Chain conformation and bioactivity of water-soluble polysaccharide extracted from Rhizoma Panacis Japonici

Materials

RPJ was purchased from Enshi (Hubei Province, China). 5-Fluorouracil (5-FU) was purchased from Sigma-Aldrich (China). 3-(4,5)-Dimethylthiazol-2,5-di-phenytetrazolium bromide (MTT) was purchased from Invitrogen Corporation (USA). All other chemical reagents were of analytical grade and commercially available. They were used as received.

Preparation of RPS3 and Fractions

Water-soluble polysaccharide RPS3 was prepared as Refs.7 and16. It was fractionated by addition of methanol as precipitant at room temperature according to the nonsolvent addition method. Briefly, RPS3 was first dissolved in water at concentration of 2% and stirred for 24 h to obtain transparent solution, and then methanol was added into the RPS3 solution dropwise under stirring at 25°C until the solution turned milky. The turbid mixture was centrifuged at 9000 rpm and the precipitated product was washed by methanol three times to obtain the first fraction. The supernate was subjected to the next step of fractionation by adding more volumes of methanol. In this way, the RPS3 was divided into six fractions coded as RPS3-1, RPS3-2, RPS3-3, RPS3-4, RPS3-5, and RPS3-6. Each fraction was dissolved in water again and precipitated by additional methanol repeatedly. The fractions were finally vacuum dried to obtain white powder.

Characterization of Molecular Parameters

GC-MS total ion current (TIC) chromatograms of methylation analyses and ¹³C NMR spectra of the fractions were performed as described in Ref.7. GC-MS was carried out on a GCT system (Macro-Mass, Waters, USA) with a capillary GC column (HP-5MS, 30 m × 0.32 mm ID × 0.25 μm, Agilent, USA) using helium (He) as carrier gas. ¹³C NMR spectra of the polysaccharides were measured on a NMR spectrometer (Inova-600, Varian, USA) in D₂O at 25°C.

The LLS intensity of the polysaccharide solution was measured at 25°C with a multiangle LLS (MALLS, Dawn EOS, Wyatt Technology Corporation, USA) equipped with a GaAs₃ semiconductor laser (λ = 690 nm) at angles (θ) of 43, 52, 60, 69, 80, 90, 100, 111, 121, and 132°. The polysaccharide solutions with concentration (c) in the range from 0.2 to 1.0 mg/ml in 0.15M NaCl aqueous solution were prepared, and optical clarification of the solution was achieved by filtration through 0.2 μm syringe filters (Whatman, England) before injecting into the scattering cell.

The size-exclusion chromatography combined with laser light scattering (SEC-LLS) measurements were carried out on the above MALLS with a differential refractive index detector (RI, Optilab DSP, Wyatt Technology Corporation, USA) and a pump (P1000, Thermo Separation Products, USA) with a SEC column (Asahipak GF-710HQ, 7.5 mm ID × 30 cm, Shodex, Japan). The polysaccharide solutions were prepared at a concentration of 1.0 mg/ml in 0.15M NaCl aqueous solution, and the injection volumes were 100 μl. The eluant was 0.15M NaCl aqueous solution purified by 0.2 μm membrane (Whatman, England) and degassed. The flow rate was 1.0 ml/min. The value of refractive index increments (dn/dc) 0.145 ml/g16 was used in LLS and SEC-LLS determination.

The intrinsic viscosity ([η]) of the polysaccharide samples in 0.15M NaCl aqueous solution was measured at 25°C ± 0.1°C by using an Ubbelohde capillary viscometer, the concentration was in the range of 5–15 mg/ml. The kinetic energy correction was assumed to be negligible. The Huggins and Kraemer equations were used to estimate the [η] value as following:

(1)

(2)

where η_sp/c is the reduced viscosity; ln η_r/c is the inherent viscosity; c is the polysaccharide concentration; k′ and k″ are the constants for given polysaccharide in desired condition.

Morphology of the polysaccharide aqueous solution at a concentration of 1 mg/ml was observed on a high-resolution transmission electron microscope (TEM, JEM-2010FEF, JEOL, Japan). Electron microscope copper grids were first coated with a thin film of carbon and dipped in the sample solution, dried under room condition for half hour, then vacuum dried for half hour. TEM images were taken at an accelerating voltage of 200 kV.

