Electrospun cellulose acetate membrane for size separating and antibacterial screening of crude polysaccharides

Abstract

This study aims to produce electrospun cellulose acetate (CA) membrane as the alternative supporting medium for a separation of crude polysaccharides by electrophoresis and a screening of their antibacterial activity. Among the tested conditions of fabrication, electrospun CA membrane at 57% porosity showed the best separation of each polysaccharide from the standard mixture and the crude extract of Aloe vera via electrophoresis. As compared with the commercial CA membrane, the produced electrospun CA membrane demonstrated more separated spots of polysaccharides. The antibacterial activity of the electrophoretic polysaccharide was also determined against Escherichia coli and Staphylococcus aureus as the inhibition zone after the bacterial culture agar was overlaid on the membrane and incubated for 24 h. The results of this study suggested the potential application of electrospun CA membrane combining with electrophoresis as a simple method for separating crude polysaccharides and screening for their antibacterial activity.

1 Introduction

Bioactive polysaccharides derived from medicinal plants have gained many interests especially for biomedical applications due to their various pharmacological activities including wound healing stimulation [1], inhibition of cancer cell growth [2], modulation of immune responses [3], reduction of inflammation [4, 5] and inhibition of bacterial growth [6]. Several methods have been used to evaluate these bioactive phytocompounds with high sensitivity, resolution and quantitative detection such as high‐performance liquid chromatography [7], gas chromatography mass spectrometry [8] and Fourier transform infrared spectroscopy (FTIR) [9]. However, these methods are not convenient for a screening experiment that needs simple, fast, robust and inexpensive approach without the requirement of complicated laboratory equipment.

Electrophoresis is a simple technique for separating biomolecules, frequently used in most laboratories. Electrophoresis on a semi‐solid medium such as agarose gel or polyacrylamide gel can separate nucleic acids and proteins by their differences of size, charge or conformation. Electrophoretic separation on a solid membrane such as cellulose acetate (CA) can be applied for clinical chemistry and haematology such as detection of haemoglobins, separation heparin and its glycosaminoglycan impurities and determination of serum protein concentration [10, 11]. The principle of this technique is based on an electrophoretic separation of charge biomolecules with different molecular masses on a porous membrane [12], thus this technique can be applied to separate charged polysaccharides by sizes.

Electrophoretic mediums are generally produced by polymer casting [13] and phase separation [14]. By casting method, the microporous CA membrane was produced by evaporation of volatile solvent, which denser porous morphology could be obtained via increasing CA concentration. Similarly, microporous structure of CA membrane was obtained via a phase separation method. Thus, these commercial CA membranes generally contained an average pore size in a micrometre scale, which their separation was limit to some biomolecules with molecular weights in this range [15]. Recently, electrospinning technique has received many interests since it is a simple technique for a production of a porous membrane with controlled fibre sizes from nanometre to micrometre scale [16]. Owing to the general features of randomly distributed pores throughout, high volume per surface area, and adjustable porosity and pore size, the electrospun membranes may be served as a potential supporting medium for electrophoresis, which has not been studied. Thus, this paper is interesting in combining the use of electrospun CA membrane and electrophoresis technique for separating crude polysaccharides of plant extracts and screening of their antibacterial activity.

2 Materials and methods

2.1 Chemicals

CA (CA, acetyl content 39.7%, Molecular Weight (MW) = 50,000), pectin (40 and 80 kDa) and p ‐iodonitrotetrazolium (INT) were purchased from Sigma Aldrich (USA). D(+)−glucose (0.18 kDa), dextran (6 kDa) and carrageenan (60 kDa) were purchased from Fluka (USA). All chemicals used were of analytical grade.

2.2 Fabrication of electrospun CA membranes

The CA membrane was produced via electrospinning technique. The electrospinning unit consisted of an infusion pump (homemade), a high‐voltage power supply (AU‐301P1 Matsusada Precision, Japan) and a grounded collector (a stainless steel drum with 13 cm external diameter and 30 cm length) [17]. CA powder was dissolved in a mixed solvent of acetic acid and water (95:5, v/v) for 13 h at ambient temperature. The solution was put into a 10 ml plastic syringe (Terumo, Belgium) with a 0.5 mm diameter metal needle. The optimised condition for electrospinning was studied to obtain the continuous fibres. The CA membranes were fabricated by adjusting polymer concentrations (5−20%, w/v), polymer volumes (1−25 ml) and voltages (15−25 kV). A flow rate was fixed at 0.8 ml/h and the distance between an injector and a rotating collector was fixed at 15 cm. Electrospun CA membranes were kept in a vacuum desiccator for 24 h at ambient temperature before using.

