Fabrication of zein-modified starch nanoparticle complexes via microfluidic chip and encapsulation of nisin.

A microfluidic chip is a micro-reactor that precisely manipulates and controls fluids. Zein is a group of prolamines extracted from corn that can form self-assembled nanoparticles in water or a low concentration of ethanol in a microfluidic chip. However, the zein nanoparticles have stability issues, especially in a neutral pH environment due to the proximity of the isoelectric point. This study was designed 1) to evaluate the effect of octenyl succinic anhydride (OSA) modified starch on the stability of zein nanoparticles formed using a microfluidic chip and 2) to apply the zein-OSA starch for encapsulation of nisin and evaluate its anti-microbial activity in a model food matrix. A T-junction configuration of the microfluidic chip was used to fabricate the zein nanoparticles using 1% or 2% zein solution and 0-10% (w/w) of OSA starch solution. The stability of the nanoparticles in various ionic strength environments was assessed. Encapsulation efficiency and anti-microbial activity of nisin in the zein nanoparticles against Listeria monocytogenes in a fresh cheese were measured. As the concentration of OSA starch increased from 0 to 10%, effective diameter increased from 117.8 ± 14.5 to 198.7 ± 13.9 nm without affecting polydispersity indexes and zeta-potential changed toward that of the modified starch indicating the zein surface coverage by the OSA starch. The zein-OSA starch nanoparticle complexes were more stable at various sodium chloride concentrations than the zein nanoparticles without OSA starch. The encapsulation efficiency of nisin was positively correlated with the OSA starch concentration. The anti-microbial activity of nisin in the fresh cheese also increased until 3-days of storage as the concentration of the OSA starch increased, which presented both a potential and challenge toward applications.

Introduction

Microfluidic chips are novel micro-reactors that are used to control the ultra-small volume of fluids within the channel dimensions of tens of micrometers. Researchers have been able to expand the application of microfluidic chips in many fields, including drug discovery, biomedical imaging, and biomolecule synthesis (Teh et al., 2008). The diversified application of microfluidics is attributed to their superior ability to precisely control a small segment of liquid, which thereby could be used as reaction confinements (Mark et al., 2010). In terms of food-related applications, the microfluidic chip can be used as the tool for emulsion (Comunian et al., 2018; Priest et al., 2011; Ravanfar et al., 2018; Zhao-Miao et al., 2018), micro-detections (Guo et al., 2015; Kant et al., 2018; Kim et al., 2015), and micro-reactors (Marze et al., 2014; Nguyen, Marquis, Anton and Marze, 2019a, 2019b). There are many advantages for the microfluidic chip: small reagent volumes, selectivity, green credentials, rapid reactions, and small footprints (Elvira et al., 2013). However, very few food-grade materials have been successfully adapted in a microfluidic chip and more studies are needed to explore the applications that require food-grade materials.

Zein is the prolamine protein in corn. More than 50% of its amino acids are hydrophobic, including leucine, proline, and alanine. According to the amino acid sequence and solubility in water, zein can be classified into four distinct types: α-zein, β-zein, γ-zein and δ-zein (Zhang et al., 2011) and among those types, α- is constituted around 70% in total (Lawton, 2002). Zein is widely used in food and pharmaceutical industries because it is generally recognized as safe (GRAS), biodegradable, and biocompatible. Zein can form self-assembled particles by anti-solvent precipitation in the solution at a low ethanol concentration (Lawton, 2002; Patel and Velikov, 2014). However, zein nanoparticles are not stable in the wide range of pH, especially having issues at the isoelectric point at around 6.8. Polysaccharides can be attached to the surface of zein nanoparticles to increase stability. The nanoparticles with protein-polysaccharide complex have gained attention in the food, personal care, and pharmaceutical industries because they are easy to prepare, biodegradable, and biocompatible (Chang et al., 2017). Protein-polysaccharide complex has been reported to encapsulate bioactive compounds and increase the encapsulation efficiency and stability of the core materials (Dai et al., 2018). Octenyl-succinic-anhydride (OSA) modified starch has been used in the food industry for more than 40 years (Hui, Qi-he, Ming-liang, Qiong, & Guo-qing, 2009) and is a candidate to form zein nanoparticle complexes. The OSA modified starch is an amphiphilic molecule obtained from the esterification reaction between starch hydroxyl groups and octenyl succinic anhydride and can be used as an emulsion stabilizer and encapsulating agent (Sweedman et al., 2013). OSA-modified starch is negatively charged in the range of pH from 3 to 8 because the branch chain of OSA-modified starch has negative charges in the wide range of pH.

