHQA, 1), and Fr F (fucoxanthin, 3) produced a significant decrease in LPS-stimulated nitrate production at both test concentrations, with Fr C being relatively less potent by only having a significant effect at the highest test concentration ( Figure 2). SHQA (1) significantly attenuated nitrate production, indicating that this compound may also be considered to possess anti-inflammatory properties. Under the conditions of the anti-inflammatory assay there was no evidence for cytotoxicity toward the RAW 267.4 cells and, thus, it can be assumed that the inhibition of nitrate production was not due to differences in the relative cytotoxicity. In naïve cells, i.e., in the absence of LPS, no response was induced by any of the samples.

Cytotoxicity of S. incisifolium Fractions Against HT-29 and Caco-2 Cancer Cells
Cytotoxicity towards HT-29 and Caco-2 colorectal cancer cell lines was evaluated for S. incisifolium extracts. The most cytotoxic S. incisifolium fractions yielded IC 50 values < 50 µg/mL on HT-29 and Caco-2 cells (Table 2). This is consistent with studies that have been done to show the selectivity of Sargassum extracts towards HT-29 and Caco-2 cell apoptosis [21]. Fr F (fucoxanthin, 3) was the most potent sample tested. Fr D, (SHQA, 1) was found to be more selective towards Caco-2 than HT-29 cells (SI = 1.97). Taken together these results indicate the presence of multiple cytotoxic compounds in the S. incisifolium extract. Dose-response curves showing the inhibition of HeLa ( Figure S10), HT-29 ( Figure S11) and Caco-2 ( Figure S12) cell proliferation by FrA-I are provided in the Supplementary Materials.
Considering that inflammation represents the predominant therapeutic target against IBD [3], the anti-inflammatory potential of the fractions was investigated using LPS activated RAW 264.7 macrophages as a model. Both SHQA (1) and fucoxanthin (3) revealed a concentration-dependent inhibition in nitrate production, indicating potential anti-inflammatory activity. Concurrent cell viability measurements confirmed that these effects were not due to cytotoxicity. The anti-inflammatory activity of fucoxanthin (3) is in accord with previous studies [20]. Similarly, research on the anti-inflammatory activity of plastoquinone derivatives isolated from natural sources also reported that SHQA (1) significantly reduced TPA-induced mouse ear oedema [23]. The precise mechanism through which SHQA (1) exerts this anti-inflammatory response however awaits further studies. Our results suggest that SHQA (1) may, at least in part, directly target macrophage function. In contrast to the anti-inflammatory activity, treatment of naïve RAW 264.7 cells did not reveal any potential risk to induce a pro-inflammatory effect, which could exacerbate inflammation.
Oxidative stress is currently considered as a contributory factor in the initiation, progression, and severity of IBD and is regarded as more than just a simple consequence of chronic inflammation associated with the disease. Although the underlying mechanisms are yet to be thoroughly elucidated, strategies to reduce oxidative stress are anticipated to improve therapeutic outcome. Antioxidants, especially those derived from natural products, have attracted attention as acceptable ingredients to target oxidative stress in IBD [4]. Compounds with dual anti-inflammatory and antioxidant activity may be particularly relevant to the treatment of IBD. The DPPH radical scavenging activity of SHQA (1) and SCA (6) has been previously documented [23]. Reduced forms of vitamin E and coenzyme Q groups, such as hydroquinones, chromanols, and chromenols normally function as protective anti-oxidants. However, the ability of such compounds to function in these capacities of electron transfer and antioxidant activity directly depend upon the oxidation potential of the compound, which is also partly dependent upon the nuclear substituents [24]. Not surprisingly, SHQA (1) showed more potent DPPH radical scavenging activity than SQA (2, EC 50 = 95.76 µg/mL), as it is generally known that hydroquinones are more potent radical scavengers than their quinone congeners. We, therefore, identify SHQA (1) as an anti-inflammatory compound with potent radical scavenging activity. In the DPPH assay, SHQA (1) was greater than 5-fold more active relative to the standard antioxidant, ascorbic acid.
