Abstract
The quantitative difference in the antibacterial response was measured for pine rosin and propolis against Staphylococcus aureus ATCC 12598. The activity was studied for fibrous networks that form entirely bio-based cellulose-acetate (CA) materials. The analysis considers the effects of bacterial input, additive dosage, solvent type, variation in preparation, as well as the effect of storage time. Based on the results, the electrospun network structure is dependent on the solvent and the concentration of rosin and propolis. Both rosin and propolis improved the cellulose acetate solution processability, yet they formed beads at high concentrations. Rosin and propolis created strong antibacterial properties when these material systems were immersed in the liquid for 24 h at room temperature. The response remained visible for a minimum of two months. The electrospun networks of water and DMAc solvent systems with 1 to 5 wt% rosin content were clearly more efficient (i.e., decrease of 4 to 6 logs in colony forming units per mL) than the propolis networks, even after two months. This efficiency is likely due to the high content of abietic acids present in the rosin, which is based on the Fourier transform infrared spectra. The results of the additional analysis and cell cultivation with dermal fibroblast cells indicated an impairing effect on skin tissue by the rosin at a 1 wt% concentration compared to the pure CA fibers.
Download PDF
Full Article
Comparison of Rosin and Propolis Antimicrobials in Cellulose Acetate Fibers Against Staphylococcus aureus
Mikko Kanerva,a,* Vsevolod Matrenichev,b Rama Layek,a Timo M. Takala,c Pekka Laurikainen,a Essi Sarlin,a Anna Maria Elert,d Vladimir Yudin,b Jani Seitsonen,e Janne Ruokolainen,e and Per Saris c
The quantitative difference in the antibacterial response was measured for pine rosin and propolis against Staphylococcus aureus ATCC 12598. The activity was studied for fibrous networks that form entirely bio-based cellulose-acetate (CA) materials. The analysis considers the effects of bacterial input, additive dosage, solvent type, variation in preparation, as well as the effect of storage time. Based on the results, the electrospun network structure is dependent on the solvent and the concentration of rosin and propolis. Both rosin and propolis improved the cellulose acetate solution processability, yet they formed beads at high concentrations. Rosin and propolis created strong antibacterial properties when these material systems were immersed in the liquid for 24 h at room temperature. The response remained visible for a minimum of two months. The electrospun networks of water and DMAc solvent systems with 1 to 5 wt% rosin content were clearly more efficient (i.e., decrease of 4 to 6 logs in colony forming units per mL) than the propolis networks, even after two months. This efficiency is likely due to the high content of abietic acids present in the rosin, which is based on the Fourier transform infrared spectra. The results of the additional analysis and cell cultivation with dermal fibroblast cells indicated an impairing effect on skin tissue by the rosin at a 1 wt% concentration compared to the pure CA fibers.
Keywords: Rosin; Propolis; Antibacterial; Cellulose acetate; Electrospinning
Contact information: a: Tampere University, Faculty of Engineering and Natural Sciences, PO Box 589, FI-33014 Tampere University, Finland; b: Institute of Macromolecular Compounds (IMC), Russian Academy of Sciences 31 Bolshoy pr. VO, Saint-Petersburg, 199004 Russia; c: Helsinki University, Department of Microbiology, PO Box 56, FI-00014 Helsinki, Finland; d: Bundesanstalt für Materialforschung und -prüfung (BAM), 12205 Berlin, Germany; e: Aalto University, School of Natural Sciences, Department of Physics, PO Box 14300, FI-00076 Aalto, Finland; *Corresponding author: mikko.kanerva@tuni.fi
INTRODUCTION
Natural additives in nano scale structures can result in synergetic features due to their complex development and effects of small size. Wood rosin is a substance that has been suggested to replace petroleum-based resins and polymerizing elements in structural polymers (Barrueso-Martínez et al. 2003; Liu et al. 2014). In addition, rosin has been used as an additive in food packaging materials due to its strong antimicrobial response (Niu et al. 2018). Rosin has also been used as an antibacterial agent in medical wound dressings (Sipponen et al. 2007). Rosin is mostly an amorphous compound that consists of a mixture of lipophilic components and phenol extracts. It can be found in tree sapwood and heartwood with different compositions (Sjöström 1993; Vainio-Kaila et al. 2017b). The interaction between various bacterial strains and rosin’s pimaric, labdane, and abietic acids is not fully clear. However, the antibacterial activity is associated with the malfunction of the cell wall energy synthesis after contact with rosin’s acids (Söderberg et al. 1990; Sipponen et al. 2009). The antibacterial activity in pure rosin has been clearly reported against Gram-positive Staphylococcus aureus and Streptococcus pneumonia strains. Additionally, it has been reported against Gram-negative Escherichia coli and Salmonella enterica strains (Söderberg et al. 1990; Vainio-Kaila et al. 2017a). The low molecular weight solvent components are usually removed to achieve a definite chemical composition (Nirmala et al. 2013).