In Vitro Cytotoxicity Assay

The cytotoxicity assay was performed with African green monkey kidney fibroblast cells (COS-7) and human hepatocareinoma (HepG2) by fluorescence spectroscopy. 5-FU was used as positive control. In brief, the cells were seeded in a 96-well plate at a density of 6000 cells/well in 100 μl RPMI 1640. After incubation for 24 h in incubator (37°C, 5% CO₂), 100 μl of RPMI 1640 containing polysaccharide aqueous solution with a particular dose was added, and the mixtures were further incubated for 48 h. Subsequently, 20 μl 3-(4,5)-dimethylthiazol-2, di-phenytetrazolium bromide (MTT, 5 mg/ml) solution was added to each well and the cells were further incubated for 4 h. After that, the medium was removed and 150 μl (CH₃)₂SO was added and shaken at room temperature. The optical density (OD) was measured at 570 nm using a microplate reader (Model 550, Bio-Rad, USA). The relative cell viability was calculated as follows:

(3)

where OD_sample is the OD obtained in the presence of polysaccharides or control and OD_cell is that obtained in the absence of polysaccharides.

The statistical calculations between two sets of data were carried out using Student's t-test software and differences were considered significant if P < 0.01. The following basic numerical characteristics were used during the statistical processing of the results: arithmetic mean, standard deviations, and standard deviations are displayed as error bars in the figures.

Fractionation

Figure 1 shows the Zimm plot and SEC-LLS chromatograms for RPS3 in 0.15M NaCl at 25°C. The weight average molecular mass (M_w) was found to be 1.24 × 10⁶ and the molecular mass distribution was wide as reported previously.16 The SEC-LLS chromatograms of the RPS3 fractions are shown in Figure 2. Despite small shoulder peak appeared at RPS3-1, RS3-5, and RPS3-6, the six fractions were eluted at different elution volumes with symmetrical and relatively narrow peaks. Clearly, the largest molecular mass fraction RPS1 was first eluted, then the second RPS2, and finally RPS3-6. The SEC chromatogram of each fraction was separated well, as a result of the wide molecular mass distribution for the original polysaccharide. This indicated that the RPS3 sample was fractionated successfully in molecular mass by such a method, and the efficiency of the fractionation was good. Figure 3 shows the GC-MS TIC chromatograms of methylation analyses and ¹³C NMR spectra of two of the fractions. It showed that the TIC peaks of the fractions were similar as that of RPS3 and the 1→4-glucose content of RPS3-2 and RPS3-3 were 58.3% and 55.4%, respectively, which were close to that of RPS3. Moreover, the NMR spectra were almost the same as RPS3, only the peak at 79 ppm changed slightly narrow and shifted a little to higher field. It suggested that the RPS3 fractions remained the structure distribution of the unfractionated RPS3 heteropolysaccharide. It was proved that the nonsolvent addition method was suitable for the fractionation of wide distribution heteropolysaccharide.

Figure 4 shows the plots of η_sp/c vs. c and ln η_r/c vs. c for the polysaccharide fractions. By extrapolating the linear relationships of η_sp/c vs. c and ln η_r/c vs. c to an infinite dilution, the experimental values of [η] of the fractions were obtained. The results of the M_w, 〈s²〉_z^1/2, polydispersity index (M_w/M_n), and [η] are summarized in Table I. The M_w/M_n values of the fractions were smaller than 2.0. This meant that the molecular mass distribution of RPS3 fractions was relatively narrow. Therefore, the fractions can be used to study on the solution properties of RPS3 in the M_w range of 5.62 × 10⁴ to 3.05 × 10⁶.

