Antimicrobial Activity of Aqueous Plant Extracts as Potential Natural Additives †

: The 21st century has brought along many changes in how consumers look at food and perceive their diets. There is an increasing awareness towards what goes into manufacturing these foods, with an important concern being drawn towards food additives. While it is known that additives are needed to preserve or change attributes of food, it is also widely understood that consumers prefer natural additives to artificial ones, and thus, the industry is as E221 against B. cereus and E. cloacae . Overall, this work proves that plant extracts obtained though “green” and low-cost technologies can be alternatives to artificial food additives, due to showing the same, or, in some cases, better antimicrobial activity. Furthermore, a mixture of these extracts can result in synergistic effects and improve the antimicrobial activities. The next step of this work, which is ongoing, will focus on the determination of the efficacy of these extracts and their incorporation in muffins. Contributions: Conceptualization, M.C. S.H.; methodology, M.C.P., J.M.U., M.I.; software, B.M.; validation, M.C.; investigation, M.C.P. and J.M.U.; data curation, B.M.; writing— original draft preparation, M.C and B.M; supervision, M.S., L.B., and I.C.F.R.F.; project administra-tion, L.B.


Introduction
The shelves on supermarkets are stocked with fresh fruit, vegetables, meats, and fish from many provenances, and prepared meals, processed goods, and other food with high freshness and quality. All of these products have to endure the journey from harvest, to processing, to the store, and finally, to the consumer's house. Once bought, they must endure on the consumer's shelf for some more time. For foods to withstand the passage of time, the industry relies on food additives, which are now one of the most widespread means of food preservation. Still, the use of additives has not been consensual over the recent decades, with food scares and scandals involving these molecules. With the turn of the century, and mainly in the past 10 years, consumers have become more aware of the food they eat, and the consequences of this food on their health. The consequence of this has been a preference for unprocessed food, and, when processed, a clear preference for the use of natural food additives [1]. Overexposure to synthetic food additives has been related to health issues, and thus, the industry now turns to nature to use the molecules it has to offer, rather than using them as an inspiration for various purposes. In this scenario, provided that the preservation of food is one of the most important tasks, the use of plant extracts has been increasing to carry it out. Specifically, rosemary extract is one of the most relevant examples, being approved by the European Food Safety Authority to be used in various foods. Other extracts have been used successfully although they have not yet been approved [2,3]. Using plants as preserving agents could be a game changer in the food industry by allowing by-products of crop production to be used as sources for these additives, contributing to a sustainable production and reducing agricultural waste.

Plant Material
The plants used in this work-oregano (Origanum vulgare L.), rosemary (Rosmarinus officinalis L.), salvia (Salvia officinalis L.), lemon balm (Melissa officinalis L.), and basil (Ocimum basilicum L.)-were all acquired from the enterprise Cantinho das Aromáticas in Portugal, having arrived at the lab in a dried form. They were subsequently ground to a 20mesh powder.

Ultrasound-Assisted Extraction (UAE)
Ultrasound-assisted extraction of bioactive compounds from the different species analyzed was performed using an ultrasonic system (QSonica, model CL-334, Newtown, Connecticut, EE. UU). The solid-to-liquid ratio was adjusted in every repetition, maintaining a constant volume between every run, and right after each extraction, samples were centrifuged (5000 rpm for 20 min at 10 °C), filtered through filter paper (Whatman n° 4), and dried at 40 °C using a vacuum rotary evaporator (Büchi R-210, Flawil, Switzerland). The aqueous fraction was then lyophilized (47 °C, 0.045 bar; FreeZone 4.5, Labconco, Kansas City, MO, USA) to obtain the powdered extract. The extracts were stored in a dry place prior to further analysis. This design was built and randomized using Design Expert 12.0.1. (Stat-Ease, Inc. Minneapolis, MN, USA. The fixed variables in this experiment were defined as X1: time (7.5 and 12.5 min), X2: solvent (0 and 80% EtOH), and X3: UAE Power (275 and 450 W). The dependent variable was expressed according to the Rosmarinic acid content (Y1). In the fractional factorial design, factors were aliased as Intercept = Intercept + ABC; A = A + BC; B = B + AC; and C = C + AB, in order to determine pertinent factors and ranges of experimentation to be used in future optimization designs. Secondary selection and evaluation of factors and levels were carried out to determine the appropriate experimental domains following the guidelines of the first trial. Independent variables included X1: time (10-300 s) and X4: temperature (20-75 °C) at 6 and 4 different levels, respectively. For each of the 2 factors, a unifactorial multilevel or general factorial was performed and analyzed.