Propolis is a complex, wax-like organic product that combines the best characteristics of plant and animal-based production in nature. The pharmacological features of propolis have been realized for centuries. Barbosa et al. (2015) extracted hydroalcoholic antioxidants from red propolis to balance the inflammatory response of artificial nerve conduits. Similarly, Awale et al. (2008) reported cytotoxic activity of propolis against tumor cells. Additionally, propolis can be used to form high quality electrospun nanofibers when blended with synthetic polymers such as polyurethane, poly-lactic acid, and polyvinylpyrrolidone to gain observable antibacterial activity (Kim et al. 2014; Sutjarittangtham et al. 2014; Asawahame et al. 2015). Propolis (sometimes called bee glue) is formulated by honeybees to fill small crevices and pin holes in the walls of the wooden environment around their nest (Savolainen 2016). Likewise, bees use propolis to embalm intruders too large to be removed from the nest. Bees digest rosin and wax-like fluids from buds of different trees and plants to prepare propolis. The main components of propolis are various flavonoids, phenolics, organic acids, alcohols, and pollen (Toreti et al. 2013). The exact chemical content of pollen depends on the environment and plants available to the bees (Fatrcova-Šramková et al. 2013).
These natural substances (rosin and propolis) are superior to most of the synthetic biocides because they are antimicrobial, and they support biodegradation or the healthy cell development in various host organisms. The carrier structure for making rosin- or propolis-containing biopolymer films and fibers could be made of cellulose acetate (CA). CA has high throughput and thermal stability when it is electrospun into high surface area to volume networks (Luo et al. 2013). Furthermore, deacetylation of CA can be used to transform CA based fibrous networks into pure cellulose type systems (Son et al. 2004). In recent studies, CA fibers and films with antibacterial activity have been prepared primarily using silver and ZnO nanoparticles (Son et al. 2006; Anitha et al. 2012). Moreover, CA films and electrospun networks have been applied to vitamin, enzyme, and drug immobilization products instead of tissue engineering products (Konwarh et al. 2013). Rosin and propolis can be used in synthetic polymers, but their comparative effects have not been surveyed especially when forming biobased material systems.
In this work, the antibacterial response of rosin and propolis were quantitatively compared. The effects of both natural substances were studied in fibrous networks to create entirely biobased CA materials. The fibrous networks were analyzed in terms of their antibacterial behavior against the Gram-positive S. aureus ATCC 12598 in the Ringer’s medium. The purpose of this study was to analyze the antibacterial activity of rosin and propolis in the CA materials while considering the effects of bacterial input, additive concentration, solvent type, variation in preparation, as well as the effect of storage time. Additionally, the effects of rosin integration on the biocompatibility, in terms of the viability of dermal fibroblast cells, was studied.
EXPERIMENTAL
Polymers and Natural Additives
The cellulose acetate (CA) was purchased from Merck (Steinheim, Germany) KGaA (Mw = 30,000 and acetyl content of 39.8 wt%). The dimethylacetamide (DMAc) was provided by Sigma Aldrich (Steinheim, Germany).
The rosin was a commercial, industrial grade extract that was provided by Forchem (Rauma, Finland). The rosin had been derived from pine (gum) rosin and was received in a crushed particle form (batch 16032017, acid value 167 mg KOH/g, softening point
74 °C).
Two different propolis products were selected for the study: nutritive capsules (Gélules Ultra, Apimab Laboratoires, Clermont-l’Hérault, France) in powder form (encapsulated) and hand collected raw propolis (Rudenko, batch 0464316, St. Petersburg, Russia) in its original form collected during the summer of 2017 (half a year prior to the fiber preparation). In this study, the Gélules Ultra is referred to as ’propolis I’ and the Rudenko as ‘propolis II’. The two selected propolis products represented the different processing levels of propolis.
Electrospinning and Film Preparation
The cellulose acetate (CA) was dissolved in two different kinds of solvent systems: water and acetone, and DMAc and acetone. In these systems, the CA concentration ranged between a 5 and 20 wt% content. The percentage of water in the H2O systems was
15 wt%, and the DMAc percentage in the DMAc systems was 33 wt%. For preparing solutions for electrospinning, the required amount (1 to 5 wt%) of rosin or propolis was added to the CA solutions. The CA-rosin/propolis solutions were stirred at room temperature until dissolution of the rosin or propolis. Finally, the systems were sonicated for 5 min.
