Abstract
Milk is a putrescible commodity that is extremely prone to microbial contamination. Primarily, milk and dairy products are believed to be easily contaminated by pathogenic microorganisms, including Listeria monocytogenes, Salmonella spp., and Staphylococcus aureus. The microbiological quality of raw milk and dairy products regarding foodborne pathogens is of paramount importance due to concern of human health. In this study 400 buffalo raw milk samples were screened for assessing the prevalence of L. monocytogenes, Salmonella spp., and S. aureus. This study implemented uniplex-polymerase chain reaction (u-PCR) and multiplex-polymerase chain reaction (m-PCR) assays for the fast simultaneous detection of these pathogens comparing to the conventional culturing methods. Raw milk samples were found contaminated with the prevalence of 2.2%, 4.0%, and 14.2% for L. monocytogenes, Salmonella spp., and S. aureus, respectively. These pathogens were detected with the optimized polymerase chain reaction assays after 6 h of enrichment. u-PCR and m-PCR demonstrated the limit of detection as 104, 102, and 10 cells/mL after 6, 12, 18, and 24 h for each culture of the pathogens. A high sensitivity (10 colony-forming unit [CFU]/mL) of the m-PCR protocol was noted. The developed protocol is a cost-effective and rapid method for the simultaneous detection of pathogens associated with raw milk and dairy industries.
Introduction
According to World Health Organization (WHO), foodborne diseases lead to 420,000 global deaths annually (Guldimann and Johler, 2018). The Centers for Disease Control and Prevention (CDC) has reported ∼73,000 annual cases of foodborne illness in 2013 leading to the hospitalization of 4200 patients and 80 deaths in the United States. High prevalence of foodborne diseases has also been reported in several developing countries (Zeng et al, 2018). Milk is regularly consumed as a nutritious food. However, unhygienic milk processing conditions make it susceptible to microbial contamination that could ultimately cause health issues in humans. The highly nutritive nature of milk facilitates rapid microbial growth. The factors such as the infection of mammary glands and poor hygienic practices, including unsterilized milk storage containers, improper hand washing, and coughing, could result in milk contamination (Regasa et al, 2019).
Listeria monocytogenes, Salmonella, and Staphylococcus aureus are primarily public health-related pathogens, which cause food poisoning of milk and dairy products (Hayashi et al, 2013; Kim et al, 2018). L. monocytogenes associated food poisoning cases have increased from 1439 in 2005 to 2530 in 2016 and in the United States. Approximately 73,000 cases of foodborne diseases occur annually, and L. monocytogenes presence in 20% of raw foods, including milk products, has been reported. In Asia, listeriosis outbreaks have only been reported in Japan, Vietnam, Turkey, China, and Thailand (Wai et al, 2020). In India, occasional human listeriosis cases have been reported but its prevalence in foods is limited (Dhama et al, 2013; Khan et al, 2014a; Khan et al, 2014b; Mary and Shrinithivihahshini, 2017; Shrinithivihahshini et al, 2011; Sreeja et al, 2016; Suriyapriya et al, 2016).
Furthermore, bacteria belonging to the genus Salmonella are known to cause 90 million cases of diarrhea-related diseases every year worldwide, and 85% of these cases are linked with contaminated food consumption, including milk and dairy products (Abulreesh, 2012). S. aureus intoxications due to the consumption of contaminated dairy products have resulted in foodborne disease outbreaks globally. The prevalence of S. aureus in 2.6% raw milk samples has been reported in a farm study of United States (Leuenberger et al, 2019).
The incidence of foodborne diseases is high in developing countries (Zeng et al, 2018). During the last decade, the prevalence of disease causing pathogens in milk and milk products has ranged between 2.0% and 11.9% in India (Kaushik et al, 2014; Singh et al, 2018; Sudhanthirakodi et al, 2016). Food safety issues have gained significant public interest during the last two decades. The food industry is now enforcing strict regulations to limit the presence of foodborne pathogens in various foods. Salmonella, L. monocytogenes, and S. aureus prevalence are not permissible per 25 mL of milk sample (FSSAI, 2011). The significance of these foodborne pathogens to food safety and public health demands the development of sensitive, accurate, and rapid detection methods for their identification in complex food matrices, including milk.