2.3 Characterisation of fabricated CA membranes

Fibre morphology, membrane thickness, pore size and porosity were characterised. Electrospun CA membranes were cut into a size of 1 cm × 1 cm, adhered onto the stub, and sputter coated with gold for 8 min before studying on a scanning electron microscope (SEM) (JSM−6010 PLUS, JEOL, USA). Fibre morphology and membrane thickness were determined from SEM images. Pore sizes and fibre diameters were evaluated from a measurement of 300 points in the taken SEM micrographs using the ImageJ open‐access software (National Institutes of Health (NIH), USA).

Porosity of each electrospun CA membrane was evaluated by a liquid displacement assay using hexane as a liquid displacement [18]. Hexane (9 ml, V1) was placed in a cylinder and CA membrane was immersed for 5 min. The increased volume was recorded as V 2. After the CA membrane was removed, the residual volume of hexane was recorded as V 3. Porosity of membrane was calculated by the following expression:

$$\mathrm{porosity}\left(\text{\%}\right)=\left[\frac{\left(V1-V3\right)}{\left(V2-V3\right)\times 100}\right]$$

Chemical characteristics of the membranes were analysed by FTIR (Vertex 70 series OPUS, Bruker, USA). FTIR spectra were acquired in a transmission mode using a Tensor 27 spectrometer (Bruker, USA) equipped with a deuterated triglycine sulfate (DTGS) detector. The sample (1 g) was mixed with 200 mg potassium bromide in an agate mortar, ground to a fine powder and pressed into a thin pellet. The spectra of samples were determined in a spectral range of 400–4000 cm⁻¹ with a resolution of 4 cm⁻¹. Spectral data were processed using the Bruker software system Opus 7.0 (Bruker, USA). Hydrophilicity of electrospun CA membranes was determined from water contact angles (FTA 1000, First Ten Angstrom, USA) as the method was modified from Wadbua et al. [19]. Water of 5 µl was dropped onto the membranes of 1 cm × 1 cm with a distance of 1 cm. At least five measurements were performed for each membrane. Water absorption was evaluated according to the method of Siri and Maensiri [20] The CA membrane (1 cm × 1 cm) was immersed in deionised water at room temperature for 24 h. The initial dried weight was recorded as W 1. After the membrane was removed from the water, its wet weight was measured (W 2). The water absorption was calculated by the following expression

$$\mathrm{water}\mathrm{absorption}\left(\text{\%}\right)=\left[\frac{\left(W2-W1\right)}{W1}\right]\times 100$$

2.4 Aloe vera extraction

Fresh A. vera leaves at 12 months old and 30−50 cm in length were collected. The extraction method was modified from Ni et al. [21]. The jelly part was cleaned with deionised water to remove yellow sap. After homogenising, it was centrifuged at 8000g for 20 min. The polysaccharide pellet was collected and changed its form to sulphate polysaccharides by sulphation reaction [22]. Ammonium sulphate (500 mg) was dissolved in 100 ml solution mixture of sulphuric acid and n ‐butanol (3:5 v/v) before 2.5 g polysaccharides was added. The mixture was incubated at 0° C for 30 min with continuous stirring. The mixture was neutralised with 1 M sodium hydroxide, dialysed against distilled water for 3 days and freeze‐dried.

2.5 Polysaccharide separation ability of CA membranes

Sugar and polysaccharide standard mixtures consisted of D(+)−glucose (0.18 kDa), dextran (6 kDa), short‐chain pectin (40 kDa), carrageenan (60 kDa) and long‐chain pectin (80 kDa). The standard mixtures (50 µg/ml) and the crude extract of A. vera (50 µg/ml) were either spotted or streaked (0.5 µl) on the CA membranes and subjected for electrophoresis for 1.5 h using 0.2 M calcium acetate (pH 7.0) buffer and 7 mA constant current with a maximum 240 V as the method was modified from Pielesz and Biniaś [9]. The electrophoretic membranes were stained with 0.5% toluidine blue in 3% acetic acid solution, destained with deionised water and air‐dried [23]. Retention factor (R _f) of each separated spot was the distance travelled by the compound divided by that of the solvent.