The electrospinning device included a high voltage generator (Chargemaster, SIMCO, Santa Clara, CA, USA), a syringe, a needle, a syringe pump, and a ground electrode. The needle was connected to the high voltage supply, which can generate direct current voltages up to 50 kV. The distance between the needle tip and the ground electrode ranged between 15 and 30 cm. Positive voltage applied to polymer solutions ranged between 10 and 16 kV. CA solutions were delivered via a syringe pump to control the mass flow rate.
The mass flow rate values ranged between 5 and 20 mL per h. The electrospinning was performed at room temperature. For the study of the preparation effects (batch dependency), a parallel series of fibers were prepared at IMC (Russia) using a Nanon-01A (MECC Co., Minneapolis, US) with the same control parameters used earlier. Additionally, the polymer films were produced by a tape casting method with the following three compositions (H2O systems): CA (10 wt%) as the reference, CA (10 wt%) plus rosin (5 wt%), and CA (10 wt%) plus rosin (1 wt%).
Characterization
Fourier transform infrared spectroscopy (FTIR) spectra of the rosin and propolis, the prepared nanofibers, and the film samples were obtained with a Tensor 27 spectrophotometer (Bruker, Billerica, US) to characterize the content of the materials. The viscosity of the solutions for electrospinning was determined with an Anton Paar viscometer (model Physica MCR 301, Graz, Austria) at 25 °C. Visual light microscopy (DM 2500 M, Leica, Wetzlar, Germany) and scanning electron microscopy (SEM) using a field emission gun electron microscope (ULTRAplus, Zeiss, Oberkochen, Germany) were used for studying the fiber networks and polished laminate cross sections. The morphology and fiber diameters were estimated by analyzing the SEM images.
AFM studies were obtained using a NanoIR2S (Bruker/Anasys Instruments, Santa Barbara, CA, USA) coupled with a multichip QCL source (MIRcat, Daylight Solutions, San Diego, CA, USA). An Au coated silicon probe (contact AFM-IR, cantilever, Anasys Instruments, Santa Barbara, US) was employed. Electrospun network samples for AFM were first embedded in epoxy by casting (Epofix, Struers, Cleveland, US). After the embedment, 100 to 150 nm thick slices were cut of the fiber network’s cross section by using a cryo-ultramicrotome (Leica Ultracut 7, Wetzlar, Germany). The slices were captured on copper grids covered by a carbon film (Electron Microscopy Sciences, Hatfield, US). The grids with the samples were mounted on metallic chips (Ted Pella, Redding, CA, USA) for AFM.
Antibacterial Activity
Antimicrobial activities of the electrospun networks were tested against the indicator bacteria Staphylococcus aureus ATCC 12598. The indicator was cultured at
37 °C in lysogeny broth (LB) with 1.5% agar for solid media. Antibacterial tests with the electrospun material were carried out in a Ringer’s solution of 1/4 strength (2.25 g NaCl, 0.105 g KCl, 0.06 g CaCl2, and 0.05 g NaHCO3 in 1 L of distilled water, Merck). The sample material (approximately 2.5 mg) was collected from approximately 1 by 1 cm samples of electrospun networks (on aluminum backing during spinning).
The indicator strain was first cultured overnight in the LB broth. The number of colony-forming units per mL of the o/n culture was determined by serial dilutions and plating onto the LB agar. From the serial dilutions of 10-3 to 10-4, 1 mL (referring to about 105 to 107 cfu per mL) was used for the antimicrobial test by mixing with the sample material (approximately 1 cm2 of network) in 2 mL Eppendorf tubes. The mixtures were incubated for 24 h at room temperature in a rotator (22 rpm). After incubation, the samples were serially diluted in 1/4 strength Ringer’s solution and plated onto the LB agar. This was performed to determine the bacterial survival by colony counting. At a minimum, two separate analysis campaigns per sample series were done, and the average was calculated for indicating the survival in the graphs.
Cultivation of Dermal Fibroblast Cells
The biological material of the skin biopsy of a young, healthy donor was used to extract the dermal fibroblasts of the skin. To extract the initial culture of dermal skin fibroblasts, a standard enzymatic procedure was used (Varga et al. 2005). The cell suspension obtained during the enzymatic treatment was cultured in a DMEM nutrient medium (PanEco, Berg am Irchel, Switzerland) with an additive of L-glutamine, 10% bovine fetal serum, and penicillin (100 units per mL) and streptomycin (100 µg per mL) (all reagents by Gibco, Waltham, MA, USA). Cultivation took place in a CO2 incubator (3423, Thermo Fisher Scientific, Waltham, MA, USA) at a temperature of 37 °C and a high humidity. The concentration of CO2 was 5%. In total, 3 to 12 passages were used for the experiment. The colorimetric assay (MTT) for the number of viable cells was carried in phases as described in Table 1.
