Establishment of ddPCR detection technology system for three contaminants in fermented milk

Abstract

Three typical contaminant bacteria, namely, Gluconobacter, Acetobacter, and Lactobacillus, were selected from fermented milk. The strong specific genes were screened as target sequences, and the droplet digital polymerase chain reaction (ddPCR) amplification system and reaction conditions were optimized. We established a ddPCR detection method for the three kinds of bacteria in fermented milk and then verified the specificity and sensitivity of the method. Absolute quantitative results were also analyzed. Results showed that ddPCR detection method for the three contaminants was specific and sensitive, and that the lowest detectable concentrations of Gluconobacter, Bacteroides, and Lactobacillus plantarum were 8.8 × 10⁰ CFU/mL, 8.9 × 10¹ CFU/mL, and 9.6 × 10¹ CFU/mL, respectively. The relationship between the quantitative analysis results and ddPCR test results was good, proving the feasibility of ddPCR for absolute quantitative detection. This manuscript reports on establishment of a ddPCR detection system for three bacterial contaminants in fermented milk. The ddPCR detection system has efficient detection of targeted bacterial contaminants in fermented milks that can be valuable to the dairy industry in terms of preventing economic loss due to microbial spoilage and associated quality defects.

Keywords:
droplet digital polymerase chain reaction; Gluconobacter; Acetobacter; Lactobacillus

1 Introduction

With the continuous development of society, peoples living standards and quality of life have been continuously improved, and healthy foods have become increasingly desired as consumers’ food selection criteria have shifted from taste and price to health-related food information such as nutrition and food materials and ingredients (Bi et al., 2020). Fermented milk is a collective term for milk and other animal dairy products and plays an important role in the formation of microbial fermentation and acidic coagulation products (Farnworth et al., 2007; Muramalla & Aryana, 2011). Fermented dairy foods, owing to their rich nutrient and probiotic content, are associated with a wide range of health benefits (Zhang et al., 2019). Fermented milk contains a lot of live probiotics (Klemm et al., 2020) and some essential nutrients, and it can alleviate lactose intolerance (Barichella et al., 2016), can assist in the treatment of constipation and bacterial diarrhea (González et al., 2019). can regulate the balance of gut microbiome (Galdeano et al., 2011). The latest research found that not only viable Lactobacillus strains but also milk products fermented by Lactobacillus can modulate the gut–bone axis (Eor et al., 2020), especially Lactobacillus plantarum B719‐fermented milk product could serve as a potential candidate for the treatment and prevention of postmenopausal osteoporosis (Lee et al., 2020). Therefore, fermented milk has a certain market demand. However,the product quality of fermented milk extremely susceptible to storage conditions, shelf-life, microbial contamination, and other factors. The effect of microbial contamination on fermented milk is the most serious and difficult to control (Nwamaka & Chike, 2010; Falenski et al., 2011; Poimenidou et al., 2009). Fermented milk production requires various fermenting bacteria, such as lactic acid bacteria, yeast and acetic acid bacteria in kefir (Saygili et al., 2021), and there are contaminating microorganisms in raw materials such as staphylococcus in raw milk (Moghadam et al., 2021). Thus, the fermentation of milk is easily susceptible to bacterial contamination. Once the contaminants began to multiply and metabolize, the fermented milk will produce whey separation, become musty, show discoloration, produce gas, clot, and have alcohol-fermented flavor or sour taste, which affects the function of fermented milk and causes fermented milk to spoil. The quality and safety of dairy products such as fermented milk has attracted more and more attention. Once the contaminated fermented milk is brought into the market, not only it will affect the health of consumers but also generate a huge loss in economic benefits to enterprises, seriously affecting the reputation of the enterprise. Different strains of fermentation had obvious effects on sensory characteristics of fermented milk (Cruz et al., 2021). Therefore, timely and effective detection of bacterial population in fermented milk is essential to ensure the quality and safety of fermented milk (Raspor & Goranovič, 2008).

Gluconobacter is an important bacterium belonging to the genus Acetobacter, which was discovered and studied previously. Gluconobacter belongs to aerobic bacteria and is common in environments with high sugar content. Acetobacter is also an important bacterium belonging to the genus Acetobacter, which is lately, initially classified as a subgenus of Acetobacter (Goodfellow et al., 2012). Acetobacter uses mostly glucose, ethanol or acetate as carbon source for its growth with the optimum growth pH of 2.5–6.0. Lactobacillus plantarum belongs to the genus Lactobacillus and are mostly isolated from plants, suggesting its name. L. plantarum belongs to the same type of lactic acid bacteria (Shah, 2000). Growth of L. plantarum requires complex nutrient requirements including amino acids, peptides, salts, fatty acids, and fermentable carbohydrates. Acetate and L. plantarum in fermentation products, wine industry, and meat industry have also caused great concern in the past few years (Bartowsky & Henschke, 2008; Ouoba et al., 2012). Fermented milk is contaminated with acetic acid bacteria, and internal microbes interact with various mechanisms, which may cause positive, neutral or negative effects directly through physical contact or through signal molecules, ultimately leading to changes in product composition and product quality (Pastink et al., 2008). The detection of acetic acid bacteria as microbial contaminants meets actual production needs and can also improve product quality and safety (Foschino et al., 1993).

