Introduction

Bacterial nanocellulose (BNC) is biosynthesized mostly by acetic acid bacteria (e.g., Komagataeibacter hansenii), where, during the fermentation, the bacterial cells first secrete nanocellulose chains which assemble and aggregate, and then bundle into nanocellulose ribbons. Aerobic bacteria are known for their active motility, hence, under static conditions, BNC membrane (BNCM) is formed at the liquid–air interface (Ullah et al. 2019). Depending on the biosynthesized conditions, BNCM exhibit a wide range of remarkable physico-chemical properties, including ultrafine, porous morphology with a high specific surface area, advantageous mechanical properties, high crystallinity and water holding capacity. Besides, it’s intrinsically high purity and biocompatibility, non-toxicity, and bio-degradability, provide applications in cosmetics, pharmacy, and biomedicine (Ullah et al. 2016; Khine and Stenzel 2020; Kupnik et al. 2020; Mbituyimana et al. 2021). Its application in the delivery of drugs to a wound in the form of a mono/multi-layer film, membrane or capsule shells to inhibit external microbial infections (Czaja et al. 2006; Jipa et al. 2012; Lazarini et al. 2016; Ullah et al. 2017), as a stabilizer (Shen et al. 2023), a hydrogel carrier (Yi et al. 2023), and as a scaffold in the form of the aerogel (Zhou et al. 2021) has already been investigated well.

Microbial infections and the upsurge in antimicrobial resistance (AMR) is recognized as a top concern and a severe threat to the global economy and public health by various health agencies (e.g., the World Health Organization (WHO), Centers for Disease Control and Prevention (CDC), the European Medical Agency (EMA)) (Bhattacharjee et al. 2023). In this context, the primary strategy has become to endow biomaterials with antibacterial properties. Considering all the advantageous features of BNC, that, nevertheless, lacks the antimicrobial capacity (Li et al. 2017), different antimicrobial substances, including nanoparticles, biomolecules, and natural or synthetic polymers, have been introduced into it (Sulaeva et al. 2015) and studied for different applications.

Currently, a large number of scientific articles report on the applications of BNC-based composites, enriched with plant extracts, while the ever-increasing push for a circular economy has led to a widespread research in reusing fruit waste (Campos et al. 2020). In this frame, BNC has already been enriched with extracts from oregano, rosemary, parsley, lovage (Bodea et al. 2022), green tea and henna leaves (Pooja et al. 2019; Azarmi et al. 2022), roselle flower petals (Indrianingsih et al. 2020), propolis (Mocanu et al. 2019; Amorim et al. 2022), dragon tree (Kamal et al. 2022), sunflower, annatto, barbatimao, pink lapacho, marigold (Neves et al. 2022), and fireweed (Perużyńska et al. 2023). Fruit wastes have been studied in abundance for increased yield and higher quality of BNC production (Urbina et al. 2021), but only two studies with extracts of pomegranate peels (Ul-Islam et al. 2023) and mangosteen peels (Taokaew et al. 2014) have been detected in the literature for the purpose of achieving biological activity of modified BNC. No studies on BNC enrichment with avocado extract have been reported yet.

The health-promoting benefits of avocado (Persea americana) fruit are certainly a dominant drawing card for consumers all over the world. Global avocado production expanded to over 8.4 m millions metric tons in 2022 and at a compound annual growth rate (CAGR) of about 7% during the last decade (Rabobank 2023). Due to the large production and avocado consumption, and the fact that an avocado seed can represent up to 26% of the entire fruit weight, there is a huge amount of waste generated (Domínguez et al. 2016). The chemical composition and bioactivity of selected compounds from avocado seeds is already well known (Bangar et al. 2022; Kupnik et al. 2023; Sánchez-Quezada et al. 2023). Avocado seed extracts (ASE) are rich in polyphenols, acetogenins, fatty acids, triterpenoids, and other compounds that exhibit excellent antimicrobial, antioxidant, antihypertensive, hypolipidemic, and larvicidal activities (Leite et al. 2009; Nwaoguikpe et al. 2011; Jimenez et al. 2021). It was noticed that the content of biologically active compounds is influenced extremely by the choice of extraction method and the selected solvent, as we have shown in our latest study (Kupnik et al. 2023). In general, ASE have been shown to be excellent growth inhibitors of some Gram-negative (G-), Gram-positive (G +) bacteria, and selected fungi. The detected high contents of hesperidin, vanillin and 2,3-dihydroxybenzoic acid in ASE contributed to the excellent antimicrobial efficiency, while the high activity of the enzyme superoxide dismutase (SOD) strengthened its antioxidant potential additionally (Kupnik et al. 2023). All the obtained ASE exhibited promising results in terms of biological activity, but it is necessary to emphasize that the ASE obtained with modern, greener, and more sustainable supercritical fluid extraction (SFE), has resulted the highest content of proteins and phenols (Kupnik et al. 2023).

The aim of this study was thus to prepare BNCMs enriched with ASE obtained by different extraction techniques, and evaluate their antibacterial properties. First, a comprehensive study was carried out to optimize the cultivation conditions for production of BNCMs by K. hansenii bacteria to achieve the highest possible yield. Then, the influence of cultivation time on the mass, thickness, cellulose content and crystallinity of BNCMs was evaluated. In addition, the influence of different drying methods, i.e., air and freeze-drying, was carried out to evaluate their morphology and swelling. In the next step, BNCMs were enriched with ASE prepared by different extraction methods. The antibacterial efficiency, bacteriostatic and bactericidal activity of BNCMs enriched with ASE were evaluated on G- Escherichia coli and G + Staphylococcus aureus. To explain and interpret possible mechanisms of antibacterial activity against different bacteria, the ASE were characterized by surface charge and hydrodynamic size, while also determining potential ASE release after 24 h of exposure. The swelling and water retention capacity of BNCMs, as well as their chemical structure and morphology, were also studied, to evaluate their further applications in biomedicine, cosmetics, and pharmaceutics. To the best of our knowledge, this is the first study investigating the physico-chemical properties, morphology, and antibacterial activity of BNCMs enriched with ASE comprehensively, comparing the ASE prepared by different extraction methods (ultrasonic/US, Soxhlet/SE, and supercritical fluid/SFE).

Experimental

Chemicals and microbial strains used

Dimethyl sulfoxide (DMSO, ≥ 99.5%) was purchased from Kemika, Zagreb, Croatia. Citric acid (≥ 99.5%), D-( +)-glucose anhydrous (≥ 97.5%), di-sodium hydrogen phosphate (Na2HPO4, ≥ 99.0%), ethanol (EtOH, ≥ 99.5%), and peptone were purchased from Merck, Darmstadt, Germany. Carbon dioxide (CO2, purity 2.5) was purchased from Messer, Ruše, Slovenia. Acetic acid (glacial, ≥ 99.7%), agar, sodium hydroxide (NaOH, ≥ 95.0%), sucrose, and yeast extract were purchased from Sigma-Aldrich, St. Louis, USA, while the Mueller–Hinton broth was from Biolife, Milano, Italy. All chemicals were used without further purification. The selected microorganisms (K. hansenii (DSM 5602), E. coli (DSM 498), and S. aureus (DSM 346)) were purchased from DSMZ-German Collection of Microorganisms and Cell Cultures GmbH from Berlin, Germany.