Nisin is a 34-amino acid antimicrobial peptide produced by Lactococcus lactis ssp. Lactis (Ibarra-Sánchez et al., 2020), and nisin has been considered as generally recognized as safe (GRAS) in the food system by the US Food and Drug Administration (FDA) (21CFR184.1538) (Özel et al., 2018). In the food industry, nisin can be used as the antibacterial agent against Gram-positive pathogens, such as Listeria monocytogenes and Staphylococcus aureus (Shin et al., 2016). However, nisin has some limitations, such as instability at high temperatures, reduced activity in low acidity and high fat-containing foods, undesirable interactions with other food components, and low water solubility (de Arauz et al., 2009; Ibarra-Sánchez et al., 2020). These limitations can be overcome by encapsulating nisin into nanocarriers, including nanoemulsion, nanoparticles, nanoliposomes, and nano-fibers (Bahrami et al., 2019). In this study, the zein-OSA modified starch nanoparticles were fabricated using the microfluidic chip, and nisin was encapsulated into the zein-OSA modified starch complex. The objectives of this study were 1) to evaluate the effect of OSA modified starch on the properties of zein nanoparticles formed using a microfluidic device and 2) to evaluate the encapsulation efficiency and antilisterial activity of nisin encapsulated in zein-OSA starch nanopartilces using Queso Fresco as a model food matrix.

Materials and methods

Materials

Zein (purified) was purchased from Sigma–Aldrich (St. Louis, MO). Ethanol (200 proof) was purchased from Decon Labs (King of Prussia, PA). And OSA-modified starch was obtained from Ingredion company (Ingredion, Westchester, IL). Nisin (Nisaplin, Danisco, New Century, KS) was extracted with 70% ethanol at pH = 3.0 overnight to prepare the stock solution.

Sample preparation

The dispersed phase was 1% or 2% zein dissolved in 70% (w/v) ethanol by stirring at 200 rpm overnight in a closed container and then centrifuged at 4000 rpm for 10 min to remove undissolved materials. The continuous phase was a OSA-modified starch solution at various concentrations: 0%, 1%, 2.5%, 5%, 7.5%, and 10% (w/w). The starch was dissolved into the deionized (DI) water by stirring at 200 rpm overnight. pH was adjusted to 3 for both phases. The sample used in this study are listed in the Table 1. The 1% or 2% zein solution was loaded into a 10 ml syringe (Hamilton Robotics, Reno, NV) and the flow rate was set at 10 ml/h which was controlled by a syringe pump (Harvard Apparatus, Holliston, MA). The different concentrations of the modified starch solutions were loaded into the 10 ml syringe (Hamilton Robotics, Reno, NV) and the flow rate was set at 30 ml/h which were controlled by a syringe pump (Harvard Apparatus, Holliston, MA). The T junction, 100 μm microfluidic chip was used to collect samples and each sample was created with triplicates. The microfluidic chip was purchased from Dolomite (Dolomite Ltd., Royston, UK). The nanoparticle dispersions were collected in 20 mL glass vials (Thermo Fisher Scientific, Inc., Waltham, MA) for analyses.

Table 1

The sample codes of zein-modified starch nanoparticles formation via microfluidic chip.