The conventional culturing methods are widely used to detect foodborne microbes. However, these methods are slow (3–5 d) and laborious especially when the results are urgently needed. Therefore, polymerase chain reaction (PCR) technique is used worldwide for the detection of target bacterial pathogens. Due to high sensitivity and specificity, PCR has rapidly emerged as a principal technique for foodborne pathogen identification. Multiplex-PCR (m-PCR) could further utilize various degenerated primers for the simultaneous amplification of multiple target genes during a single reaction. m-PCR could swiftly detect pathogenic microorganisms to reduce workload and save time compared to uniplex-PCR (u-PCR) (Ding et al, 2017).
The present study applies u-PCR and m-PCR assays for the fast detection of three important foodborne pathogens (Salmonella spp., L. monocytogenes, and S. aureus) simultaneously. Furthermore, the overall efficiency of PCR assays based on the sensitivity, specificity, and precision in quantitative analyses was systematically monitored in comparison with the conventional culturing methods.
Materials and Methods
Collection of food samples
Four hundred and fifty buffalo raw milk samples were collected from the local dairy farms and retail market dairy shops of Bareilly, Uttar Pradesh, India according to Andrews and Hammack (2003). All the samples were aseptically collected from milk bucket in screw-top (50 mL) sterile glass bottles and placed in an ice-box for transferring to the laboratory without exposure to direct sunlight. The conventional culturing methods as described by Hitchins et al (2003), Amaguaña et al (1998), and Bennett and Lancette (1998) were initiated on the same day for the identification of target bacteria.
Reference strains and bacterial cultures used in this study
The standard bacterial strains used in this study and their sources are listed in Supplementary Table S1. All strains and cultures were investigated for morphological, biochemical, and virulence features. Bacterial strains were maintained on Brain-heart infusion (BHI) broth/agar (HiMedia) by subculturing according to Cappuccino and Sherman (1992).
Isolation and culture identification of target bacteria from raw milk
Target bacteria were isolated by following the standard methods of USFDA/BAM/CFSAN (Hitchins et al, 2003). A biochemical assay based on carbohydrate utilization, urease, citrate, H2S production on triple sugar iron (TSI) agar slants, lysine decarboxylase assay, and IMViC test further confirmed the presumptive Salmonella isolates (Andrews and Hammack, 2003). Gray-black shiny convex colonies surrounded by a zone of clearing were considered presumptive S. aureus on Baird Parker agar (HiMedia, India). To confirm S. aureus, a biochemical assay utilizing glucose, oxidase, catalase, mannitol, coagulase, growth in 10% NaCl, and DNAse was performed according to the USFDA/BAM/CFSAN method as reported by Bennett and Lancette (1998).
All isolated bacteria were maintained on BHI broth and stored with 20% glycerol at −20°C.
Identification of bacterial pathogens by u-PCR
Genomic DNA extraction
Phenol Chloroform extraction protocol was followed for DNA extraction (Sambrook and Russel, 2001). hlyA, invA and oriC, and nuc genes amplification from the bacterial DNA template of L. monocytogenes, Salmonella, and S. aureus: The details of primers utilized for the detection of bacterial pathogens are depicted in Table 1. To detect L. monocytogenes, the hlyA gene (234 bp) was amplified using PCR primers (Furrer et al, 1991). PCR cycling conditions are provided in the Supplementary Table S2. The reference/standard culture of L. monocytogenes (MTCC 657) served as positive control, while Listeria ivanovii (NCTC 11846), Listeria innocua (NCTC 11288), S. aureus (MTCC 1145), Rhodococcus equi (MTCC 1135), Bacillus subtilis (MTCC 121), Shigella sp. (SM-07), Citrobacter sp. (EN-06), and Escherichia coli (MTCC 443) served as negative control depicted in Table 2.