2.6 Antibacterial activity of separated polysaccharides

Antibacterial activity of the separated polysaccharides on electrospun CA membrane was evaluated by the developed method modified from a bioautographic assay [24]. The electrophoretic membranes were rinsed with deionised water, overlaid with soft agar medium (Mueller Hinton broth with 0.6% agar) containing 10⁶ colony forming unit of Staphylococcus aureus or Escherichia coli, and incubated at 37°C for 12 h. Then, the membranes were sprayed with 2 mg/ml of p ‐iodonitrotetrazolium and further incubated at 37°C for 1−2 h in a dark container. The bacterial growth inhibition zone was the clear zone on a pink to red background.

2.7 Statistical analysis

Data are expressed as the means of at least three replicates ± standard deviation. For statistical analysis, a one‐way analysis of variance (ANOVA) was used to compare the means of different data sets with Statistical Package for the Social Sciences (SPSS) 18.0 for windows software (SPSS, USA). To identify significant difference among data groups, means comparison was evaluated employing a Tukey's multiple comparison post hoc ANOVA test. The differences with a value of P < 0.05 were considered statistically significant.

3 Results and discussion

3.1 Fabrication and characterisation of electrospun CA membranes

In this paper, we developed the electrospun CA membrane as the solid medium for separating polysaccharides by sizes via electrophoresis and screening for their antibacterial activity. CA membranes were fabricated using electrospinning technique by adjusting CA concentrations (5–20%, w/v) and voltages (15–25 kV), while the flow rate was fixed at 0.8 ml/h and the distance between an injector and a rotating collector was kept at 15 cm. To obtain the optimised condition for electrospinning, each polymer solution at different concentrations was used, while a minimal flow rate was applied and a voltage was adjusted to obtain continuous fibres. Then, a flow rate was adjusted to obtain the best condition. Transmission Electron Microscope (TEM) images of all electrospining conditions that can produce continuous fibres were compared and the optimised condition was chosen from one yielding the smooth, homogenous fibres without any beads and drops of polymer. The optimised condition to produce electrospun CA membranes was to use 17% CA solution, 25 kV voltage, 0.8 ml/h flow rate and 15 cm distance. Voltage is a key player for controlling fibre size and bead formation [25]. With increasing of voltage, a fibre diameter is reduced but a formation of larger bead is increased. When voltage is increased, polymeric solution is more charged, causing a faster moving of polymeric droplets and jet to a collector; thus, a fibre diameter is smaller but larger beads are formed. For polymeric concentration, it affects bead formation, bead density, fibre morphology and size [26]. Low viscosity is favour a formation of bead, thus as the concentration increases, beads disappear. However, with too high polymeric concentration, a formation of fibre cannot be obtained due to too viscous solution. For a flow rate, increasing flow rate causes larger fibre sizes. It is due to a decrease of electrostatic density, thus a fabrication is not completely developed, yielding larger fibre size [27]. If a flow rate is too high, a formation of fibre cannot happen because the solution is dropped by a gravity force.

To produce different porosities and pore sizes, electrospun membranes were fabricated from different volumes (3, 7, 12, 15, 20 and 25 ml) of CA solution, which referred to as CA−3, CA−7, CA−12, CA−15, CA−20 and CA−25 membranes, respectively. SEM image of the representative electrospun CA−25 membrane is shown in Fig. 1, which the membrane consisted of smooth, non‐woven fibres and contained many small pores caused by interconnected fibres. Electrospun CA membranes made of different volumes of CA solution showed no morphological difference. The average fibre diameters of the electrospun CA membranes were similar in a range of 367.6−458.0 nm (Table 1). With increasing CA solution volumes, the fabricated CA membranes increased in their thickness from 0.84 to 2.78 mm (Table 1), which was a result of deposited fibres in more layers.

Fig. 1.

Electrospun CA−25 membrane

a SEM image

b Histogram of fibre size distribution

Table 1.