Table 1. The Preparation and Analysis Steps of the MTT Assay for CA Films
RESULTS AND DISCUSSION
Spinnability and Fiber Size
The spinnability of the rosin and propolis containing CA networks was studied via an extensive electrospinning test matrix. The summary of the network quality is given in Table 2, and the SEM images of the representative network samples are shown in Figs. 1 and 2.
The concentration of CA in the water and acetone solvent systems has a crucial effect on the spinning process. With an increase of the CA content (beyond 10 wt%), very fast evaporation of the solvent occurred, and significant residue of the CA tended to block the needle. This prevented the electrospinning process from taking place in making successful fibers. Lower CA concentrations ranging between 1 and 8 wt%, resulted in very even and high-quality fibers, but they did not allow the receival of continuous fibers with a diameter more than approximately 100 nm. Naturally, a low fiber diameter makes the fiber network difficult to work with in practice. Similarly, Son et al. (2004) reported an optimum CA and water concentration of 10 wt% and 15 wt%, respectively in their study of electrospun ultrafine CA-acetone-water fibers. Blocking was reported in lower water contents (higher CA).
Table 2. Electrospinning Capability and Fiber Quality of Rosin and Propolis in a Solution Based on Acetone with either H2O or DMAc
Fig. 1. Electrospun pure CA fiber networks based on either DMAc or H2O solution: a) CA (10%) in H2O basis; b) CA (5%) in DMAc basis
Fig. 2. Electrospun CA fiber networks with various rosin or propolis concentrations: a) CA (10%) rosin 5% in DMAc basis; b) CA (10%) propolis II 5% in DMAc basis; c) CA (10%) rosin 1% in DMAc basis; d) CA (10%) rosin 5% in DMAc basis
In general for the DMAc systems, thicker fibers were associated with higher CA concentrations. The spinning process in terms of the flow and even formation of networks was determined to be successful in all the applied CA concentrations. For the highest concentrations (20 wt%), slight accumulation emerged at the tip of the spinning needle during the electrospinning process. This made it necessary to clean the tip frequently. Luo et al. (2013) found a CA range of 12 to 15 wt% for their CA-acetone-dimethylformamide (DMF) and Anitha et al. defined a range of 6 to 14 wt% for CA-acetone-DMF (Anitha et al. 2012). These ranges agree with the CA concentrations of the DMAc systems found in this study.
The rosin and propolis integration (both propolis types) was shown to affect the electrospinning process by making it stabilized and led to a more uniform product. The addition of either rosin or propolis allowed for an increase in the processing speed (up to 5 to 7 mL per h), whereas without these additives, the speed was only 1 to 2 mL per h (higher rates led to dripping and frequent needle clean up). Along with the higher rates, rosin and propolis made the spinning process more steady, i.e., the jet of the solution was more stable. However, the addition of rosin and propolis tended to lead to the formation of beads in the nanofiber network (see Fig. 3). In agreement with the literature, mixtures with CA and carbon nanotubes (Luo et al. 2013) as well as CA and zinc oxide (ZnO) (Anitha et al. 2012) tend to form aggregates or beads since precipitation or aggregation occurs during spinning. When only rosin is used (in a DMF solution), additives have been reported to improve the quality of electrospun fibers and decrease their diameter (Nirmala et al. 2013). Here, the beads in the CA-rosin systems were studied using AFM and their content tended to be solid and homogenous. The beaded cross sections suggested that pure, phase separated rosin does not form the beads, but the rosin merely hinders the electrospinning process and the solution flow when the formation of beads occurs. The chemical composition and stiffness variation over individual beads inside the networks and fibers is an ongoing challenge. Therefore, it was not surveyed here.
Fig. 3. AFM analysis of the CA-rosin electrospun network and the fiber and beaded cross sections. The measurements were done in contact mode using an Au coated silicon probe. The ultra-microtomed slices including the electrospun network sample were laid on TEM grids with a carbon film.
Chemical Changes by FTIR
In order to support successful integration of rosin, propolis I, and propolis II into the CA fibers, the FTIR spectra of the pure rosin, pure propolis I, pure propolis II, pure CA fibers, rosin integrated CA fibers, and propolis integrated CA fibers (propolis I and II) were analyzed (Figs. 4 through 5). Figure 4a shows the FTIR spectra of pure rosin, pure
propolis I, and pure propolis II, where the rosin spectrum indicates a high content of abietic acid (Matsuyama et al. 2019).
