Dairy Spoilage Bacteria and Their Lipases: Prevalence Patterns and Molecular Traits

Abstract

Dairy product spoilage is primarily caused by microbial contamination and lipase activity. Systematically assessing contamination risks and elucidating the structural and functional characteristics of key lipases form a crucial foundation for achieving precise prevention and control. This article integrated 79 studies (2003–2025) to preliminarily explore the geographical distribution characteristics of spoilage bacterial contamination. The combined contamination rates were relatively high in South America (89.5%), North America (88.3%) and Asia (87.3%), followed by Europe (86.0%) and Oceania (82.2%), while Africa showed a lower rate (57.6%) based on limited data. Product-level risk differences were also identified. Considering data robustness, in addition to raw milk (92.6%), milk powder also showed a high contamination rate (86.8%). The spoilage bacterial community was highly concentrated, with Bacillus (40.7%) and Pseudomonas (20.3%) as the dominant genera, and Pseudomonas fluorescens (21.1%) and Bacillus licheniformis (20.1%) as the core spoilage species. Furthermore, structural analysis of 37 key lipases demonstrated that these enzymes universally possess a hydrophobic surface, high lipid affinity, and a conserved α/β hydrolase folding conformation. Based on these predictions, we propose a structural hypothesis: these characteristics may be related to their retained activity after thermal processing. Unlike traditional meta-analyses that merely report pollution rates, and unlike independent bioinformatics studies that analyze lipases from arbitrary strains, this study ensures that the lipases analyzed have epidemiological relevance. This work provides scientific grounds for monitoring, early warning, and risk assessment of dairy spoilage. It also lays a theoretical foundation for developing novel inhibitors targeting thermostable lipases.

1. Introduction

Dairy products are rich in protein, fat, carbohydrates, various minerals, and vitamins. They are globally recognized as an important source of nutrition and play a key role in promoting human growth and development, maintaining bone health, and supporting basal metabolism [1]. With global economic development and heightened public health awareness, dairy consumption continues to rise, reaching approximately 840 million tonnes worldwide in 2022 [2,3]. However, rich nutritional content and favourable water activity of dairy products also render them an ideal substrate for microbial growth and enzymatic reactions. This makes them highly susceptible to spoilage during processing, storage, and distribution, leading to significant economic losses and food safety risks [4]. According to statistics, in the United States alone, economic losses from microbial spoilage of dairy products at the retail and consumer levels reach approximately $165.6 billion, representing about 17% of total food sector losses [5]. Furthermore, food safety incidents triggered by spoilage microorganisms, such as the 2010 “blue cheese” outbreak in Italy, inflict long-term damage upon industry reputation and consumer confidence [6].

Spoilage microorganisms are the primary biological factors responsible for the deterioration of dairy product quality [7]. Among these, spoilage bacteria are particularly significant in terms of both contamination rates and severity of harm. These bacteria and the enzymes they secrete (such as proteases and lipases) can directly cause deterioration in the color, odor and texture of dairy products [8]. For instance, Pseudomonas fluorescens, a typical psychrophile, can grow abundantly and secrete lipases even at refrigeration temperatures (4–7 °C) [9]. Lipases, also known as glycerol ester hydrolases, catalyse the hydrolysis of triacylglycerols, thereby promoting the release of free fatty acids and glycerol [10]. Such enzymes retain its activity after pasteurization or even ultra-high temperature sterilization (UHT). It continuously catalyzes the hydrolysis of milk fat during storage, releasing short-chain fatty acids such as acetic acid and caproic acid. Although acetic acid is an important source of yogurt flavor, caproic acid can cause irreversible sensory deterioration such as acidification and off-odors during storage, leading to the “post-sterilization spoilage” phenomenon of dairy products, which severely restricts the shelf life and commercial value of dairy products [11]. Therefore, systematically understanding the contamination patterns of dairy spoilage bacteria and elucidating the molecular characteristics of their lipases at the molecular level form the foundation for achieving precise prevention and control of dairy spoilage.

Although numerous studies have investigated the contamination of spoilage bacteria in dairy products from various perspectives, these investigations exhibit certain limitations in both scope and depth. From the perspective of research, the existing works mostly focus on a single bacterial strain or product type. For instance, Marcela et al. focused on the Pseudomonas genus in Brazilian and Chinese milk [12]; Roberto et al. investigated the Bacillus cereus in the cheese processing environment [13]. Such isolated studies have resulted in highly fragmented knowledge in this field. Currently, there is a lack of meta-analyses covering the prevalence patterns and dominant bacterial communities of spoilage bacteria in dairy products from multiple regions and different categories. In terms of research depth, most studies remain confined to phenotypic descriptions such as strains identification and contamination rate statistics [8,14], with a notable absence of systematic investigations into the molecular characteristics of lipases produced by these key spoilage bacteria. This knowledge gap at the molecular mechanism level severely constrains the development of precise prevention and control strategies based on enzyme structure and function.

Therefore, this study follows a two-step progressive approach. Firstly, we employed a meta-analysis to integrate research on spoilage bacterial contamination in dairy products from 2003 to 2025. This assessed the overall prevalence of spoilage bacteria in dairy products, revealed geographical distribution and product type variations, and identified the predominant spoilage bacterial strains present. Building directly on these meta-analysis results, we then conducted systematic bioinformatics analysis specifically on the lipases secreted by the top eleven identified spoilage bacteria. This design ensures that lipase we analysis are epidemiologically relevant rather than arbitrarily selected. Our analysis encompassed conserved domains, physicochemical properties, and three-dimensional structures, aiming to elucidate the molecular-level structural and functional commonalities of these lipases. By establishing this causal chain from contamination patterns to molecular traits, this research provides a reliable theoretical foundation and a solid data basis for monitoring and controlling spoilage risks in dairy products.

2. Materials and Methods

2.1. Search Strategy and Literature Screening Process

This study strictly adheres to the standard systematic process established by the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines (http://www.prisma-statement.org/ for details (accessed on 2 May 2026)) [15]. The study followed the operational steps outlined in Figure 1: (a) database searching and categorization of potentially relevant articles; (b) assessment of article relevance; (c) quality evaluation; (d) data extraction. This paper systematically screened article concerning the contamination of dairy products by spoilage bacteria. The databases searched included PubMed, Web of Science, and China National Knowledge Infrastructure (CNKI), covering the time period from 1 January 2003 to 31 December 2025. Chinese search terms include “spoilage bacteria”, “dairy products”, “milk”, “yogurt”, “cheese”, “butter”, “milk powder”, “cream”, “condensed milk”, “ice cream”, and “milk beverages”; English search terms include “spoilage bacteria”, “dairy products”, “milk”, “cheese”, and “yogurt”. For detailed search strategies, please refer to Table A1.

Figure 1.

Flow diagram according to the PRISMA method.

The initial search yielded a total of 3751 articles, including 232 from the CNKI, 840 from the PubMed, and 2679 from the Web of Science. An additional 18 articles were obtained by searching reference lists. After deduplication, 2885 articles remained. Based on title and abstract screening, 2312 articles were excluded as irrelevant to the study theme. Full-text retrieval and re-screening were conducted on the remaining 573 articles. Inclusion criteria were: (1) studies investigating the prevalence of spoilage bacteria in dairy products; (2) provision of extractable sample sizes and positive detection data; (3) original research articles (excluding reviews, conference abstracts, etc.). Ultimately, 79 articles were included for subsequent analysis.

