Agaro-oliogsaccharides were prepared by acid hydrolysis. In order to evaluate the difference of DP of oligosaccharides on bioactivity, hydrolysis solution was fractionated by activated carbon column. After loading the hydrolysate onto column, the column was washed with 2 liters water to remove salts and monosaccharides. Followed this step, the agaro-oligosaccharides fraction was eluted sequentially with 8 %, 15 % and 25 % hydroalcoholic solution. Each fraction from the column was concentrated under reduced pressure and lyophilized.

The average molecular weight of three fractions was measured as described by Somogyi et al. 14.

The nuclear magnetic resonance (NMR) spectra were acquired on an AVANCEDMX-500-NMR spectrometer. Samples were dissolved in D₂O. ¹³C NMR spectra of 4% (w/v) solutions were recorded at 35°C under 100.69 MHz. Proton decoupled ¹³C NMR chemical shifts were measured in parts per million. For ¹H-NMR, samples (7–10 mg) were dissolved in D₂O (0.5 ml), and spectra were recorded at room temperature using a spectral width of 5.7 kHz, 90° pulse, an acquisition time of 4.4 s for 144 scans.

Mass spectrometry analysis was performed on a Bruker Reflex III MALDI-TOF-MS (Bruker-Daltonik, Germany) in the delayed extraction and positive mode. An accelerating voltage and a reflectron voltage were set at 20 kV of 22.8 kV, respectively, during the measurements. 2, 5-Dihydroxybenzoic acid was used as matrix (20 mg/ml; 3:2 water/MeCN) and approximately 10–100 pg of the DP-H agaro-oligosaccharide mixture was deposited as a mixture together with the matrix on a stainless steel target, and subsequently dried under reduced pressure. During the experiments, the laser power was adjusted to a level just above the threshold for formation of observable ions. The results from 20 to 100 laser shots were summed for sample.

Intracellular oxidant stress was monitored by measuring changes in fluorescence resulting from intracellular probe oxidation.

Human hepatocyte L-02 purchased from Chinese Institute of Biochemistry and Cell Biology was cultured in RPMl-1640 medium with 20 % fetal bovine serum. Viable cells (10⁵/ml) were plated into a 96-well for 1 day. On the day of the experiments, after removing the medium, the cells were washed with PBS for three times and then incubated with different doses of agaro-oligosaccharides in 5 % CO₂ at 37°C for 2 h. After incubation, 20 μM DCFH-DA was added for another 45 min. The DCFH-DA was removed by washing the cells with PBS. 100 μM H₂O₂ were added into cells for 45 min and the fluorescence change was monitored by fluorescence spectorphotometer at λ_ex = 475 nm, λ_em = 525 nm 15.

The cell viability was quantified using MTT assay. Briefly, 1 × 10⁴ cells were seeded in each well of microtiter plate and allowed to attach overnight. Cells were treated with various doses of agaro-oligosaccharides for different period according to the experiment purpose. For cytotoxicity test, the hepatocyte L-02 was treated for 48 h. But for the detection of protective effect of agaro-oligosaccharides on H₂O₂ damage, the L-02 was only treated for 2 h, and then 100 μM H₂O₂ was added for another 2 h. MTT in PBS was added to each well, followed by incubation for 4 h at 37°C. The formazan crystals were dissolved in DMSO. The optical density was determined with a microculture plate reader at 492 nm 16.

Mature Wistar rats weighing 150 ± 20 g were supplied by the animal center of Hangzhou, China. The animals were housed in a room with a 12 h light/dark cycle at about 22°C and fed on standard diet with ad libitum access to drinking water. All treatments were conducted between 9:00 am and 10:00 am to minimize variations. In this study, rats were randomly divided into six groups. Group 1 (control, n = 8): water for 10 days followed by administration of liquid paraffin only; group 2 (CCl₄, n = 8): water for 10 days followed by administration of CCl₄ on the final day; group 3 (positive control, n = 8): vitamin C (200 mg/kg) + CCl₄; group 4 to 6 (n = 8): agaro-oligosaccharides (200, 400, 600 mg/kg, respectively) + CCl₄. Rats were injected i.p. with vitamin C or agaro-oligosaccharides for ten consecutive days. On the final day, all animal except control group were administered with 20 % CCl₄ in liquid paraffin at a dose 5 ml/kg to induce hepatotoxicity. Previous studies demonstrated that the OS indexes could reach a maximum at 48 h after CCl₄ i.p. administration 17, therefore, in this work rats were sacrificed by collecting the blood from the carotid artery after 48 h of administration. Two organs (liver and heart) were excised immediately.

Serum was separated by centrifugation at 1000 × g at 4°C for 10 min. 10 % organ homogenates including liver and heart were prepared in ice-cold isotonic physiological saline. The GSH-Px, MDA, SOD, AST and ALT levels of tissue and serum were measured by spectrophotometric methods as described in the assay kits.

