Novel nano composites from Citrus limon and Citrullus colocynthis agricultural wastes for biomedical applications

In recent years, academic and industrial research has focused on using agro-waste for energy and new material production to promote sustainable development and lessen environmental issues. In this study, new nanocomposites based on polyvinyl alcohol (PVA)-Starch using two affordable agricultural wastes, Citrus limon peels (LP) and Citrullus colocynthis (Cc) shells and seeds powders with different concentrations (2, 5, 10, and 15 wt%) as bio-fillers were prepared. The nanocomposites were characterized by Dielectric Spectroscopy, Fourier-Transform Infrared (FTIR), Scanning Electron Microscopy (SEM), Transmission Electron Microscopy (TEM), and water swelling ratio. The antimicrobial properties of the nanocomposites against Escherichia coli, Staphylococcus aureus, and Candida albicans were examined to investigate the possibility of using such composites in biomedical applications. Additionally, the biocompatibility of the composites on human normal fibroblast cell lines (HSF) was tested using MTT (3-[4,5-dimethylthiazol-2-yl]-2,5 diphenyl tetrazolium bromide) assay. The results demonstrate that the filler type and concentration strongly affect the film's properties. The permittivity ε′, dielectric loss ε″ and conductivity σdc increased by increasing filler content but still in the insulators range that recommend such composites to be used in the insulation purposes. Both bio fillers control the water uptake, and the samples filled with LP were more water resistant. The polyvinyl alcohol/starch incorporated with 5 wt% LP and Cc have antimicrobial effects against all the tested microorganisms. Increasing the filler content has a negative impact on cell viability.


Materials
PVA is supplied by Quali-kems from India.Wheat starch (St) provided by Fluka Company from USA.Sigma-Aldrich provides Glutaraldehyde (50%) from Germany, and Glycerol is provided by Fisher from Germany.

Methods
Collection and preparation of the fillers Lemon (Citrus limon) was collected from the Egyptian markets.Then the lemon was washed thoroughly with water, peeled off, the peels dried in a hot air oven (70 °C) overnight, and finally grinded into fine powder.
Citrullus colocynthis was purchased from Egyptian markets.The plant samples were cleaned with water and allowed to dry in the air for two weeks.The dried fruit (the shell and seeds) homogenized with an electric grinder to fine powder, and stored in sealed flasks for further use 24 .
Figure S1 is a representation of the whole fruit and the used parts from it.
Preparation of the PVA/starch blend Typically, the casting procedure is used to create particular PVA/starch samples at a ratio of (50:50) wt%.Firstly, 3.5 g of PVA was dissolved in 75 ml of distilled water at 95 °C for 30 min to create a PVA solution.Next, for the PVA/Starch blend, mix starch, and glycerol at a ratio (70/30) wt% with water by using a magnetic stirrer for 10 min; following this, as a starch crosslinking agent, glutaraldehyde was applied at a concentration of 10% (based on starch weight).After that, starch was added to the PVA solution while stirring continuously to create a homogenous mixture.The produced mixture solution was put to clean Petri dishes and air-dry overnight.After the films were taken out of the Petri dishes and placed in vacuum desiccators to ensure that all of the water had been completely removed, they were used for further investigation 25,26 .
Preparation of LP or Cc/PVA/starch reinforced nanocomposites 0.07gm (2%) of LP or Cc was added to the PVA/starch solution and stirred with a magnetic stirrer for two hours To achieve a homogeneous distribution.Ultimately, the mixture was poured onto clean, dry Petri dishes and heated in an oven for 24 h at 60 °C to create the film.The previously described process was repeated for 5, 10, and 15 wt%, and it is the same method employed for creating blend film from LP or Cc with PVA/starch 27 .

Techniques Transmission electron microscope (TEM)
The particle size of the utilized bio-fillers was measured using TEM model: JEM-HR 2100 and accelerating voltage 200 kV, Japan.

