Microbial load and antibiotic resistance in raw beef preparations from northwest Spain

Abstract Beef preparations (meatballs, minced meat, hamburgers, white sausages, and red sausages) from northwest Spain were tested. Microbial counts ranged from 0.70 ± 0.00 log10 cfu/g (enterococci) to 9.57 ± 0.37 log10 cfu/g (psychrotrophs). In 73.3% of cases, total aerobic counts were higher than the microbiological limits set for the end of the manufacturing process in the European Union (EU Regulation 2073/2005). Forty Escherichia coli isolates were tested against thirteen clinically important antibiotics (disk diffusion method; CLSI). Three of the strains (7.5%) were susceptible to all the antibiotics, four (10.0%) showed resistance or reduced susceptibility to one antibiotic, and 33 (82.5%) were multiresistant (with resistance or reduced susceptibility to between two and eight antibiotics), with an average of 1.85 resistances per strain. The highest rates of resistance were observed for two antimicrobials widely used on cattle farms (cefalotin and tetracycline). The findings in this research emphasize the need to correctly handle beef preparations with the aim of reducing risks to consumers.

fecal coliforms, and enterococci) that are interesting to evaluate for meat because they act as indicators of its microbiological quality and the level of hygiene in the processes of production, handling, and maintaining an unbroken cold chain, also helping to predict the potential shelf life of products (Alonso-Calleja, Martínez-Fernández, Prieto, & Capita, 2004;Álvarez-Astorga, Capita, Alonso-Calleja, Moreno, & García-Fernández, 2002).
Bacterial resistance to antibiotics has grown at an alarming rate over the last few years and has been described as one of the greatest threats to public health, and consequently one of the main challenges for humanity in the twenty-first century. It has been considered as one of the biggest problems for health systems in Europe, as a worldwide pandemic, and even as a potential global health catastrophe .
In 2015, the European Union (EU) and the European Economic Area (EEA) saw more than 33,000 deaths attributable to infections caused by bacteria resistant to antibiotics, implying a loss of 874,541 disability-adjusted life-years (DALYs; Cassini et al., 2019). Infections by resistant bacteria lead to 23,000 deaths annually in the United States (Cecchini, Langer, & Slawomirski, 2015).
Moreover, these figures are on a rising trend, with estimates that in three decades' time ten million deaths will occur each year worldwide as a result of antibiotic-resistant bacteria, a higher figure than the 8.2 million deaths that will be caused by cancer. These figures should be contrasted with the 700,000 deaths attributable to antibiotic resistance in 2014 (O'Neill, 2016). The financial consequences of resistance to antibiotics are also of considerable weight, with estimates that every year these infections cost the healthcare systems of EU and EEA countries 1.1 thousand million euros (OECD, 2019).
The presence of antibiotic-resistant bacteria in foods is a direct risk for consumers due to the potential of these microorganisms to cause hard-to-treat foodborne infections. There is also an indirect risk of horizontal transfer of resistance genes to pathogenic microorganisms, including among unrelated genera, at various points along the food chain such as in the animals reared for consumption, in foodstuffs, on surfaces and equipment in the food-processing industry, and so on . Escherichia coli has a striking ability to acquire antibiotic resistance genes as a result of the efficient horizontal transfer mechanisms these microorganisms have developed over time. Hence, strains of this bacterial group act as reservoirs of resistance genes, which is a worrying fact in the context of public health since there is a high likelihood of gene transfer to other, pathogenic, bacteria. Moreover, this circumstance allows this bacterial group to be used as sentinel for resistance to antibiot-

ics (SCENIHR, 2009).
Monitoring resistance to antibiotics is essential not only to obtain information about the magnitude of this problem and trends within it, but also to plan and monitor the effectiveness of any control measures introduced. Monitoring antimicrobial resistance in indicator E.
coli from food-producing animals and food products of animal origin has been mandatory under EU legislation since 2014 (EFSA & ECDC, 2019). However, there is very limited information about resistance to antibiotics in strains of E. coli from beef and beef products, especially in northwest Spain.
The aims of the present work were to gain awareness of the microbiological quality of beef preparations purchased in three different retail establishments in northwest Spain, to compare levels of contamination in various types of beef preparations and to determine patterns of antibiotic resistance in strains of E. coli isolated from these foodstuffs.