Media, culture conditions and production of BNCMs

The well-known complex Hestrin and Schramm (HS) medium (Urbina et al. 2021) was selected as the growth medium, which was prepared as follows: 20 g/L glucose, 5 g/L peptone, 5 g/L yeast extract, 1.15 g/L citric acid and 2.7 g/L Na2HPO4.

HS medium was also selected as a production medium, and further, optimization of the media and culture conditions was carried out using a single-variable optimization method. In order to obtain the highest BNCM yield, glucose and sucrose were tested as carbon sources. Furthermore, the concentration (2, 4, or 6% w/v) of the added carbon source was optimized, followed by the concentration (0.5, 2, or 4% w/v) of peptone as a nitrogen source. In addition, different pH values (5.5, 6.0, and 6.5) of the medium were also tested, to achieve the highest possible yield of BNCMs` production. All the experiments were performed in triplicates.

One loopful of K. hansenii was inoculated in 50 mL of growth HS medium, and grown at 27 ± 1 °C on a rotary shaker (150 rpm) for 48 h. The resulting suspension was used as an inoculum (5% v/v, 1–5 × 106 CFU/mL) for the next experiments. The inoculated flasks were incubated under static conditions at 27 ± 1 °C for 21 days to obtain BNCMs. After cultivation and production, the prepared BNCMs were harvested from the production medium, washed thoroughly with distilled water, and immersed in 1 N NaOH solution at 80 ± 5 °C for 2 h to remove the bacterial cells and medium components embedded in the membranes. Lastly, the BNCMs were soaked and rinsed repeatedly with distilled water until a neutral pH was attained in the washed liquid. The BNCMs` production yield was measured as the wet weight of BNCM recalculated per volume of the culture medium [g/L].

Evaluation of the BNCMs` wet mass, thickness, and dry weight cellulose content

The influence was studied of different incubation times (7, 14, 21, 28, and 35 days) on the BNCMs` production. The BNCMs` mass was measured in a wet state, and its thickness was evaluated at different points (i.e., near the edges and at central positions) using a digital micrometer (3109 Series, INSIZE Co., Ltd, Spain); the values are presented as an average of 10 obtained points (± Standard Deviation/SD). Furthermore, the BNCMs were frozen at around -18 °C, and then lyophilized in a Mini Lyotrap laboratory freeze drying unit (LTE Scientific Ltd., UK) for about 48 h, and reweighed. The dry mass of cellulose content was calculated from the following equation:

$$Dry\;Mass\;Cellulose\;Content\;\left[\%\right]=\frac{m_{dry\;BNCM}}{m_{wet\;BNCM}}\times100$$
(1)

where mdry BNCM and mwet BNCM denote the weight of BNCM before (mwet BNCM) and after (mdry BNCM) the drying process, respectively. All the experiments were performed in triplicates, and the results are reported as mean value ± SD.

Effect of the drying method on the structure of the BNCMs

Two different drying methods were used to evaluate and compare the morphology of the obtained BNCMs. According to the air-drying method, the BNCMs were air dried until they reached a constant mass under controlled conditions (temperature of 20 ± 2 °C and 65 ± 5% humidity). In the case of the freeze-drying method, the samples were frozen at around -18 °C, and then lyophilized in a Mini Lyotrap laboratory freeze drying unit (LTE Scientific Ltd., UK) for about 48 h. About 96.39 ± 0.10% of water was removed by both drying techniques.

X-ray diffraction (XRD) analysis of the BNCMs

The crystallinity of the BNCMs was characterized by XRD collected on a Bruker D2 Phaser (MA, USA) diffractometer with Cu Kα radiation (λ = 1.54 Å) at 30 kV and 10 mA. Scattered radiation was detected in the range of 2θ from 5 to 70° at a scan rate of 0.3 s/step. The crystallinity index (CI) was calculated using the following equation, developed by Segal and coworkers (Segal et al. 1959):

$${C}_{I}\;\left[\%\right]=\frac{{I}_{110}-{I}_{AM}}{{I}_{110}}\times 100$$
(2)

where IAM and I110 denote, respectively, the intensity of diffraction pattern at 2θ around 18° corresponding to the amorphous regions in the cellulose, and the intensity of the diffraction peak at 2θ = 22.7° which is associated with the crystalline region of cellulose (French and Santiago Cintrón 2013; French 2014; Singhsa et al. 2018).

Using Bragg equation, the d-spacing of cellulose samples was calculated:

$$d\;\left[nm\right]=\frac{n\lambda }{2\;{\text{sin}}\;\theta }$$
(3)

where n denote order of reflection, λ is the wavelength of the X-ray source (0.154 nm) and θ is the Bragg angle corresponding to the plane. In order to determine the dominant crystal allomorph (Iα or Iβ) in different BNCMs samples, the calculation by Wada (Wada et al. 2001) was adopted. This calculation proposes the function applied to discriminate between the Iα and Iβ rich types:

$$Z=1693{d}_{1}-902{d}_{2}-549$$
(4)

where d1 and d2 denote, respectively, the d-spacing of (100) plane and the d-spacing of (010) plane. If Z > 0, the BNCM is classified as Iα rich, while Z < 0 indicate Iβ rich type.

Scanning electron microscopy (SEM) imaging of the BNCMs

The morphology of the dried BNCMs was examined by a high-resolution FEI Sirion 400NC microscope (FEI, Hillsboro, OR, USA). Approximately 1 cm2 pieces of dried BNCMs were cut, placed on a holder and sputtered with a gold to ensure conductivity and prevent charging effects.

Preparation of the ASE

The obtained avocado seeds (Hass variety) and further ASE were prepared by different extraction methods, as detailed in our publication (Kupnik et al. 2023). Briefly, ultrasound extraction (US) was performed for 3 h at 20 °C and 40 kHz, using H2O as a solvent. Soxhlet extraction (SE) with ethanol (EtOH) as a solvent was carried out for four recycles to be completed, while supercritical fluid extraction (SFE) was conducted for 2 h at 40 °C and 20 MPa, using CO2 (2 mL/min) as a solvent and EtOH (0.5–0.8 mL/min) as co-solvent. The solvents were evaporated with a rotary evaporator (Büchi Rotavapor R-114, Flawil, Switzerland) at 40 °C and reduced pressure. All the obtained ASE were stored at around -20 °C before their further usage.