Fourier-transform infrared (FTIR)
The spectrum was recorded with a JASCO FT/IR 300 E FTIR Spectrometer (Tokyo, Japan).

Dielectric measurements
The capacitance C and loss tangent tan δ were measured in the frequency range of 100 kHz to 120 MHz with the LCR HiTESTER 3535 Model.The MC-100 Dielectric Cell was used for such measurements.The permittivity ε′ was calculated from the well-known relation ε′ = C/Co where C is the capacity in the presence of the sample, and Co is that in the presence of air.The dielectric loss ε″ was obtained from the multiplication of the ε′ values by the loss tangent tan δ.

Electrical conductivity measurements
The volume resistivity ρ was measured by Super Meg Ohm Meter Hioki SM7110 (Hioki, Japan).The samples were placed in shielding box type Hioki SME3811 with outer and inner electrode diameters of 24 mm and 19 mm, respectively.This shield box is guarded and tuned for applying DC potential up to 1000 V.The Super MegOhm Meter Hioki SM7110 can measure small currents up to 1pA peco Amperes in resolution of 1fA femto Amperes.
From which the electrical conductivity σ was calculated.The electrical conductivity σ was calculated from the measured resistivity ρ according to the well-known relation σ = 1/ρ.

Swelling ratio
Film squares of size 1 cm × 1 cm were used for the water swelling test.The film squares were dried in an oven until constant weight was reached (W i ) and placed in petri dishes.50 ml distilled water was then poured into each dish.After 60 min and 120 min of immersion at 37 °C, the excess water was removed by gently wiping film squares with filter paper, and the weights were recorded (W f ).The water swelling percent is calculated as follows; where: Wi and W f are weights of films, initially and after time t, respectively.

Microbiological analysis
Using the shake flask method, the antimicrobial activity of specific samples was tested against two types of pathogenic bacteria: Gram-positive (Staphylococcus aureus), Gram-negative (Escherichia coli), and pathogenic yeast (Candida albicans).The reduction of colony forming units (CFU) was determined by measuring the optical density (OD) at 600 nm.Using this method, the pathogenic strains' inoculum size was prepared from fresh working stoke cultures and adjusted to approximately 0.5 McFarland standard (1.5 × 108 cfu/ml).Additionally, 25.0 μl of both bacterial and fungal suspensions were poured under sterile conditions into 100.0conical flasks containing 20.0 mL of a nutrient broth medium (NB).Each sample was applied separately on each of these inoculated flasks, and the suspensions were incubated for 24 h at 37 °C with 150.0 rpm rotating shaking conditions [28][29][30] .
The antimicrobial activity was measured throughout the relative [OD (%)] reduction of these pathogenic strains present in flasks containing the treated samples compared to the control flasks that contained pathogenic strains only without any treatments.All results were expressed according to the following equation: where: A: The CFU of pathogenic strains only without any treatments present on the control flask.B: The CFU of pathogenic strains on the flask contains the treated sample.

Cytotoxic effect on fibroblast cell lines (HSF)
The mitochondrial-dependent reduction of yellow MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide) to purple formazan was used to measure the viability of the cells.The following procedures were performed in a sterile environment with a Class II A2 Laminar Flow Biosafety Cabinet (Manufactured by Labconco).
After batch culture for ten days, cells were seeded at a density of 10 × 10 3 cells /well in freshly prepared growth medium in 96-well plastic plates.The plates were then kept at 37 °C for 24 h under 5% CO 2 , either without any drugs (negative control) or with various drug concentrations to obtain the final concentration of (1000, 500, 250, 125, 62.5, 31.25,15.625, 7.812 µg/ml 31,32 .
Next, the absorbance was determined at 595 nm using a reference wavelength of 620 nm and a microplate multi-well reader (Bio-Rad Laboratories Inc., model 3350, Hercules, California, USA. (1)    FTIR analysis is an essential tool for understanding the interactions at the interfaces of the matrix and the incorporated fillers.The FTIR spectrum of the PVA/starch (Blank) polymer blend is displayed in Fig. S2.The assignment of the spectrum is listed in Table 1.From Fig. S2, the presence of both polymers was evident in the blank polymer blend's spectrum.

