Process optimization, antioxidant, antibacterial, and drug adjuvant properties of bioactive keratin microparticles derived from porcupine (Hystrix indica) quills

Materials

Chloroform (95%), urea (granular), 2-Mercaptoethanol, 2, 2′-azino-bis(3-ethylbenzothiazoline-6-sulfonic acid) (ABTS), Muller Hinton Agar, were purchased from Sigma Aldrich (St. Louis, MO, USA). Methanol, polyvinyl Alcohol (PVA), acetone, sodium dodecyl sulphate (SDS), anhydrous sodium sulphate (), and dimethyl sulfoxide (DMSO) were purchased from Merck (Rahway, NJ, USA). All chemicals were supplied by local suppliers.  The erythromycin 250 mg (Indus Pharma, Pakistan) was purchased directly from a pharmacy shop in Muzaffarabad, Azad Kashmir.

Collection of porcupine quills

Samples of both fresh and old porcupine quills were collected from different sites of porcupine habitat in the district Muzaffarabad (34.3551°N, 73.4769°E) of the State of Azad Jammu and Kashmir, Pakistan. Through a fortnightly field survey, we collected the quills that had been shed naturally, without human contact with the animal. The quills were kept in airtight containers labeled with information. These samples were kept in a refrigerator at a low temperature (4 °C).

Extraction of keratin protein

The keratin extraction method with some modifications was used from earlier published research work (Yamauchi et al., 1996). The brief detail of the method with modifications is as follows. Samples of quills were cleaned overnight with agitation in a 1:1 mixture of methanol and acetone, then washed twice with Milli-Q water. Cleaned quills were allowed to air dry for the entire next day. Approximately, 25 g of the chopped quills were soaked in a solution containing 7 M of urea, 0.05 M of SDS, and 1.1 M of 2-mercaptoethanol to extract the keratin. The extraction of the keratin from quills was continued then at 60 °C for 5 h. The solution was then centrifuged at 10,000 rpm to extract the supernatant after being filtered through a 50-mesh stainless steel grid and redissolved in above solution for further 1 h. After addition of 2N HCl, the pH of hydrolysate solution was changed from 5 to 3, then a thick layer of precipitates started to settle at the bottom of the test tube (Gigliotti, Jaczynski & Tou, 2008). The precipitated keratin was washed with water and centrifuged three times for ten minutes at 12,000 rpm to eliminate salts and other contaminants. The precipitates of the keratin were then air dried and labeled as refined keratin. The yield of the keratin was measured according to following Eq. (1); (1)${\displaystyle \text{Yield of Keratin Protein}\left(\text{\%}\right)=\frac{\text{Amount of Extracted Keratin Protein}\left(\mathrm{g}\right)}{\text{Initial Amount of Porcupine’s Quills (g)}}\times 100}$

Extraction of lipid

A modified version of the process described by Bligh & Dyer (1959) was used to extract the lipids from small pieces of porcupine quills. Distilled water, methanol, and chloroform were added in the ratios of 1:2:0.8 v/v. In order to separate the two phases, chloroform and water were added in a proportion of one-third of the total volume. A separating funnel was used to separate the lower chloroform phase, which was then collected in a screw-capped septum vial with anhydrous sodium sulphate and kept at 4 °C for later analysis. The amount of lipid was measured in mg/g of dry weight of porcupine’s quills. The average amount of lipids extracted was 4.12 mg/g.

Synthesis of bioactive keratin microparticles

In Fig. 1, the detailed scheme is elaborated which was applied to prepare bioactive keratin microparticles. Keratin protein 50 mg mL⁻¹ were added into PVA solution (3:1 w/w) and mixed for a minimum of 4 h at room temperature before casting into a petri dish and allowed to dry at room temperature. The resultant film was then washed with water to solubilize the PVA and to achieve the formation of keratin microparticles. These keratin microparticles were re-dispersed three times in Milli-Q water and centrifuged in order to obtain the pellet.

These keratin microparticles were turned bioactive by adding 3 w/v% of lipid extracted from porcupine quills. The lipid was added to keratin microparticles under continuous stirring at a low temperature, which was further followed by air drying. The 5% solution of drug erythromycin (antibiotic) treated with bioactive keratin particles for 18 h for drug loading. The following abbreviations were used for representing different compositions.