2.2. Data Extraction and Quality Assessment

The following information was extracted from each included article: title, authors, country/region of study, year of publication, type of dairy product, sample size, number of positive samples for spoilage bacteria, spoilage bacterial strains, and detection methods. The Newcastle-Ottawa Scale (NOS) (https://www.ohri.ca/programs/clinical_epidemiology/oxford.asp (accessed on 2 May 2026)) was used to assess the quality of the included articles. The scale evaluates studies across three dimensions (sample selection, comparability, and outcome assessment) and consists of nine items (Table A2). Based on the number of items, the article was classified into three quality levels: studies meeting ≥ 7 items were considered high quality, those meeting 4–6 items were of moderate quality, and those meeting ≤3 items were classified as low quality.

2.3. Data Analysis

Adhering to the fundamental principles and standardized procedures of meta-analysis, this study developed a systematic data extraction form to collect information from the included articles. All statistical analyses were performed using Microsoft Excel. In this study, Cochran’s Q test, the I² statistic, and 95% confidence intervals were used to assess heterogeneity. Generally, p ≥ 0.1 in Cochran’s Q test indicates no heterogeneity, while p < 0.1 indicates the presence of heterogeneity among studies. The I² statistic measures the percentage of total variation in effect sizes that is attributable to true heterogeneity between studies (rather than chance factors). Conventionally, I² values of 25%, 50%, and 75% are used as thresholds for low, moderate, and high heterogeneity, respectively. An I² greater than 50% is considered to indicate substantial heterogeneity, and an I² greater than 75% is considered to indicate considerable heterogeneity. Since this is a meta-analysis of single-group rates with no control group and therefore no positive or negative outcomes, the results are largely descriptive. Accordingly, a random-effects model was employed for all meta-analyses [16]. Comprehensive Meta-Analysis 3.0 (CMA 3.0) software was employed to calculate the pooled effect size, 95% confidence interval, p value, and Q value.

2.4. Publication Bias Test

Publication bias refers to the phenomenon where research results with statistical significance are more likely to be reported and published compared to those with non-significant or null findings. It constitutes one of the primary factors affecting the validity of meta-analysis outcomes [17]. In meta-analyses, funnel plots are frequently employed to visually assess publication bias. However, funnel plot asymmetry is not solely attributable to publication bias; heterogeneity is also recognized as a significant factor contributing to such asymmetry. When studies exhibit high heterogeneity (I² > 75%), relying solely on funnel plots to assess the applicability of publication bias is of limited value. Therefore, to mitigate the subjectivity inherent in graphical interpretation, this study employs Fail-safe N analysis (N_fs) [18] combined with Egger’s regression test [19] for publication bias testing, thereby enhancing the objectivity and accuracy of the results.

• (1)Fail-safe N analysis

Fail-safe N analysis aims to assess the number of unpublished studies required to nullify the statistical significance of the current meta-analysis results, thereby evaluating the robustness of the findings. The N_fs values are calculated at significance levels of α = 0.05 and α = 0.01, using the following formula:

$${{\mathrm{N}}_{fs}}_{0.05}={\left(\sum {\displaystyle \frac{z}{1.64}}\right)}^2-K$$ (1)

$${{\mathrm{N}}_{fs}}_{0.01}={\left(\sum {\displaystyle \frac{z}{2.33}}\right)}^2-K$$ (2)

Among these, z denotes the z-score for each individual study and K represents the number of included articles. Should N_fs exceed the critical threshold (5K + 10), the meta-analysis findings are considered more reliable with reduced publication bias influence; conversely, insufficient stability indicates potential bias risks in the conclusions.

• (2)Egger’s regression test

Egger’s regression test identifies publication bias by examining the symmetry of funnel plots. This method establishes a linear regression model with precision as the independent variable and Standard Normal Deviate (SND) as the dependent variable:

$$SND=A+B\times Precision$$ (3)

Among these, A denotes the intercept, reflecting the asymmetry of the funnel plot; B represents the slope, indicating the magnitude of the effect size. Should the p value for the intercept term in the regression equation be less than 0.05, this indicates the presence of significant publication bias; conversely, a higher p value suggests a lower risk of bias. Both tests may be conducted using the Comprehensive Meta-Analysis statistical software.

2.5. Screening and Physicochemical Characterization of Lipase from Spoilage Bacteria

To obtain the sequences and physicochemical information of lipases from predominant spoilage bacteria in dairy products, we first searched the NCBI Protein Database using the names of predominant spoilage bacterial strains (identified through meta-analysis) in combination with the keyword “Lipase” to retrieve amino acid sequences. Subsequently, the ProtParam within the ExPASy online platform (http://web.expasy.org/protparam/ (accessed on 2 May 2026)) was employed to analyze the physicochemical properties of the obtained lipase sequences. This yielded key parameters including relative molecular weight, theoretical isoelectric point (pI), aliphatic index, and average hydrophilicity (GRAVY), thereby establishing a data foundation for subsequent structural and functional investigations.

2.6. Domain and Functional Analysis of Lipase from Spoilage Bacteria

Based on the lipase amino acid sequence obtained from NCBI, three-dimensional structure prediction was performed using the SWISS-MODEL online homology modelling platform (https://swissmodel.expasy.org/ (accessed on 2 May 2026)). This platform automatically aligns the target sequence against a database of known protein structures, selecting the template with the highest similarity to construct a three-dimensional structural model of lipase. To facilitate subsequent structural analysis, the resulting model was visualized using the PyMOL 3.1 molecular graphics software.

3. Results

3.1. Characteristics and Quality Assessment of Included Studies

A total of 79 articles that were ultimately included in the analysis (Table A3) are summarized in terms of their basic characteristics in Table 1. These articles spanned 27 countries across six continents. Based on the processing technology and product form, the dairy products included in this study were classified into five categories: (1) Liquid milk, including raw milk, pasteurized milk, ultra-high temperature milk (UHT milk), refrigerated milk, extended shelf life milk (ESL milk), bactofuged milk, and yogurt; (2) Milk powder, including milk powder and dried colostrum; (3) Milk fat products, including cream and butter; (4) Ice cream; and (5) Cheeses. According to the assessment using the Newcastle-Ottawa Scale (NOS) (Table A2), the overall methodological quality of the included studies was relatively high, with over 90% of the studies receiving good scores in terms of sample selection and quality reporting. However, 6.33% of the studies had deficiencies in reporting sample size, suggesting that future research should further improve the completeness and transparency of methodological descriptions.

Table 1.

Characteristics of included studies.

Data Features		Number of Articles
Study region	North America	12
	Oceania	6
	Africa	4
	South America	15
	Europe	24
	Asia	18
Product type	Liquid milk	79
	Milk powder	13
	Milk fat products	4
	Cheeses	21
	Ice cream	2
Key spoilage bacteria	Bacillus cereus	35
	Bacillus licheniformis	27
	Pseudomonas fluorescens	30
	Bacillus subtilis	21
	Bacillus pumilus	18
	Pseudomonas fragi	14
	Pseudomonas putida	14
	Lactococcus lactis	13
	Pseudomonas lundensis	11
	Pseudomonas aeruginosa	8
	Enterococcus faecalis	8
	Bacillus circulans	6
	Enterococcus faecium	6
	Pseudomonas gessardii	7
	Pseudomonas psychrophila	6
	Pseudomonas veronii	6

3.2. Geographic Distribution Characteristics of Spoilage Bacteria in Dairy Products

The meta-analysis results showed significant regional heterogeneity in the contamination rate of spoilage bacteria in dairy products (Figure 2, Table 2). The highest contamination level was found in South America (89.5%), followed by North America (88.3%), Asia (87.3%), Europe (86.0%) and Oceania (82.2%), while Africa had a relatively lower contamination rate (57.6%).

Figure 2.

Geographical distribution of spoilage bacteria in dairy products.

Table 2.

Detection rate of spoiled dairy products by region.