The traditional method for detecting acetic acid bacteria is gradient culture analysis method, which is a method of combining traditional purification and culture by using physiological and biochemical identification reactions (Zhou et al., 2017). Phenotypic analysis is simple in principle, simple in operation, and low in cost. However, due to the complex and wide variety of conditions for the separation of acetic acid bacteria, the detection sensitivity is low, the detection period is long, and the separation and screening are easily disturbed by other strains in physiological and biochemical reactions. The characterization of the acetic acid bacteria will affect the accuracy of the test (Gibson et al., 1996; Heid et al., 1996; Zipper et al., 2004). At the end of the 20th century, Vogelstein & Kinzler (1999) proposed the concept of digital polymerase chain reaction (dPCR), which distributes a sample reaction system evenly into a large number of reaction units. Each reaction unit does not contain or contains one to multiple nucleic acid sequences. The number is consistent with the Poisson distribution. The target molecules were independently subjected to polymerase chain reaction (PCR) amplification in each reaction unit. At the end of the amplification, the fluorescence signal of each reaction unit is detected, and the copy number of the target nucleic acid sequence is calculated based on the ratio of the Poisson distribution and the number of reaction units positive for the fluorescence signal to the total number of reaction units. Currently, commercialized dPCR technology can be divided into two categories, namely, microtiter digital PCR or droplet dPCR (ddPCR) and chip-type dPCR (cdPCR) technology. The dPCR exhibits the advantages of high sensitivity, high precision, high tolerance, and absolute quantification. The dPCR is an emerging technology with great potential. The dPCR is widely used in rare mutation detection, copy number variation analysis, and complex sample gene expression detection (Margulies et al., 2005; Sanders et al., 2011; Dube et al., 2008).

In this study, we established a ddPCR detection system for three kinds of contaminant bacteria and carried out the specificity, sensitivity, and absolute quantitative feasibility of the three bacteria, Gluconobacter, Gluconacetobacter and L. plantarum.

2 Materials and methods

2.1 Experimental strains

The strains used in this experiment (Table 1) were stored at -80 °C in the laboratory. All strains were activated with their indicated medium, incubated at the appropriate temperature, and transferred thrice. Except for Gluconobacter, Gluconacetobacter and Lactobacillusplantarum, the rest were control samples.

Table 1
Experimental strains.

Strain number	Strain name (Latin)	Culture conditions/Gram-character	Culture medium
CGMCC1.0637	Gluconobacter oxydans	28 °C, Aerobic, G−	Acetate culture medium
CGMCC1.3748	Pseudomonas aeruginosa	28 °C, Aerobic, G−	Acetate culture medium
CICC10357	Glucose bacteria	28 °C, Aerobic, G−	Acetate culture medium
CICC21682	Acetobacter aceti	30 °C, Aerobic, G−	Acetate culture medium
CICC21684	Acetobacter aceti	30 °C, Aerobic, G−	Acetate culture medium
CICC21683	Acetobacter aceti	30 °C, Aerobic, G−	Acetate culture medium
CICC22519	Acetobacter aceti	30 °C, Aerobic, G−	Acetate culture medium
CGMCC1.24	Acetobacter liquefaction	30 °C, Aerobic, G−	Acetate culture medium
CGMCC1.811	Acetobacter spp	30 °C, Aerobic, G−	Acetate culture medium
CICC21102	Sucrose GluconAcetobacter	30 °C, Aerobic, G−	Acetate culture medium
CICC6076	Lactobacillus plantarum	37 °C, Anaerobic, G+	Lactobacillus culture medium
CICC6224	Lactobacillus rhamnosus	37 °C, Anaerobic, G+	Lactobacillus culture medium
CICC6243	Lactobacillus acidophilus	37 °C, Anaerobic, G+	Lactobacillus culture medium
CICC6024	Lactobacillus helveticus	30 °C, Facultative anaerobic, G+	Skim milk powder medium
CICC10389	Escherichia coli	37 °C, Facultative anaerobic, G−	Nutrient broth
CICC23471	Slow Staphylococcus	37 °C, Facultative anaerobic, G+	Nutrient broth
CICC10398	Staphylococcus epidermidis	37 °C, Facultative anaerobic, G+	Nutrient broth
CICC22941	Staphylococcus saprophyticus	37 °C, Facultative anaerobic, G+	Nutrient broth
CICC23431	Staphylococcus aureus	37 °C, Facultative anaerobic, G+	Nutrient broth
CICC10373	Hemolytic Streptococcus	37 °C, Facultative anaerobic, G+	Blood agar medium
CICC22937	Campylobacter jejuni	37 °C, Micro-aerobic, G−	Campylobacter medium
CICC6223	Streptococcus thermophilus	42 °C, Facultative anaerobic, G+	MRS medium
CICC6077	Lactobacillus delbrueckii	37 °C, Facultative anaerobic, G+	MRS medium
CICC6069	Bifidobacterium infantis	37 °C, Anaerobic, G+	Bifidobacterium culture medium
CICC21633	Listeria monocytogenes	37 °C, Facultative anaerobic, G+	Brain infusion medium
CICC21617	Vibrio parahaemolyticus	30 °C, Facultative anaerobic, G−	3.5% sodium chloride broth
CMCC63301	Bacillus cereus	30 °C, Facultative anaerobic, G+	MYP medium
NCTC12900	Escherichia coli O157:H7/NM	36 °C, Facultative anaerobic, G+	Nutrient broth
ATCC14028	Salmonella typhimurium	36 °C, Facultative anaerobic, G+	Nutrient broth
CMCC50071	Salmonella enterica subsp enterica	36 °C, Facultative anaerobic, G+	Nutrient broth
CMCC49027	Proteus vulgaris	36 °C, Facultative anaerobic, G+	Nutrient broth
CMCC51592	Shigella sonnei	36 °C, Facultative anaerobic, G+	Nutrient broth
CICC10041	Bacillus cereus	30 °C, Facultative anaerobic, G+	Beef Ingredients Trace Element Medium
CICC21669	Yersinia enterocolitica	26 °C, Facultative anaerobic, G−	Nutrient broth