L—Lipid

E—Erythromycin

K—Keratin microparticles

LK—Lipid coated keratin microparticles

LKE—Lipid coated keratin microparticles (bioactive) loaded with erythromycin

Response surface optimization of yield and size of the keratin microparticles

Box-Behnken design was adopted for response surface analysis of the independent variables i.e., 2-mercaptoethanol, urea, reaction time and temperature for optimizing yield of the keratin and particle size of the keratin microparticles. Four independent variables were tested at three levels of inputs for each of 2-mercaptoethanol (1.5, 2.0, 2.5%), urea (6, 7.5 and 9%), reaction time (12, 18, 24 h) and temperature (90, 100, 110 °C) to map the response of dependent variables i.e., yield (%) and particle size (µm) of the keratin. The Design-Expert 6.0.8 portable version (Stat-Ease, Inc. Minneapolis, MN, USA) was used to process the data of the dependent variables by fitting into the quadratic polynomial model. The analysis of variance (ANOVA) was selected to find out the significant difference in the interaction among the independent variables which significantly affected the response of the dependent variables.

Surface analysis

Images captured under the optical microscope (Leica S6D Stereo Zoom Microscope, Wetzlar, Germany) were used to investigate the surface of L, K, LK, and LKE compositions. After applying 1 µl of each sample on the glass slide, samples were dried under laminar flow at room temperature. ImageJ software is a Java-based image processing application accessible from the National Institutes of Health webpage (https://imagej.nih.gov/ij/) was used for processing of the microscopy images. The size of the keratin microparticles was calculated after calibration of the scale bar on the captured microscopy images, and a bar chart with a distribution curve was used to display the variation in the size and distribution of the keratin microparticles.

Chemical composition

Infrared spectra were measured after 50 µl of each sample mixed with KBr powder. Mixture was mixed well with pestle and mortar and pressed under pressure to obtain compact disc. The spectra of each composition was recorded using fourier transform infrared (FTIR) spectrometer (Shimadzu-8400S, Kyoto, Japan). All spectra were recorded in the range from 4,000 to 400 cm⁻¹ with a resolution of 4 cm⁻¹, and 32 scans.

Scavenging activity

Scavenging activity of the keratin microparticles was assessed using ABTS (Mistry, Keum & Kim, 2016). Approximately, 180 mL of ABTS solution was mixed up with 20 mL of keratin microparticles’ solution with different doses of 0.1, 1, 10, 100, and 1,000 mg mL⁻¹ and then incubated for 10 min in the dark environment. The absorbance at 734 nm was then determined through UV-Visible spectroscopy (T60UV-Visible Spectrophotometer, PG Instruments, Wibtoft, UK). Ascorbic acid served as the reference standard to evaluate the antioxidant activity of the keratin microparticles. The scavenging activity (%) was calculated from following Eq. (2) (2)${\displaystyle \text{Scavenging activity}\left(\text{\%}\right)=\frac{\text{Absorbance of standard}\times \text{Absorbance of sample}}{\text{Absorbance of standard}}\times 100}$ The scavenging activity of the keratin microparticles was compared with scavenging activity of the ascorbic acid and the minimum inhibitory concentration (IC₅₀) was calculated using following standard Eq. (3). (3)${\displaystyle {\text{IC}}_{50}=\left(50-\mathrm{b}\right)/\mathrm{a}.}$ Where b is the intercept and a is the slope of the given curve.

In vitro antibacterial activity

LKE (load of erythromycin @ 5%) along with other compositions i.e., K, L, and LK were tested through well diffusion method for in vitro antibacterial potency against the bacterial species—Staphylococcus aureus (gram-positive) and Escherichia coli (gram-negative). Antibacterial activity tested in triplicates for each composition.

Different concentrations of each composition were set to 20, 40 and 100 mg L⁻¹ dissolved in DMSO and were loaded into wells formed on the Mueller Hinton Agar petri plates. Then, petri plates were incubated at 37 °C for 24 h. The zone of inhibition (ZOI) was measured by using vernier caliper (Mitutoyo, Japan). DMSO was used as a control in reference wells.