FTIR analysis for LP and PVA/starch/LP nanocomposites
The spectrum of the LP, and the main bands identification are 3286 cm −1 refered to hydroxyl groups OH of macro molecular association (pectin and cellulose).2921 cm −1 : asymmetric stretching of -CH group.2850 cm −1 symmetric stretching vibration of -CH group.1608 cm −1 ; stretching vibration of carboxylate ions (COO-) of pectin.1730 cm −1 : stretching vibration of ester carbonyl group(C=O).1020-1300 C-O group.The spectrum is in accordance with others reported for lemon peel 36,37 .Some changes were observed in the spectra after incorporating LP into the PVA/starch matrix.It could be observed that the peak of hydroxyl groups at 3281 cm −1 has shifted to higher wave numbers, and their intensities were lowered compared to the spectrum of the blank.The 759 cm −1 peak shifted to a higher wave number.There were also changes in the intensity of nearly all other peaks.These changes may be attributed to modified PVA/starch association or the formation of new hydrogen bonds between the PVA, starch, and lemon peels 17,38 .
After incorporation of Cc in the PVA/starch composites, the spectra revealed the presence of numerous changes; there was a decrease in intensities of some bands such as 3281 cm −1 , 1715, 1647, 1377and 1246 cm −1 and the bands at 1078, 1019 and 924 cm −1 were shifted to higher wave number.The appearance of a new band at 1538 cm −1 which present in the spectrum of the Cc due to the phenol ring.All these changes predict the presence of new bonds that takes place between Cc and the PVA/starch blend.These bonds are mainly due to the presence of many functional groups in the Cc structure, as seen in the spectrum in Fig. 2.

Swelling ratio
In the swelling test, after immersion of the samples in distilled water at 37 °C, unfilled PVA/starch film dissolved in water within the first hour, which indicates poor resistivity to water.This was expected and consistent with previous studies since PVA and starch are hydrophilic polymers 25 .
The effect of the type and concentration of the nano-fillers on the water swelling behavior of PVA/starch nanocomposites loaded with LP and Cc is represented in Fig. 3. From the figure, the swelling percentage decreases by increasing the LP and Cc filler concentration, improving the filled nanocomposites' water resistance.Moreover, it is also observed that, the nanocomposites containing LP were more resistant to water absorption than that containing Cc.
The controlled swelling behavior of the filled films may be linked to the crosslinking effect of the fillers and PVA/starch matrix, which is confirmed by FTIR results, thus depreciating the possibility of water uptake.

Scanning electron microscope
SEM micrographs of PVA/starch, PVA/starch/LP, and PVA/starch/Cc nanocomposites with filler concentrations 5 and 15 wt% for the sake of brevity were presented in Fig. 4a and b with two magnifications.The pristine PVA/ starch blend exhibited a relatively heterogeneous structure, which suggested intermediate miscibility of PVA with starch in the blend.
After incorporation of the nano-fillers, LP and Cc dramatic changes in the blend films were observed.Increasing the concentration of the filler makes the matrix denser and more compact, which appears clearly in the nanocomposites containing 15 wt% for both fillers as an interlinked structure.This is considered a reasonable interpretation for the decrease in the water uptake by increasing the filler content, and the high swelling of the unfilled polymer blend is a result of its loose network compared to the reinforced ones.