Oligonucleotide Primers Used in This Study
Prevalence of Listeria monocytogenes, Salmonella spp., and Staphylococcus aureus in Raw Milk
invA and oriC genes were amplified for the detection of Salmonella as described by Cocolin et al (1998) and Widjojoatmodjo et al (1991); PCR cycling conditions are provided in the Supplementary Table S2. The reference/standard cultures of Salmonella typhimurium (MTCC 98) and Salmonella enteritidis (E2094) served as positive control, while E. coli (MTCC 443), S. aureus (MTCC 1145), B. subtilis (MTCC 121), Shigella sp. (SM-07), Citrobacter sp. (EN-06), and Campylobacter jejuni (NCTC 11168) served as negative control depicted in Supplementary Table S4.
S. aureus was detected by amplifying the nuc gene (270 bp) using PCR primers according to Brakstad et al (1992). PCR cycling conditions are provided in Supplementary Table S2. The reference/standard culture of S. aureus (MTCC 1145) was used as positive control, whereas S. epidermidis (MTCC 435), L. monocytogenes (MTCC 657), and R. equi (MTCC 1135), B. subtilis (MTCC 121), Shigella sp. (SM-07), Citrobacter sp. (EN-06), and C. jejuni (NCTC 11168) cultures served as negative controls for testing the PCR specificity (Supplementary Table S4). PCR products were separated through agarose gel electrophoresis (1.5%), stained with ethidium bromide (Sigma-Aldrich), and visualized under ultraviolet (UV) light.
Sensitivity of u-PCR for determining the limit of detection in spiked (artificially inoculated) milk samples
The methodology of Alarcón et al (2004) with minor modifications was adopted to prepare the inoculated raw milk samples for determining u-PCR assay sensitivity.
False positive results may obtain from other indigenous bacterial population present in raw milk. For that purpose u-PCR assay was checked for its specificity, using different bacterial pathogens listed in Supplementary Table S4. Each aliquot was examined for the presence of specific bacterial pathogen using cultural methods also. A measure of 1.0 mL bacterial culture was spread on respective selective solid media using spread plate method for examining bacterial colonies.
DNA from spiked samples was extracted by heat lysis method (Arora et al, 2006). Initially, the spiked samples (100 μL) were centrifuged at 5000 rpm for 5 min. The pellet was resuspended in 100 μL in a sterile microcentrifuge tube, and the same procedure was repeated thrice. The resulting pellet was completely dissolved in phosphate buffered saline and heat treated for 10 min in a boiling water bath followed by immediate ice treatment for 10 min. The obtained cell lysate was centrifuged at 5000 rpm for 5 min, and the supernatant was collected in a fresh tube to use as a PCR template. PCR products were subjected to 1.5% (w/v) agarose gel electrophoresis, and an uninoculated aliquot served as control.
Multiplex-PCR
The m-PCR assay was performed for the amplification of respective genes of L. monocytogenes (hlyA, 234 bp), S. typhimurium (oriC, 163 bp), and S. aureus (nuc, 270 bp) according to Alarcón et al (2004). PCR cycling conditions are provided in the Supplementary Table S3. L. monocytogenes (MTCC 657), S. typhimurium (MTCC 98), and S. aureus (MTCC 1145) served as a positive control for testing the PCR specificity, whereas L. ivanovii (NCTC 11846), L. innocua (NCTC 11288), E. coli (MTCC 443), B. subtilis (MTCC 121), Shigella sp. (SM-06), Citrobacter sp. (EN-06), C. jejuni (NCTC 11168), S. epidermidis (MTCC 435), and R. equi (MTCC 1135) served as negative control (Supplementary Table S4). PCR products were separated through agarose gel electrophoresis (1.5%), stained with ethidium bromide (Sigma-Aldrich), and visualized under UV light. PCR products were stored at −20°C till further analysis.