Dynamic light scattering (DLS) analysis of ASE

The hydrodynamic size and zeta potential (ZP) of the ASE at a concentration of 0.05 wt% (dispersed in milliQ), were assessed at 25 ± 0.1 °C by DLS using a Zetasizer Nano ZS ZEN360 (Malvern Instruments Ltd., Malvern, UK). The DTS1070 disposable folded capillary cell was used, implementing the following parameters: a dispersion refractive index of 1.330 (milli-Q H2O) and a viscosity of 0.8872 cP (milli-Q H2O). A field of 150 V was implemented across the nominal electrode spacing of 16 mm. The samples (1 mL) were measured immediately after being dispersed at 10.000 rpm for 3 min using an Ultra-turrax T25 (IKA GmbH, Staufen, Germany). At least six individual measurements were performed, and the mean values ± SD were calculated.

Modification of the BNCMs with ASE

For modification of the BNCMs, the ASE obtained via different extraction methods were dispersed individually in sterilized deionized water (dH2O) or a 5% (v/v) DMSO solution, to prepare 1 or 10 wt% concentrated ASE solutions, respectively. The exhaust method (Sakthivel et al. 2016; Sharma and Bhardwaj 2019) was applied, which is based on the transfer of bioactive compounds from the bulk medium (i.e., the ASE solutions) to the surface of the substrate (i.e., the BNCMs) and further diffusion into the substrate. For that purpose, the wet BNCMs were simply immersed in different ASE solutions for 48 h. The magnetic stirrer (Tehtnica Rotamix 550 MMH, Železniki, Slovenia) of 150 rpm was used to facilitate diffusion of the ASE into the BNCMs. After adsorption, the modified BNCMs enriched with ASE were taken out and soaked over a filter paper to remove the excess extract.

Attenuated total reflectance–fourier transform infrared (ATR-FTIR) analysis of BNCMs enriched with ASE

The ATR-FTIR was employed for chemical characterization of differently ASE enriched BNCMs. Approximately 1 cm2 pieces of dried samples were used for recording the spectrum. The IR spectra were performed using a Perkin-Elmer spectrum one FTIR spectrometer (Waltham, MA, USA) with ATR crystal accessories at ambient conditions, and a 4 cm−1 resolution from collecting 16 scans per spectrum for each sample at a region of 4000–450 cm−1. The air spectrum subtraction and background were performed in parallel. The Spectrum 5.0.2 software was used for the data analysis. All the measurements were performed in duplicate.

Antibacterial activity of BNCMs enriched with ASE

To assess the antibacterial ability qualitatively against the G- E. coli and G + S. aureus of the prepared BNCMs, the agar diffusion assay was applied, following the protocol described in our previous studies (Kupnik et al. 2021a, b) with slight modifications. The BNCMs were exposed to UV light overnight before testing to avoid possible contamination. 100 μL of the prepared bacteria solution (1–5 × 106 CFU/mL) was spread evenly over Mueller–Hinton agar plates. Then, approx. 1 cm2 pieces of sterilized BNCMs were put directly on the inoculated agar plates. The incubation of the inoculated agar plates took place under optimal growth conditions for both bacteria (37 ± 1 °C for 24 h). The unmodified/pure BNCMs were taken as a control in each case. The results are reported as the diameter (in mm) of the inhibition zone formed around the samples. All the experiments were performed in triplicates, and the results are reported as the mean value ± SD.

The bacteriostatic and bactericidal activity of the modified BNCMs with ASE were assessed using the plate count method. The BNCMs were exposed to UV irradiation overnight to avoid possible contamination. The sterilized approx. 1 cm2 pieces of BNCMs were put into the test tubes containing 10 mL of bacterial suspension with the concentration of 1–5 × 106 CFU/mL and incubated under optimal growth conditions for both bacteria (37 ± 1 °C for 24 h). After incubation, 100 μL of the final and serially diluted bacterial suspensions were spread over the Mueller–Hinton agar plates and incubated for 24 h at 37 ± 1 °C. The number of colonies were counted under a colony counter (PBI F4, Milano, Italy). The antibacterial activity of BNCMs enriched with ASE was determined by calculating the log reduction factor (log R), the percentage of reduction (R), and bactericidal efficacy using the following equations:

$$Log\;Reduction\;Factor\;\lbrack\%\rbrack={log}_{10}(\frac{S_p}{S_m})$$
(5)
$$Bacterial\;Reduction\;Rate\;\lbrack\%\rbrack=\frac{S_p-S_m}{S_p}\times100$$
(6)
$$Bacterial\;Killing\;Efficacy\;\left[\%\right]=\frac{{CFU}_c-{CFU}_s}{{CFU}_c}\times100$$
(7)

where Sp and Sm denote, respectively, the number of viable bacterial colonies from pure BNCM (unmodified) and the number of viable bacterial colonies from modified (ASE reach) BNCM. The CFUc value signifies the number of colonies forming units per mL of the control sample (no BNCM sample added, only E. coli or S. aureus), and the CFUs value means the number of colony forming units per mL of BNCMs enriched with ASE. All the experiments were performed in triplicates, and results are reported as the mean value ± SD.

Swelling and water retention properties of BNCMs enriched with ASE

To determine the swelling ratio (SR), dried BNCMs with ASE (in size of approx. 2 cm2) were weighed (mdry BNCM) and immersed into 50 mL of dH2O. After specific time periods, the BNCMs were weighed (mwet BNCM) after removing the surface water by filter paper. To assess the water retention (WR) capacity, the dried BNCMs enriched with ASE were weighed (mdry BNCM), and then soaked into dH2O for 24 h. The BNCMs were taken out, whipped witha filter paper to remove the surface water, and placed on an open plate to evaluate the water retention. The weights of the BNCMs (mwet BNCM) were measured at certain time periods. The WR and SR were determined using the following equation (Lin et al. 2013b):

$$WR\;or\;SR\;\left[\%\right]=\frac{m_{wet\;BNCM}-m_{dry\;BNCM}}{m_{dry\;BNCM}}\times100$$
(8)

where mwet BNCM and mdry BNCM denote, respectively, the weight of the wet BNCM and the weight of the dried BNCM. All the experiments were performed in triplicates, and results are reported as the mean value.

Release of ASE from BNCMs

To determine the release of different ASE from the prepared BNCMs, wet and air-dried BNCMs enriched with ASE (in size of approx. 1 cm2, wet weight of approx. 0.2 g) were placed in a vial containing 5 mL of dH2O at 37 ± 1 °C with constant stirring (100 rpm). After 24 h, the amount (mg/mL) of released ASE was monitored using a UV–VIS spectrophotometer, and determined using pre-prepared calibration curves for the dilution series of the corresponding ASE.