Continent	Country	Number of Articles	Pooled Detection Rate (95% CI)	Heterogeneity
				p Value	I2 (%)
North America		12	88.3% (0.67–0.97)	0.00	97.82
	United States	11	89.1% (0.71–0.97)	0.00	98.80
	Canada	1	75.0% (0.59–0.86)	1.0 0	0.00
Oceania		6	82.2% (0.60–0.94)	0.00	93.33
	Australia	4	70.9% (0.43–0.89)	0.00	94.52
	New Zealand	2	98.8% (0.70–1.00)	0.08	68.11
Africa		4	57.6% (0.30–0.81)	0.00	87.55
	Ghana	1	46.7% (0.30–0.64)	1.00	0.00
	Tunisia	3	67.3% (0.26–0.93)	0.00	91.46
South America		15	89.5% (0.78–0.96)	0.00	85.20
	Argentina	2	39.9% (0.20–0.64)	0.01	77.66
	Brazil	12	94.9% (0.82–0.99)	0.00	79.18
	Uruguay	1	83.3% (0.19–0.99)	1.00	0.00
Europe		24	86.0% (0.75–0.94)	0.00	91.32
	Belgium	2	98.6% (0.91–1.00)	0.55	0.00
	France	2	98.7% (0.91–1.00)	0.57	0.00
	Germany	3	69.1% (0.40–0.88)	0.11	44.15
	Ireland	1	92.0% (0.85–0.96)	1.00	0.00
	Italy	6	66.1% (0.38–0.86)	0.00	88.16
	Malta	1	90.0% (0.33–0.99)	1.00	0.00
	Poland	1	34.0% (0.23–0.47)	0.00	80.13
	Russia	1	71.3% (0.36–0.92)	0.88	0.00
	Slovakia	1	83.3% (0.19–0.99)	1.00	0.00
	Spain	2	97.8% (0.86–1.00)	0.98	0.00
	Sweden	1	83.3% (0.69–0.92)	1.00	0.00
	Turkey	3	28.8% (0.21–0.38)	0.26	25.99
Asia		18	87.3% (0.72–0.95)	0.00	87.72
	China	13	91.7% (0.66–0.98)	0.00	82.00
	India	1	35.1% (0.25–0.47)	0.02	67.77
	Iraq	1	99.0% (0.86–1.00)	1.00	0.00
	Iran	1	99.5% (0.93–1.00)	1.00	0.00
	Japan	1	93.8% (0.46–1.00)	1.00	0.00
	Saudi Arabia	1	96.2% (0.60–1.00)	1.00	0.00

Fifteen studies from South America involved Argentina, Brazil, and Uruguay. The overall prevalence in this region was 89.5% (95% CI: 0.78–0.96), with high heterogeneity (I² = 85.20%). Brazil contributed the largest number of studies (12 studies), with a detection rate of 94.9% (95% CI: 0.82–0.99), making it the main contributor to the South American data. The detection rates for Argentina and Uruguay differed significantly, at 39.9% (95% CI: 0.20–0.64) and 83.3% (95% CI: 0.19–0.99), respectively. However, due to the limited number of studies included from these two countries, their statistical precision is limited, and the results should be interpreted with caution.

Among the 12 studies included from North America, the vast majority originated from the United States, with a pooled detection rate of 89.1% (95% CI: 0.71–0.97). The extremely high heterogeneity (I² = 98.80%) suggests substantial variation in contamination levels across different cities or dairy product types within the United States. Only one study from Canada reported a detection rate of 75.0%. Given that the studies were predominantly from the United States, the representativeness of the North American data is limited.

A total of six studies were included from Oceania (four from Australia and two from New Zealand), with a pooled detection rate of 82.2% (95% CI: 0.60–0.94) but extremely high heterogeneity (I² = 93.33%). New Zealand had a higher detection rate than Australia, with lower heterogeneity (I² = 68.11%). However, since all studies from Oceania came only from Australia and New Zealand, lacking data from other countries, this pooled detection rate has limited representativeness for Oceania as a whole.

A total of 18 studies were included from Asia, with an overall prevalence of 87.3% (95% CI: 0.72–0.95) and high heterogeneity (I² = 87.72%). Among these, Iran had the highest contamination rate at 99.5% (95% CI: 0.93–1.00), while India had the lowest at 35.1% (95% CI: 0.25–0.47). Of note, the Asian data were mainly concentrated in China, thus having limited representativeness for other regions.

The overall prevalence of spoilage bacteria in European dairy products was 86.0% (95% CI: 0.75–0.94), with Italy contributing the largest number of studies (six studies). Most countries showed relatively high detection rates, such as Malta (90.0%, 95% CI: 0.33–0.99), Belgium (98.6%, 95% CI: 0.91–1.00), and Spain (97.8%, 95% CI: 0.86–1.00). Notably, the detection rates in Poland and Turkey were significantly lower than those in other European countries, at only 34.0% (95% CI: 0.23–0.47) and 28.8% (95% CI: 0.21–0.38), respectively, indicating substantial regional differences in contamination levels within Europe.

A total of four studies were included from Africa, with a pooled detection rate of 57.6% (95% CI: 0.30–0.81). Research in this region was mainly concentrated in Tunisia (three studies), and heterogeneity was high (I² = 91.46%).

The following limitations should be noted when interpreting the above results. First, the number of included studies from some continents is limited (e.g., only four from Africa and six from Oceania), resulting in wide confidence intervals for the relevant estimates and insufficient statistical precision. Second, the geographical distribution of studies is highly concentrated: Most North American studies come from the United States, 80% of South American studies came from Brazil, 75% of African studies came from Tunisia, Oceania studies came only from Australia and New Zealand, and Asian studies were mainly from China. This data concentration means that the pooled estimates have very limited representativeness for their respective continents as a whole. Third, heterogeneity across studies is generally high (I² > 75%), suggesting that contamination levels may be strongly influenced by factors such as local climate, production conditions, and detection systems. In summary, the intercontinental comparisons presented in this study should be regarded as exploratory findings rather than definitive conclusions regarding contamination levels across continents. Future monitoring studies are needed in more geographical regions to validate these findings.

3.3. Differences in Contamination Risks Among Various Dairy Product Types

Significant variations in spoilage bacteria contamination rates were observed across different dairy product types (Table 3). Among liquid milk, raw milk showed the highest contamination rate at 92.6%. It should be noted that raw milk naturally contains indigenous milk lipases, whose activity can also contribute to dairy spoilage. Therefore, the contamination rate for raw milk serves only as a reference for microbial contamination levels and should not be directly compared with other dairy products. Among other liquid milk products, ESL milk had a contamination rate of 87.5%, refrigerated milk 83.9%, yogurt 71.1%, and pasteurized milk 57.5%. UHT milk and bactofuged milk showed relatively low contamination rates (28.9% and 10.0%, respectively). In addition, one study reported a contamination rate of 90.0% for fluid milk without specifying the type. The combined contamination rate of milk powder is 86.8%, while that of dried colostrum is 93.8%. In milk fat products, the contamination rates of cream and butter are 95.8% and 20.0% respectively, but the data volume is very small and should be regarded as preliminary results. The contamination rate of ice cream is relatively low (29.2%); the contamination rate of cheese is at a medium level (46.5%). However, only a few study types (such as ESL milk, refrigerated milk, bactofuged milk, butter, ice cream, dried colostrum) with statistical values should be regarded as preliminary results and need to be verified by future studies with larger sample sizes. Furthermore, included studies rarely reported the fat content of the tested dairy products (e.g., full-fat, semi-skimmed, skimmed). Given that fat content is an important factor influencing lipolytic spoilage, its impact on contamination rates could not be analyzed due to the limitations of the original data. Overall, high-moisture, non-frozen products (e.g., liquid milk) are more likely to serve as vehicles for microbial proliferation, while processing and storage conditions (e.g., drying, freezing) can significantly influence spoilage levels. This reflects that different dairy products exhibit distinct patterns of microbial contamination risk due to variations in raw material characteristics, processing techniques, and storage conditions [20,21]. Furthermore, the microbial standards in the EU Regulation EC 2073/2005 mainly set limits for hygiene indicator bacteria (such as Enterobacteriaceae) and pathogenic microorganisms (such as Coagulase-positive staphylococci) in products like pasteurized milk, milk powder, ice cream, and cheeses [22]. This study, however, focuses on the contamination levels of spoilage bacteria. Therefore, it is possible to consider a comprehensive assessment of the microbial quality of dairy products in the future (taking into account both pathogenic and spoilage bacteria).