Keratin extraction and microparticles synthesis

The protein keratin, which was taken out of porcupine quills, is depicted in its raw state in Fig. 2A. This keratin protein is a large, bulky protein mass that is whitish in color. Maximum keratin yield from quill waste was achieved at 120 °C, 7% urea, 18 h reaction duration, and 2% mercaptoethanol. As a result, the procedure and testing conditions were appropriate for obtaining keratin protein from porcupine quills.

The keratin microparticles, which were made from keratin protein that had been crudely purified, are shown in Fig. 2B. After the addition of 50 mg mL⁻¹ of keratin protein into the PVA solution (3:1 w/w), and stirring for 4 h, the keratin was successfully dispersed. The isolated keratin protein was successfully fine-tuned into well-dispersed microparticles after the keratin dispersion in PVA solution. Microparticles with a minimum size of 69 µm were produced at conditions of 120 °C, 7% urea, 18 h reaction time, and 2% mercaptoethanol, whereas microparticles with a maximum size of 115 µm were seen at conditions of 18 h, 110 °C, 2% mercaptoethanol, and 7% urea. The extraction of keratin as a biopolymer from quill waste was therefore successfully proven in this work.

Effect of extraction conditions

Temperature

The effects of varying the temperature settings (90, 100, 110, and 120 °C) on the extraction of crude keratin protein from porcupine quills and its impact on the size of keratin microparticles are depicted in Fig. 3A. The yield of keratin protein was discovered to increase generally with an increase in the reaction vessel’s temperature. The outer shell of the quill effectively protects the keratin located deeper within the cortical part of the quill. As a result, maintaining of the higher temperature may increase the disruption process and more efficiently release the keratin protein. At 80 °C, the keratin yield was 30%; at 90–100 °C, it increased further up to 35%, which is a 5% improvement. The increase in the keratin yield of 39.5%, was noticeable upon the temperature increase of 110 °C. Overall 9.5% improvement in the yield of the keratin protein from the porcupine’s quills was observed after gradually increasing the temperature under extraction conditions. The factors like, i) the difference in the discharge of amorphous and crystalline forms of the keratin from the cortex of the porcupine quills; and ii) possibly poor heat transfer at temperatures above 120 °C would have possibly caused barrier towards achieving higher yield than the estimated in this study.

The keratin protein’s microparticle size positively covaries with the extraction of the keratin. Figure 3A shows the change in the size of the keratin microparticles in response to the temperature changes. The increase of temperature from 80 to 110 °C, the size of the keratin microparticle increases. At 110 °C, the largest size of keratin microparticle was measured to be 103.5 µm. Keratin extracted at different temperatures was dispersed in PVA solution and noticed the difference in the sizes of the keratin microparticles. This difference demonstrates temperature affectes the keratin protein’s structural integrity during the extraction process. Due to the difference in the structure of the keratin, size variation was noticed during the formation of the keratin microparticles. PVA functions as a dispersant that lowers the interfacial tension between water and insoluble polymers (i.e., keratin). The size of the microparticles increases in case of the keratin extracted at the higher temperature. Increase in the yield of the keratin at higher temperature may have involved irregular and bigger size of the keratin, which were dispersed poorly in the PVA solution. Hence, consequently, large size of keratin microparticles are generated. At 120 °C, the decrease in the size of the keratin microparticle was noticed. In such cases, higher temperature (i.e., 120 °C) may have dissociated the more labile amorphous portion of the keratin protein from its more structured crystalline portion, which reduced the size of the keratin microparticles.

Response surface analysis

Keratin yield

The lowest proportion (27.26%) of keratin protein was produced by the independent variables mercaptoethanol 1.5%, urea 7.5%, reaction time 18 h and temperature 100 °C. The reaction mixture of 2.5% mercaptoethanol, 7.5% urea, 18 h at 110 °C, produced the best yield of keratin (42.25%). The important elements that contributed to the increase in keratin yield were the change in temperature from 100 to 110 °C and the increase in mercaptoethanol concentration from 1.5 to 2.5%. Mercaptoethanol, 2.5%; urea, 7.5%; reaction time, 24 h; temperature, 100 °C; and the largest size of the keratin microparticles achieved was 118.87 µm. The smallest size of the keratin microparticles, 60.5 µm, was produced with 1.5% of mercaptoethanol , 6% of urea, and 18 h of reaction period and 100 °C of temperature. The size of the keratin microparticles was affected by the concentration of mercaptoethanol, urea, and reaction time. Less stringent conditions probably helped to produce the smallest keratin microparticles.