All data are expressed as mean ± SD. In cell based assay, the control and agaro-oligosaccharides treated cells were compared by student t-test. In animal assay, the statistical tests were one-way ANOVA followed by post-hoc Newman-Keuls multiple comparisons test. A probability level of 0.05 was considered statistically significant.

Activated charcoal column has been performed as saccharide isolation tool for decades. Depending on this technology, we successfully achieved to isolate three fractions of agaro-oligosaccharides with average molecular weight of 619, 1126 and 1631, respectively, eluted by 8 %, 15 % and 25 % aqueous alcoholic solution. We use these three fractions for the following experiments, designated as DP-L, DP-M and DP-H, according to their differences in molecular weight.

Since agar is a linear copolymer of galactose (G), alternated with 3, 6-anhydrogalactose (A), the structural difference of agaro-oligosaccharides are mainly related with the degree of polymerization. In this report, the ¹H-NMR and ¹³C-NMR spectra of agaro-oligosaccharides was studied using acid hydrolyzed fragments, and typical deshielded ¹H-NMR and ¹³C-NMR signals corresponding to the anomeric hydrogens and carbons were obtained and presented in Fig. 1. Assignments were based on the close similarity with literature values, and the interpretation of these signals was indicated in Table 1. The spectra give out twelve distinctive major anomeric carbon signals which were expected for the major disaccharide repeat unit. The presence of these signals demonstrates the presence of floridean starch in this fraction, because all of the signals illustrated the galactose ring structures present in seaweed galactans 1819. The ¹³C NMR spectrum of oligosaccharide are very consistent with those previously published for neoagarose series, with chemical shifts of carbons of unit G'-1α and G'-1β appeared at identically 92.4 and 96.4 ppm, having intensities in the ratio of 1 : 2. While, from the result, we didn't observe any signal for 3, 6-anhydrogalactose at the non reducing ending (a peak at 91.4 ppm) 202122. These results reflect the presence of galactose units at the reducing ends of the reaction products, but no 3, 6-anhydrogalactose at non-reducing end.

We applied MALDI-TOF-MS in order to know the structural information of composition and DP in the agaro-oligosaccharide fractions which were obtained from hydrolysis. The results shown in Fig. 2 indicated that a large number of well-regulated peaks are present, and these agaro-oligosaccharide ions could be identified as series of sodium molecular ions with relatively high intensities corresponded to 509, 815, 1121, and so on. It can be seen, by comparing these ions, that the molecular mass difference between every two adjacent ion is the same as 306 Da. This molecular mass difference of 306 Da is the exact molecular mass of agaro-biose (GA), the basic structural unit, which is 324 Da, minus 18, which is the number of H₂O's molecular weight, therefore, it is clearly observed that the agaro-oligosaccharides had very regular molecular structures with gradient increase of its chain length with the polymerization unit of agarobiose. Further calculation for m/z 509, the lowest high intensity ion observed in the mass spectrum, found that m/z 509 corresponds to the sodium adduct of agarotriose (GAG) [M_tri+Na]⁺. Based on this information, the ion at m/z 815, 1121, 1427.... corresponds to agaro-oligosaccharides for n = 5, 7, 9..., respectively with galactose at both reducing end and non-reducing end. For agaro-oligosaccharides, two forms of saccharides exist depending on the end sugar moiety, namely, neoagaro-series with 3, 6-anhydro-galactose at the non-reducing end and agaro-series with galactose at the non-reducing end. The results obtained here indicated that our sample obtained belong to agaro-series with odd numbers of sugar unit.

We firstly investigated the antioxidant activities of agaro-oligosaccharides in the cellular system. DCFH-DA, which can be conversed from non-fluorescence into fluorescence through oxidation, was used as fluorescent probe to monitor the changes of oxidative stress in hepatocyte L-02 induced by addition of H₂O₂. In our experiment, all the measurements were carried out at the steady stage (incubation time, 60 min) in order to minimize variations, because it has been reported that treatment of H₂O₂ will lead to the abruption of ROS in few minutes, and then decrease to a steady stage 23.

Fig. 3 showed that addition of agaro-oligosaccharides caused concentration-dependent attenuation of DCF fluorescence. Three groups of agaro-oligosaccharides showed almost no inhibitory activity at the 125 μg/ml. When the concentration increased, DP-H expressed highest activity, followed by DP-M, much weaker for DP-L group, which indicating that the antioxidant bioactivity in vitro improves with the higher degree of polymerization of agaro-oligosaccharide. Fig. 4 is a typical fluorescent microscopic picture of the DCF fluorescence in hepatocyte L-02 treated with DP-H. It is obvious that H₂O₂ lead to the production of ROS, which transformed the DCFH into DCF (Fig. 4D), showing more fluorescent cells than untreated cells (Fig. 4A). DP-H additions decreased the free radical formation. Fig. 4B clearly illustrated that DP-H at concentration of 1 mg/ml could inhibit the oxidation of DCFH significantly. While with the concentration decreased to 125 μg/ml (Fig. 4C), the number of fluorescent cells was also increased, and which means that the antioxidant activity of agaro-oligosaccharides acts in a concentration-dependent manner.