Sensitivity of m-PCR for determining the limit of detection in spiked (artificially inoculated) milk samples
The methodology of Alarcón et al (2004) with minor modifications was adopted to prepare the inoculated raw milk samples for determining m-PCR assay sensitivity. False positive results may be obtained from other indigenous bacterial population present in raw milk. For that purpose m-PCR assay was checked for its specificity, using different bacterial pathogens listed in Supplementary Table S4. Each aliquot was examined for the presence of specific bacterial pathogen using cultural methods also. A measure of 1.0 mL bacterial culture was spread on respective selective solid media using spread plate method for examining bacterial colonies.
The heat lysis method of Arora et al (2006) was followed for the DNA extraction from the spiked samples as mentioned above.
Results and Discussion
Determination of u-PCR specificity for the targeted bacterial pathogens
Specific primers were used to verify the specificity of u-PCR assay against different bacterial organisms as shown in Supplementary Figure S1. The results indicated the specificity of primer pairs for the corresponding target organism. Specific PCR amplification products were achieved for each bacterial pathogen. Nonspecific products were not observed in any u-PCR assay. Similarly, m-PCR assay of L. ivanovii (NCTC 11846), L. innocua (NCTC 11288), E. coli (MTCC 443), and R. equi (MTCC 1135) only generated pathogen-specific bands (Supplementary Fig. S2).
Prevalence of L. monocytogenes, Salmonella spp., and S. aureus in raw milk samples
The prevalence of different bacterial pathogens in raw milk is depicted in Table 2. S. aureus prevalence was noted to be highest at 14.2% (n = 57/400) among all the bacterial pathogens followed by Salmonella spp., 4.0% (n = 16/400), and L. monocytogenes 2.2% (09/400). A wide range of S. aureus prevalence (15–80%) has been previously reported in raw milk (Toubar et al, 2018). A broad range of Salmonella prevalence in raw milk has also been reported globally. In Egypt, different studies have reported higher Salmonella prevalence of 14.0–52.0% and 12.0–24.0% in raw milk (El-Baz et al, 2017; Omar et al, 2018). Similarly, Singh et al (2018) have demonstrated a high (11.9%) Salmonella prevalence in the raw milk samples of India. Approximately in line with our findings, L. monocytogenes prevalence has been reported from different countries, including India (1.09%), Morocco (5.9%), Sweden (1.0%), United States (6.5%), Latvia (1.4%), Iran (1.7–5.0%), and Ethiopia (2.0%) (Akrami-Mohajeri et al, 2018; Mansouri-Najand et al, 2015; Seyoum et al, 2015; Sharma et al, 2017).
The presence of bacterial pathogens in raw milk could be attributed to the polluted environment of dairy farms and milk shops. The unsatisfactory pre- and posthandling of raw milk and inadequate cleaning further facilitate the pathogen dissemination. Pathogen invasiveness and toxin-associated pathogenicity of raw milk-related bacteria could pose serious health hazards to the consumers. Several studies have also elaborated that direct contact could transfer bacterial pathogens in milk handlers through fecal-oral routes (El-Baz et al, 2017). Different pathogen prevalence rates found during this study might be due to the unhygienic milk processing and penurious sanitation of animals, dairy farms, and milk shops (Sonnier et al, 2018). These findings advocate the unawareness of milk handlers with basic raw milk collection guidelines.