Average fibre diameters, water contact angles and thicknesses of the electrospun CA membranes fabricated from CA solution of 3, 7, 12, 15, 20 and 25 ml, designated as CA−3, CA−7, CA−12, CA−15, CA−20 and CA−25, respectively

Electrospun CA membranes	Diameter, nm (n = 300)	Water contact angle, ° (n = 3)	Thickness, mm (n = 20)
CA−3	426.0 ± 76.2c	130.2 ± 1.5a	0.84 ± 0.11a
CA−7	426.7 ± 93.9c	132.6 ± 0.9a	0.99 ± 0.16b
CA−12	367.6 ± 75.5a	131.1 ± 1.0a	1.14 ± 0.18c
CA−15	411.3 ± 95.5b,c	131.1 ± 1.8a	1.48 ± 0.12d
CA−20	458.0 ± 128.7d	129.8 ± 1.2a	2.43 ± 0.12e
CA−25	400.3 ± 112.2c	127.7 ± 1.7a	2.78 ± 0.14f

^a‐f Statistically significant difference (P < 0.05) among the membranes.

Surface hydrophilicity and water absorption of the produced electrospun CA membranes were also evaluated. Surface hydrophilicity of the membranes can be evaluated from water contact angles, which the value <90° corresponds to high‐surface wettability of the membrane and the value >90° indicates its low water wettability [28]. In this paper, all produced CA membranes exhibited low‐surface hydrophilicity as determined from the water contact angles in a range of 127.7°–132.6°. This may be resulted from the acetyl groups in the CA membrane and the membrane surface that contains very small pores, thus delaying penetration of water drop [29]. Though, the fabricated CA membranes had low‐surface hydrophilicity, they could well absorb water (71.9–83.1%). Water absorption capacity was greater in the CA membranes derived from more volumes of CA solution. With increasing volumes of CA solution, the obtained membranes contained higher fibre layers, more empty space and higher surface area, thus retain more water in their structure [30].

Porosity and average pore size of the produced membranes are shown in Fig. 2. Increasing volumes of CA solution for fabricating the membranes resulted in the gradual reduction of both porosity and pore size. Pore sizes of CA−3, CA−7, CA−12, CA−15, CA−20 and CA−25 membranes were 675.7 ± 72.9, 588.6 ± 47.4, 522.2 ± 61.3, 483.8 ± 84.7, 386.49 ± 80.46 and 169.9 ± 89.4, respectively. Porosity of CA membranes were 77.8 ± 1.6, 77.3 ± 5.2, 74.2 ± 6.1, 67.8 ± 5.1, 64.2 ± 2.3 and 56.9 ± 6.5, respectively. With increasing volumes of CA solution, non‐woven fibres were more produced and denser packed, causing smaller pore size and reduced porosity [31, 32]. For the commercial CA membrane (Helena laboratory, USA), it had the larger pore size (571.3 ± 462.5 nm) and the lower porosity (72 ± 4.6 nm) as compared with the CA−25 membrane. Thus, it was hypothesised that CA−25 membrane might have a better separation ability of smaller molecules than the compared commercial membrane.

Fig. 2.

Electrospun CA membranes derived from CA solution of 3, 7, 12, 15, 20 and 25 ml

a Pore size of the membranes derived from different volumes of CA solution

b Porosity of the membranes derived from different volumes of CA solution

The FTIR spectra of the produced membrane and CA powder are shown in Fig. 3. The spectrum of the produced membrane showed the absorption peaks that are the characteristic peaks of CA including hydroxyl group at 3470 cm⁻¹, acetate group at 1739 cm⁻¹ [33, 34], symmetric bending of methylene groups at 1370 cm⁻¹, asymmetric stretching of carboxylate group at 1230 cm⁻¹ and C–O stretching of primary alcohols at 1038 cm⁻¹ [33, 35]. These absorption peaks were similar to those of CA powder, suggesting no change of CA chemical structure caused by electrospinning process.

Fig. 3.

FTIR spectra of CA powder and electrospun CA membrane in a range of 4000–400 cm⁻¹