Table 3.

Detection rates of various spoiled dairy product types.

Classification	Type	Number of Articles	Pooled Detection Rate (95% CI)	Heterogeneity
				p Value	I2 (%)
Liquid milk	Raw milk	45	92.6% (0.87–0.96)	0.00	94.59
	Pasteurized milk	20	57.5% (0.41–0.73)	0.00	96.68
	UHT milk	4	28.9% (0.11–0.58)	0.00	90.75
	Refrigerated milk	2	83.9% (0.03–0.99)	0.00	92.05
	ESL milk	1	87.5% (0.27–0.99)	1.00	0.00
	Bactofuged milk	1	10.0% (0.05–0.18)	1.00	0.00
	Yogurt	5	71.1% (0.40–0.90)	0.09	50.73
	Type not specified	1	90.0% (0.33–0.99)	1.00	0.00
Milk powder	Milk powder	12	86.8% (0.66–0.96)	0.00	88.06
	Dried colostrum	1	93.8% (0.46–1.00)	1.00	0.00
Milk fat products	Cream	3	95.8% (0.78–0.99)	0.28	20.98
	Butter	1	20.0% (0.07–0.40)	1.00	0.00
Ice cream	Ice cream	2	29.2% (0.23–0.37)	0.51	0.00
Cheeses	Cheese	21	46.5% (0.36–0.58)	0.00	84.25

3.4. Analysis of Primary Spoilage Bacterial Genera and Strains in Dairy Products

This study identified 77 genera of spoilage bacteria through statistical analysis. As shown in Figure 3A, the distribution of spoilage bacteria in dairy products exhibits a highly concentrated pattern. Bacillus, accounting for 40.7%, emerged as the predominant spoilage bacterial group, followed by Pseudomonas at 23.0%. Together, these two genera accounted for over 60% of the total, constituting the dominant bacterial community responsible for dairy product spoilage. Other genera such as Paenibacillus (7.7%), Anoxybacillus (2.9%), and Enterococcus (2.8%) were detected but each accounted for less than 10.0%. Notably, among non-dominant genera, 62 genera exhibited very low proportions. Consequently, this study grouped them under the broad category of “others”, which accounted for 8.6% of the total. Figure 3B further reveals the core status of Bacillus. In addition, Pseudomonas and Paenibacillus showed relatively high abundance in liquid milk, Clostridioides were more prominent in cheeses, and Anoxybacillus accounted significantly in milk powder. These findings indicate that product characteristics significantly influence the composition of spoilage bacterial communities.

Figure 3.

Composition and distribution of spoilage bacterial genera in dairy products. (A) Genus-level composition of spoilage bacteria in dairy products. (B) Clustering heatmap of spoilage bacterial genera in different dairy product categories.

This study conducted a comprehensive analysis of spoilage bacteria strains reported in 79 literature sources, encompassing a total of 290 bacterial strains (Table A4). To enhance the representativeness of findings, analysis focused on spoilage bacteria strains documented in more than five studies. The detection rate data for bacterial strains in Table 4 further clarifies the contamination characteristics of dairy products. The combined detection rates for Pseudomonas fluorescens and Bacillus licheniformis were highest at 21.1% (95% CI: 0.15–0.29) and 20.1% (95% CI: 0.16–0.25) respectively, making them the most prevalent spoilage bacteria in dairy products. Both are closely associated with dairy product deterioration (e.g., proteolysis, fat oxidation), and their high heterogeneity (p < 0.05, I² > 90%) suggests contamination levels may be influenced by geographical location, product type, or detection methods. Bacillus cereus (19.4%, 95% CI: 0.14–0.26) and Bacillus subtilis (11.7%, 95% CI: 0.08–0.17) were also prevalent, further highlighting the predominance of the genus Bacillus in dairy spoilage.

Table 4.

Detection rates of key spoilage bacteria in dairy products.

Spoilage Strain	Number of Articles	Pooled Detection Rate (95% CI)	Heterogeneity
			p Value	I2 (%)
Bacillus cereus	35	19.4% (0.15–0.28)	0.00	95.88
Bacillus licheniformis	27	20.1% (0.16–0.25)	0.00	94.11
Pseudomonas fluorescens	30	21.1% (0.15–0.29)	0.00	96.09
Bacillus subtilis	21	11.7% (0.08–0.17)	0.00	92.20
Bacillus pumilus	18	8.2% (0.05–0.13)	0.00	91.74
Pseudomonas fragi	14	10.2% (0.07–0.15)	0.00	83.53
Pseudomonas putida	14	8.3% (0.04–0.15)	0.00	93.74
Lactococcus lactis	13	10.5% (0.05–0.23)	0.00	94.72
Pseudomonas lundensis	11	7.5% (0.03–0.17)	0.00	95.06
Pseudomonas aeruginosa	8	4.1% (0.02–0.09)	0.00	85.25
Enterococcus faecalis	8	6.5% (0.11–0.30)	0.00	96.70
Bacillus circulans	6	3.7% (0.03–0.05)	0.53	0.00
Enterococcus faecium	6	7.0% (0.05–0.10)	0.34	12.20
Pseudomonas gessardii	7	5.0% (0.02–0.11)	0.00	80.29
Pseudomonas psychrophila	6	3.7% (0.01–0.11)	0.00	81.67
Pseudomonas veronii	6	1.8% (0.01–0.03)	0.89	0.00

Additionally, within the genus Pseudomonas, Pseudomonas fragi (10.2%, 95% CI: 0.07–0.15), Pseudomonas putida (8.3%, 95% CI: 0.04–0.15), and Pseudomonas lundensis (7.5%, 95% CI: 0.03–0.17) also exhibited detectable rates, reflecting the genus’s broad adaptability in dairy processing and storage stages. Lactococcus lactis (10.5%, 95% CI: 0.05–0.23), Enterococcus faecalis (6.5%, 95% CI: 0.11–0.30), and Enterococcus faecium (7.0%, 95% CI: 0.05–0.10) exhibited relatively lower detection rates. However, their confirmed spoilage potential (e.g., gas production and acid spoilage) warrants attention in Yoghurt and cheese.

In summary, the eleven predominant spoilage bacteria identified in dairy products in this study are: Pseudomonas fluorescens, Bacillus licheniformis, Bacillus cereus, Bacillus subtilis, Lactococcus lactis, Pseudomonas fragi, Pseudomonas putida, Bacillus pumilus, Pseudomonas lundensis, Enterococcus faecium and Enterococcus faecalis. To illustrate the biochemical mechanisms by which these predominant spoilage bacteria cause dairy spoilage, a schematic diagram is presented in Figure 4.

Figure 4.

Biochemical mechanisms of dairy spoilage caused by bacterial lipases.

To assess the extent to which the meta-analysis results of this study were influenced by publication bias, tests were conducted for each analysis group, and the results are shown in Table A5. Fail-safe N analysis indicated that all groups had values far exceeding the critical thresholds, suggesting that the overall results were minimally affected by publication bias and exhibited high stability. Egger’s regression test revealed potential publication bias in four groups related to the conclusions (the prevalence of Lactococcus lactis, Pseudomonas putida, Pseudomonas lundensis, and Enterococcus faecalis). However, Fail-safe N for these four groups were all far above the critical values, indicating that even in the presence of publication bias, the current conclusions remain highly robust. Considering the results of both methods together, the influence of publication bias on the main conclusions of this study is generally manageable, and most of the analytical results demonstrate good stability.