Dielectric measurements
The permittivity ε′ and dielectric loss ε″ were measured over a frequency range 10 5 to 10 8 Hz at a room temperature of about 25 °C.The obtained data are given in Fig. 5a and b.
The ε′ values of a matter can be a good indicator of the possibility of its capacity to store up energy in the incidence of the electric field.Figure 5a and b show that ε′ are high at low frequencies for nanocomposites containing both fillers.The increase in ε′ values at lower frequencies could be explained according to the space charge effects 39,40 and the interfacial polarization 41 .This rise in governed space charge polarization, due to the formation of the confidential mobility of the bound dipole carriers, is reliable for the formation of orientation polarization 42 .This increase was quiet down and became linear, showing no further logical change at high-frequency ranges.
The ε″ explains the attendant loss or the dissipation of power 43,44 .With the same condition as ε′, the dielectric loss ε″ for LP and Cc nanocomposite systems was measured as a function of the applied f, and the data are pointed up graphically in Fig. 5a and b.The ε″ is taken as an indicator of the losses in energy due to the survival of the movement of dipoles between the PVA/starch matrix and the used filler.At that frequency range, the losses may be due to the moving dipoles due to the presence of filler 45,46 .To recognize the effect of these fillers on both ε' and ε″ and how they are exaggerated either by the type or content of the used fillers, both values are demonstrated in Fig. S3 at fixed frequency f = 100 kHz.That figure reflects that both ε′ and ε″ increased by increasing filler content; in addition, nanocomposite containing Cc as a filler possess higher values of ε′ and ε″ than those of LP-filled ones.
Figure 5a and b show that the curves concerning εʺ and the applied f are so convoluted and are a sign of the presence of extra than one relaxation procedure.So, it was worthwhile to analyze such curves to obtain the various relaxation processes.The analyses of such curves after subtraction of the losses due to the electrical www.nature.com/scientificreports/conductivity were done using two Frohlich terms with distribution parameter P = 4 and Havriliak Negami function according to the equations given elsewhere 47 .An example of the analyses for PVA/starch/ 10% LP was given in Fig. S4a.The obtained data of both τ 1 and τ 2 are given in Fig. S4b.The obtained relaxation times τ 1 and τ 2 could be ascribed to some local molecular motions of the side chain and attached groups rather than the main chain motion as it is expected to be frozen since the measurements were carried out at 25 °C, i.e., lower than the glass transition T g of the polymers under investigations 48,49 .

Electrical conductivity
The calculated electrical conductivity σ was illustrated graphically versus the applied voltage V for both PVA/ starch/LP and PVA/starch/Cc nanocomposites, Fig. 6a and b.The Ohmic dc conductivity σ dc was calculated from the linear relationship between σ dc and the applied volt at low voltage values.The obtained data are illustrated graphically versus filler content in Fig. 6c.σ dc values increased by increasing filler concentration for both fillers and Cc values were higher than composites containing LP.
Figure shows that σ dc values are in the order of 10 -10 S/cm, which highly recommends such composites to be used for insulating purposes.

Cytotoxicity effect on fibroblast cell lines (HSF)
Cytotoxicity testing is a favorable first step to assure the biocompatibility of a biomaterial.In this study, despite Citrullus colocynthis being a herbal medicine, it has some toxic effects.Shafaei et al. investigated the toxic effects of C. colocynthis on albino rats.The rats given seed extract at doses of 100 or 200 mg/kg/day showed relatively mild intestinal damage.Unlike seed extract, C. colocynthis pulp extract has the potential to be lethal to rabbits.Thus, for medicinal applications, seed extract might be the best option 50 .
This study performed and suggested these new composites for topical applications so the toxicity study was of great concern.
The samples were tested against the Normal Human Skin Fibroblast.Samples concentration range between (1000 to 31.25 µg/ml) using MTT assay.
Figure S5 represents the viability versus different concentrations for 2 and 5 wt% LP-loaded nanocomposites for the sake of brevity, and Table 2 displays the viability, cytotoxicity, and IC 50 of the nanocomposites containing LP.
Figure S6 represents the viability versus different concentrations for 2 and 5 wt% Cc-loaded nanocomposites for the sake of brevity, and    The results indicated that the nanocomposites incorporated with fillers have better viability than unfilled ones; the decreased viability of the blank sample may be a result of using the chemical crosslinker glutaraldehyde.It is reported that chemical crosslinking may have a negative effect on cell proliferation 51 .
According to the ISO 10993-5 in-vitro cytotoxicity standard, a cytotoxic effect is defined as a "reduction of cell viability by more than 30%" 52,53 .
Figure S7 represents a comparison between the 2 fillers at different concentrations.It can be noticed that the nanocomposites containing 2 and 5 wt% of LP demonstrate high viability of cells (more than 70%), which is considered safe for normal skin fibroblast.However, for the nanocomposites containing 2 and 5 wt% of Cc, Cell vitality values reached the minimum ISO requirements (70.5 and 69.07), which is the border between cytotoxic and non-cytotoxic.