2.2 DNA extraction

Reagent D treatment of bacterial culture medium was performed as follows. Passage was shaken thrice in a bacterial culture medium, and 100 μL enrichment solution was added for absorption in a 2 mL sterile transparent centrifuge tube. Then, the bacteria from Reagent D were incubated and removed at -20 °C and cooled to room temperature. Liquid of 4-fold volume of Reagent D solution was added to the bacterial solution and incubated at room temperature for 5 min. The medium was placed in a low-temperature centrifuge tube and placed under a 500 W halogen lamp below 15-20 cm. After 5 min, the bacteria were centrifuged at 7400 × g for 5 min at room temperature and carefully removed the supernatant with a pipette. We added 100 μL of sterile deionized water to the centrifuge tube and mixed well for genomic DNA extraction. The extraction of bacterial genomic DNA was carried out according to the method of extracting the bacterial DNA kit by magnetic bead method using the automatic DNA extraction instrument (Thermo Company, USA).

2.3 ddPCR detection method

The factors affecting the results of ddPCR were optimized according to the previous stage, and the optimum amplification conditions of each strain were finally determined.

The total volume of the ddPCR reaction system was 20 μL, consisting of 10 μL 2 × ddPCR Super Mix, 1.2 μL upstream primer, 1.2 μL downstream primer, 0.4 μL probe, 4.4 μL template, and water. The well-mixed PCR reaction system was transferred to the droplet-generating card, and 70 μL of droplet generation oil was added to the droplet-generating card. Then, the droplet-generating card was placed in the droplet generator for reaction. The resulting droplets were subsequently transferred to 96 wells of ddPCR, and the 96-well plate was sealed with a sealing film to prepare a PCR reaction.

The ddPCR reaction procedure was performed as follows: 95 °C predenaturation for 10 min; 94 °C denaturation for 1 min; 56 °C annealing, 45 s; 40 cycles, and 98 °C for 10 min. The annealing temperature of methotrexate was 54 °C. The annealing temperature of Acetobacter spp. was 52.4 °C, and the annealing temperature of L. plantarum was 52.7 °C.

After the end of the ddPCR reaction, the 96-well plate was placed in the QX200 Droplet Reader, and the sample information was sequentially recorded. After the start of the test, the instrument automatically identified the droplets of each sample in sequence, and the droplets were detected by the two-color detection. The positive and negative results were determined based on the intensity of the fluorescence signal emitted by the droplets, and the number of positive and negative droplets per sample was recorded. After the signal acquisition is complete, the software Quantasoft was used to calculate the final result and was imaged.

2.4 Primer and probe design

In combination with the literature, the primers were selected from the sequences published in the GenBank database for the specificity and relatively conserved sequence of the strains of Gluconobacter, Gluconacetobacter, and L. plantarum.. The target genes were glucose bacteria KF896260.1, glucose Bacillus sp. HQ677466.1, and L. plantarum AJ579541.1. The primers were designed using Primer Premier 6.0 and verified by Oligo 7.37, followed by BLAST online alignment. Finally, the primers of the above strains were determined by experimentation (Table 2).

Table 2
Primers and probe sequences.

Target strain	Target gene	Primers and probe sequences (5’→3’)
Gluconobacter	ITS	Upstream primers: CAATCCGGAACAAGCTGAAAA
		Downstream primer: TTCGTGCTGCGAGTTTGAGA
		Probe: FAM-CAGTCGTACGTA-BHQI
Glucose-acetobacte	16S	Upstream primers: CAATCCGGAACAAGCTGAAAA
		Downstream primer: TTCGTGCTGCGAGTTTGAGA
		Probe: FAM-CAGTCGTACGTA-BHQI
Lactobacillus plantarum	scrB	Upstream primers: CTACTACGCGCTAATCGG
		Downstream primer: CGTTCAACTCGCCTTGACA
		Probe: FAM-ACGACTACTGAGATCGGTGCT-BHQI

2.5 Specificity and sensitivity testing

The DNAs of the strains listed in Table 1 were extracted and diluted to 103 copies. Specificity test was carried out on the optimal reaction conditions for the specific test of Gluconobacter, Glucoseacetobacte and L. plantarum.

Fermented milk purchased from the local supermarkets was not contaminated by Gluconobacter, Glucoseacetobacte and L. plantarum as confirmed by the national standard test. The three kinds of bacteria were added into the fermented milk, and the concentration of Gluconobacter, Glucoseacetobacter and L. plantarum in the samples were 10⁰ CFU/g, 10¹ CFU/g, 10² CFU/g, 10³ CFU/g, 10⁴ CFU/g, 10⁵ CFU/g, 10⁶ CFU/g, 10⁷ CFU/g and 10⁸ CFU/g. Then, the contaminants were extracted by magnetic beads, and genomic DNA of bacteria was used for the validation test for the sensitivity by using ddPCR assay method.