| Meat samples
A total of thirty samples of beef preparations produced in three retail establishments in the city of León in northwest Spain were analyzed. The establishments involved were two butcher's shops (A and B) and a supermarket (C). Samples from meatballs, minced meat, hamburgers, white sausages, and red sausages were taken on site, transported immediately to the laboratory, and kept under refrigeration (4 ± 1°C) for a maximum of four hours prior to the start of the analyses. Six samples of each type of product were investigated.

| Microbiological analysis
Using sterile tweezers and a scalpel, 25 g of product were taken from each sample and placed in a homogenization bag together with 225 ml of sterile 0.1% peptone water (Oxoid Ltd.). These samples were homogenized (Masticator, IUL Instruments) for two minutes. Subsequent decimal dilutions were performed in test tubes with 9 ml of the same diluent. Table 1 shows the culture media, incubation conditions, and references used for each of the microbial groups studied. All inoculations were carried out in duplicate. Plates with between 25 and 250 colonies (spread-plate technique) or between 30 and 300 colonies (pour-plate technique) were counted, and mean counts were calculated. All culture media were purchased from Oxoid.

| Isolation and identification of Escherichia coli isolates
Between four and six colonies were taken from each violet red bile agar (VRBA) plate for later identification. The strains were streaked onto plates of tryptone soy agar (TSA) and then incubated for 24 hr at 44 ± 1°C to obtain pure cultures. The resultant pure cultures were examined for colony and cell morphology, Gram stain, and oxidase and catalase activities. Presumptive E. coli strains were confirmed using a miniaturized E. coli test system (Liofilchem s.r.l., Teramo, Italy) in accordance with the manufacturer's instructions. A total of forty E. coli isolates (eight strains from each type of product) were selected for later antibiotyping. The strains were stored at −50°C in tryptone soy broth (TSB) with 20% glycerol.

| Antibiotic resistance study
The susceptibility of the strains was tested against a panel of thirteen clinically important antimicrobials. The following antibiotic disks (Oxoid) were used to perform antibiograms by means of the disk diffusion method on Mueller-Hinton agar: gentamicin (CN;

| Statistical analysis
The microbial counts were transformed to log 10 cfu/g. The data were subjected to analysis of variance (ANOVA) techniques using Duncan's multiple range test to separate averages. Significant differences were established for a probability level of 5% (p < .05).
All the tests were carried out with the Statistica® 8.0 package (Statsoft Ltd.).

| Microbial load in beef preparations
Analysis of variance (ANOVA) was used to investigate the three factors, microbial group (G), establishment (E), and type of beef preparation (T). This highlighted the influence (p < .001) of the three factors and their interactions. Table 2 shows the average counts for all the microbial groups evaluated. Mean values (log 10 cfu/g) ranged from 0.70 ± 0.00 for enterococci in minced meat and hamburgers from establishment B to 9.57 ± 0.37 for psychrotrophs in meatballs from establishment C.
All forty beef samples harbored E. coli strains.
The levels of each microbial group by establishment and type of product are shown in Figures 1 and 2, respectively. The highest counts were recorded in establishment A for TACs, psychrotrophs, fecal coliforms, and enterococci, and in establishments A and C for enterobacteria. The lowest microbial levels were observed in establishment B. Similarly, the kind of meat preparation impacted on results, with red sausages having the lowest counts for all microbial groups except for enterococci ( Figure 2).

| Antibiotic resistance in E. coli isolates from beef preparations
Only three E. coli strains (7.5%) were susceptible to all the antibiotics, four (10.0%) were resistant or intermediate to one antibiotic, and thirty-three (82.5%) were multiresistant (resistant or intermediate to two or more antibiotics). Of these latter strains, six (15%) showed resistance or reduced susceptibility to two antibiotics, eight (20%) to three antibiotics, five (12.5%) to four antibiotics, seven (17.5%) to five antibiotics, two (5%) to six antibiotics, three (7.5%) to seven antibiotics, and two (5%) to eight antibiotics. The average number of resistances per strain was 1.85. Grouping together the strains with resistance and those with reduced susceptibility, the number of resistances per strain was 3.60. Figure 3 shows the number of strains resistant to each of the antibiotics examined. Noteworthy among the results is the high percentage of isolates with resistance or reduced susceptibility to cefalotin (33 strains, 82.5% of isolates) and tetracycline (22 strains, 55.0% of isolates).