Evaluation of limit of detection of u-PCR and m-PCR in comparison to bacterial culture methods in spiked (artificially inoculated) milk samples
Culture-based microbiological investigations have served as gold standard methods by offering unique benefits such as pathogen identification, contributions in outbreak management, and generation of antibiotic susceptibility patterns. However, the disadvantages such as time-consuming protocols and the need for multiple enrichment culturing broths and solid media hinder their applications when results are required rapidly. During this study, the efficiency of conventional culture methods was evaluated at different enrichment times, which detected the pathogens after 12 h (Table 3). The limit of detection (LOD) of the methods revealed almost similar results for each pathogen while detecting through culturing methods (Table 4). The culture methods of each pathogen demonstrated LOD as 105, 102, and 10 cells/mL after 12, 18, and 24 h. Culturing methods could not detect the LOD at 6 h (Table 4). The sensitivity and efficiency of PCR assays were also evaluated during this study in comparison to conventional culture methods.
Efficiency of Uniplex-Polymerase Chain Reaction Assay for the Detection of Listeria monocytogenes (hlyA), Salmonella spp. (invA), and Staphylococcus aureus (nuc) in Artificially Contaminated Raw Milk Samples After 0, 6, 12, 18, and 24 h of Enrichment
BPW, buffered peptone water; u-PCR, uniplex-polymerase chain reaction.
Sensitivity of Uniplex-Polymerase Chain Reaction and Multiplex-Polymerase Chain Reaction Assay for the Detection of Listeria monocytogenes (hlyA), Salmonella spp. (invA), and Staphylococcus aureus (nuc) in Artificially Contaminated Raw Milk Samples After 0, 6, 12, 18, and 24 h of Enrichment in Buffered Peptone Water
m-PCR, multiplex-PCR.
PCR assays successfully detected the pathogens even at 0 h of enrichment (Table 3). u-PCR and m-PCR demonstrated the LOD as 107, 106, 102, and 10 cells/mL after 0, 6, 12, 18, and 24 h of enrichment for each pathogen (Supplementary Figs. S3 and S4). Culture methods could not produce specific results with 108, 107, 106, 105, and 104 cells/mL after 6 h, which reduced to 104–103 cells/mL after 12 h. On the other side, u-PCR and m-PCR demonstrated these results as 105 and 104 cells/mL after 6 h. No false positive results were obtained during u-PCR assay checked for its specificity using different bacterial pathogens (Supplementary Fig. S1).
The superiority of the PCR technique over culturing methods in pathogen detection from artificially inoculated milk samples has been documented. The bacterial cultures in antibiotics containing media with significantly reduced bacterial growth and lower production of biochemical compounds are extremely difficult to detect through the conventional techniques (Gebretsadik et al, 2011; Karmakar et al, 2016; Percival et al, 2004; Priyanka et al, 2016; Vidic et al, 2019). PCR-based pathogenic identification is a highly efficient and specific technique. However, the contaminants such as traces of calcium ions, proteinases, collagen, few polyphenolic substances, and complex polysaccharides could lead to inhibition of Taq polymerase and degradation of DNA thus reducing the efficiency of PCR assays (Bélec et al, 1998; Kim et al, 2001; Law et al, 2014; Upadhyay et al, 2010; Vidic et al, 2019; Widjojoatmodjo et al, 1991). Despite these issues, several studies have demonstrated PCR-based techniques as rapid, sensitive, efficient, and reliable for bacterial identification in food samples (Beaubrun et al, 2012; Liu et al, 2019; O'Grady et al, 2009; Wang et al, 1992).
Rapid and simultaneous identification of multiple pathogens in food products could significantly ensure food safety. Therefore, m-PCR is widely used for microbial detection. It could simultaneously detect multiple pathogens in a single PCR. The standardization and optimization of the m-PCR protocol is a complicated process but once established it is a comparatively inexpensive and fast technique for microbial detection. Different combinations of foodborne pathogens have been efficiently detected using m-PCR, which include E. coli O157:H7, L. monocytogenes, Salmonella spp., S. aureus, and Vibrio cholerae (Chen et al, 2012), L. monocytogenes, S. aureus, and Salmonella spp. (Ding et al, 2017), S. aureus, E. coli O157:H7, Salmonella enterica, and L. monocytogenes (Parichehr et al, 2019). During the current study, a m-PCR assay was optimized to simultaneously detect L. monocytogenes, Salmonella, and S. aureus from the raw milk samples, which revealed high sensitivity of m-PCR protocol (10 colony-forming unit [CFU]/mL).