The use of dependent and independent variables enabled statistical data processing to establish the response’s dependence on the independent variable. Based on the statistically analysed data for keratin yield and keratin microparticle size, Fig. 4 displays the analysis of variance (ANOVA). According to Table 1, the quadratic model of the keratin yield experiment had a p-value of 0.0001 (less than 0.05), indicating that it was significant. The model also had an F-value of 10.61, meaning that there was only a 1.79% possibility that a value this large could be the result of noise. This F-value highlighted the relevance of the model. The model had a robust enough signal to be used for optimization, as evidenced by the sufficient precision measure of 16.403 (more than 4) for the sample. Since they both had p-values of 0.0001, the mercaptoethanol and urea were significantly altering the keratin yield.

Table 1:

Box-Behnken experimental design runs, factor levels and response variables.

Run	Factors (X)				Response (Y)
	X1	X2	X3	X4	Y1	Y2
	Mercaptoethanol (%)	Urea (%)	Reaction time (h)	Temperature (°C)	Keratin yield (%)	Particle size (µm)
1	2	6	12	100	34.26	82.23
2	2	7.5	24	110	40.25	83.25
3	2	7.5	12	110	38.54	85.87
4	2.5	9	18	100	40.92	118.35
5	1.5	6	18	100	29.54	60.65
6	2	6	18	90	35.25	74.84
7	1.5	7.5	18	90	28.11	66.86
8	1.5	7.5	12	100	27.36	63.36
9	1.5	9	18	100	33.25	71.87
10	2	7.5	18	100	39.58	87.85
11	2	7.5	24	90	41.68	86.23
12	2	6	24	100	36.32	76.17
13	2	7.5	18	100	40.00	88.24
14	2	6	18	110	37.20	76.22
15	2	9	18	90	40.50	91.01
16	2.5	7.5	18	110	42.25	114.47
17	1.5	7.5	24	100	31.00	67.47
18	1.5	7.5	18	110	31.50	61.50
19	2.5	7.5	18	90	41.25	116.25
20	2	9	24	100	42.51	93.89
21	2	7.5	12	90	39.00	84.87
22	2	9	12	100	41.00	86.87
23	2.5	7.5	12	100	40.00	111.58
24	2.5	7.5	24	100	40.90	118.87
25	2	9	18	110	41.50	92.11
26	2	7.5	18	100	39.00	87.50
27	2.5	6	18	100	37.56	109.00
28	2	7.5	18	100	40.00	88.44
29	2	7.5	18	100	40.12	89.11

DOI: 10.7717/peerj.15653/table-1

The results of an ANOVA with interactions between several independent variables and the dependent variable, keratin microparticle sizes, are given in Fig. 5. A p-value of 0.0001 or less than 0.05 indicated the quadratic model’s relevance. Additionally, the model had an F-value of 16.62, which indicated that there was only a 0.78% possibility that a value this large could be the result of noise. This F-value highlighted the relevance of the model. Since they exhibited p-values of 0.0001, mercaptoethanol and urea were significantly impacting the keratin yield and keratin microparticle size response. Figure 6 shows that both experimental and predicted data for yield and size of the keratin microparticle fit well through applying the quadratic model through the response surface method.

Surface morphology of keratin microparticles

The shape, size and distribution of keratin microparticles were studied using optical microscopy (Fig. 7). Figure 7A, round shape of the lipid was used as reference of the shape assessment for keratin microparticles. In Fig. 7B, the shape assessment revealed that keratin microparticles were heterodisperse. This shape showed the difference from rounded to multifaceted morphology in the aqueous solution. The dried form of keratin microparticles showed the characteristics of specially dispersed and randomly loosely packed form. These observations are supported by earlier work (Rad, Tavanai & Moradi, 2012; Zhang et al., 2013). The size and distribution of the particles ranged up to 100 µm. However, the highest keratin microparticles count of 25% was noticed for the size of ≤200 µm.