Microbial group
Culture medium

| Microbial loads in beef preparations
Total aerobic count (TAC) has been used as a criterion to predict the shelf life of meat, given that the presence of this microbial group in large numbers may cause rapid alterations in the product. Counts between 10 7 and 10 8 cfu/g are generally associated with the beginning of changes in organoleptic properties such as appearance, smell, and taste (Nychas, Skandamis, Tassou, & Koutsoumanis, 2008). TACs can also be indicators of inappropriate processing, so determining their presence is a way of monitoring good manufacturing practices (GMP). While high levels of these microorganisms do not necessarily imply potential risks for human health, the importance of TAC lies in the fact that it is an expression of the hygienic quality of foodstuffs (Del Río et al., 2007). Note: Averages in the same row sharing one or more superscript letters show no significant differences between them (p ≥ .05). Averages in the same column for the same microbial group sharing one or more subscript letters show no significant differences between them (p ≥ .05).

F I G U R E 1
Microbial counts (log 10 cfu/g) from beef-based preparations (combined values for the different types of products examined) from three establishments (A, B and C). Values for the same microbial group sharing one or more letters show no significant differences between them (p ≥ .05) the average values for TAC in the present study, 73.3% exceeded 6.7 log units (Table 2) Dainty and Mackey (1992) or those of Griffiths, Phillips, and Muir (1981), who indicate that levels of 6 to 8 log 10 cfu/g of microorganisms are sufficient to produce off-odors and appearance defects in meat. Stanbridge and Davies (1998) also state that levels of psychrotrophs over 7 to 8 log units trigger strange smells and surface sliminess in meat. Contrarily, the results presented here are in agreement with the findings of El-Leithy and Rashad (1989), who report that none of the samples of minced meat and mincemeat products analyzed as having more than 8 log 10 cfu/g had offodors. Goepfert (1976) also found that aerobic plate counts equal to or greater than 7.7 log 10 cfu/g did not produce any organoleptic alteration in meat.
The samples investigated showed similar figures for TACs (incubated at 30°C) and psychrotrophic microorganisms (incubated at 7°C). These findings do not coincide with those in an earlier work  -Astorga et al., 2002), where higher levels of psychrotrophic microorganisms than TACs were recorded in meat and meat preparations stored under refrigeration. In fact, psychrotrophs are the microorganisms of choice for assessing the microbiological quality of refrigerated meat, and the fact that similar levels of these two microbial groups were observed in the present work suggests that storage temperatures were not sufficiently low.
Most of the enterobacteria and fecal coliforms found in meat come from contamination with feces, so their presence in large numbers may indicate poor hygiene in the slaughterhouse from which the meat is sourced, insufficiently hygienic handling, inappropriate storage, or a combination of these (Andritsos et al., 2012). It should be noted that all the counts for these microbial groups were within the guideline microbiological criteria, which states that this should not exceed 2 (maximum 4) log units/g (ICMSF, 2011;IFST, 1997) or 3.5 (maximum 4.5) log units/g (Uyttendaele et al., 2010). Other authors have noted similar counts to those in the present study regarding the presence of enterobacteria and fecal coliforms in beef (Lunning et al., 2011). The detection of E. coli strains in all the meat analyzed is also a frequent finding for red meat and poultry samples (Álvarez-Fernández, Cancelo, Díez-Vega, Capita, & Alonso-Calleja, 2013).
The presence of a large number of enterococci in foodstuffs implies inadequate hygiene practices or exposure of the food to conditions that allow undesirable bacteria to excessively multiply (Franz, Holzapfel, & Stiles, 1999). These microorganisms can play an important role as indicators of deficiencies in cleaning and disinfection practices, given their considerable resistance not only to drying out, but also to both high and low temperatures and to detergents and disinfectants. This strong resistance also explains why this microbial group is not a valid indicator of fecal contamination (Thian & Hartman, 1981). The levels of enterococci recorded in this study are similar to those previously observed in prepared poultry products (Buzón-Durán, Capita, & Alonso-Calleja, 2017).
When the data were grouped by establishment, the highest microbial counts were found in establishment A, followed by establishment C. Establishment B's results were the most satisfactory.
Other authors have also found differences between types of establishment. To this effect, Andritsos et al. (2012) compared the microbiological quality in specialist butcher's shops and the meat sections of supermarkets. They reported deficient outcomes in all the supermarkets and marginally acceptable results in the butcher's shops.
In the present work, establishment A was a small butcher's shop with staff with many years' experience, but little specific training in food hygiene and little engagement with the Hazard Analysis and Critical Control Points (HACCP) system in force. Furthermore, there were no set microbiological specifications with the slaughterhouse supplying the meat. These may be explanatory factors as to why high microbial counts were observed in this establishment. Establishment B was a medium-sized butcher's specializing in the sale of high-quality products, whose personnel had less experience but had undergone more specific training in food hygiene, and was highly committed to implanting the HACCP system. Moreover, the slaughterhouse supplying establishment B with meat was small-scale and belonged to a meat brand of recognized prestige where hygiene quality is a highly controlled aspect. Establishment C was the meat section of a supermarket. In this kind of establishment, all staff are trained in food hygiene by the quality control department of the business prior to taking up employment, and considerable resources are allocated to implanting the HACCP system. Furthermore, the chain of supermarkets to which this establishment belongs had agreed upon some microbiological specifications with its supplier and any failure to comply with them would result in rejecting the meat.
Differences in microbiological quality were observed among the various kinds of meat preparations. Noteworthy among the results were the low counts for most of the microbial groups in red (spiced) sausages, the exception being enterococci. This finding points to the possibility that the spices and other additives used in the production process, for instance paprika, may have a bactericidal and/or bacteriostatic effect on certain microorganisms, as has been suggested by other researchers (Gottardi, Bukvicki, Prasad, & Tyagi, 2016).