Results and discussion

Optimization of cultivation conditions for BNCMs` production

Komagataeibacter (formerly Gluconacetobacter or Acetobacter) is well known in the production of BNC, but cultivation conditions vary with the strains. K. hansenii is characterized by a very high cellulose production rate (Bimmer et al. 2023) and high crystallinity (Güzel and Akpınar 2020), therefore, it was used and tested for optimal production in a complex HS medium and under static culture conditions (Sperotto et al. 2021) by varying nutritional (e.g., the source and concentration of carbon and nitrogen) and physical (e.g. pH) parameters. The studied parameters` influence and obtained production yield of BNCM are summarized in Table 1. When the glucose and sucrose were compared in terms of their efficiencies in the production yield of BNCM, the sucrose was more efficient, resulting in 15.48 ± 1.86 g/L, compared to 13.54 ± 1.39 g/L BNCMs with glucose. It was already established by other studies that, for some isolates, sucrose can serve as the best carbon source for high production of BNC, and that after reaching the optimal concentration, the yield of BNC production begins to decrease with increasing the concentration of the carbon source (Zeng et al. 2011; Rani and Appaiah 2011). However, according to the literature (Keshk and Sameshima 2005; Pourramezan et al. 2009; Mohite et al. 2013) 2 w/v% of sucrose added to the culture medium shell resulted in the most effective production. On the contrary, in our case, the addition of 2 w/v% of sucrose did not result in the formation of BNCMs, while the addition of 4 w/v% resulted in an almost 2.5 times higher production yield (45.49 ± 3.98 g/L) than by adding 6 w/v% of sucrose. As peptone and yeast extract are preferred nitrogen sources and the basic components of the complex medium (Sperotto et al. 2021), the effect of peptone concentration was evaluated on the BNCMs` production yield. It was observed that the lowest tested concentration (0.5 w/v%) did not result in the formation of BNCMs, while a 2 w/v% concentration of peptone gave a 2 times higher BNCM yield (51.86 ± 4.22 g/L) compared to 4 w/v%. The optimal pH range for BNCMs` production for Acetobacter sp. was found to be between 4.0–7.0 (Fontana et al. 1990), while it differed between the strains and variables on the culture conditions. It is considered that most strains produce BNCMs most effectively in the pH range of 5.0–6.0 (Lima et al. 2017; Cielecka et al. 2020, 2021), which is in agreement with our study. When applying static conditions, it is important to note that the pH value decreases during cultivation (usually to pH 3.5) because of the respiratory metabolism of bacteria (Lin et al. 2013a). The initial pH value of 6.0 was demonstrated to be optimal for K. hansenii, as the BNCM production yield was as much as 5.1 times higher (54.48 ± 4.08 g/L) than at pH 5.5 (10.66 ± 1.82 g/L), and approx. 1.7 times higher than at pH 6.5 (31.52 ± 3.56 g/L). As compared to other studies (Rani and Appaiah 2011; Uzyol and Saçan 2017), the initial pH value is very important for bacterial cell growth and high BNC production. The HS fermentation medium modified with 4 w/v% of sucrose and 2 w/v% of peptone at an initial pH of 6.0 was thus optimal medium for K. hansenii and provided the highest BNCM yield ( 54.48 ± 4.08 g/L). The purified BNCM obtained under optimal growth conditions is shown in Fig. 1b.

Table 1 Influence of different growing parameters on BNCM production by K. hansenii in HS medium at 27 ± 1 °C after 21 days
Full size table
Fig. 1
figure 1

a BNCMs after 35 days of cultivation in HS medium at 27 °C resulting in the formation of 3 layer membranes (1st, 2nd, 3rd) (b) Purified BNCM after 21 days of cultivation in HS medium using 4 w/v% of sucrose and 2 w/v% of peptone at pH 6.0 and 27  ± 1 °C

Full size image

After obtaining the optimal cultivation conditions for the production of BNCMs, different cultivation times have been evaluated in terms of the BNCMs` physical characteristics. Based on visual appearance, after 7 days of fermentation the BNCM had not yet formed, while, after 14 days, a visible thin layer had already started to form at the air–liquid interface. According to other studies (Mikkelsen et al. 2009; Ruka et al. 2012), sucrose requires an additional metabolic step to catalyze sucrose into fructose and glucose itself. Therefore, in the sucrose-based medium, initially, low levels of cellulose are produced, and, after a longer time, the levels of cellulose increase, due to an increased lag period for cellulose production. The 1st compact BNCM was visible after 21 days of cultivation. Interestingly, after 28 days of fermentation, two layered BNCMs were formed, while, after 35 days, 3 layered BNCMs were present in the Erlenmeyer flasks, as shown in Fig. 1a.

The multiple layers of BNCMs were thought to be caused by the formation of a new membrane due to the preceding membrane sinking. At the beginning of cultivation, the fermentation medium provides full nutritional value, yet the bacterial concentration is very low at that time and the BNCM starts to form slowly. Further, as the BNCM is formed at the air–liquid interface, its thickness and mass increase, so the first BNCM may sink with time. The formation of a BNCM is related directly to the surface at the air–liquid interface. When the concentration of bacterial cells is high and the nutrient value of the cultivation medium remains adequate and sufficient, the sub-fibrils of cellulose are extruded constantly from the linearly ordered pores at the surface of the bacteria. These then crystallize into microfibrils, and form intertwined and overlapping cellulose ribbons, forming disorganized, parallel planes. Hence, on the surface of the medium the cellulose can assemble into a new BNCM (Auta et al. 2017; Wang et al. 2019).

As illustrated in Fig. 2, the cultivation time certainly has an influence on all the studied characteristics of BNCMs. The first important factor is the BNCMs` overall yield (Fig. 2a). It was 1.9 and 1.2 times higher (117.8 ± 9.67 g/L) after 35 days of cultivation than after 21 (63.18 ± 7.46 g/L) and 28 (96.31 ± 8.13 g/L) days. The masses of the 1st BNCMs were comparable, ranging from 6.214 ± 0.918 g to 6.475 ± 0.431 g. The mass of the 2nd and 3rd BNCMs, which were formed after a longer cultivation time, decreased, which can be attributed to less nutritional fermentation medium. A similar trend can also be detected in the thickness (Fig. 2b) and cellulose content (Fig. 2c). The maximum thickness was reached by the 1st BNCMs (up to 1.876 ± 0.243 mm), while the thickness of the 3rd BNCM after 35 days was determined as 0.483 ± 0.152 mm. Similarly, the cellulose content ranged in-between 2.998 ± 0.241 and 3.203 ± 0.137 wt% in the 1st BNCMs, while the 2nd and 3rd formed BNCMs reached only half the cellulose content (up to 1.631 ± 0.121 wt%) compared to the 1st one.

