Microbiological and Chemical Quality of Portuguese Lettuce—Results of a Case Study

In addition to environmental pollution issues, social concerns about the sustainability, safety, and quality of conventionally grown fruits and vegetables have been increasing. In order to evaluate if there were any microbiological differences between samples of organic and conventional lettuce, a wide range of parameters were tested, including pathogens and indicator organisms: the enumeration of Escherichia coli; the detection of Salmonella spp.; the detection/enumeration of Listeria monocytogenes; the enumeration of lactic acid bacteria, Pseudomonas spp. yeasts and molds, and Enterobacteriaceae. This study also evaluated the chemical safety of the lettuce samples, quantifying the nitrate concentration and 20 pesticides (14 organochlorine and 6 organophosphorus pesticides). Significant differences (p < 0.05) between the conventional and organic samples were only detected for the counts of total microorganisms at 30 °C. Pathogens were absent in all the samples. The analytical method, using the quick, easy, cheap, effective, rugged, and safe (QuEChERS) approach for pesticide extraction, was suitable for detecting the targeted analytes; the limit of quantification (LOQ) was between 0.6 and 1.8 µg/kg (lower than the Maximum Residue Levels (MRLs) established by EU legislation). In three organic lettuce samples, one organochlorine pesticide (α-HCH) was observed below the MRLs. For the samples analyzed and for the parameters investigated, except for the total mesophilic counts, the organic and conventional lettuces were not different.


Introduction
"Fruits and vegetables are important components of a healthy diet", and their daily consumption reduces the risk of severe illnesses, such as cardiovascular diseases and certain types of cancer [1,2]. Despite these scientifically validated health benefits, contaminated produce (by virus, bacteria, or parasites) has been linked to major cases and outbreaks of foodborne diseases in recent years, and has led to some of the biggest food recalls [3,4]. Among the top 10 multistate outbreaks ranked by the number of illnesses in 2019 in the USA, six were linked to the consumption of raw produce [5]. In addition to biological threats, contamination by chemicals from anthropogenic and natural sources is another global food safety issue [6]. These chemicals belong to several groups, with pesticides receiving the most attention. The consumption of pesticide-contaminated food via one's daily diet is a major

Sampling
Ten lettuces from certified organic production and 10 from conventional production were purchased in supermarkets or farmers' markets located in Porto between 11 March and 6 May 2019. All the samples were from Romaine lettuces grown in Portugal. Whenever possible, at each location and date of purchase the same number of biological and conventional samples were purchased. Samples were transported to the laboratory at room temperature and analyzed in less than 24 h after being collected.
For chemical safety analysis, the samples were separately frozen at −18 • C. Each sample was cold-homogenized using a high-performance blender (Vorwerk, Portugal). The samples were then stored at −20 • C. The homogenized samples were stored in the freezer before being thawed just prior to extraction.

Chemicals, Materials and Standards
The chromatographic-grade acetonitrile (MeCN) was from Carlo Erba (Val de Reuil, France), and the n-hexane and sodium acetate were from Merck (Darmstadt, Germany). The lead acetate and lead oxide were from Sigma (Steinheim, Germany).

Nitrate Concentration
The determination of nitrate concentration in lettuce samples was performed according to the potentiometric method described by Lima et al. [34], with some modifications. Briefly, two grams of fresh lettuce leaves were dried (100 • C) and powdered; then, approximately 50 mg of lettuce powder was mixed with the 10 mL of nitrate extraction solution (ISA-Pb) composed of 1 × 10 −2 M lead acetate, 5 × 10 −3 M lead oxide, and 1 × 10 −2 M sodium acetate; shaken for 5 min; and filtered. Determinations were performed over this extract solution in a nitrate ion-selective electrode. This was the method performed because its efficiency had already been proved, as well as cost-effectiveness and quickness.