The sample code	OSA-Modified Starch Concentration% (w/w)	Zein Concentration% (w/w)
MS0-Zein1	0	1
MS1-Zein1	1
MS2.5-Zein1	2.5
MS5-Zein1	5
MS7.5-Zein1	7.5
MS10-Zein1	10
MS0-Zein2	0	2
MS1-Zein2	1
MS2.5-Zein2	2.5
MS5-Zein2	5
MS7.5-Zein2	7.5
MS10-Zein2	10

Particle size and polydispersity index determination

All the samples were diluted at 1:40 using the DI water and the effective diameter was measured by the Dynamic Light Scattering (DLS) particle size analyzer (Brookhaven Instruments, Holtsville, NY). The effective diameters were reported as the surface average diameter (D3,2) and the equation was expressed as follow:where ni is the number of particles with diameter di. The three replicated readings were collected for each sample. The polydispersity index (PDI) was also obtained by the DLS and three replicated readings were averaged.

Zeta-potential measurements

The zeta-potential of the freshly prepared samples were measured by the zeta-potential analyzer (Brookhaven Instruments, Holtsville, NY) at the various pH values from 3 to 8. The suspensions were diluted by 80 times to minimize the electrokinetic potential not a contribution from inter-colloidal interaction.

Interactions related to the formation of the particle

The contributions of hydrogen bonding, hydrophobic interaction, and disulfide bonding for particle formation were evaluated by the urea, sodium dodecyl sulfate (SDS), and dithiothreitol (DTT), respectively according to the previously published method (Akbari and Wu, 2016). The zein-nanoparticle solutions with different concentrations of the OSA-modified starch were mixed with the agents at various concentrations to assess the driving force for the complex formation. For the urea, the concentration was varied from 1M to 7 M, from 0.25% to 1% (w/w) for SDS, and from 10 mM to 60 mM for DTT. After 6 h, the turbidity of the solutions was measured by the UV–visible spectrophotometer (Genesys5, Thermo Scientific Co., Waltham, MA) at 600 nm wavelengths. T0 was the initial turbidity of the solutions and T was the turbidity of the solutions after adding the dissociation agents. The ratio of the T/T0 was used to indicate the dissociation of the particles.

The stability of the particles in the salt solution

The stability of the zein-nanoparticles was measured by adding the particles to various concentrations of sodium chloride (NaCl) solutions and the sedimentation was observed (H. Chen and Zhong, 2015). The same volume of the NaCl solution was combined with the freshly prepared zein-nanoparticles solution with different concentrations of the modified starch (0%, 1%, 2.5%, 5%, 7.5%, and 10%). The final NaCl concentrations in the mixture were 0 mM, 25 mM, 50 mM, and 125 mM.

Circular dichroism (CD) spectroscopy

The CD spectra were collected using JASCO J-715 spectropolarimeter (Jasco Inc., Easton, MD) and the method in a previous publication was used with some modifications (Feng and Lee, 2017). The 100 times diluted samples were placed in an 1 cm path length quartz cell, and the wavelength ranged from 200 to 250 nm at 25 °C with the scanning speed of 50 nm/min. The ratio of the α-helix, β-sheet, and the random coil was calculated by the BeStSel software (Micsonai et al., 2015).

The encapsulation efficiency of nisin

Encapsulation of nisin

Nisin was dissolved to reach 0.5 mg/ml in 1% or 2% zein in 70% (w/v) ethanol solution as the dispersed phase by stirring at 200 rpm overnight and then centrifuged at 4000 rpm for 10 min to remove unsolvable materials. The continuous phase was prepared by dissolving OSA-modified starch into the DI water to reach various concentrations (0%, 5%, and 10% (w/w)) by stirring the solution at 200 rpm overnight to ensure complete hydration of the modified starch. pH was adjusted to 3 for both phases. The 1% or 2% zein solution with nisin was loaded into the 10 ml syringe (Hamilton Robotics, Reno, NV) and the flow rate was 10 ml/h which was controlled by a syringe pump (Harvard Apparatus, Holliston, MA). The modified starch solutions were loaded into the 10 ml syringe (Hamilton Robotics, Reno, NV) and the flow rate was 30 ml/h which were controlled by a syringe pump (Harvard Apparatus, Holliston, MA). The T junction 100 μm microfluidic chip was used to collect samples with three batch replicates. The nanoparticle dispersions were collected in 20 mL glass vials (Thermo Fisher Scientific, Inc., Waltham, MA) for analyses.