Fig. 2
figure 2

Effect of K. hansenii cultivation time in HS medium at 27  ± 1 °C on the (a) mass, (b) thickness, (c) cellulose content, and (d) XRD patterns of the formed BNCMs with crystallinity index (CI), d-spacing and Z values

Full size image

The diffractograms of the so biosynthesized BNCMs, presented in Fig. 2d, confirmed the formation of a semi-crystalline structure of cellulose I, which was composed of amorphous and crystalline regions that occur in two distinct allomorphs, Iα and Iβ, respectively (Atalla and VanderHart 1984). The more intense and sharper peak means higher crystallinity of the sample (Khamrai et al. 2017). All the diffractograms depict three main diffraction signals at 2θ of around 14.6°, 16.8°, and 22.7°, thus corresponding to a cellulose Iα one-chain triclinic unit cell with Miller indices for the peaks of 100, 010, and 110, which coincided with the peaks 1 \(\overline{1 }\) 0, 110, and 200 of cellulose Iβ (French 2014). As reported (Aleshina et al. 2019; Heydorn et al. 2023), diffraction signals at around 20.5° are assigned to multiple overlapping Iα crystalline signals 11 \(\overline{2 }\), 01 \(\overline{2 }\), and 10 \(\overline{2 }\). In addition, it is necessary to discuss the XRD pattern of BNCM after 21 days of cultivation, where additional peaks that are not from cellulosic material were observed. As can be seen from the literature (George et al. 2008; Sheykhnazari et al. 2011; Mohammadkazemi et al. 2015b; Fan et al. 2016), additional peaks can be attributed to components from the culture media and may be present due to not thoroughly purified BNCM sample. Thus, the components of the culture medium can interfere with the aggregation of BNC fibrils, which in turn lowers the crystallinity. In particular, the presence of a peak at 2θ around 19° could slightly interfere with that at around 18° corresponding to the amorphous region in the cellulose and used to calculate crystallinity index (CI) in Segal equation. Due to the co-existing peak positions of Iα and Iβ allomorphs, it is difficult to differentiate between them by determining only the XRD peak positions. Nevertheless, from d-spacing, the Z values (Fig. 2, inserted table) were calculated according to a method described by Wada (Wada et al. 2001). Calculated values enable the dominant classification of cellulose Iα (when Z > 0, triclinic form) and Iβ (when Z < 0, monoclinic form). Presented results imply the domination of the triclinic structure (Iα type) in all BNCMs, regarding the Z values between 2.05 and 7.81. It was also observed that the higher the crystallinity, the higher the Z value was, indicating a higher domination of Iα-rich type cellulose in BNCMs. Previous results also confirmed that cellulose produced by bacteria and algae is predominantly enriched with Iα crystalline phase (Digel et al. 2023). However, a higher intensity of peak at around 14.6°, as that at around 16.8°, indicates the preferred orientation of crystalline segments in the membranes, due to the reflection mode on the X-ray diffractometer. According to the literature (Fang and Catchmark 2014), observation of the BNCMs in transmission mode with the incident X-ray beam from perpendicular to the membrane surface would show an inverted ratio of the mentioned peaks. Furthermore, the calculated crystallinity values fell within the range of 67.33–82.93%. The highest crystallinity was observed for both BNCMs harvested after 28 fermentation days, and its slight decreasing after further production. The results are in line with other studies (Sheykhnazari et al. 2011), indicating that the bacterial strain, media composition and fermentation time used had a great impact on its degree (Dayal et al. 2013). For comparison, cultivation of K. hansenii in HS medium resulted in a crystallinity of 58% (Cielecka et al. 2021), cultivation of G. xylinus on sucrose resulted in BNC with a crystallinity of 63.2% (Mohammadkazemi et al. 2015a), while the crystallinity of the BNC produced by different Komagataeibacter strains emerged in the range of 64–80% (Vigentini et al. 2019; Gupte et al. 2021) and in the range of 73–90%, depending on the choice of carbon source (Wang et al. 2018).

Effect of the drying method on the BNCMs` microstructure and ASE integration

Air-drying and freeze-drying methods were applied in order to obtain BNCMs with different microstructure, fibrils entanglement/arrangement, water absorption capacity, and accessibility of hydroxyl groups (Pinto et al. 2012; Hu et al. 2014; Zeng et al. 2014) for an effective integration (diffusion and interaction) of ASE. Figure 3 depicts air-dried BNCM (a) and freeze-dried BNCM (b) with corresponding SEM images of the surface (c, d) and cross-sections (e, f). It can be seen that the drying method had no important influence on the surface structure, while there is a difference in their cross-sections. Both cross-section images of the BNCMs show a layered fibrillated structure, although, when the BNCM was air-dried, the thermodynamically slow moisture removal resulted in shrinkage due to irreversible interactions (i.e., hydrogen bonds) (Fatima et al. 2022) and thus the formation of a compact, denser and more homogeneous structure with high adhesion between the fibrils (Fig. 3e). On the other hand, the freeze-dried BNCM resulted in a well-oriented, but more open and micro-porous network with a less dense nature (Fig. 3f), similar to a previous report (Illa et al. 2019), corresponding to the creation of ice-crystals from water molecules at the contact points where cryogelation occurs (Tang et al. 2010).

Fig. 3
figure 3

a Air-dried and (b) freeze-dried (optimally produced) BNCMs visualized by SEM imaging on their surfaces (c and d; voltage 15.0 kV, magnification 500x), and in the cross-sections (e and f; voltage 15.0 kV, magnification 10000x)

Full size image

During the development of modified BNCMs (Fig. 4a) through the ex-situ exhaust method, the solutions, suspended particles, and pure liquid diffused into the porous BNCM network and interacted with the available hydroxyl groups by inter- and intra-molecular hydrogen bonding interactions. The nature of additives and their particle size are main factors that can influence the degree of penetration/diffusion and attachment greatly. Figure 5 presents the SEM images of BNCMs enriched with 1 wt% of differently extracted ASE. The traces of dried ASE attached to the fibrils` surface are visible, and appear as layers in the BNCMs` network structure. The SOX-EtOH (Fig. 5c) and scCO2 + EtOH (Fig. 5d) extracted ASE are incorporated homogeneously throughout the densely packed air-dried BNCMs. On the contrary, US-H2O ASE (Fig. 5b) particles accumulated and integrated in the BNCMs, both on the surface and in the bulk (as seen from the cross-sections of the BNCMs), which can most likely be explained by the poorer dispersibility of this type of extract, in line with results from its hydrodynamic size (Table 2). The remarkably hydrodynamically larger in size (around 1.34 µm) and highly polydisperzed (PDI of 0.65) US-H2O ASE are visible from the SEM micrographs, while they cannot be observed in case of the much smaller (0.1–0.11 µm) and size-homogeneous (PDI of 0.17–0.14) SOX-EtOH and scCO2 + EtOH ASE enriched BNCMs. However, the morphological features, in aspect of layered fibrillated homogenous structure with open and micro-porous network, of BNCMs enriched with ASE are comparable with pure BNCM.