3.2 Ability to separate polysaccharide mixtures of CA membranes

CA−3, CA−7, CA−12, CA−15, CA−20 and CA−25 membranes, which have different porosities and pore sizes, were applied as the electrophoretic supporting media for separating the standard mixtures consisted of D(+)−glucose (0.18 kDa), dextran (6 kDa), short‐chain pectin (40 kDa), carrageenan (60 kDa) and long‐chain pectin (80 kDa). The standard mixture was dropped on each electrospun CA membrane as a single spot and electrophoresed for 1.5 h in 0.2 M calcium acetate buffer (pH 7.0) using 7 mA constant current with a maximum 240 V. Polysaccharide migration in each membrane is shown in Fig. 4. CA−3 membrane showed no ability to separate the standard mixtures. CA−20 and CA−25 membranes showed the separated five spots of the standard mixtures, while the other membranes showed no complete separation of the standard mixtures. These results showed that the CA membrane with lower porosity and pore size demonstrated better separation of each polysaccharide, since low porosity and small pore size allow a better separation of closed molecular weight polysaccharides [36]. Through controlling porosity and pore size of electrospun CA membrane, it would be the advantage on separation of high and close molecular weight polysaccharides apart via this simple technique. Among all fabricated membranes, the CA−25 membrane, which has lowest porosity and pore size, showed the best separation result, thus it was chosen for separating polysaccharide mixtures of crude plant extract. Electrophoresis allows separation of charged molecules based on their size and shape. Pore size of the medium matrix plays an important role as a sieve for size and shape separation of molecules, while a porosity and voltage control a migration rate [37]. Thus, with low pore size and porosity, a better separation of smaller molecules is obtained. In this paper, it is very likely that a separation ability of CA–25 membrane is due to its pore size and porosity, not thickness of the membrane. Increased thickness of the membrane can only hold more loaded samples on the membrane. In this paper, the crude extract of A. vera was used as the model because it was known to contain many sizes of polysaccharides such as aloeferon (70 kDa), polyuronide (275−374 kDa), acemannan (5000 KDa), arboran A (12 kDa), arboran B (52 kDa), glucomannan (450 kDa), mannan (15 kDa), mannose (0.18 kDa) [38, 39] and some polysaccharides possess antibacterial activity such as acemannan [40].

Fig. 4.

Ability of CA−3, CA−7, CA−12, CA−15, CA−20 and CA−25 membranes (1−6) to separate five polysaccharides from the standard mixture

3.3 Separation and screening for antibacterial activity of A. vera extract

CA−25 membrane was evaluated for separation and screening for antibacterial activity of A. vera extract and the standard mixtures, which were deposited on the membrane as a streaked line. Separated samples were reported as the retention factor (R _f), which is the travelled distance of the sample relative to the solvent front [41]. Fig. 5 a showed the separation of the standard mixtures of five bands (R _f of 0.82, 0.78, 0.67, 0.62 and 0.26) and A. vera extract of six bands (R _f of 0.82, 0.78, 0.66, 0.48, 0.33 and 0.21) according to size. To evaluate whether the separated polysaccharides on the membrane are actively against bacterial growth, the agar containing E. coli or S. aureus was overlaid on the membrane and incubated for 24 h at 37°C. Fig. 5 b revealed the single inhibition zone against E. coli on the fabricated CA membrane at the R _f 0.33 band of A. vera extract. The R _f 0.33 band was also active against S. aureus (Fig. 5 c). In this paper, the standard polysaccharides were used as the negative control, thus they showed no antibacterial activity against both bacterial strains. The results of this paper demonstrated that the electrospun CA membrane was effective to separate polysaccharides via electrophoresis technique and the antibacterial activity of the separated polysaccharides could also be analysed.

Fig. 5.

Ability of CA−25 membrane to separate and screen for antibacterial activity of mixed polysaccharides

a Electrophoretic polysaccharides of the standard mixture (1) and the A. vera extract (2)

b Growth inhibition against E. coli of the electrophoretic polysaccharides of the standard mixture (1) and the A. vera extract (2)

c Growth inhibition against S. aureus of the electrophoretic polysaccharides of the standard mixture (1) and the A. vera extract (2)

4 Conclusion

This paper showed the development of electrospun CA membrane for rapid separating and screening for antibacterial activity of polysaccharide mixture. Optimised condition to produce electrospun CA membrane was 17% (w/v) CA solution in 95% acetic acid, 25 kV voltage, 0.8 ml/h flow rate and 15 cm distance between injector and collector. To produce different porosities and pore sizes of electrospun CA membranes, different volumes of CA solution (3, 7, 12, 15, 20 and 25 ml) subjected to electrospin. With increasing volumes of CA solution for electrospinning, fabricated CA membranes exhibited the decreased porosity and pore size. Among the fabricated CA membranes, the CA−25 membrane, which has lowest porosity and pore size, showed the best efficacy to separate the standard mixtures. It was applied to separate the polysaccharide mixture of A. vera extract and showed well separation of six bands. The separated bands also could be directly assayed for antibacterial activity and one band showed antibacterial activity against both E. coli and S. aureus. The results of this paper clearly demonstrated the application of the electrospun CA membrane as a solid electrophoretic membrane for rapid separating polysaccharide mixture by the sizes and screening for its antibacterial activity.