Fig. 4. FTIR spectra of the raw materials used and electrospun CA fibers: a) raw materials; b) electrospun CA fibers with rosin (H2O basis)
First, it was observed that the carboxylic group of rosin showed a characteristic peak for the >C=O stretching (Moustafa et al. 2017) vibration band and the O–H stretching vibration at 1690 cm-1 and 3390 cm-1, respectively. By contrast, the >C=O stretching vibration for propolis I and propolis II were indicated at 1703 cm-1 and 1735 cm-1 and the O–H stretching (Hussein et al. 2017) vibration appeared at 3276 cm-1 and 3324 cm-1, respectively. For propolis, the >C=O peak was seen rather weak which could indicate the presence of hydroxycinnamic acids typical in propolis. For rosin, due to the C–H asymmetric stretching vibration of –CH3 group and symmetric C–H stretching vibration of –CH2 group, the characteristic bands (Moustafa et al. 2017) appeared at approximately 2921 and 2862 cm-1, respectively. In the data of propolis I and propolis II, the C–H asymmetric and symmetric stretching (Hussein et al. 2017) vibration peaks appeared at 2921 cm-1 and 2850 cm-1, respectively
The data in Fig. 4b shows the FTIR spectra of pure CA fibers (water), CA fibers with rosin 1 wt% of water and CA fibers with rosin 5 wt% of water. The pure CA fibers (water) clearly show the characteristic peak at 3486 cm-1 for the O–H stretching vibration of hydroxyl groups (Fei et al. 2017), whereas this peak in the CA fibers with rosin 1 wt% of water and CA fibers with rosin 5 wt% of water appear at 3482 cm-1 and
3477 cm-1, respectively. The CA fibers (water) show a >C=O stretching vibration peak at 1742 cm-1 due to the ester group (Fei et al. 2017) of CA. These peaks are shifted to
1737 cm-1 and 1736 cm-1 in CA fibers with rosin 1 wt% of water and CA fibers with rosin 5 wt% of water, respectively. The slight shift in the O–H and >C=O stretching vibration peak in the CA fibers with rosin (water) from the CA fibres (water) is due to the H-bonding interaction between the functional groups of rosin and functional groups of CA. Besides this, CA fibers (water) show a characteristic peak at 2940 cm-1 and 2870 cm-1 due to the C–H asymmetric stretching vibration of the –CH3 group and symmetric stretching (Fei et al. 2017) vibration of the –CH2 group of CA.
In rosin-integrated CA fibers (water), these peaks appear more clearly along with a higher intensity due to the merging of the C–H stretching of rosin with CA. Figure 5a shows the FTIR spectra of the CA fibers (water), the CA fibers with propolis I at 4 wt% and CA fibers with propolis II at 4 wt%. From Fig. 5a, it can be observed that a slight shift occurs in some of the peak positions for propolis I and propolis II integrated CA fibers with respect to the CA fibers. This can be attributed to the interfacial interaction between CA with propolis I and propolis II. The >C=O stretching vibration peak in both the CA fibers with propolis I 4 wt% of water and also CA fibers with propolis II 4 wt%, shifted to
1736 cm-1. However, the O–H stretching vibration peak shifted to 3480 cm-1 and
3477 cm-1 in CA fibers with propolis I 4 wt% of water and CA fibers with propolis II
4 wt%, respectively. This suggests that the H-bonding interaction between propolis II and the CA fibers is slightly stronger compared to propolis I and the CA fibers. Though the pure propolis I and pure propolis II show the C–H stretching vibration (asymmetric and symmetric) peak at the same wave number (2921 cm-1 and 2850 cm-1), the propolis I-CA fibers (water) and propolis II-CA fibers (water) show a shift in these peak positions.
The C–H asymmetric and symmetric stretching vibration in CA fibers (water) (2940 cm-1 and 2870 cm-1) shifted to 2921 cm-1 and 2852 cm-1 in propolis II-CA fibers (water) and 2940 cm-1 and 2880 cm-1 in propolis I-CA fibers (water). These peaks are clearly more intense in propolis I-CA fibers (water) and propolis II-CA fibers (water) compared to the pure CA fibers (water). This can be attributed to the merging of the C–H stretching of propolis I and propolis II with CA fibers. In general, rosin and propolis addition led to single-color and essentially transparent solutions (see an example in Fig. 4b) i.e. fully dissolved systems, and not to dispersions. However, during electrospinning, solvent is being extracted upon fiber formation and certain (nano scale) phase separation could occur. On a nano scale, based on AFM (Fig. 3) imaging, no evident phase separation was observed in the samples.