Fig. 4
figure 4

a Photo images of wet BNCMs enriched with different concentrations (1 and 10 wt%) of ASE, obtained by US-H2O, SOX-EtOH, and scCO2 + EtOH extractions, and corresponding (b) ATR-FTIR spectra of freeze-dried samples

Full size image
Fig. 5
figure 5

SEM images of freeze-dried (a) pure BNCM and BNCMs enriched with (b) US-H2O, (c) SOX-EtOH, and (d) scCO2 + EtOH extracts with the concentration of 1 wt%. Subscripts 1 indicate the surface of the BNCMs (voltage 15.0 kV; magnification 200x), while 2 (voltage 15.0 kV; magnification 200x) and 3 (voltage 15.0 kV; magnification 5000x) indicate the cross-sections of the BNCMs enriched with ASE

Full size image
Table 2 Zeta potential, hydrodynamic size and polydisperse index of 0.05 wt% ASE, as well as their MIC90 values after 24 h (Kupnik et al. 2023)
Full size table

The FTIR spectral profiles of the freeze-dried BNCMs were analyzed to provide an insight on the alteration of the BNCMs` chemical composition and interactions after the incorporation of ASE. As seen from Fig. 4b, all the spectra exhibited characteristic absorption bands of pure bacterial cellulose, i.e. peaks at 3270–3280 cm−1 (O–H stretching), 2920 cm−1 (C-H stretching of alkane and asymmetric CH2 stretching), 1640 cm−1 (H–O-H bending band of adsorbed H2O), 1500 cm−1 (CH2 groups), 1450 cm−1 ( H-C-H and O-C-H stretching of carbonyl groups), 1160 cm−1 (C–O–C stretching), 1035–1060 cm−1 (C-O stretching) (Ghozali et al. 2021). No significant difference between the FTIR spectra of differently dried BNCMs can be detected, which suggests that the drying process did not change the intermolecular composition of the BNCMs, which is in agreement with the results of other authors (Mohamad et al. 2022).

The presence of ASE did not result in the appearance of many specific spectral characteristics that could be attributed to the molecular structure of ASE. The peak at 3343 cm−1 is present in all the spectra, and could be assigned to the stretching vibration of the -OH bonds in the cellulose (Yassine et al. 2016). The mentioned absorption band is more intense for BNCMs enriched with ASE compared to pure BNCM, as it can be associated to the vibration of the -OH stretching in alcohols, carboxylic acids, and phenols (Rodríguez et al. 2016) from ASE. The higher intensity of bands around 2930 cm−1 for BNCMs enriched with SOX-EtOH and scCO2 + EtOH ASE may correspond to the CH3 or CH2 vibrations of carboxylic acids, such as salicylic acid (Neves et al. 2022). This is in accordance with our already published results (Kupnik et al. 2023), where salicylic acid was determined in both the mentioned extracts, but not in US-H2O ASE. The band at 1640–1600 cm−1 could be due to the stretching vibration of C = C, aromatic ring deformations, asymmetric bending vibration of N–H and stretching vibration of C = O and C = C from flavonoids and amino acids, and C = O stretching from caffeic acid (Oliveira et al. 2016). The bands between 1280 and 1274 cm−1 could be attributed to the C–C-O group of phenolic compounds and C-O vibration of -OH of flavonoids (Neves et al. 2022). According to the literature, absorption bands in the spectra of BNCMs enriched with ASE at 3343 cm−1, around 2930 cm−1, between 1315–1360 cm−1 and at approx. 1030 cm−1 may correspond to the -OH stretching band, H-C-H symmetric stretching of alkanes, N = O asymmetric stretching (nitrate), C–C-N symmetric/asymmetric bending and C-N stretching, respectively, which are characteristic peaks for phytochemicals from ASE (Bogireddy and Agarwal 2019).

Swelling and water retention properties of BNCMs enriched with ASE

The drug or extract release activity (kinetic) from the materials can be influenced by many factors, including drug/extract concentration, its physico-chemical characteristics (size and surface chemistry), the kinetic of release and solubility/dispersibility in the surrounding media (environment), as well as the swelling properties of the material (Kalkhoran et al. 2018). Therefore, BNCMs enriched with ASE were evaluated for swelling (SW) and water retention (WR) properties, while also determining their release after 24 h of exposure. The results of SR and WR are shown in Fig. 6. The gravimetric method determined that, in the case of BNCMs` immersion in 1 or 10 wt% solution of ASE, an average of 1.46 ± 0.69 mg and 14.05 ± 4.21 mg of ASE per 1 g of wet BNCM were incorporated, respectively, regardless of the chosen extraction method. Furthermore, the release of the extracts, determined by a spectrophotometric method, was found to be of 85.76 ± 6.89 (from air-dried BNCMs) and 95.62 ± 2.71% (from wet BNCMs), respectively.

Fig. 6
figure 6

a Swelling ratio (SR) and (b) Water retention (WR) rate of air-dried pure BNCM and BNCMs enriched with ASE (extracted by US-H2O, SOX-EtOH, and scCO2 + EtOH processes) concentration of 1 wt%

Full size image

The modifications in the physical appearance of BNCMs are likely to bring changes to their physical characteristics. The SR and WR are the most important properties which are connected directly in the biomedical wound dressing applications (Islam et al. 2012). The proper moist environment can promote the release of antibacterial substances, therefore, it provides protection against contamination, while, on the other hand, ensures a proper wound moisture by adsorbing and retaining it, which accelerates the healing procedure and enables a painless and easy dressing change without damage to the healed or newly formed skin (Shezad et al. 2010). The results from the presented study indicate that all BNCMs were able to absorb a high amount of water. The SW of pure BNCM and BNCMs enriched with ASE were around 600 and 490–530%, respectively, comparable, as reported earlier, for bacterial cellulose hydrogels (i.e., 400–600% SR) (Treesuppharat et al. 2017). It can be inferred that the integrated extracts leads to a lowering of the SR, which can be attributed to the hydrogen bonding interaction between the phytomolecules of ASE and the OH groups on the cellulose surface, blocking the availability of cellulosic hydroxyl groups for water adsorption (Ul-Islam et al. 2023), as evidenced from the FTIR analysis. On the other hand, this also depends on the arrangement of its microstructure, as closely arranged fibrils bind water molecules more efficiently because of the stronger hydrogen bonding interactions (Shezad et al. 2010; Islam et al. 2012), thus resulting in quantitatively lower ASE adsorption, as well as their release. In the case of pure BNCM and BNCMs enriched with ASE it took about 48 h for the near complete evaporation of the adsorbed water. When comparing the results of BNCMs with differently extracted ASE, the sample containing ASE prepared by scCO2 gave kinetically the lowest SW as well as WR, which can be related to the hydrodynamically smaller ASE particles interacting more extensively with the fibrillated structure, which slows down the adsorption of water and its excretion. As reported earlier for BNCMs enriched with natural antibacterial agents (Lin et al. 2013b), it can be concluded that BNCMs enriched with ASE maintain a suitable moisture environment for wound healing applications for low- to mid-range exudates.