Based on the above analysis, we observed consistently high I² values (>75%) in most subgroups. In a large-scale meta-analysis covering 27 countries, spanning over twenty-two years and with diverse study designs, this level of high heterogeneity was in line with expectations. The high heterogeneity may stem from: geographical and climatic differences; variations in dairy processing techniques; differences in detection methods; and variations in study quality and sample size. More importantly, the random-effects model used in this study provides more conservative pooled estimates (with wider confidence intervals). Additionally, the higher safety factor also indicates that, despite the heterogeneity, our pooled estimates remain robust.

3.5. Analysis of the Conserved Domains of Lipase Produced by Spoilage Bacteria in Dairy Products

Based on the aforementioned findings, this study identified 11 predominant spoilage bacterial strains. By cross-referencing with the NCBI database, 37 lipase sequences associated with these spoilage bacteria were further screened (Table A6). Analysis of the conserved domains within the screened lipase sequences revealed a total of 18 distinct conserved domains. Twenty-eight lipases contained only one conserved domain, eight lipases contained two conserved domains, and only one lipase contained three conserved domains. Among these, lipases containing the Abhydrolase superfamily conserved domain were the most prevalent (11 items). This domain belongs to the α/β hydrolase fold superfamily and is a potential regulator of lipid metabolism and signal transduction [23]. The second most prevalent was the CVT17 superfamily (6 items), a domain playing a pivotal role in autophagosome-lysosome degradation [24], primarily involved in intracellular transport and vesicular secretion.

Five lipases contain the conserved SGNH_hydrolase superfamily domain. Enzymes within this family possess flexible active sites capable of undergoing conformational changes upon substrate binding [25]. Four lipases contain the Aes superfamily conserved domain, while three possess the EstA superfamily conserved domain. The EstA superfamily domain exhibits specificity for short-chain acyl substrates [26], and both domains participate in lipid transport and metabolism. Lipases containing conserved domains of the DAP2 superfamily, Lip2 superfamily, or MenH superfamily (two from each) are primarily involved in the transport and metabolism of amino acids, lipids, and coenzymes, respectively.

Additionally, lipases containing the CusA superfamily, FrmB superfamily, KxYKxGKxW_sig superfamily, HemolysinCabind superfamily, PldB superfamily, COG2931 superfamily, COG4782 superfamily, Axe1 superfamily, CwlO1 superfamily, and CHAP superfamily were the least abundant (one each). The CusA superfamily comprises Cu and Ag efflux pumps. It is primarily involved in inorganic ion transport and metabolism. The FrmB superfamily has a defense function. The PldB superfamily shares the same function as the EstA superfamily and Lip2 superfamily. The COG2931 superfamily consists of calcium-binding proteins associated with RTX toxins. Its function is similar to that of the Axe1 superfamily, primarily participating in the synthesis, transport, and degradation of secondary metabolites. The CHAP superfamily exhibits amidase-like activity and is involved in bacterial cell wall metabolism. The functions of the KxYKxGKxW_sig superfamily, HemolysinCabind superfamily, CwlO1 superfamily, and COG4782 superfamily remain uncharacterized.

3.6. Analysis of the Physicochemical Properties of Lipase Produced by Spoilage Bacteria in Dairy Products

According to Table A6, the physicochemical properties of the lipases showed the following characteristics: The molecular weight distribution was broad, ranging from 24.12 to 86.82 kDa. The amino acid sequence lengths varied significantly, from 212 to 776 a.a. The isoelectric points also showed a wide range, from 4.10 to 9.68. Most of the lipases had acidic or neutral isoelectric points. The average hydrophilicity of all 36 lipases is negative, indicating a hydrophobic surface characteristic. This structural feature facilitates interaction with lipid substrates. Lipophilic coefficients are generally high (69.11–100.87), further confirming the strong affinity of this enzyme class for lipid matrices. Furthermore, lipases exhibited marked variations in instability coefficients, a property potentially impacting their tolerance and stability within food processing environments.

3.7. Structural Prediction Analysis of Lipase Produced by Spoilage Bacteria in Dairy Products

This study employed the Swiss-Model online tool to perform homology modelling on the aforementioned 37 lipase sequences, with visualization completed using PyMOL (Figure A1). The homology modelling results revealed that each of the 37 lipases matched distinct templates. All lipases exhibited a characteristic α/β-folded structure, wherein β-sheets formed a central backbone surrounded by multiple α-helices. These β-sheets predominantly adopted a parallel arrangement, maintaining consistent orientation and high degree of twist. Lipases from different sources exhibit variations in the number and positioning of α-helices, as well as the quantity of β-strands, thereby conferring distinct catalytic properties [9,27]. For instance, the lipase from Pseudomonas fragi (GenBank: VFB20546.1) possesses 32 β-strands and 19 α-helices; conversely, the lipase from Bacillus cereus (GenBank: ETT75504.1) possesses only 10 β-sheets and 11 α-helices. Within the three-dimensional models of lipases, the N-terminal structures exhibit diversity, comprising either helical or ring-like conformations. We observed that the lipases from Pseudomonas cereus (except GenBank: ETT75504.1) and Lactococcus lactis, which exhibit lower instability coefficients, both possess extended ring-like conformations at their N-termini (Figure 5). The maintenance of overall conformation depends on the integrity of the hydrophobic core and the hydrogen bond network. However, the flexibility of the N-terminus may affect protein stability to some extent. Future studies could consider truncating or modifying this region to verify its specific contribution to stability. It should be noted that the above structural observation results are derived from homology modeling and predictions, rather than experimental determination. The identified α/β hydrolase fold and hydrophobic surface features provide a reasonable structural explanation for thermal stability. However, these remain hypotheses and do not constitute experimental evidence. Future experimental studies are needed to validate this hypothesis.

Figure 5.

Three-dimensional structural models of lipases from Bacillus cereus, Lactococcus lactis and Pseudomonas fragi. (A) Homology model of lipase from Bacillus cereus. (B) Homology model of lipase from Lactococcus lactis. (C) Homology model of lipase from Pseudomonas fragi. The secondary structure elements are color-coded: cyan for α-helices, red for β-sheets, and purple for loops.

4. Discussion

Through an integrated approach combining meta-analysis with bioinformatics, this study systematically characterizes the prevalence patterns of spoilage bacteria in dairy products and the shared molecular features of their key lipases. Findings indicate that contamination by spoilage bacteria in dairy products exhibits significant geographical and product-specific heterogeneity, with spoilage bacterial communities predominantly concentrated within the genera Bacillus and Pseudomonas. Further analysis indicates that the lipases secreted by these dominant strains possess conserved hydrophobic interfaces and stable three-dimensional folding structures. This series of discoveries, spanning from macro-level prevalence patterns to micro-level molecular characteristics, provides new perspectives and evidence for deepening our understanding of dairy product spoilage mechanisms and developing precise prevention and control strategies.

First of all, the contamination and distribution of spoilage bacteria in dairy products exhibit significant geographical and product heterogeneity. Meta-analysis results showed that the pooled detection rates were higher in South America (89.5%), North America (88.3%), and Asia (87.3%), while the pooled detection rate in Africa was lower (57.6%). However, the interpretation of these geographical differences is subject to limitations. First, the number of included studies and the geographical coverage across continents are unbalanced: Almost all North American studies came from the United States; twelve of the fifteen South American studies came from Brazil; the six Oceania studies came only from Australia and New Zealand; and three of the four African studies came from Tunisia. Therefore, these pooled detection rates essentially reflect the situation in these few countries and have limited representativeness for their respective continents as a whole. Second, high heterogeneity is commonly observed across studies from different countries, suggesting that regional climate, production practices, and detection systems may jointly influence contamination levels [28]. However, due to the limitations of the available data, the specific roles of these influencing factors require further investigation. In summary, the geographical differences presented in this study should be regarded as exploratory findings rather than definitive conclusions regarding contamination levels across continents.