Sample Preparation
The OCP and OPP were extracted from the lettuce samples based on the previously reported QuEChERS approach with dispersive solid-phase extraction (d-SPE) cleanup [35]. A schematic illustration of the method is shown in Scheme 1 and included the five steps: (1) 5.0 g of a homogenized sample or spiked sample was weighed into a 50 mL polypropylene tube, 10 mL of MeCN was added, and the tube was directly vortexed; (2) original QuEChERS powders (4 g of anhydrous MgSO 4 and 1 g Foods 2020, 9, 1274 4 of 10 of NaCl) were added, the tubes were vortexed for 1 min, and centrifuged for 5 min at 4500 rpm at room temperature; (3) 1 mL of the upper layer was transferred to the d-SPE cleanup tube (150 mg of MgSO 4 , 50 mg of PSA, and 50 mg of carbon), and the tubes were vortexed for 5 min at 4500 rpm at room temperature; (4) 800 µL of the cleaned extract was transferred to a labeled vial, the extract was dried by exposing it to a low flow of nitrogen gas, and it was re-dissolved in n-hexane; finally, (5) the sample extract was vortexed and 2 µL of the extract was used for the gas chromatogrphy (GC) analysis.
The OCP and OPP were extracted from the lettuce samples based on the previously reported QuEChERS approach with dispersive solid-phase extraction (d-SPE) cleanup [35]. A schematic illustration of the method is shown in Scheme 1 and included the five steps: (1) 5.0 g of a homogenized sample or spiked sample was weighed into a 50 mL polypropylene tube, 10 mL of MeCN was added, and the tube was directly vortexed; (2) original QuEChERS powders (4 g of anhydrous MgSO4 and 1 g of NaCl) were added, the tubes were vortexed for 1 min, and centrifuged for 5 min at 4500 rpm at room temperature; (3) 1 mL of the upper layer was transferred to the d-SPE cleanup tube (150 mg of MgSO4, 50 mg of PSA, and 50 mg of carbon), and the tubes were vortexed for 5 min at 4500 rpm at room temperature; (4) 800 µ L of the cleaned extract was transferred to a labeled vial, the extract was dried by exposing it to a low flow of nitrogen gas, and it was re-dissolved in n-hexane; finally, (5) the sample extract was vortexed and 2 μL of the extract was used for the gas chromatogrphy (GC) analysis. Scheme 1. Schematic illustration of the extraction procedure.

Method Validation
For analytical validation, the following parameters were evaluated: the linearity in solvent and matrix, the recoveries, the limit of detection (LOD), the limit of quantification (LOQ), and the method accuracy and precision. A matrix blank extract was used for the preparation of a set of experimentsnamely, 7 matrix-matched calibration standards (3-150 µ g/kg) and three sets of matrix spiking at low, medium, and high levels (25; 50; 100 µ g/kg) for all pesticides. Precision was calculated in terms of the intraday repeatability RSD% (n = 3) to three spiking levels for all the analytes.

GC Analysis
The gas chromatographic analysis was performed using three different types of equipment, (Table 1) Table 1.

Method Validation
For analytical validation, the following parameters were evaluated: the linearity in solvent and matrix, the recoveries, the limit of detection (LOD), the limit of quantification (LOQ), and the method accuracy and precision. A matrix blank extract was used for the preparation of a set of experiments-namely, 7 matrix-matched calibration standards (3-150 µg/kg) and three sets of matrix spiking at low, medium, and high levels (25; 50; 100 µg/kg) for all pesticides. Precision was calculated in terms of the intraday repeatability RSD% (n = 3) to three spiking levels for all the analytes.

GC Analysis
The gas chromatographic analysis was performed using three different types of equipment, (Table 1) Table 1.

Statistical Analyses
To establish a comparison between organic and conventional products regarding the microbiology parameters and nitrate concentration, the results were analyzed using a one-way analysis of variance (ANOVA) with Tukey's post-hoc test (SPSS, Version 23.0, Inc., Chicago, IL, USA).

Microbiological Analysis
The results of the microbiological analyses are shown in Table 2. In general, variability was observed between samples. For most of the parameters analyzed, the mean counts were higher for conventional than for organic samples, but with the exception of total counts at 30 • C, no other differences were statistically significant (p > 0.05). In general, the microbial counts observed in the present study were higher than those previously reported [22,24,36]. It should be highlighted that, while in the present study, samples were purchased without any indication of collection dates because they were unpacked at the point of purchase, in the study by Oliveira et al. [22] the samples were collected directly from farms. Salmonella spp. and L. monocytogenes were absent in all the samples analyzed; E. coli was always below the detection limit of the enumeration technique. These pathogens have been occasionally found in lettuce samples from organic and conventional origin [22,24,36]. As in previous studies [22,24], no differences between organic and conventional lettuce were observed regarding the microbiological safety. It is important to highlight that most samples were purchased in stores of large supermarket chains with stringent food quality and food safety standards.