Phenolic Fingerprinting and Quantification, and Statistical Analysis
The chromatographic data was acquired using a Dionex Ultimate 3000 UPLC (Thermo Scientific, San Jose, CA, USA) coupled to a diode array detector (280, 330, and 370 nm) and an electrospray ionization mass detector (Linear Ion Trap LTQ XL, Thermo Finnigan, San Jose, CA, USA), working in negative mode. The chromatographic separation was performed using a Waters Spherisorb S3 ODS-2 C18 (3 μm, 4.6 mm × 150 mm, Waters, Milford, MA, USA) column at 35 °C. The compounds were identified considering the retention time, UV-Vis, and mass spectra in comparison with available standards and with literature data. For the fractional factorial design, Pareto and analysis of variance (ANOVA) was carried out to determine the principal effects and their magnitudes, while, for the general factorial designs, comparison of different samples was carried out through the application of one-way analysis of variance (ANOVA) using an F-test with p-value = 0.05, along with the least significant difference (LSD) Fischer's test. Both analyses were performed using Statgraphics Centurion XVI software (StatPoint Technologies, Inc. Warrenton, VA, USA).

Antimicrobial Activity
The extracts were dissolved in 5% DMSO and screened against the microorganisms listed above. These microorganisms were obtained from the mycological laboratory of the Institute for Biological Research "Sinisa Stanković" in Serbia. Minimum inhibitory concentrations (MIC) were determined by a serial microdilution method, and the rapid piodonitrotetrazolium violet (INT) colorimetric assay, following the protocol described by Soković et al. (2010) and Soković and van Griensven (2006) [4]. For the negative control, 5% DMSO was used, while positive controls were streptomycin and ampicillin for the antibacterial activity, and ketoconazole and bifonazole for the antifungal activity, with the results expressed as MICs, MBCs, and MFCs in mg/mL.

Extraction Optimization Studies
Factor selection had been completed on previous optimization design, performed within the research group and available literature in order to reduce time and maximize yields. The choice for using UAE was highlighted; therefore, the most relevant factors to include were the time of extraction, potency, and percentages of ethanol and water used as an extraction solvent. A decision was made to work with the lowest UAE power in order to save resources, as well as with a high ethanol concentration in the extraction solvent. The efficiency of extraction with less polar solvents on the yielding of rosemary phenolic compounds has been also described by other authors (Oliveira et al., 2016; Sik et al., 2020), which reinforce this decision [5,6]. Regarding Figure 1a, it is clear that lemon balm is the plant which yields not only more phenolic compounds, but also the highest content of rosmarinic acid. In total, four peaks were positively identified by comparison with com-  In Figure 1b, the effect of temperature is shown for all plants, plotting it against the amount of extract obtained from the UAE extractions. Interestingly, the behavior of each plant was distinct. Rosemary showed a decrease of phenolic content with the increase of heat, while lemon balm increased the quantity of polyphenols with the increase of heat. The yields of sage, basil, and oregano were not affected by the variation in temperature. Overall, considering the cost of heating water at an industrial level, it does not seem be significantly beneficial to use heat to extract compounds from these plants with ultrasounds. Figure 2 shows the variation of extractive capacity of the two plants with the highest amounts of total phenolic compounds for the different extraction times, namely 10, 15, and 30 s, and 1, 2, and 5 min. Considering rosemary (Figure 2a), it is quite clear that extending the extraction time did not significantly increase the yield in phenolic compounds. For instance, 10 s yielded almost the same amount of total phenolic compounds as 5 min. A very similar effect was seen for lemon balm (Figure 2b), in which there was a higher amount of total phenolic compounds for 2 and 5 min. Considering the cost of operation and the extraction time, the yield of 10 s seems to be quite interesting, especially due to the very small variation in total phenolics when compared to the other times.

Antimicrobial Analysis
In Figure 2, the antimicrobial analysis is represented, with the positive controls on the right of each group of results in dark grey and black, the synthetic preservatives in light green and cyan, and the plant extracts in darker colors. In Figure 3a, the MIC of the foodborne bacteria are shown. None of the plant extracts of the synthetic compounds showed values as low as the positive controls, although rosemary and salvia did show some promising effects, namely being better than or having an equal effect as sodium benzoate for all bacteria except Bacillus cereus. A very similar effect was sought for MBC (Figure 3b), where these two plant extracts showed similar effects to sodium benzoate in killing the bacteria. Still, overall, potassium metabisulfite proved to be very effective against all but Bacillus cereus. Concerning the fungi, Figure 3c shows the MICs. Overall, the plant extracts seem to have a higher effect against fungi than bacteria. All plant extracts showed better antimicrobial activity than both synthetic preservatives; in fact, the values of the plant extracts were very close to the positive controls. Specifically, for A. fumigatus, P. funiculosum, and P. verrucosum, salvia and rosemary showed values under or equal to the positive controls, revealing an excellent capacity to be used in food to hinder fungi growth. The same tendency is seen for MFC (Figure 3d), where the extracts have lower values for all fungi when compared to sodium benzoate, but also better than potassium metabisulfite for, once again, A. fumigatus, P. funiculosum, and P. verrucosum.

Conclusions
In conclusion, UAE seems to be a promising method to obtain polyphenolic extracts from Lamiaceae plants, namely rosemary, lemon balm, sage, oregano, and basil. Rosemary and basil showed the highest amounts of polyphenols of the studied plants. UAE is a very easy and rapid method to obtain these plant extracts, at any temperature and in brief bursts of time. The obtained extracts could be used as natural antimicrobials for foods, due to their interesting antimicrobial activity, comparable in some instances to commercial artificial counterparts.