Secondly, from the perspective of product type, the contamination rates of different processing methods of liquid milk vary significantly: with increasing heat treatment intensity, the contamination rate dropped from 57.5% in pasteurized milk to 28.9% in UHT milk. This trend indicates that the heat treatment intensity is a key factor in controlling microbial contamination in dairy products [29]. Additionally, the contamination rate of milk powder reached 86.8%, suggesting that low water activity alone cannot completely control the risk of spoilage. Other key influencing factors such as storage conditions and pH also need to be considered [30]. In conclusion, differentiated control should be implemented based on product characteristics. It should be noted that factors such as fat content of products, type of cheese (such as raw material and ripeness), and whether exogenous lipases are added during processing all affect the risk of spoilage. However, most of the article does not report this information, and it is recommended that future research should include this.

Thirdly, the structure of the spoilage bacterial community in dairy products is highly concentrated. Bacillus and Pseudomonas are the dominant genera, together accounting for more than 60% of the total. This finding corroborates multiple regional studies and review conclusions in recent years [4], reflecting the broad adaptability and ecological competitive advantage of these two bacterial groups within the dairy product environment. Among these, Pseudomonas fluorescens (21.1%) and Bacillus licheniformis (20.1%) emerged as the predominant strains, indicating their pivotal role in the spoilage process. From a physiological perspective, Pseudomonas, as a typical psychrophile, maintains metabolic activity under refrigerated conditions and secretes thermostable lipases and proteases. This leads to acid spoilage in milk during low-temperature storage [31]. Relying on their heat-resistant spores, Bacillus species can withstand conventional thermal processing and persist in products. They thus become major spoilage factors in low-moisture products such as milk powder [32]. This finding provides a clear target for targeted monitoring and sterilization process optimization. It is essential to evaluate the efficacy of existing thermal sterilization protocols in inactivating spores and to strengthen the entire cold chain to inhibit the proliferation of psychrophilic bacteria.

More importantly, this study identified the common structural characteristics of lipases from predominant spoilage bacteria at the molecular level. Based on these computational predictions, we propose a hypothesis: these lipases generally possess hydrophobic surfaces, high lipid affinity, and a conserved α/β hydrolase folding conformation, which may be related to their thermal stability. Three-dimensional modeling revealed that the Abhydrolase superfamily lipases exhibit a typical α/β hydrolase folding conformation, with a “lid” structure composed of α-helices present above the catalytic triad [33]. We speculate that this structure may affect the enzyme’s interface activation ability and thermal stability by regulating the active center. It should be emphasized that the above structure-function relationship is derived from homology modeling and prediction, not experimentally proven. Therefore, it should be regarded as a hypothesis that needs to be verified. This study expands the understanding of spoilage mechanisms from the microbial survival level to a new dimension of enzyme molecular stability. Furthermore, the lipase sequences were obtained from database reference strains, not directly sequenced from the isolates in the included studies. Future research should consider direct sequencing of lipase genes from spoilage isolates.

5. Conclusions

This study conducted a meta-analysis on the prevalence of spoilage bacteria in dairy products across different regions and categories. We identified the predominant spoilage bacterial strains and revealed the conserved structural features of their lipases through bioinformatics analysis. Appropriate processing methods (such as heat treatment) are indeed one of the key factors in reducing the prevalence of spoilage bacteria in dairy products: the contamination rate of pasteurized milk is significantly higher than that of UHT milk. However, relying solely on a single processing method cannot completely control spoilage; multiple factors such as water activity, storage temperature, fat content, and pH should be comprehensively controlled. The hydrophobic surface and high lipid affinity of the predominant spoilage bacterial lipases may explain their thermal stability. This finding offers a potential structural explanation for the “post-sterilization spoilage” phenomenon. Based on these findings, we suggest designing specific rapid detection techniques targeting the conserved structural domains and folding conformations of lipases, and developing targeted inhibition techniques based on conformational induction of inactivation. These data will provide important theoretical value for future research in this field and offer multi-level strategic insights for the monitoring and prevention of dairy product spoilage risks.

Appendix A

Table A1.

Summary of search queries.

Database	Search Query
CNKI	(Topic: Dairy Products) OR (Title/Abstract/Keyword: Dairy Products + Milk + Yogurt + Cheese + Butter + Milk Powder + Cream + Condensed Milk + Ice Cream + Dairy Beverage (Precise)) AND (Topic: Spoilage Bacteria) OR (Title/Abstract/Keyword: Spoilage Bacteria + Spoilage Microorganisms (Precise)) AND Publication Date Between (1 January 2003, 31 December 2025)
PubMed	((((((((“Dairy Products” [Mesh]) OR (Dairy Products [Title/Abstract])) OR (Dairy Product [Title/Abstract])) OR (Products, Dairy [Title/Abstract])) OR (Product, Dairy [Title/Abstract])) OR (Milk [Title/Abstract])) OR (Cheese [Title/Abstract])) OR (Yogurt [Title/Abstract])) AND (Spoilage Bacteria) AND ((“1 January 2003” [Date–Publication]: “31 December 2025” [Date–Publication]))
Web of Science	(TS = (Dairy Products) OR AB = (Dairy Products OR Dairy Product OR Products, Dairy OR Product, Dairy OR Milk OR Cheese OR Yogurt)) AND (TS = (Spoilage Bacteria) OR AB = (Spoilage Bacteria)) AND DOP = (1 January 2003/31 December 2025)

Table A2.

The quality of Meta-analysis in article.

Number	NOS Evaluation Entry	Yes/Article (%)	No/Article (%)
1	Are the research objectives clearly defined?	79 (100.00%)	-
2	Do the samples meet the inclusion criteria?	79 (100.00%)	-
3	Do sampling locations be recorded?	79 (100.00%)	-
4	Has the sample size for the study been specified?	74 (93.67%)	5 (6.33%)
5	Are the sample types specified?	79 (100.00%)	-
6	Do the strain information be provided?	79 (100.00%)	-
7	Is standardized, credible authentication technology employed?	79 (100.00%)	-
8	Was appropriate statistical analysis employed?	79 (100.00%)	-
9	Are the results adequately presented?	79 (100.00%)	-

Table A3.

Sample information collection form.