Shake flask method (dynamic method)
The antimicrobial activity of PVA/starch blank and nanocomposites containing 5 wt% of each filler was chosen as the optimum filler concentration to be tested against 3 selected microorganisms Gram-negative bacteria, Escherichia coli, Gram-positive bacteria Staphylococcus aureus and candida albicana.The results are presented in Table 4.As indicated in the table, PVA/starch blank film did not show any antimicrobial activity against tested microorganisms.On the other hand, both types of nanocomposites positively affect the tested organisms, and higher values were recorded for nanocomposites containing LP.
The antimicrobial potency of these nanocomposites is believed to be due to phenolic compounds, flavonoids, and essential oils contained in the bio-fillers, which added biologically active values and consequently enhanced the antimicrobial activities of the nanocomposites.

Conclusion
In this study, nanocomposites based on PVA/starch incorporated with two different types of waste bio-fillers namely Citrus limon (peels) and Citrullus colocynthis (seeds and shells) powders have been prepared using solution casting technique.The target was to get rid of such waste in one hand and on the other hand to obtain eco-friendly composites useful for various applications such as biomedical and or anti static applications.The dielectric properties and water resistance improved after the incorporation of both fillers.Considering the testing findings, PVA/starch/LP composites were more water resistant.
Nanocomposites containing higher concentrations of the biofillers (10 and 15 wt.%) show cytotoxic effects on normal fibroblast cell lines, so they are excluded in biomedical applications, but they still have advantages in use in insulating applications as described by the dielectric results.
In a final conclusion, Lemon peels and Citrullus colocynthis powders in the 5 wt% filler ratio act as antimicrobial and reinforcing agents for the PVA/starch matrix.Nevertheless, the nanocomposites containing lemon peels were preferable in biomedical applications.They recommended undergoing an in vivo study to confirm their efficiency for intended use as skin contact material as patches or wound healing material.On the other hand, the values of the σ dc are in the order of 10 −10 S/cm which highly recommends such composites to be used in the insulating purpose that wide up the applications of such composites.

Table 4.
Relative reduction [OD reduction (%)] of the pathogenic strains after 24 h incubation using the shake flask method.

Figure 1
Figure1shows the TEM images of lemon peel powder (LP) and crushed Citrullus colocynthis (Cc) particles.The micrograph shows ultrafine particles in the nano range with diameter ranges 7-9 nm for lemon peel particles and from 19 to 25 nm for Cc.
Figure 2 represents the FTIR spectra of the nanocomposites along with individual LP and Cc nanofillers.

Figure 3 .
Figure 3.Comparison between swelling ratios of the nanocomposites containing different concentrations of the nano-fillers.

Table 2 .
Cell viability, cytotoxicity, and IC50 for PVA/starch/LP nanocomposites at 125 µg/ml sample concentration.IC50 (concentration of the sample which causes the death of 50% of cells in 48 h). PVA/

Table 3 .
Cell viability, cytotoxicity and IC 50 for PVA/starch/Cc nanocomposites at 125 µg/ml sample concentrations.IC 50 (concentration of the sample which causes the death of 50% of cells in 48 h).