IR spectroscopy

The chemical makeup of the keratin microparticles and their various interactions with lipid and drug were determined using FTIR (Fig. 8). The results show that the chemical structure of keratin microparticles and their formulation with lipid and drug have changed. All of the examined materials’ absorbance spectra displayed the same pattern of keratin-specific peaks as in earlier studies (Ma et al., 2016).

The keratin samples displayed spectral bands that were designated as Amide A, Amide I–III and corresponded to peptide bonds (–CO–NH) (Xu et al., 2014; Sharma et al., 2017). Stretching vibration of –O–H and –N–H (Amide A) is responsible for the wide vibration band region between 3,400 cm⁻¹ and 3,250 cm⁻¹(Pavia et al., 2014). Stretching vibration of –NH bonds is linked to the Amide A band, which was observed at 3,280 cm⁻¹. The stretching bonds in –CH are connected to the transmission bands that range from 3,000 to 2,700 cm⁻¹. The –C =O stretching (Amide I) is responsible for the strong transmission band in the region of 1,628 cm⁻¹ (Kakkar, Balaraman & Shanmugam, 2016). The C–H stretching and N–H bending are attributed to the Amide II vibration band in the range of 1,533 cm⁻¹ (Eslahi, Dadashian & Nejad, 2013). Due to C–N stretching and N–H bending, the weak band between 1,315 cm⁻¹ and 1,263 cm⁻¹ corresponds to the Amide III band (Vasconcelos, Freddi & Cavaco-Paulo, 2008). The asymmetric and symmetric S–O stretching vibrations (cysteine-S-sulfonated residues), which are generated due to the interaction of sulphides and cysteine in the protein extraction process, respectively, are responsible for the absorption bands at 1,045 cm⁻¹ (Aluigi et al., 2014). The successful coating of the keratin microparticles concealed the keratin signal in LK samples. Overlapping lipid and drug signals were seen in LKE. Thus, it was discovered that the surface of keratin microparticles had been successfully coated with drug-loaded in lipids.

Antioxidant activities

The extracted keratin microparticles were evaluated for the antioxidant activities by using ABTS radical free scavenging bioassay that showed the maximum absorbance at 734 nm The activity was investigated at different concentrations such as 10, 20, 40, 60, 80 and 100 mg mL⁻¹ as shown in Fig. 9. Overall, highly significant antioxidant efficacies were observed for extracted keratin microparticles. The keratin microparticles had IC₅₀ values of 29.83 mg mL⁻¹ in ABTS bioassays, as compared to ascorbic acid value of 16.19 mg mL⁻¹. The efficacies of keratin microparticles as antioxidant agents were higher as compared to ascorbic acid.

Antibacterial activity and adjuvant effect of the keratin microparticles on drug efficacy

In Table 2, lipid extracted from the porcupine’s quills showed the ZOI values of 25.65 ± 0.54 mm against the E. coli which was higher than the ZOI noticed for S. aureus which was 24.54 ± 0.41 mm. Erythromycin showed almost the same ZOI values for E.coli and S. aureus. The keratin microparticles showed ZOI of 16.5 ± 0.71 mm for E. coli which was higher than the ZOI values of 15.75 for S. aureus. The keratin microparticles after lipid coating showed an activity of 23.50 ± 0.70 mm for E. coli and 25.75 ± 0.35 mm for S. aureus. The lipid coated keratin microparticles with loaded erythromycin have shown the activity of 43.00 ± 0.70 mm and 40.25 ± 0.35 mm for E. coli and S. aurues respectively. It is an interesting fact that the lipid coating of keratin microparticles and then loading of drug to lipid coated keratin microparticles increased the inhibitory effect due to the adjuvant effect of both the lipids and keratin microparticles on the drug erythromycin. Inhibitory activity of the keratin microparticles shifted approximatley 16.00 to 36.00 mm after lipid coating of the keratin microparticles. Therefore, a difference of 20 mm (55%) in the inhibitory activity which is a significant contribution. Similarly, the lipid coated keratin inhibitory activity was approximately 24 mm which increased to 42 mm after loading of drug erythromycin. This difference of inhibition was 18 mm (43%).

Table 2:

The antibacterial activity of keratin microparticles with different compositions against clinical pathogenic bacteria E. coli and S. aureus.

DOI: 10.7717/peerj.15653/table-2