Sample Name	Positive Samples	Total Samples	Location	Testing Method	Research
Raw milk	8	8	United States	API 20 NE	[34]
Pasteurized milk	12	12	United States	API 50 CH and 16S rDNA	[35]
Raw milk	18	18	Belgium	16S rRNA	[36]
Cheese	15	50	Argentina	API 50 CH	[37]
Cheese	35	42	Sweden	16S rDNA	[38]
Raw milk	39	39	United States	16S rDNA	[39]
Pasteurized milk	11	11
Cream	7	7	Japan	PCR-denaturing gradient gel electrophoresis	[40]
Raw milk	43	43	Brazil	API 50 CH and API Coryne	[41]
Refrigerated milk	40	40	Tunisia	API 20 NE	[42]
Cheese	16	16	Argentina	Bergey’s Manual	[43]
Pasteurized milk	29	95
Yoghurt	4	4	China	API	[44]
Pasteurized milk	4	4
Fluid milk	4	4
Raw milk	8	8	Italy	RAPD-PCR	[45]
Pasteurized milk	5	5	United States	PCR-DGGE, 16S rRNA	[46]
Raw milk	62	62	Belgium	16S rRNA	[47]
Milk powder	92	100	Ireland	API 50 CHB	[48]
Cheese	/	140	Turkey	Gram staining	[49]
Cheese	2	2	United States	16S rDNA	[50]
Milk powder	22	22	Uruguay	16S rDNA	[51]
Milk powder	22	22	China	16S rDNA	[52]
Dried colostrum	7	7	United States	16S rRNA	[53]
Raw milk	201	1228	United States	16S rRNA	[54]
Pasteurized milk	1002	1228
ESL milk	3	3	Germany	16S rRNA	[55]
Raw milk	/	3	Australia	16S rDNA	[56]
Pasteurized milk	/	3
Cheese	/	3
Milk powder	/	3
Raw milk	1	1	Germany	FT-IR spectroscopy and partial gene sequencing	[57]
Cheese	2	5
Yoghurt	2	3
Milk powder	8	11
UHT milk	0	2
Pasteurized milk	3	4	Brazil	Bactray commercial kit	[58]
Raw milk	22	22	Spain	16S rRNA	[59]
Raw milk	/	/	China	16S rRNA	[60]
Raw milk	19	30	Tunisia	16S rRNA	[61]
UHT milk	10	30
Pasteurized milk	11	20
Raw milk	/	56	Italy	16S rRNA	[62]
Pasteurized milk	30	55	India	Selective medium culture combined with biochemical identification	[63]
Milk powder	18	52
Ice cream	10	40
Cheese	33	90
Butter	5	25
Raw milk	7	7	Australia	16S rRNA	[64]
Raw milk	3	3	Brazil	16S rDNA	[65]
Raw milk	203	211	Australia	The method of Manero and Blanch	[66]
Pasteurized milk	5	24
Raw milk	20	20	Germany	16S rRNA	[67]
Raw milk	3	3	China	16S rDNA	[68]
Raw milk	61	61	Brazil	16S rRNA	[69]
Pasteurized milk	62	104	Australia	MALDI-TOF MS	[70]
Raw milk	92	154
Raw milk	1	1	Slovakia	MALDI-TOF MS	[71]
Cheese	20	20	Italy	16S rRNA	[72]
Milk powder	6	6	New Zealand	16S rDNA	[73]
Raw milk	28	100	Tunisia	16S RNA	[74]
Bactofuged milk	10	100
Pasteurized milk	16	100
UHT milk	13	100
Refrigerated milk	33	100
Raw milk	10	10	United States	16S rRNA	[75]
Raw milk	93	120	United States	16S rRNA	[76]
Raw milk	14	30	Ghana	PCR	[77]
Cheese	10	31
Yoghurt	12	28
Raw milk	/	22	China	16S rRNA	[78]
Ice cream	38	125	Turkey	PCR	[79]
Milk powder	7	7	China	PCR	[80]
Raw milk	20	20	Brazil	PCR	[81]
Raw milk	20	20	Brazil	RFLP	[82]
Raw milk	20	20	Brazil	16S rRNA	[83]
Raw milk	46	125	China	16S rRNA	[84]
Raw milk	4	4	Malta	16S rRNA	[85]
Pasteurized milk	226	280	United States	16S rDNA	[86]
Raw milk	56	56	United States	16S rRNA	[87]
Cheese	2	2	Russia	DNA barcoding and DNA metabarcoding	[88]
Milk powder	2	3
Yoghurt	2	3
Milk powder	9	30	Poland	ISO 7932 [89]	[90]
Cheese	3	35
Pasteurized milk	18	60
Cheese	42	80
Cheese	76	175
Raw milk	4	4	Italy	16S rRNA	[91]
Pasteurized milk	2	8
Cheese	3	20
Raw milk	240	240	New Zealand	MALDI-TOF MS	[92]
Raw milk	50	50	Brazil	16S rRNA	[93]
Milk powder	21	21	France	16S rRNA	[94]
Cheese	89	340	Italy	ISO 7932:2005 [95]	[96]
Raw milk	27	27	Brazil	16S rRNA	[97]
Cream	64	64	France	Aamplified panC gene sequence analysis	[98]
Raw milk	6	6	China	Automatic microbial identification system	[99]
Milk powder	56	56	China	16S rRNA	[100]
UHT milk	53	100	Brazil	PCR	[101]
Cheese	32	32	Italy	MALDI-TOF MS	[102]
Raw milk	25	25	China	PCR	[14]
Raw milk	12	35	Turkey	PCR	[103]
Cheese	5	30
Raw milk	3	3	Brazil	16S rRNA and 23S rRNA	[104]
Raw milk	100	100	Iran	16S rRNA	[105]
Yoghurt	12	12	Saudi Arabia	PCR	[106]
Cheese	2	2	China	16S rRNA	[107]
Raw milk	5	5	Brazil	Pseudomonas Agar Base supplemented with Cetrimide	[108]
Raw milk	50	50	Iraq	16S rRNA	[109]
Pasteurized milk	106	363	China	16S rDNA	[110]
Pasteurized milk	27	36	Canada	MALDI-TOF MS	[111]
Cheese	13	31
Raw milk	21	21	Spain	PCR	[112]
Pasteurized milk	3	5
Cheese	10	13
Cream	2	2
Raw milk	260	260	China	16S rRNA	[113]

Table A4.

Strains information collection form.