Nitrate Concentration
There were no significant differences (p > 0.05) between the nitrate content of organic and conventional lettuce (Table 3). Although we do not have information on the growth conditions of the Foods 2020, 9, 1274 6 of 10 samples analyzed (e.g., fertilization regimes), the fact that all the samples were of the same variety and were probably grown under similar light conditions may justify this observation [37]. According to the EFSA [38], the maximum levels for nitrate in fresh lettuce (Lactuca sativa) are between 2500 and 4500 mg/kg. It was possible to observe that, in organic lettuce, the nitrate values varied between 1400 and 1700 mg/kg and, on the other hand, the values in conventional lettuce varied between 1200 and 1900 mg/kg, which means that all the samples are within the recommended limits. According to Laia et al. [39], in Portugal the values of nitrate concentration for lettuce samples through the year of 2018 varied between 528 and 2369 mg/kg, meaning that our results are within the same range.

Method Evaluation
The analytical methodology was validated according to the EU Commission's DG Health & Food Safety (SANTE) guidelines [40]. Table 4 showed a summary of the analytical validation parameters. Suitable matrix-matched calibration curves were obtained. Appropriate coefficients of determination (R 2 ) were obtained for the scope of the targeted pesticides, with R 2 in the range of 0.9900-0.9994. The LOQ values were similar to or better than those of previously published works [19][20][21]. The LOQ ranged from 0.77 to 1.79 µg/kg for OCP and 0.57-1.57 µg/kg for OPP. The LOQ values are lower than the MRLs set by the European Commission (shown in Table 4). The recoveries of the method were evaluated at three concentration spiking levels, and were within acceptable ranges as set by European Union SANCO guidelines (70-120%). Except for malathion, which presented a recovery percentage of between 55% and 65%, the remaining pesticides showed values of between 72% and 97%, and with suitable precision and relative standard deviation (RSD) values below 20%. These results demonstrated that the developed method had a good analytical performance to apply in food safety analysis.

Sample Analysis
The validated method was applied to determinate the pesticide residues in 10 conventional lettuce and 10 organic lettuce samples. Only one organochlorine pesticide (α-HCH) was detected in three of the organic farming lettuce samples analyzed at a concentration of between (Org7 and Org9) and 0.11 µg/kg (Org6). The detectable concentrations were below the MRLs established by the European Commission. Despite being banned several years ago in the developed world, levels of these pollutants have been detected by researchers in Portugal and other countries [41,42] in various matrices, including environmental [43,44], food [42,45,46], and human samples [47]. These results contribute once again to reinforce the fact that these compounds due to their chemical characteristics persist in the environment. In addition, despite the illegal use of OCP in Europe, they were not reported in Portugal, but their extensive use remains ongoing in the African continent [41]. Indeed, these compounds have been found in vegetables (tomato, cabbage, and lettuce), and the authors of this study [42] reported that the most detected OCP in a greater concentration were the total HCHs (mean ΣHCHs = 15.115 µg/kg), followed by the total DDTs (mean ΣDDTs = 0.451 µg/kg). The presence of OCP was also reported in organic-farmed strawberries [45]. Regarding OPP, no detection was observed in the studied samples. However, in recent studies, the presence of OPP (e.g., chlorpyrifos) has been reported in organic lettuce [21] and conventional strawberries [46]. The results obtained in the present study showed that there are concerns with OPP in regards to chemical safety.

Conclusions
Regardless of the type of production, no harmful organisms nor chemical hazards were found in the lettuce samples analyzed. Despite the low number of samples analyzed, which we recognize as a limitation of this study, it is important to highlight that most of the samples were purchased in big retailers that are represented all over the country. These results tend to support the idea that big retailers demand high quality and safety standards from their suppliers.
Regarding pesticide analysis, the analytical parameters showed that the method presented a suitable analytical performance, including satisfactory LOQ, recoveries, and precision for the target analytes. The present work demonstrated that the developed method was reliable, simple, and environmentally friendly. Additionally, the method was robust and was successfully applied to detect pesticides in a total of 20 organic and conventional lettuce samples. The results were consistent with previous findings, which showed that organic foods contain few residues. It is therefore proposed to perform continuous monitoring studies of pesticide residues in vegetables, even though they are from organic production.