It is well recognised that patients with IBD show a higher incidence of developing colon cancer [4], primarily believed to be the result of chronic intestinal inflammation. Subsequently, many IBD patients also develop the requirement for cancer therapy, which is accompanied by unique challenges associated with this comorbidity. Often chemotherapeutic drugs damage the intestine, resulting in the remission of IBD. It thus follows that although cytotoxicity towards cancer cells may have an advantage in cancer treatment, it is also at risk of aggravating intestinal inflammation and thus, IBD. Evaluation of the fractions obtained from S. incisifolium revealed significant toxicity towards colon cancer cell lines Caco-2 and HT-29 with fucoxanthin (3) being the most potent. This is consistent with previous studies which report fucoxanthin (3) to inhibit the proliferation of HT-29 and Caco-2 cells through inducing cell cycle arrest in the Go/G1 phase at low concentrations (25 µM) and apoptosis at higher concentrations [25]. SHQA (1) was also significantly cytotoxic towards these colon cancer cells.
The adipocyte has been described as "a dynamic cell that plays a fundamental role in energy balance and overall body homeostasis" [26]. The formation of adipocytes (adipogenesis) is a differentiation process governed by transcriptional cascades involving a regulated set of gene expression events [22]. The peroxisome proliferator-activated receptor gamma (PPAR-γ) has been termed as the 'master regulator' of adipogenesis sufficient to differentiate fibroblasts into mature adipocytes [27]. The PPAR-γ is therefore not only crucial for adipogenesis but is also a requirement for the maintenance of the differentiated state. Hence, a compound or extract that stimulates the differentiation of preadipocytes into mature adipocytes is considered as a PPAR-γ agonist. The PPAR-γ is predominantly expressed in adipose tissue with lower levels of expression in other tissues, such as cardiac, renal, and hepatic tissues [28]. The association of PPAR-γ activation and consequent adipogenesis with an increase in tissue insulin sensitivity provided the basis for the development of thiazolidinediones as a class of anti-diabetic drugs. Other known PPAR isoforms include α and β/δ, for which dual and pan agonists have been identified.
The high relative expression of PPAR-γ in the colon has stimulated many studies on the role of PPAR-γ in gut health. While early studies focused heavily on the involvement of PPAR-γ in the process of colonic tumor suppression, more recently, research has expanded to include intestinal inflammation and fibrosis, major factors in the pathogenesis of IBD. Identification of the direct involvement of PPAR-γ in the mechanism of action of mesalazine, a clinically effective drug often used to treat ulcerative colitis, has highlighted the anti-inflammatory role of PPAR-γ and renewed the search for novel PPAR-γ agonists to treat IBD [9]. Fibrosis, excessive deposition of extracellular matrix components including collagen, is a common complication of IBD, leading to obstruction and loss of function of the intestine. PPAR-γ agonists can diminish fibrogenesis through the antagonist effects on TGF signalling. Given that the anti-inflammatory drugs currently used to treat IBD are unable to attenuate intestinal fibrosis, new therapeutic approaches are sought with PPAR-γ agonists holding considerable promise. Taken together, it is clear that PPAR-γ has again emerged as an important therapeutic target for the development of new drugs to treat IBD.
The structural derivatives were also evaluated for antioxidant activity and cytotoxicity against HeLa derivative cells. SNQA (5) showed a dramatic decrease in radical scavenging activity while SCA (6) antioxidant activity remained essentially unchanged. Cytotoxicity towards HeLa derivative cells, a cell line previously shown to be devoid of PPAR-γ protein [30], was unchanged relative to SHQA (1).
To our knowledge, this is the first report on the PPAR-γ-mediated activity of SNQA (5). In an attempt to improve the side effect profiles of current PPAR-γ agonists, research has explored the replacement of the thiazolidine ring with other 'acidic head groups', which have lesser side effects. Such examples include a study performed by Sundriyal et al. in which, after replacement of the thiazolidine ring with a 1,4-naphthoquinone moiety, the newly synthesized compounds still retained PPAR-γ activity comparable to pioglitazone [31]. This shows that the 1,4-naphthoquinone, SNQA (5), is a potential PPAR-γ agonist similar to the well-known thiazolidinediones and supports our findings and potential for the treatment of IBD. Additionally, 1,4-naphthoquinones are commercially available, less costly, and easily derivatized.

Conclusions
SHQA (1) is identified as a promising lead compound due to its effects on multiple therapeutic targets relevant to IBD. Derivatization to SNQA (5) significantly improved the PPAR-activity. However, this dramatically reduced its antioxidant activity and had minimal effect on cytotoxicity.

Conflicts of Interest:
The authors declare no conflict of interest.