Oxidative stress is an important factor to induce the cell death. Cell viability assay showed that the presence of H₂O₂ (100 μM) resulted in cell death ratio increasing to 60 % after 2 h of treatment (Fig. 5). Compared to H₂O₂ alone, cell death was reduced obviously when exposed to each agaro-oligosaccharide group at the higher concentrations (from 500 μg/ml to 1 mg/ml). The cell viability significantly increased to 64.26 % for DP-M treated cells at the concentration of 1 mg/ml (Fig. 5). At low concentrations (125 μg/ml to 500 μg/ml), there was almost no variation observed between the agaro-oligosaccharide treated cells and the control, except DP-M treated group showing some weak cell protective effect. These results demonstrated that the antioxidant activities of agaro-oligosaccharides were positively correlated with the improvement of the cell viability.

In order to test whether agaro-oligosaccharides affected the growth of human hepatocyte L-02 without H₂O₂ treatment, cell proliferation was assessed by direct MTT assay. The cells were incubated with various amounts of agaro-oligosaccharides for 48 h, and the change in cell number was determined by analyzing the values of cells treated with agaro-oligosaccharide versus that of control (Fig. 6). From result, we found that compared with control group, the agaro-oligosaccharides exhibited very slight effects on the cell growth. After 48 h of treatment, the growth is slightly inhibited as of 14.18 % for DP-H at 1 mM, while for DP-M and DP-H, the corresponding cell proliferation ratio was >100 % with concentration ≤ 250 μM, which means that, at proper concentration, the agaro-oligosaccharides can promote the proliferation of L-02 cells. Therefore, the cell survival effect of agaro-oligosaccharide alone in the antioxidation cellular assay can be considered almost naught because the cells were only treated for 2 h, so we concluded that agaro-oligosaccharides can effectively protect the cells from oxidation induced death through scavenging intracellular oxidative damage induced by ROS.

We further studied the in vivo antioxidant effects of agaro-oligosaccharides. It was not uncommon that compounds possessing in vitro activity, however, fail to maintain the activity when administrated into body. We established an oxidative animal model by CCl₄ injection. Considering the proliferation and antioxidant effects of agaro-oligosaccharides on hepatocyte, we used the mixture of DP-M and DP-H as our sample for animal test. The effects of agaro-oligosaccharides on oxidative stress in rats were estimated by determining the activities of MDA, SOD, GSH-Px, ALT and AST in serum and tissues.

MDA level is a main marker of endogenous lipid peroxidation 24. In CCl₄ treated group, the MDA level increased significantly in liver (F = 2.087, P < 0.05), but little difference was observed in serum, which confirmed that the toxicity of CCl₄ is focused in the liver. By contrast, MDA level in the agaro-oligosaccharides treated groups decreased significantly compared with CCl₄ treated group. At 400 mg/kg, the MDA level reduced at least 44 % and 21 % in liver (F = 4.274, P < 0.05) and heart, respectively, versus the CCl₄ treated group. Actually, the MDA level of agaro-oligosaccharides treated groups showed almost the same as the blank control group (Table 2). It provided the information that exhibiting a very successful block of lipid oxidation.

SOD and GSH-Px are intracellular antioxidant enzymes that protect against oxidative process 25. As show in Table 3 and 4, a single high dose injection of CCl₄ induced severe oxidative damage and the SOD and GSH-Px level decreased markedly. While various concentrations of agaro-oligosaccharides could effectively normalize the enzyme activities and the two indexes were even higher than Vitamin C group. In liver and serum, the SOD level reached to highest at 400 mg/kg (F = 3.878, P < 0.05; F = 9.363, P < 0.05). Similar results were obtained in case of the GSH-Px activities.

Serum levels of transaminases (ALT, AST) were used as indicators to evaluate the attribution of agaro-oligosaccharides to the structure damage of the liver 2627. In this experiment, the enzyme assays of serum transaminases showed that a toxic dose of CCl₄ significantly raised the levels of ALT and AST to 687 (F = 3.761, P < 0.05) and 415 U/l (F = 4.204, P < 0.05). Agaro-oligosaccharides could inhibit the enzyme activities effectively. The ALT level reached to minimum when the sample concentration was 400 mg/kg (22.16 % less than the control group). For AST, the agaro-oligosaccharides reduce it in a dose dependent manner. At the highest concentration (600 mg/kg), AST level decreased to 222 U/l. However, it is strange to find that Vitamin C didn't reduce AST but raised it to 32 % versus the control without CCl₄ treatment (Fig. 7).