Nisin encapsulation efficiency

To quantify nisin in the nanocapsules suspensions, 300 μL of nanocapsule suspension was mixed with 700 μL of 100% ethanol to disrupt nanocapsules and release nisin for total nisin, or mixed with 700 μL of deionized water to maintain nanocapsules integrity to measure free nisin in the supernatant. Samples were passed sequentially through 0.45, 0.22, and 0.1 μm filters (PES Syringe Filter, Sartorius) to separate polymer fractions or microcapsules from the solution before injecting into an HPLC for nisin quantification. The encapsulation efficiency was calculated as follows:

Nisin quantification with HPLC

Nisin was quantified utilizing HPLC as previously described with minor modifications (Feng et al., 2019). Briefly, a Waters 2695 Alliance HPLC system (Waters, Milford, MA) equipped with a Hewlett-Packard series 1050 photodiode array detector (Hewlett-Packard, Palo Alto, CA) was used. The analytical column was a reversed-phase Hypersil GOLD C18 (175 A, 250 × 4.6 mm, 5 μm) (Thermo Scientific, San Diego, CA). Solvent A was 0.1% (v/v) trifluoroacetic acid (TFA) in ultra-pure water, and solvent B was 90% (v/v) HPLC-grade acetonitrile (Thermo Fisher Scientific) containing 0.1% TFA (v/v) in ultra-pure water. The column temperature was maintained at 40 °C, sample injection volume was 20 μL, and samples were measured at UV absorption of 214 nm. A linear gradient from 31% B to 43% B over 16 min was run at a flow rate of 1.0 mL/min. The standard curve was plotted using 10, 50, 100, 200, and 300 μg/mL nisin (Nisaplin®, 2.5% nisin w/w), and the amount of nisin was calculated from the area of the peak at 214 nm. HPLC analysis was performed in duplicate for each sample.

The antimicrobial test

Microorganisms and culture conditions

Listeria monocytogenes strains (Agricultural Research Service Culture Collection Northern Regional Research Laboratory strains B-33104, B33419, B-33420, B-33424, and B-33513) were recovered from frozen glycerol stock (−80 °C) and grown in brain heart infusion (BHI; Difco, Becton Dickinson and Co., Sparks, MD) broth, at 37 °C for 24 h with 250-rpm agitation to obtain cell concentrations of ∼9 Log CFU/mL. L. monocytogenes cocktails were prepared by combining equal volumes of stationary phase cultures of the five different foodborne outbreak-associated strains. The serial dilutions of the L. monocytogenes cocktail were further prepared in PBS (KCl 200 mg/L; KH2PO4, 200 mg/L; NaCl, 8 g/L; Na2HPO4, 1.15 g/L, pH 7.2) to obtain the desired cell concentrations. Listeria enumeration was carried out on PALCAM Listeria-Selective agar (EMDMillipore) supplemented with 20 mg/mL ceftazidime (Tokyo Chemical Industry Co. Ltd.) and incubated for 48 h at 37 °C.

Antilisterial properties in Queso Fresco

The antimicrobial activity of free and encapsulated nisin was evaluated by their addition to Queso Fresco (QF) as previously described with minor modifications (Van Tassell et al., 2015). Aliquots of microcapsule suspensions containing 37.5 μg nisin were vacuum concentrated in a Vacufuge Concentrator 5301 (Eppendorf North America, Westbury, NY, USA) for 45 min at room temperature. Batches of QF were prepared with Nisaplin® and concentrated microcapsules suspension to an equivalent amount of 37.5 μg of nisin/mL of milk added into the milk before renneting. This corresponded to approximately 250 μg of nisin/g of cheese, the maximum permissible concentration in the United States. Sample cheeses were inoculated with a five-strain L. monocytogenes cocktail directly into the curd before pressing, for a final concentration of approximately 3.5 Log CFU/g. All cheeses were stored at 4 °C for up to 14 days until sampled for Listeria enumeration. Cheeses were individually homogenized and serially diluted in PBS and spread plated on PALCAM Listeria-Selective agar supplemented with 20 μg/mL ceftazidime to enumerate L. monocytogenes. Plates were incubated at 37 °C for 48 h.