2.6 Absolute quantification of contaminants

From the NCBI, the theoretical copy numbers of the corresponding target gene of Gluconobacter KF896260.1, Gluconacetobacter HQ677466.1, and L. plantarum AJ579541.1 were found in the entire genome of the strain. According to the ddPCR, the number of copies, the measured sensitivity, and the deviation rate, which are all present in 20 μL of the system, can be calculated as follows:

Measured sensitivity = 20 μL copy number × 18.18 (DNA extraction volume/the volume of the template in the system)/theoretical copy number,

Deviation rate = (measured sensitivity − count sensitivity)/count sensitivity × 100%.

To verify the feasibility of ddPCR detection method, ddPCR was detected by different concentrations of Gluconobacter, Gluconacetobacter, and L. plantarum,, and the linear relationship between the counting results and the ddPCR test results was analyzed.

2.7 Actual sample test

53 different fermented dairy products were selected from the market. Three fermented milk samples were randomly selected from 53 kinds of fermented milk to add three target strains with the concentration of 10² CFU/mL as the experimental group and the other fermented milk as the control group.

3 Results

3.1 Specificity of PCR method

The genomic DNA of the test strains listed in Table 1 was amplified by PCR with the amplification primers and probes of Lactobacillus Gluconacetobacter and L. plantarum, followed by detection with a droplet detector. Figure 1 shows that the three strains of glucose bacteria (a), two glucose acids (b), and L. plantarum (c) were able to detect positive droplet signals, whereas negative and other nonSalmonella (d) were not amplified, indicating that the study of the establishment of contamination bacterial detection method for the three kinds of bacteria to determine their specificity is good.

Figure 1
Three kinds of contaminant bacteria determined by ddPCR method and the specific test results. (a) Glucose bacteria, (b) Gluconacetobacter, (c) Lactobacillus plantarum, (d) negative and other nonSalmonella. (a) 1: Glucose bacteria (CGMCC1.0637); 2: Pseudomonas aeruginosa (CGMCC1.3748); 3: Glucose bacteria (CICC10357); 4: Acetobacter aceti (CICC21682); 5: Acetobacter aceti (CICC22519); 6: Acetobacter aceti (CICC21683); 7: Acetobacter liquefaction (CGMCC1.24); 8: Acetobacter spp. (CGMCC1.811); 9: sucrose GluconAcetobacter (CICC21102); 10: Salmonella enteritidis (CICC 21482); 11: Kentucky brucei (CICC 21560); 12: Shigella flexneri (CICC 21534); 13: Staphylococcus aureus (CICC10384); 14: Bifidobacterium infantis (CICC6069); 15: Listeria monocytogenes (CICC21633); 16: Escherichia coli (CICC 10389); 17: Bacillus cereus (CICC10041); 18: Streptococcus thermophilus (CICC6223). (b) 1: Acetobacter liquefaction (CGMCC1.24); 2: Acetobacter spp. (CGMCC1.811); 3: Glucose bacteria (CICC10357); 4: Acetobacter aceti (CICC21682); 5: Salmonella enteritidis (CICC 21482); 6: Kentucky brucei (CICC 21560); 7: Shigella flexneri (CICC 21534); 8: Staphylococcus aureus (CICC10384); 9: Bifidobacterium infantis (CICC6069); 10: Listeria monocytogenes (CICC21633); 11: Escherichia coli (CICC 10389); 12: Bacillus cereus (CICC10041); 13: Streptococcus thermophilus (CICC6223); 14: Lactobacillus plantarum (CICC6076); 15: Campylobacter jejuni (CICC22937); 16: Lactobacillus delbrueckii (CICC6077); 17: Lactobacillus helveticus (CICC6024). (c) 1: Lactobacillus plantarum (CICC6076); 2: Lactobacillus rhamnosus (CICC6224); 3: Lactobacillus acidophilus (CICC6243); 4: Acetobacter spp. (CGMCC1.811); 5: Lactobacillus delbrueckii (CICC6077); 6: Streptococcus thermophilus (CICC6223); 7: Salmonella enteritidis (CICC 21482); 8: Kentucky brucei (CICC 21560); 9: Shigella flexneri (CICC 21534); 10: Escherichia coli (CICC 10389); 11: Bacillus cereus (CICC10041); 12: Staphylococcus aureus (CICC10384); 13: Yersinia enterocolitica (CICC 21669); 14: Bifidobacterium infantis (CICC 6069); 15: Acetobacter aceti (CICC22519).

3.2 Sensitivity of PCR method

The results showed that the separation of negative and positive droplets occurred from the third reaction, and the results of ddPCR method showed that the concentration of the negative and positive droplets increased from the third reaction to the 8th reaction time. The number of positive microbes that can be detected is very small when the template concentration is 8.8 × 10⁰ CFU/g. The number of copies reported by the droplet analyzer is 0.12 copies/μL. Thus, ddPCR method detected the sensitivity of the bacteria at 8.8 × 10⁰ CFU/g (Figure 2a). The sensitivity of ddPCR detection method showed that when the dilution of the original template was 8.9 × 10⁰ CFU/g, the droplet detector was determined to be negative. The droplet analyzer reported the copy number of 0.20 copies/μL. Thus, ddPCR method was used to detect the glucose-acetate bacterial count sensitivity of 8.9 × 10¹ CFU/g (Figure 2b). The plate count of L. plantarum was 9.6 × 10⁷ CFU/g, and the sensitivity of the ddPCR assay showed that it can be detected in the 8th reaction when the concentration of bacteria was 9.6 × 10¹ CFU/g. The number of positive microbes was very small, and the copy number was 0.63 copies/μL. Therefore, ddPCR method was used to detect the sensitivity of L. plantarum of 9.6 × 10¹ CFU/g (Figure 2c).