Similar to u-PCR, LOD values also varied at different enrichment times with m-PCR assay. However, the simultaneous pathogen detection with m-PCR assay reduced the time by revealing the LOD of 10 and 10 cells/mL even after 6 and 18 h of enrichment (Table 4). These findings confirmed the high sensitivity of the established protocol. Lower LOD could be attributed to the poor quality of extracted DNA from food samples (Qamar et al, 2017). The reference method of Alarcón et al (2004) adopted in this study has documented a very close m-PCR assay sensitivity in different beef samples. The lowest pathogen quantity was detected for L. monocytogenes (5.7 CFU/mL), S. typhimurium (7.9 CFU/mL), and S. aureus (26 CFU/mL). Similarly, various studies have depicted lower pathogen detection with u-PCR assay at different enrichment times (Amagliani et al, 2007; Bailey, 1998; Sowmya et al, 2012). The detection limit could be influenced by the sample source, amount of DNA, contaminants in the extracted food sample DNA, and the presence of nontarget DNA interfering in the amplification reaction.
To conclude, m-PCR assay for the simultaneous identification of three important foodborne pathogenic bacteria (L. monocytogenes, Salmonella spp., and S. aureus) was successfully developed and validated during this study. The m-PCR technique coupled with heat lysed DNA extraction from pre-enrichment broth revealed the raw milk LOD of 10 cells/mL after 18 h. The simple DNA extraction method and proposed assay significantly save time in pathogen identification compared to traditional culturing methods. Therefore, this technique could play an imperative role in the quality control of the dairy industry. Furthermore, except PCR apparatus, this method does not require sophisticated equipments for the pathogen identification and evaluation. Therefore, it is suitable for both large and small scale dairy industries.
Ethics Approval
Not required as no human participant or animal subject was used in this research.
Data Availability
All data of this research have been used in the article.
Footnotes
Acknowledgments
The authors are grateful to Director, IVRI, for providing permission and facilities during this investigation. The authors also express their sincere gratitude to Dr. SVS Malik (Ex-Head VPH) for his support throughout work at VPH division, IVRI, and for providing reference strains of Listeria spp., S. epidermidis, and R. equi. The authors are also thankful to Dr. R.K. Agarwal (Incharge of National Salmonella Centre, IVRI, Bareilly, India) for providing reference strains of Salmonella and continuous support during the investigation. The authors also appreciate Dr. K.N. Bhilegaonkar for his guidance on laboratory procedures and facilities during this work. The authors are also thankful to Dr. Mayank Rawat (Principle Scientist), Division of Biological Standardization, IVRI, Izatnagar, India for providing reference strain S. aureus.
Authors' Contributions
Conceptualization: J.A.K. and R.S.R. Data curation: R.G., F.M.H., and L.A.N., P.A. Formal analysis: J.A.K., I.A., and F.M.A. Investigation: J.A.K. Methodology: J.A.K., I.A., and R.S.R. Resources: M.A., T.A., A.A.-K., and J.A. Supervision: I.A.. Validation: I.A., F.M.H., and K.E. Writing—original draft: J.A.K., I.A., and H.H.A. Writing—review and editing: H.H.A.
Disclosure Statement
No competing financial interests exist.
Funding Information
The authors are grateful to the Deanship of Scientific Research, King Saud University for providing funds through Vice Deanship of Scientific Research Chairs. The research work was only supported by the institutional research budget.
Supplementary Material
Supplementary Table S1
Supplementary Table S2
Supplementary Table S3
Supplementary Table S4
Supplementary Figure S1
Supplementary Figure S2
Supplementary Figure S3
Supplementary Figure S4
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