| Antibiotic resistance in E. coli isolates from beef preparations
The forty strains of E. coli isolated were tested against a panel of thirteen clinically important antimicrobials. An alarming 92.5% of the strains were resistant or intermediate to one or more antibiotics, with over 80% of the strains presenting multiresistance (with resistance or reduced susceptibility to two or more antibiotics).
Bacteria resistant to antibiotics can cause infections in consumers through the consumption of foodstuffs in instances of cross-contamination or when food is insufficiently cooked. Given that clinical treatment for infections is rendered complex when strains are resistant to antibiotics, the presence of multiresistant bacteria in food is a cause for great concern .
Along these lines, some years ago it was demonstrated that the use of antibiotics modifies the resistance genes present in bacterial communities (the resistome). The effects on this set of genes persist for years even in the absence of any contact with antibiotics (Agga et al., 2019;Johnsen et al., 2011;Sommer & Dantas, 2011). Smith et al. (2007 found that successive exposures to antibiotics create a resistance that is stable over time and that resistant strains can compete with susceptible strains even when there is no selective pressure. Hence, rates of resistance to antibiotics on a farm need not be directly linked to their use in this environment (Luangtongkum et al., 2006).

| CON CLUS IONS
The samples of beef preparations studied presented high levels of microorganisms. In 73.3% of cases, total aerobic counts were higher than the limits set in microbiological criteria for the end of the manufacturing process. Striking differences in the microbial counts obtained were seen depending on the establishment where the products were purchased and on the type of meat preparation involved. The results from the present study provide evidence that strains of E. coli in beef preparations pose a major potential risk (both direct and indirect) to consumers, given the considerable rates of resistance or reduced susceptibility to antibiotics that were found. This is a worrying fact from the viewpoint of public health, pointing to a need to take measures to reduce the rates of resistance to antibiotics in the bacteria present in these foodstuffs.