Figure 2
The results of ddPCR method to detect the sensitivity of three contaminants. (a) Gluconobacter: 8.8 × 10⁷, 8.8 × 10⁶, 8.8 × 10⁵, 8.8 × 10⁴, 8.8 × 10³, 8.8 × 10², 8.8 × 10¹, 8.8 × 10⁰, 8.8 × 10⁻¹; (b) Acetobacter: 8.9 × 10⁷, 8.9 × 10⁶, 8.9 × 10⁵, 8.9 × 10⁴, 8.9 × 10³, 8.9 × 10², 8.9 × 10¹, 8: 8.9 × 10⁰. (c) Lactobacillus plantarum: 9.6 × 10⁷, 9.6 × 10⁶, 9.6 × 10⁵, 9.6 × 10⁴, 9.6 × 10³, 9.6 × 10², 7: 9.6 × 10¹.

3.3 Sensitivity of PCR method

The measured sensitivity was calculated according to the theoretical copy number of the bacteria and the actual test results of the genus Lactobacillus acidophilus and L. plantarum. When the measured sensitivity was greater than the counting sensitivity, the deviation rate displayed positive deviation. When the measured sensitivity was less than the counting sensitivity, the deviation rate showed negative deviation. The results showed that the sensitivity of the other two strains was slightly higher than the counting sensitivity (Table 3), and the deviation rate of ddPCR was lower than that of the counting sensitivity of 25% (Table 4). The absolute quantification of ddPCR is feasible in practical applications.

Table 3
Relation of copy number and sensitivity.

Name of strain	Theoretical copy number (copies)	Measured copy number (copies/μL)	20 μL copy number (copies)	Measured sensitivity (CFU/mL)
Glucose bacteria	4	0.12	2.4	10.91
Acetobacter spp	1	0.20	4	72.72
Lactobacillus plantarum	2	0.63	12.6	114.53

Table 4
Deviation rate of sensitivity.

Name of strain	Measured sensitivity (CFU/mL)	Count sensitivity (CFU/mL)	Deviation rate (%)
Glucose bacteria	10.91	8.80	23.95
Acetobacter spp	72.72	89.00	−18.29
Lactobacillus plantarum	114.53	96.00	19.31

In this study, ddPCR was detected by different concentrations of Streptomyces griseus, Acetobacter, and L. plantarum. The linear relationship between the counting results and the ddPCR test results was analyzed. As shown in Figure 3, the standard curve generated by the correspondence between the copy number and the number of colonies of the three strains showed that R² was > 0.99, which are 0.9933, 0.9998, 0.9954 respectively, indicating that the linearity of the sensitivity of the ddPCR to the bacteria in yogurt was good. This finding further proved the effectiveness of the method in detecting bacteria.

Figure 3
Linear relationship between detection sensitivity of Glucose bacteria (a), Gluconacetobacter (b) and Lactobacillus plantarum (c) by using ddPCR method.

3.4 Actual sample test

The actual samples were amplified by PCR with amplified primers and probes of Lactobacillus gluconate and Lactobacillus plantarum, and then detected by drop detector. Figure 4 shows three kinds of fermented milk added with the target strain, all of which show positive results. The remaining samples without target strains showed negative results. The results showed that the food matrix had no significant effect on the detection method of contaminated bacteria established in this study. And the detection method can be expanded to more fermented products for further verification, so as to expand the application scope of the method. Now some companies have begun to look for other non-dairy fermentation alternatives, such as development of vegetal probiotic beverage of passion fruit (Passiflora edulis Sims), yam (Dioscorea cayenensis) and Lacticaseibacillus casei, which can not only promote the survival of probiotics (Guedes et al., 2021), but also reduce the consumption of dairy raw materials. microbial contamination issues.

Figure 4
Test results of actual fermented milk with target strain (a) and without target strain (b) by ddPCR method.

According to the theoretical copy number of bacteria and the actual detection results of Lactobacillus acidophilus and Lactobacillus plantarum, the concentration of target bacteria was calculated and measured. When the measured concentration is greater than the added concentration, the deviation rate shows a positive deviation. When the measured concentration is less than the added concentration, the deviation rate is negative. The results (Table 5 & Table 6) show that the deviation of target bacteria concentration is no more than 25% in actual sample detection, which indicates that the method is feasible in practical application. Obviously, it is not just polluting bacteria. In the future, we can choose fermented bacteria that have a positive impact on sensory characteristics of fermented milk, and establish their detection methods, which can more accurately ensure the sensory quality of fermented milk and attract consumers.

Table 5
Relation of copy number and sensitivity of actual samples.