Molecular Parameters

The power law function 〈s²〉_z^1/2 = k M_w ^ν can be estimated from many experimental points in the SEC-LLS chromatogram for RPS3 or the LLS data for its fractions. Figure 5 shows plots of 〈s²〉_z^1/2 versus M_w for RPS3 and the fractions in double logarithmic coordinates, respectively. The two fitting lines were almost parallel. From the straight line fitting of the experimental points by SEC-LLS for RPS3, the relation of 〈s²〉_z^1/2 to M_w was established as 〈s²〉_z ^1/2 = 0.44 M_w^{0.33 ± 0.01}. Moreover, the relationship from the straight line of 〈s²〉_z^1/2 versus M_w for RPS3 fractions by LLS in the M_w range of 5.62 × 10⁴ to 3.05 × 10⁶ was obtained as 〈s²〉_z ^1/2 = 1.5 M_w^{0.27 ± 0.02}. The ν values obtained from two methods were accordant to be at average value of 0.30 ± 0.03. According to the theory of polymer solutions, the exponents of 0.33, 0.50–0.60, and 1.0 reflect the polymer molecular shape as sphere, random coil, and rigid rod, respectively.17-19 The low ν value of RPS3 confirmed that RPS3 indicated a compact globular chain conformation.

The exponent value (α) of Mark-Houwink equation is usually related to the shape and the nature of macromolecules in solution. In general, α value of 0.5 suggests that the polymer molecules behave as a dense sphere, whereas that in the range of 0.6–0.8 indicates a flexible chain.20, 21 The value of α depends on the DB as well. The typically α values vary between 0.34 and 0.20 for hyperbranched glycopolymers with high DB values, the bigger the DB value, the smaller the α value.22 Figure 6 shows the [η] dependence on the M_w for RPS3 fractions in 0.15M NaCl at 25°C. The Mark-Houwink equation obtained from the fitting line for RPS3 fractions in the M_w range from 5.62 × 10⁴ to 3.05 × 10⁶ was established as [η] = 0.71 M_w^{0.27 ± 0.01}. The experimental α value of 0.27 for RPS3 was noticeably low, which suggested a relative compact sphere-like structure. The low α value in Mark-Houwink equation for RPS3 was in good agreement with that in 〈s²〉_z ^1/2 = 1.5 M_w^{0.27 ± 0.02} equation, as a result of its branched structure. It is similar to the low α value of Mark-Houwink equation (α = 0.30) of a hyperbranched polysaccharide from Pleurotus tuber-regium in 0.25M LiCl/Me₂SO solution.23

The fractal dimension (d_f) is another important parameter to characterize the conformation. It describes the usually noninteger (fractal) power of mass increasing with radius for polymers in solution. It can further help understand the characteristic structure. Usually, for a rigid rod, d_f value is 1; for linear polymers with a Gaussian coil nature, d_f value is in the range of 5/3–2; and for a three-dimensional object with a homogeneous density, d_f is 3.24-26 The larger the value of the fractal dimension is, the more compact the polymer structure tends to be. Scherrenberg et al. have found that d_f of some dendritic polymer was 3.20 The fractal dimension can be calculated from the exponent ν of the relationship between 〈s²〉_z^1/2 and M_w and it is defined as the inverse of the exponent27:

(4)

The d_f value was calculated to be 3.0 from ν values of the RPS3 polysaccharide. The result further confirmed that RPS3 was in hard globular chain conformation.

In addition, the d_f value of polymers can also be derived from the Mark-Houwink equation by following equation28:

(5)

The d_f value of RPS3 was calculated as 2.4 according to Eq. (5). The result was lower than that from the measurements of ν, but it lied in the range for hyperbranched polymers. The d_f value is not integer but fractal dimension, which is common for polymers derived from their nonuniform of inside structure. In view of the results, the water-soluble hyperbranched polysaccharide RPS3 existed as spherical conformation in 0.15M NaCl aqueous solution.

From the data of [η] and M_w for the fractions, the unperturbed chain dimension (A = (〈r²〉_o/M)^1/2) can be simply evaluated from the slope B of the plot of [η]/M_w^1/4 versus M_w^1/2 put forward by Zimm and Kilb,29 which is applicable for hyperbranched polymer:

(6)

(7)

where Φ_o is a constant with a value of 2.87 × 10²³ g/ml. The plot of [η]/M_w^1/4 versus M_w^1/2 is shown in Figure 7. An equation [η]/M_w^1/4 = 0.93 + 4.8 × 10⁻⁵ M_w^1/2 was obtained from the linear plot. The A value for RPS3 was calculated by Eq. (7) from the slope to be 1.48 ± 0.05 Å, which is the same as that 1.48 Å of a hyperbranched glycoprotein extracted from sclerotia of Pleurotus tuber-regium.30