Statistical analysis

All the experimental results were conducted in triplicate. The results were expressed as mean ± standard deviation (n = 3). The significant differences of the results among the different treatments were conducted by ANOVA (P< 0.5), and the averages were compared using Tukey's test.

Results and discussion

Particle size and polydispersity index (PDI)

The particle size and polydispersity index (PDI) of the 1% and 2% zein nanoparticles with the various concentration of the OSA-modified starch (0%, 1%, 2.5%, 5%, 7.5%, and 10%) are shown in Fig. 1. As the concentration of OSA-modified starch increased, the particle size of the zein-nanoparticles increased. The particle size of the zein-nanoparticle without the modified starch was around 117.8 nm, which is in good agreement with the previously reported results (H. Chen and Zhong, 2015; Gonçalves da Rosa et al., 2020). As the concentration of OSA-modified starch increased from 1% to 10%, the effective diameter increased from 117.8 ± 14.5 to 198.7 ± 13.9 nm with the 1% of zein solution and from 150.53 ± 3.48 to 317.6 ± 3.4 nm with the 2% of zein solutions. The effective diameter increased as the concentration of OSA-modified starch increased because the addition of modified starch attached to the surface of zein nanoparticles to enlarge the particle size. Although there were statistical differences in the PDI in the zein-nanoparticle with various concentrations of the modified starch, there was no clear trend. The PDI describes the degree of “non-uniformity” of the distribution, which indicated the addition of modified starch did not change the uniformity of the zein nanoparticles fabricated in the microfluidic chip.

Fig. 1

The effective diameter (a) and PDI (b) of the 1%, and 2% zein nanoparticles with various concentrations of modified starch.

Zeta-potential

The zeta-potential of the 1% and 2% zein nanoparticles with the various concentrations of OSA-modified starch are shown in Fig. 2. For the zein nanoparticles without OSA-modified starch, as the pH increased from 3 to 8, the zeta-potential changed from the positive to the negative, and the isoelectric point (pI) of the zein was around 6.4, which is in good agreement with the published results (H. Chen and Zhong, 2015; Hu and McClements, 2015; Li et al., 2019). The zeta-potential of OSA-modified starch is negative in the range of pH from 3 to 8 because the branch chain of OSA-modified starch has negative charges in the wide range of pH (Lin et al., 2018). The concentration of modified starch can be categorized in three levels: low concentration (1% and 2.5%), medium concentration (5% and 7.5%), and high concentration (10%). For the 1% zein solution with the low concentration of modified starch, as the pH increased, the zeta-potential changed from positive to negative, and the pI was around 4 which is lower than the zein only nanoparticles. With the medium and high concentration of modified starch, the zeta-potential was negative in the entire pH range from 3 to 8. For the 2% zein solution with the low and medium concentration of modified starch, as the pH increased, the zeta-potential changed from positive to negative, and the pI was around 3.3. With the high concentration of modified starch, the zeta-potential was negative in the entire pH range from 3 to 8. At the low modified starch concentration, the modified starch only partially covered the surface of the zein nanoparticles. As the starch concentration increased, the modified starch increased the coverage of the zein surface of the zein nanoparticles. Eventually, the zein nanoparticle surface was fully covered by the starch. The results also suggest that the electrostatic interaction was one of the main driving forces to form the zein OSA-modified starch nanoparticles (S. Chen et al., 2020).

Fig. 2

The zeta-potential of zein nanoparticle with different concentrations of modified starch.(a) 1% zein nanoparticles; (b) 2% zein nanoparticles.

Different interactions such as hydrogen bonding, electrostatic interaction, disulfide bonding, and hydrophobic interactions could affect the zein OSA-modified starch nanoparticles formation. To understand the driving force between zein and OSA-modified starch, different dissociation agents were used. The SDS disrupted the hydrophobic interactions among the zein-modified starch, and the urea and DTT tested the hydrogen bonding and disulfide bonding, respectively. The ratio of the T/T0 was used to indicate the dissociation particles. As SDS and urea were added, the zein OSA-modified starch dissociated and the turbidity of the suspension decreased due to the disrupted interactions. As shown in Fig. 3 with the concentration of the SDS and urea increased, the T/T0 decreased. However, as the concentration of the DTT increased, the T/T0 did not change. The results indicated that the hydrophobic interactions and hydrogen bonding were the main driving force to form the zein-modified starch nanoparticles via microfluidic chip (Akbari and Wu, 2016).