Name of strain	Theoretical copy number (copies)	Measured copy number (copies/μL)	20 μL copy number (copies)	Measured sensitivity (CFU/mL)
Glucose bacteria	4	1.55	31	140.90
Acetobacter spp	1	0.28	5.6	101.81
Lactobacillus plantarum	2	0.80	16	145.44

Table 6
Deviation rate of sensitivity of actual samples.

Name of strain	Measured sensitivity (CFU/mL)	Count sensitivity (CFU/mL)	Deviation rate (%)
Glucose bacteria	140.90	120.00	17.42
Acetobacter spp	101.81	120.00	−15.16
Lactobacillus plantarum	145.44	120.00	21.2

4 Conclusions

Based on the specific primers of Lactobacillus, Gluconobacter, Bacteroides, and L. plantarum, primers were used to detect the bacteria and pathogens in fermented milk by the droplet digital polymerase chain reaction, and the genomic deoxyribonucleic acid was extracted by Reagent D. The optimal reaction conditions were optimized, and the droplet digital polymerase chain reaction method for the quantitative detection of three kinds of bacteria and pathogens in fermented milk was established.

The droplet digital polymerase chain reaction method is useful for the detection of bacteria in fermented milk. The detection sensitivity of the method was determined after artificial contamination of the fermented milk. The detection sensitivity of Bacillus subtilis was 89 CFU/g, whereas the sensitivity of L. plantarum was 96 CFU/g.

The difference between the number of copies and the number of counted colonies of Lactobacillus, Acetobacter, and L. plantarum were within 25%, and 10⁴ CFU/g displayed a good linear relationship with the conventional counting results (the standard curve R² was greater than 0.99), proving absolute quantitative accuracy of the droplet digital polymerase chain reaction detection in yogurt caused by bacterial contamination and pathogens.

Therefore, this study established an effective and feasible the droplet digital polymerase chain reaction technology system for the quantitative detection of bacteria in fermented milk and provided the theoretical basis and technical guarantee for the wide application of droplet digital polymerase chain reaction in food safety detection.

• Practical Application: The research contributes to the establishment of a standardized ddPCR method for the detection of target bacterial contaminants in fermented milk. The establishment of the ddPCR method will help to improve the identification level of contaminants in fermented milk, standardize the operation of food safety market supervision.

• Funding

  This research was funded by Science and Technology Plan Project of the State Administration for Market Regulation, (2019MK008).