Antibacterial activity of BNCMs enriched with ASE

Until now, no studies have been detected in the reviewed literature with the enrichment of BNCMs with ASE for the purposes of antibacterial efficacy for further applications. Therefore, the presented study is the first reporting the antibacterial effectiveness of the prepared BNCMs, the influence of differently prepared extracts on their antibacterial activity against G- E. coli and G + S. aureus (as already confirmed in our previous research (Kupnik et al. 2023) given MIC90 between 210–280 μg/mL for E.coli and 140–2780 μg/mL for S. aures; Table 2) and, further, on BNCMs enriched with ASE, providing additional information and insight for potential applications.

The qualitative agar diffusion method was used to evaluate the antibacterial efficiency of BNCMs enriched with ASE, and the results are shown in Table 3 and Fig. 7.

Table 3 Qualitatively determined antimicrobial activity of BNCMs enriched with different ASE concentrations against G- E. coli and G + S. aureus, swelling ratio (SW), and release of ASE from wet and air-dried BNCMs after 24 h
Full size table
Fig. 7
figure 7

Inhibition zone against G- E. coli by BNCMs enriched with (a) 1 wt% and (b) 10 wt% scCO2 + EtOH ASE, determined by an agar diffusion assay (the concentration of E. coli was 1–5 × 106 CFU/mL) after 24 h of microbial growth at 37  ± 1 °C. Pure BNCM was applied as negative control

Full size image

The growth of both G- E. coli and G + S. aureus was inhibited effectively by BNCMs immersed in highly concentrated 10 wt% ASE solutions, due to the consequent higher adsorption of extracts (around 14.05 mg/g vs. 1.46 mg/g), and, therefore, also 10 times higher release (around 30.1 mg/mL vs. 0.30 mg/mL). BNCMs with 1 wt% US-H2O ASE did not inhibit the growth of any of the tested bacteria, while the addition of 1 wt% SOX-EtOH ASE inhibited both bacteria by 17 ± 3 mm. BNCMs with the addition of 10 wt% ASE solutions proved to be good growth inhibitors of both the tested bacteria, with inhibition zones formed in the range between 13 ± 2 and 26 ± 1 mm. BNCMs enriched with scCO2 + EtOH ASE proved to be the most promising antibacterial agents, due to the best results compared to BNCMs modified with US-H2O and SOX-EtOH ASE. Besides, the difference in the inhibition of bacteria by BNCMs with differently concentrated solutions (1 and 10 wt%) of scCO2 + EtOH ASE was small, resulting in 23–26 ± 1 mm and 16–18 ± 3 mm inhibition zones for E. coli and S. aureus, respectively.

The antibacterial activity of BNCMs enriched with ASE extracts was quantified further using the plate count method. The reduction rate (R) and corresponding log reduction factor (log R) were determined, along with the bactericidal/killing effect for both G- E. coli (Fig. 8a) and G + S. aureus (Fig. 8b) bacteria, as associated with inflammatory-related skin diseases (Madawi et al. 2023). Regarding the results presented in Fig. 8, all the BNCMs enriched with ASE exhibited a decrease in bacterial population after 24 h of exposure.

Fig. 8
figure 8

Bacterial reduction rate, bactericidal effect, and log reduction of BNCMs enriched with US-H2O, SOX-EtOH, and scCO2 + EtOH ASE of different concentrations on (a) G- E. coli and (b) G + S. aureus determined by the plate count method (the concentration of both microorganisms was 1–5 × 106 CFU/mL) after 24 h of microbial growth

Full size image

The highest, 100% reduction against G- E. coli, was attained using 10 wt% SOX-EtOH ASE (log R 2.33) and both scCO2 + EtOH ASE (log R of 2.48 (1 wt%) and 2.56 (10 wt%), respectively) BNCMs, followed by 85% reduction using 10 wt% US-H2O ASE (log R 0.94). A strong bactericidal effect was also detected for the mentioned BNCMs. The BNCMs with 1 wt% US-H2O ASE resulted in a lower bacteriostatic potential (i.e., the bacterial strain survived, but did not reproduce) with minor bactericidal effects (72.04% production rate and log reduction of 0.55). E. coli was the most resistant to the addition of BNCMs modified with 1 wt% SOX-EtOH ASE, which resulted in around a 22.22% reduction rate, log reduction of 0.1 and 18.42% killing efficiency. The release of 1wt% SOX-EtOH ASE resulted in 0.26 mg/mL, while the minimum inhibitory concentrations (MIC90) value for SOX-EtOH ASE against E. coli was determined at 0.28 mg/mL, which is the main reason for the lower bactericidal and bacteriostatic activity.

In the case of the G + S. aureus, BNCMs enriched with 10 wt% of SOX-EtOH and scCO2 + EtOH ASE, a strong bactericidal effect was detected, which was confirmed with a 100% reduction of the bacterial cells and log reduction of 1.95. This is associated with the hydrodynamically smallest (0.1 μm) and well homogenized extract particles, as well as the relatively high (484.91%) SW, as small particles can easily penetrate through the outermost layer of the bacterial cell envelope and thus contribute to a better antibacterial effect. BNCMs modified with 1 wt% scCO2 + EtOH ASE exhibited somewhat lower bacteriostatic (88.89%) and bactericidal (88.89%) effects, corresponding to a log reduction of 0.95. Comparatively, BNCMs enriched with 1 wt% US-H2O, 1 wt% SOX-EtOH, or 10 wt% US-H2O ASE resulted to about half less effective bacterial reduction (44.39–55.62%) and killing efficiency (40.51–52.53%), with a log reduction of 0.25–0.45. Therefore, they can be considered as materials with minor bactericidal and bacteriostatic effects against S. aureus. The results correspond with the release studies, and relate with the MIC90 values determined in our previous study (Kupnik et al. 2023). The MIC90 for US-H2O ASE was determined at 2.78 mg/mL, while the release of US-H2O extracts after 24 h was 0.27 and 2.71 mg/mL for 1 and 10 wt%, respectively.

The extraction method has a significant influence on the content of biologically active compounds present in ASE, and, consequently, their bioactivity (Alexandre et al. 2018). Compared to UE and SE, SFE enables faster, near-ambient temperature, more sustainable and green extraction, which results in the recovery of pure biologically active extracts (Brglez Mojzer et al. 2016). Phytochemical screening of the ASE exhibited the existence of different secondary metabolites, like alkaloids, flavonoids, phenolics, saponins, and tannins, and is well-documented in the literature (Setyawan et al. 2021; da Silva et al. 2022; Bangar et al. 2022; Nyakang’i et al. 2023). Differing bioactive compounds show distinctive modes of actions in the case of antimicrobial activity. For example, alkaloids interact with microbial DNA in order to promote antimicrobial action, while flavonoids disrupt bacterial cell walls, as they have the ability to form complexes with extracellular/soluble proteins of microbes. On the other hand, phenolics and tannins can bind to bacterial cell wall receptors, inhibit the enzymes which are required for cellular metabolism, and cause membrane disruption (Singh 2017; Wink 2022). In this manner, it is clear that the various phytochemicals in ASE contribute synergistically to imparting antibacterial activity.