Fig. 3

The effect of the Urea, SDS, and DDT on change of the turbidity of the zein-modified starch nanoparticle. (a), (c), (e) 1% zein nanoparticles; (b), (d), (f) 2% zein nanoparticles.

Stability of the particles in salt solution

Fig. 4 shows the visual observation of the zein-modified starch nanoparticles in a series of NaCl solutions. As the concentration of the NaCl increased, the stability of the nanoparticles decreased and aggregated more. For the same concentration of NaCl, as the concentration of the modified starch increased, the stability of the zein nanoparticles increased and less aggregates were observed. The modified starch could increase the stability of the zein nanoparticles within 0–125 mM of NaCl solution. The increased stability in elevated NaCl concentration provides expands application of the nanoparticles as many food products such as fresh cheeses and salad dressings.

Fig. 4

The pictures of the zein-modified starch nanoparticles in a series of NaCl concentrations.

The circular dichroism spectroscopy of the 1% and 2% zein nanoparticles with different concentrations of the modified starch (0%, 5%, and 10%) are shown in Table 2 and Fig. 5. In the CD spectra, individual zein nanoparticles and zein OSA-modified starch exhibited two troughs at 210 nm and 225 nm. For the untreated zein nanoparticles, the fractions of α-helix, β-sheet, and random coil were around: 14%, 43%, and 43% which is in good agreement with the literature (Feng and Lee, 2017). Compared with the zein nanoparticles, there was no significant difference in the ratio of α-helix, β-sheet, and random coil among the various concentration of the zein OSA-modified starch complex. The results showed that the modified starch did not change the secondary structure of zein in the nanoparticles.

Table 2

The CD Spectroscopy results of the zein-nanoparticle with 0, 5, and 10% modified starch.

The sample	α-helix (%)	β-sheet (%)	Random coil (%)
MS0-Zein1	14.8 ± 1.35a	43 ± 2.69a	42.2 ± 2.35a
MS5-Zein1	12.8 ± 1.07a	44.2 ± 1.52a	43 ± 1.65a
MS10-Zein1	11 ± 0.96a	43.5 ± 0.78a	45.5 ± 2.17a
MS0-Zein2	14 ± 0.86 a	41.2 ± 1.45a	44.8 ± 1.07a
MS5-Zein2	13.6 ± 0.93 a	47.9 ± 2.96b	38.5 ± 2.78b
MS10-Zein2	11.9 ± 0.65 a	42.6 ± 1.02a	45.5 ± 1.98a

Fig. 5

The CD Spectra of the zein-nanoparticle with 0, 5, and 10% the modified starch.

The encapsulation efficiency of the nisin in the 1% and 2% zein nanoparticles with the various concentration of modified starch (0%, 5%, and 10%) was showed in Fig. 6. As the concentration of the modified starch increased, the encapsulation efficiency of nisin increased. For the 1% zein nanoparticles, the encapsulation efficiency enhanced from 10% to 40%, and for the 2% zein nanoparticles, the encapsulation efficiency increased from 23% to 50%. The results are probably due to the increased wall thickness on the zein surface which is indirectly presented with particle size (Fig. 1), zeta potential (Fig. 2), and CD (Table 2) data. When the OSA-modified starch attached to the surface of zein nanoparticles, the particle size increased and the thicker layer of OSA-modified starch provides more space as well as a barrier for nisin to be encapsulated thus encapsulation efficiency increased.

Fig. 6

The encapsulation efficiency of the nisin encapsulated zein nanoparticle with 0, 5, and 10% modified starch.