References

• Barichella, M., Pacchetti, C., Bolliri, C., Cassani, E., Iorio, L., Pusani, C., Pinelli, G., Privitera, G., Cesari, I., Faierman, S. A., Caccialanza, R., Pezzoli, G., & Cereda, E. (2016). Probiotics and prebiotic fiber for constipation associated with Parkinson disease: an RCT. Neurology, 87(12), 1274-1280. http://dx.doi.org/10.1212/WNL.0000000000003127 PMid:27543643.
  » http://dx.doi.org/10.1212/WNL.0000000000003127
• Bartowsky, E. J., & Henschke, P. A. (2008). Acetic acid bacteria spoilage of bottled red wine—a review. International Journal of Food Microbiology, 125(1), 60-70. http://dx.doi.org/10.1016/j.ijfoodmicro.2007.10.016 PMid:18237809.
  » http://dx.doi.org/10.1016/j.ijfoodmicro.2007.10.016
• Bi, Y., Choi, S., & Kim, I. (2020). Visitors’ motives for attending a healthy food exhibition. International Journal of Environmental Research and Public Health, 17(8), 2703. http://dx.doi.org/10.3390/ijerph17082703 PMid:32326398.
  » http://dx.doi.org/10.3390/ijerph17082703
• Cruz, M. F., Rocha, R. S., Silva, R., Freitas, M. Q., Pimentel, T. C., Esmerino, E. A., Cruz, A. G., Fidalgo, T. K. S., & Maia, L. C. (2021). Probiotic fermented milks: children’s emotional responses using a product-specific emoji list. Food Research International, 143, 110269. http://dx.doi.org/10.1016/j.foodres.2021.110269 PMid:33992370.
  » http://dx.doi.org/10.1016/j.foodres.2021.110269
• Dube, S., Qin, J., & Ramakrishnan, R. (2008). Mathematical analysis of copy number variation in a DNA sample using digital PCR on a nanofluidic device. PLoS One, 3(8), e2876. http://dx.doi.org/10.1371/journal.pone.0002876 PMid:18682853.
  » http://dx.doi.org/10.1371/journal.pone.0002876
• Eor, J. Y., Tan, P. L., Son, Y. J., Lee, C. S., & Kim, S. H. (2020). Milk products fermented by Lactobacillus strains modulate the gut–bone axis in an ovariectomised murine model. International Journal of Dairy Technology, 73(4), 743-756. http://dx.doi.org/10.1111/1471-0307.12708
  » http://dx.doi.org/10.1111/1471-0307.12708
• Falenski, A., Mayer-Scholl, A., Filter, M., Göllner, C., Appel, B., & Nöckler, K. (2011). Survival of Brucella spp. in mineral water, milk and yogurt. International Journal of Food Microbiology, 145(1), 326-330. http://dx.doi.org/10.1016/j.ijfoodmicro.2010.11.033 PMid:21167618.
  » http://dx.doi.org/10.1016/j.ijfoodmicro.2010.11.033
• Farnworth, E. R., Mainville, I., Desjardins, M. P., Gardner, N., Fliss, I., & Champagne, C. (2007). Growth of probiotic bacteria and bifidobacteria in a soy yogurt formulation. International Journal of Food Microbiology, 116(1), 174-181. http://dx.doi.org/10.1016/j.ijfoodmicro.2006.12.015 PMid:17292991.
  » http://dx.doi.org/10.1016/j.ijfoodmicro.2006.12.015
• Foschino, R., Garzaroli, C., & Ottogalli, G. (1993). Microbial contaminants cause swelling and inward collapse of yoghurt packs. Le Lait, 73(4), 395-400. http://dx.doi.org/10.1051/lait:1993437
  » http://dx.doi.org/10.1051/lait:1993437
• Galdeano, C. M., Núñez, I. N., LeBlanc, A. M., Carmuega, E., Weill, R., & Perdigón, G. (2011). Impact of a probiotic fermented milk in the gut ecosystem and in the systemic immunity using a non-severe protein-energy-malnutrition model in mice. BMC Gastroenterology, 11(1), 64. http://dx.doi.org/10.1186/1471-230X-11-64 PMid:21615956.
  » http://dx.doi.org/10.1186/1471-230X-11-64
• Gibson, U. E., Heid, C. A., & Williams, P. M. (1996). A novel method for real time quantitative RT-PCR. Genome Research, 6(10), 995-1001. http://dx.doi.org/10.1101/gr.6.10.995 PMid:8908519.
  » http://dx.doi.org/10.1101/gr.6.10.995
• González, S., Fernández-Navarro, T., Arboleya, S., Reyes-Gavilán, C. G., Salazar, N., & Gueimonde, M. (2019). Fermented dairy foods: impact on intestinal microbiota and health-linked biomarkers. Frontiers in Microbiology, 10, 1046. http://dx.doi.org/10.3389/fmicb.2019.01046 PMid:31191465.
  » http://dx.doi.org/10.3389/fmicb.2019.01046
• Goodfellow, M., Kämpfer, P., Busse, H.-J., Trujillo, M. E., Suzuki, K.-I., Ludwig, W., & Whitman, W. B. (2012). Bergey’s manual® of systematic bacteriology (Vol. 5). New York: Springer http://dx.doi.org/10.1007/978-0-387-68233-4
  » http://dx.doi.org/10.1007/978-0-387-68233-4
• Guedes, C. K. R. M., Guedes, A. F. L. M., Silva, J. R., Silva, E. B. B., Santos, E. C. M., Stamford, T. C. M., & Stamford, T. L. M. (2021). Development of vegetal probiotic beverage of passion fruit (Passiflora edulis Sims), yam (Dioscorea cayenensis) and Lacticaseibacillus casei. Food Science and Technology, 41(Suppl. 2), 619-626. http://dx.doi.org/10.1590/fst.66120
  » http://dx.doi.org/10.1590/fst.66120
• Heid, C. A., Stevens, J. K., Livak, K. J., & Williams, P. M. (1996). Real time quantitative PCR. Genome Research, 6(10), 986-994. http://dx.doi.org/10.1101/gr.6.10.986 PMid:8908518.
  » http://dx.doi.org/10.1101/gr.6.10.986
• Klemm, P., Dischereit, G., & Lange, U. (2020). Adult lactose intolerance, calcium intake, bone metabolism and bone density in German-Turkish immigrants. Journal of Bone and Mineral Metabolism, 38(3), 378-384. http://dx.doi.org/10.1007/s00774-019-01070-4 PMid:31802224.
  » http://dx.doi.org/10.1007/s00774-019-01070-4
• Lee, C. S., Lee, S. H., & Kim, S. H. (2020). Bone‐protective effects of Lactobacillus plantarum B719‐fermented milk product. International Journal of Dairy Technology, 73(4), 706-717. http://dx.doi.org/10.1111/1471-0307.12701
  » http://dx.doi.org/10.1111/1471-0307.12701