The observed differences in antibacterial activity between different ASEs may thus be related to the extraction method and the resulting bioactive substances, the amount and availability of bioactive ASE on the surface of BNCMs, the release properties of the BNCMs, as well as the hydrodynamic size of the ASEs and their surface chemistry (Kokol et al. 2023a), which are crucial for their interaction with the bacteria outer cell layer and cell damage (Malanovic and Lohner 2016; Pajerski et al. 2020). The G- bacteria are composed of the inner cytoplasmic membrane, a thin peptidoglycan cell wall, and an outer membrane containing lipopolysaccharides. Contrarily, G + bacteria lack the protective outer membrane, as they consist only of the inner cytoplasmic membrane and a thick peptidoglycan cell wall (Rajagopal and Walker 2017; Rojas et al. 2018). In general, G + bacteria are known as more susceptible to the addition of antibacterial agents. However, the outer membrane of G- bacteria is permeable to small molecules due to porin channels, and impermeable to large molecules (Kokol et al. 2023b). Differently structured and large ASEs thus play an important role in the interaction with the outermost layer of the bacterial cell envelope, which may affect cell interaction and sensitivity to various ASEs.

Zeta potential (ZP) is a measure of the magnitude of the electrostatic attraction (or charge), or the repulsion between the functional groups or particle surfaces (Liu et al. 2023). The ZP of avocado seed ingredients in the range between -49.7 and -23.3 mV was evaluated (Sánchez-Quezada et al. 2023), which is in accordance with the ZP of the ASE extracted in our study, determined as average values in between -34.3 to -22.2 mV (see Table 2). In addition, bacterial surface charge, an electrochemical property of the cell surface, has often been described with ZP, and the net surface charge for most bacteria is negative. The average ZP for S. aureus and E. coli were found to be around -35.6 and -44.2 mv, respectively (Halder et al. 2015). Bacterial exposure to the antibacterial agents may be governed by electrostatic interactions, which can result in a ZP change. Two main reasons for altering ZP are possible, bacterial injury, or binding of the antibacterial agent to the bacterial surface (Ferreyra Maillard et al. 2021). These assumptions are supported by many researches including positively and negatively charged substances (Lopez-Romero et al. 2015; Ong et al. 2019; Ferreyra Maillard et al. 2019; 2021). As ASE are negatively charged, the resulting ZP of bacteria are probably shifted to more negative values. On that point, accumulation on the bacterial surface is the first step to cause bacterial injury. Regardless of the direction of ZP change, the addition of ASE affects cellular physiology, thus leading to the bacterial growth inhibition and/or cell death (Ferreyra Maillard et al. 2021). The more noticeable changes in ZP correlate with higher antibacterial activity (Ong et al. 2019), which is in line with the results observed for ASE. SOX-EtOH ASE and scCO2 + EtOH ASE resulted in more negative ZP, leading to more intense lowering of ZP for E. coli and S. aureus, exhibiting better antibacterial efficiency in comparison with US-H2O ones.

As already demonstrated well, the antibacterial properties of materials increase when prepared in the form of nanoparticles, as a large specific surface and the availability of a higher number of surface atoms can enhance their bioactivity (Shameli et al. 2011). Herein, a distinct difference was observed in the analyzed hydrodynamic sizes. Again, SOX-EtOH ASE and scCO2 + EtOH ASE exhibited more than 11.7 times lower hydrodynamic size (approx. 100 nm) compared to US-H2O ASE (i.e., 1340.7 nm). Consequently, ASE particles accumulated on the bacterial surface, the cell wall was disrupted and their leaching began into the highly permeable plasma (Khurana et al. 2014). On the other hand, the average particle size of the US-H2O extract was higher, therefore, the interaction between the bacterial cell and the US-H2O ASE particles was limited severely, causing lower reduction of bacterial growth, and/or death. Additionally, PDI visualizes particle size distribution, and as the PDI value is lower, the uniformity degree is higher (Sitti et al. 2018). As the PDI of SOX-EtOH ASE and scCO2 + EtOH ASE were significantly low (i.e., 0.17 and 0.14, respectively), indicating monodisperse samples, the production was expected of more homogeneous biocomposites. However, the PDI of US-H2O ASE was higher (i.e., 0.65), indicating its poor uniformity and polydispersity. An inhomogeneous penetration and attachment of US-H2O ASE may be the main reason for the interpretation of its lower antibacterial efficacy against G- E. coli and G + S. aureus compared to the other two ASEs, which is also in line with the results from the ZP values, hydrodynamic size and PDI of ASEs.

In addition, it is important to note that PDI values below 0.2 are acceptable in practise for polymer-based nanoparticle materials, while, in drug delivery, PDI values of 0.3 or below are considered as acceptable (Danaei et al. 2018). Hence, SOX-EtOH and scCO2 + EtOH ASE are applicable for incorporation into polymer-based nanomaterials, and suitable for further applications in drug delivery with antibacterial effect.

Conclusions

In the presented study, an HS fermentation medium modified with 4 w/v% of sucrose and 2 w/v% of peptone at an initial pH value of 6.0 was proved to be an optimal medium for BNCMs` production with a high degree of crystallinity (up to 83%) using K. hansenii (DSM 5602). For the first time, the composites of BNCMs enriched with antimicrobially active ASE, extracted by different methods, were studied for applications in the health sector. Although all BNCMs enriched with ASE exerted growth reduction or complete inhibition of G- E. coli and G + S. aureus, BNCMs enriched with SOX-EtOH ASE and scCO2 + EtOH ASE showed excellent bacteriostatic (up to 100% reduction with a log reduction up to 2.56) and bactericidal (up to 100%) activity. Such activity can be attributed to the hydrodynamically smaller (0.1–0.11 μm), homogeneous and more surface negative (-33 mV) ASE particles in the latter, favorable for interactions with the bacteria, and, as such, also suitable for applications in the delivery of antibacterial compounds. On the other hand, high swelling (up to 530%) and water retention properties demonstrated the ASE-enriched BNCMs suitability for versatile applications in biomedicine, cosmetics, and pharmacy related with skin regenerative medicine and wound healing. The biotechnological process of obtaining BNCMs should also be taken into consideration to be more efficient, what could happen due to alternative sources of carbon for BNCMs biosynthesis. However, the utilization of avocado seeds` waste thus enables and encompasses a broader strategy towards a circular economy that can be converted into sustainable value-added products.