Nisin is known to inhibit Gram-positive bacteria, such as L. monocytogenes, a foodborne pathogen that has been associated with listeriosis outbreaks linked to fresh cheeses (Ibarra-Sánchez et al., 2018). Nisin antimicrobial activity relies on its ability to form pores on Gram-positive bacteria cell wall: nisin inhibits peptidoglycan synthesis by forming a complex with Lipid II, and nisin insertion permeabilizes the cytoplasmic membrane, which allows the efflux of cytoplasmic content and subsequent death of bacteria (Gharsallaoui et al., 2016). The antimicrobial activity of nisin encapsulated in zein-modified starch against L. monocytogenes was evaluated in Queso Fresco (Fig. 7). L. monocytogenes counts in control cheeses reached approximately 5.5 Log CFU/g by day 14 of storage at 4 °C, an increase of about 2 Log CFU from the initial inoculum. Similar to our previous studies (Feng et al., 2019; Ibarra-Sánchez et al., 2021; Ibarra-Sánchez et al., 2018; Martínez-Ramos et al., 2020), a discreet reduction in the pathogen counts was observed in cheeses added with free nisin compared to control cheeses. The limited antilisterial efficacy of nisin in Queso Fresco is the result of nisin's instability at near-neutral pH, and interactions with milk fat, caseins, and cations (Ibarra-Sánchez et al., 2020). Nisin was encapsulated to overcome the limited anilisterial efficacy of free nisin in fresh cheeses. All cheeses added with nisin encapsulated in zein-modified starch exhibited overall lower pathogen counts during storage. The antilisterial effect of the encapsulated nisin was evident compared to the free nisin until day3. From day 5 until day 14, a difference of approximately 0.5 and 0.8 Log CFU/g relative to control cheeses, was observed in cheeses with nisin in 1% and 2% zein nanoparticles, respectively. However, different concentrations of modified starch (0%, 5%, and 10%) had a negligible impact on enhancing the efficacy of encapsulated nisin to reduce L. monocytogenes counts in Queso Fresco samples. Based these observations, it was hypothesized that the location of encapsulated nisin will significantly impact the efficacy of the nisin on Listeria in fresh cheese. Although we observed that increasing the concentration of modified starch increased the encapsulation efficiency of nisin in zein nanoparticles (Fig. 6), it is possible that nisin encapsulated on or near the surface (layer of OSA modified starch) was released quickly (e.g. during cheesemaking), and nisin encapsulated into the deeper section of zein nanoparticles were released later to be active against L. monocytogenes in Queso Fresco. This hypothesis need further studies to be tested.

Fig. 7

Antimicrobial activity nisin encapsulated zein nanoparticle with 0, 5, and 10% modified starch. (a) 1% zein nanoparticles; (b) 2% zein nanoparticles.

Conclusions

Zein, a food grade biopolymer, was able form self-assembled nanoparticles using a microfluidic device, which may expand the application of microfluidic devices to the food industry. Specifically, zein OSA-modified starch nanoparticles were prepared by a microfluidic chip with a fixed flow rate at pH 3. As the concentration of the modified starch increased, the particle size and turbidity increased, so the particle size can be controlled by adjusting the formulation. The impact of zein or modified starch concentrations on the PDI was not significant. The main driving forces to form the complex nanoparticles were hydrophobic interaction, electrostatic interaction, and hydrogen bonding. The modified starch did not change the secondary structure of the zein nanoparticles during formation. The incorporation of OSA modified starch enhanced the stability of zein nanoparticles which is a critical properties for applications in foods. The concentration of the modified starch showed a significant effect on the encapsulation efficiency of the nisin but did not show a significant impact on the antimicrobial activity against L. monocytogenes tested in Queso Fresco samples. However, the antilisterial effect of the encapsulated nisin was still limited to storage of up to 3 days. So, achieving prolonged antiliterial effect should be the focus of future studies by characterizing the detailed structure of nanocapsules and enhanced release control.

CRediT authorship contribution statement

Xuanbo Liu: The role and contribution of the authors are. Luis Alberto Ibarra-Sánchez: design and perform experiments, collect and analyze data, and prepared the manuscript, Formal analysis, Data curation. Michael J. Miller: design and perform experiments on fresh cheese, collect and analyze microbial data, and contribute to manuscript preparation, Formal analysis, Data curation. Youngsoo Lee: advise researchers and manage fresh cheese experiments and microbial data analysis as well as review the draft manuscript, Writing – original draft, Formal analysis, Data curation.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.