• Margulies, M., Egholm, M., Altman, W. E., Attiya, S., Bader, J. S., Bemben, L. A., Berka, J., Braverman, M. S., Chen, Y.-J., Chen, Z., Dewell, S. B., Du, L., Fierro, J. M., Gomes, X. V., Godwin, B. C., He, W., Helgesen, S., Ho, C. H., Irzyk, G. P., Jando, S. C., Alenquer, M. L. I., Jarvie, T. P., Jirage, K. B., Kim, J.-B., Knight, J. R., Lanza, J. R., Leamon, J. H., Lefkowitz, S. M., Lei, M., Li, J., Lohman, K. L., Lu, H., Makhijani, V. B., McDade, K. E., McKenna, M. P., Myers, E. W., Nickerson, E., Nobile, J. R., Plant, R., Puc, B. P., Ronan, M. T., Roth, G. T., Sarkis, G. J., Simons, J. F., Simpson, J. W., Srinivasan, M., Tartaro, K. R., Tomasz, A., Vogt, K. A., Volkmer, G. A., Wang, S. H., Wang, Y., Weiner, M. P., Yu, P., Begley, R. F., & Rothberg, J. M. (2005). Genome sequencing in microfabricated high-density picolitre reactors. Nature, 437(7057), 376-380. https://doi.org/10.3410/f.1027324.334968. PMid:16056220.
• Moghadam, M. A. M., Anvar, S. A., Amini, K., & Khani, M. (2021). The effect of Lactococcus lactis and Bifidobacterium bifidum probiotics cell free supernatants on the expression of HDC and TDC genes in Staphylococcus strains isolated from milk. Food Science and Technology, 1-7. Ahead of print. http://dx.doi.org/10.1590/fst.03221
  » http://dx.doi.org/10.1590/fst.03221
• Muramalla, T., & Aryana, K. J. (2011). Some low homogenization pressures improve certain probiotic characteristics of yogurt culture bacteria and Lactobacillus acidophilus LA-K. Journal of Dairy Science, 94(8), 3725-3738. http://dx.doi.org/10.3168/jds.2010-3737 PMid:21787909.
  » http://dx.doi.org/10.3168/jds.2010-3737
• Nwamaka, N. T., & Chike, A. E. (2010). Bacteria population of some commercially prepared yoghurt sold in Enugu state, eastern Nigeria. African Journal of Microbiological Research, 4(10), 984-988.
• Ouoba, L. I. I., Kando, C., Parkouda, C., Sawadogo‐Lingani, H., Diawara, B., & Sutherland, J. P. (2012). The microbiology of bandji, palm wine of Borassus akeassii from Burkina Faso: identification and genotypic diversity of yeasts, lactic acid and acetic acid bacteria. Journal of Applied Microbiology, 113(6), 1428-1441. http://dx.doi.org/10.1111/jam.12014 PMid:22979949.
  » http://dx.doi.org/10.1111/jam.12014
• Pastink, M. I., Sieuwerts, S., Bok, F. A. M., Janssen, P. W. M., Teusink, B., van Hylckama Vlieg, J. E. T., & Hugenholtz, J. (2008). Genomics and high-throughput screening approaches for optimal flavour production in dairy fermentation. International Dairy Journal, 18(8), 781-789. http://dx.doi.org/10.1016/j.idairyj.2007.07.006
  » http://dx.doi.org/10.1016/j.idairyj.2007.07.006
• Poimenidou, S., Belessi, C. A., Giaouris, E. D., Gounadaki, A. S., Nychas, G.-J. E., & Skandamis, P. N. (2009). Listeria monocytogenes attachment to and detachment from stainless steel surfaces in a simulated dairy processing environment. Applied and Environmental Microbiology, 75(22), 7182-7188. http://dx.doi.org/10.1128/AEM.01359-09 PMid:19767476.
  » http://dx.doi.org/10.1128/AEM.01359-09
• Raspor, P., & Goranovič, D. (2008). Biotechnological applications of acetic acid bacteria. Critical Reviews in Biotechnology, 28(2), 101-124. http://dx.doi.org/10.1080/07388550802046749 PMid:18568850.
  » http://dx.doi.org/10.1080/07388550802046749
• Sanders, R., Huggett, J. F., Bushell, C. A., Cowen, S., Scott, D. J., & Foy, C. A. (2011). Evaluation of digital PCR for absolute DNA quantification. Analytical Chemistry, 83(17), 6474-6484. http://dx.doi.org/10.1021/ac103230c PMid:21446772.
  » http://dx.doi.org/10.1021/ac103230c
• Saygili, D., Döner, D., İçİer, F., & Karagözlü, C. (2021). Rheological properties and microbiological characteristics of kefir produced from different milk types. Food Science and Technology, 1-6. Ahead of print. http://dx.doi.org/10.1590/fst.32520.
• Shah, N. P. (2000). Probiotic bacteria: selective enumeration and survival in dairy foods. Journal of Dairy Science, 83(4), 894-907. http://dx.doi.org/10.3168/jds.S0022-0302(00)74953-8 PMid:10791807.
  » http://dx.doi.org/10.3168/jds.S0022-0302(00)74953-8
• Vogelstein, B., & Kinzler, K. W. (1999). Digital PCR. Proceedings of the National Academy of Sciences of the United States of America, 96(16), 9236-9241. http://dx.doi.org/10.1073/pnas.96.16.9236 PMid:10430926.
  » http://dx.doi.org/10.1073/pnas.96.16.9236
• Zhang, K., Dai, H., Liang, W., Zhang, L., & Deng, Z. (2019). Fermented dairy foods intake and risk of cancer. International Journal of Cancer, 144(9), 2099-2108. http://dx.doi.org/10.1002/ijc.31959 PMid:30374967.
  » http://dx.doi.org/10.1002/ijc.31959
• Zhou, W., Zhang, Y., Wang, S., Li, Y., Zhang, J., Zhang, C., Wang, Z., & Zhang, Z. (2017). LAMP, PCR, and real-time PCR detection of Acetobacter aceti in yogurt. Food Science and Biotechnology, 26(1), 153-158. http://dx.doi.org/10.1007/s10068-017-0020-7 PMid:30263522.
  » http://dx.doi.org/10.1007/s10068-017-0020-7
• Zipper, H., Brunner, H., Bernhagen, J., & Vitzthum, F. (2004). Investigations on DNA intercalation and surface binding by SYBR Green I, its structure determination and methodological implications. Nucleic Acids Research, 32(12), e103. http://dx.doi.org/10.1093/nar/gnh101 PMid:15249599.
  » http://dx.doi.org/10.1093/nar/gnh101

Publication Dates

• Publication in this collection
  06 May 2022
• Date of issue
  2022

History

• Received
  03 Nov 2021
• Accepted
  05 Feb 2022