The characteristic ratio (C_∞) is independent of length scale and pertinent to discuss the global conformation of hyperbranched RPS3. C_∞ is a parameter that presents the chain extension resulted from restriction of bond angle and steric hindrance. The larger the C_∞ is, the more rigid the chain is. It is expressed by,

(8)

where n is the repeated units, l is the average length of repeated units, and M_o is the molecular mass of a repeated unit. The l is 0.485 nm by Ref.34, and M_o is 162. The C_∞ of RPS3 in the aqueous solution was calculated to be 15.1 ± 1.5. It was relatively larger than that of a linear chain.31 This could be explained that RPS3 was α-(1→4)-D-glucan heteropolysaccharide with 37% DB, and the chain density is much denser at branch-linked points, leading to larger C_∞.32 Furthermore, the hyperbranched glucan with expanded branching chains led to spheric conformation.

Molecular Morphology

The conformation parameters of RPS3 deduced from the polymer solution theory indicated that it existed as spherical chain conformation in aqueous solution. To provide direct evidence of the polysaccharide chain conformation, TEM was used to observe the molecular morphology. Figure 8 shows the TEM image of RPS3 and fraction RPS3-3 in dilute aqueous solution. The spherical shapes of the RPS3-3 fraction were the same as RPS3 in dilute aqueous solution. The diameters of the particle of RPS3 visualized by TEM were in the range of 20–100 nm with the average diameter 50 nm, and the diameter of the RPS3-3 molecules was almost the homogeneous to be 60 nm, which were in good agreement with that determined by LLS. The molecular morphology observed by TEM proved directly that the macromolecules of RPS3 and its fraction existed as spherical conformation in aqueous solution. The shape of RPS3-3 fraction was the same as RPS3, also indicating that the nonsolvent addition method did not change the conformation of the fraction during fractionation. Therefore, the conformation parameters for the complex hyperbranched polysaccharides in the aqueous solution were reliable.

Interaction of RPS3 and Fractions to Cells In Vitro

The cytotoxicities of RPS3 and its fractions were evaluated on the COS-7 and HepG2 cells by MTT assay. Figure 9 shows the influence of dose of RPS3 and its fractions on the proliferation of COS-7 cells, the relative errors were less than 10%. The cell viability of COS-7 cells in RPS3 and its fractions were significantly higher than that of 5-FU, indicating the low cytotoxicity of RPS3 and its fractions, it was especially true at doses less than 100 μg/ml, above that concentration some of the fractions resulted in about 20% loss of cell COS-7 viability. This suggested that the polysaccharide cytotoxicity to COS-7 had dose dependence. Compared with the fractions, RPS3 exhibited less proliferation on COS-7 cells, suggesting that polysaccharide with narrow molecular mass distribution was benefit to cell growth of COS-7. The cell viability of COS-7 of the RPS3 fractions having M_w from 10 × 10⁴ to 60 × 10⁴ was higher than others. Figure 10 shows the influence of dose of RPS3 and its fractions on the proliferation of HepG2 cells. There was obvious dose dependence of polysaccharide fractions on the cell viability of HepG2. Compared with 5-FU, RPS3 and its fractions at different doses had little cytotoxicity to HepG2 cells too. Except RPS3-6 with the smallest molecular mass, the fraction with smaller molecular mass was less cytotoxic to HepG2 cells. RPS3-1 and RPS3-2 appeared to exert a cytotoxic effect on HepG2 cells, i.e., they limit proliferation. Interestingly, this did not appear to be the case in COS-7 cells. It is meaningful that the different responses to the two cell lines at dose less than 10 μg/ml, namely, the COS-7 cell viability increase with the dose decrease, whereas the HepG2 cell viability decrease. It could be concluded that the fractions could inhibit the tumor cells and showed no harm to normal cells at low dose and this bioactivity increased in a moderate range of molecular mass, which was accordant to polysaccharide from Pleurotus tuber-regium and polysaccharide from Poria cocos.33, 34