Bacteria Strains	Positive Bacterial Strains	Total Bacterial Strains	Research
Pseudomonas fluorescens	172	338	[34]
Pseudomonas putida	47	338
Bacillus lentus	1	53	[35]
Bacillus cereus	6	53
Bacillus licheniformis	11	53
Bacillus mycoides	1	53
Bacillus stearothermophilus	3	53
Bacillus pumilus	3	53
Microbacterium lacticum	16	53
Kocuria varians	4	53
Bacillus subtilis	1	53
Bacillus barbaricus	2	159	[36]
Aneurinibacillus thermoaerophilus	2	159
Bacillus smithii	2	159
Bacillus fordii	2	159
Virgibacillus proomii	6	159
Brevibacillus agri	8	159
Ureibacillus thermosphaericus	11	159
Brevibacillus borstelensis	12	159
Bacillus pallidus	25	159
Bacillus fordii	2	159
Bacillus licheniformis	37	159
Bacillus subtilis	2	159
Bacillus cereus	15	50	[37]
Bacillus pumilus	20	50
Clostridium tyrobutyricum	25	124	[38]
Clostridium sporogenes	40	124
Clostridium cochlearium	15	124
Paenibacillus borealis	2	12	[39]
Paenibacillus amylolyticus	1	12
Bacillus pumilus	1	12
Bacillus weihenstephanensis	1	12
Bacillus licheniformis	1	12
Bacillus cereus	1	12	[40]
Bacillus amyloliquefaciens	1	12
Bacillus subtilis	1	12
Enterococcus faecium	1	12
Enterococcus faecalis	1	12
Lactobacillus paracasei	1	12
Lactococcus lactis	1	12
Leuconostoc mesenteroides	1	12
Staphylococcus saprophyticus	1	12
Staphylococcus warneri	1	12
Achromobacter denitrificans	1	12
Pseudomonas fluorescens	1	12
Pseudomonas fluorescens	94	250	[41]
Pseudomonas putida	3	250
Aeromonas caviae	1	250
Aeromonas sobria	1	250
Burkholderia cepacia	12	250
Aeromonas hydrophila	20	250
Klebsiella oxytoca	10	250
Erwingella americana	7	250
Chryseomonas luteola	3	250
Alcaligenes faecalis	1	250
Sphingomonas paucimobilis	1	250
Bacillus coagulans	1	250
Bacillus lentus	2	250
Pseudomonas fluorescens	52	258	[42]
Aeromonas hydrophila	41	258
Pseudomonas putida	15	258
Pseudomonas cepacia	34	258
Chryseomonas luteola	13	258
Bacillus polymyxa	3	17	[43]
Bacillus macerans	2	17
Leuconostoc mesenteroides	5	17
Clostridium tyrobutyricum	4	17
Lactobacillus fermentum	3	17
Pseudomonas fluorescens	1	3	[44]
Kocuria varians	1	3
Burkholderia cepacia	1	3
Pseudomonas fluorescens	1	66	[45]
Serratia marcescens	2	66
Klebsiella ornithionolytica	1	66
Pseudomonas trivialis	1	66
Pseudomonas reactans	1	66
Pseudomonas putida	1	66
Citrobacter freundii	1	66
Pseudomonas fragi	4	66
Pseudomonas lundensis	1	66
Pseudomonas gessardii	1	66
Hafnia alvei	3	66
Staphylococcus pasteuri	1	66
Bacillus thuringiensis	1	66
Lactococcus lactis	2	66
Corynebacterium variabilis	1	66
Staphylococcus epidermidis	1	66
Rodococcus erythropolis	1	66
Carnobacterium piscicola	1	66
Lactococcus piscium	2	66
Staphylococcus warneri	1	66
Pseudomonas migulae	1	25	[46]
Pseudomonas putida	5	25
Pseudomonas fragi	5	25
Pseudomonas fluorescens	7	25
Clostridioides difficile	1	25
Streptococcus agalactiae	1	25
Buttiauxella izardii	1	25
Buttiauxella brennerae	1	25
Pseudomonas stutzeri	1	25
Pseudomonas brassicacearum	1	25
Buttiauxella warmboldiae	1	25
Pseudomonas fragi	25	103	[47]
Pseudomonas lundensis	68	103
Bacillus cereus	59	100	[48]
Bacillus licheniformis	52	100
Bacillus subtilis	38	100
Bacillus pumilus	10	100
Bacillus mycoides	3	100
Bacillus megaterium	5	100
Bacillus polymyxa	2	100
Bacillus sphaericus	7	100
Bacillus circulans	3	100
Bacillus coagulans	3	100
Pseudomonas fluorescens	1	140	[49]
Pseudomonas pseudoalcaligenes	22	140
Pseudomonas alcaligenes	4	140
Pseudomonas aeruginosa	1	140
Pseudomonas fluorescens	17	26	[50]
Bacillus cereus	1	26
Lactococcus lactis	8	26
Bacillus licheniformis	88	601	[51]
Bacillus megaterium	10	207
Bacillus pumilus	6	207
Anoxybacillus flavithermus	86	207
Bacillus subtilis	10	207
Bacillus licheniformis	223	801	[52]
Anoxybacillus flavithermus	164	801
Geobacillus stearothermophilus	99	801
Bacillus lichenformis	5	29	[53]
Bacillus badius	2	29
Kocuria rhizophila	2	29
Bacillus cereus	2	29
Bacillus subtilis	9	29
Pseudomonas fluorescens	4	29
Enterococcus faecium	1	29
Bacillus massiliensis	1	29
Microbacterium hydrocarbonoxydans	1	29
Bacillus pumilus	108	1288	[54]
Bacillus licheniformis	188	1288
Bacillus subtilis	23	1288
Bacillus cereus	27	1288
Paenibacillus odorifer	506	1288
Paenibacillus graminis	46	1288
Paenibacillus amylolyticus	101	1288
Bacillus cereus	12	162	[55]
Microbacterium lacticum	26	162
Bacillus subtilis	29	162
Bacillus licheniformis	2	162
Bacillus pumilus	48	162
Bacillus circulans	2	162
Lactococcus lactis	1	162
Enterococcus casseliflavus	1	162
Acinetobacter johnsonii	2	162
Chryseobacterium haifense	2	162
Paenibacillus borealis	1	162
Bacillus flexus	1	162
Brevibacillus agri	8	162
Lysinibacillus fusiformis	3	162
Paenibacillus turicensis	1	162
Moraxella osloensis	2	162
Psychrobacter glacincola	2	162
Streptococcus thermophilus	19	162
Bacillus licheniformis	60	237	[56]
Bacillus subtilis	25	237
Bacillus pumilus	20	237
Bacillus cereus	20	237
Anoxybacillus flavithermus	19	237
Paenibacillus polymyxa	1	237
Bacillus megaterium	8	237
Bacillus clausii	1	237
Bacillus subtilis	38	379	[57]
Bacillus licheniformis	72	379
Geobacillus stearothermophilus	30	379
Bacillus cereus	88	379
Bacillus circulans	18	379
Bacillus pumilus	13	379
Pseudomonas fluorescens	12	100	[58]
Pseudomonas aeruginosa	4	100
Pseudomonas mendocina	1	100
Pseudomonas pseudoalcaligenes	1	100
Streptococcus thermophilus	10	164	[59]
Lactococcus lactis	115	164
Lactococcus garvieae	28	164
Enterococcus durans	1	164
Enterococcus faecalis	2	164
Streptococcus agalactiae	6	164
Lactococcus raffinolactis	4	164
Leuconostoc pseudomesenteroides	1	164
Streptococcus salivarius	1	164
Bacillus subtilis	1	5	[60]
Bacillus licheniformis	1	5
Bacillus stearothermophilus	1	5
Bacillus sporothermodurans	25	100	[61]
Bacillus cereus	22.5	100
Bacillus subtilis	12.5	100
Bacillus licheniformis	12.5	100
Bacillus pumilus	10	100
Brevibacillus brevis	10	100
Bacillus sphaericus	7.5	100
Pseudomonas fluorescens	85	178	[108]
Pseudomonas aeruginosa	13	178
Pseudomonas putida	34	178
Pseudomonas fluorescens	19	80	[62]
Pseudomonas aeruginosa	15	80
Pseudomonas fragi	3	80
Pseudomonas putida	7	80
Pseudomonas rhodesiae	2	80
Pseudomonas fulva	4	80
Pseudomonas libanensis	1	80
Hafnia alvei	5	80
Serratia marcescen	9	80
Bacillus cereus	55	100	[63]
Bacillus cereus	33	100
Pseudomonas fluorescens	4	58	[64]
Pseudomonas lurida	1	58
Pseudomonas gessardii	3	58
Pseudomonas proteolytica	1	58
Pseudomonas gingeri	1	58
Pseudomonas poae	3	58
Pseudomonas veronii	1	58
Pseudomonas valomonii	1	58
Pseudomonas psychrophila	2	58
Pseudomonas fragi	4	58
Bacillus pumilus	2	58
Bacillus licheniformis	5	58
Bacillus safensis	1	58
Hafnia alvei	5	58
Hafnia paralvei	2	58
Bacillus weihenstephanensis	3	58
Acinetobacter guillouiae	1	58
Serratia liquefaciens	1	58
Stenotrophomonas maltophilia	1	58
Pseudomonas lundensis	2	58
Pseudomonas fluorescens	4	8	[65]
Pseudomonas putida	2	8
Serratia liquefaciens	2	8
Enterococcus casseliflavus	7	211	[66]
Enterococcus faecalis	157	211
Enterococcus faecium	18	211
Enterococcus durans	6	211
Enterococcus asini	2	211
Enterococcus malodoratus	1	211
Enterococcus gallinarum	1	211
Enterococcus mundtii	1	211
Enterococcus pseudoavium	1	211
Enterococcus raffinosus	1	211
Enterococcus sulfureus	1	211
Enterococcus saccharolyticus	1	211
Enterococcus hirae	18	211
Lactococcus lactis	11	100	[67]
Pseudomonas proteolytica	9	100
Pseudomonas fragi	4	100
Pseudomonas lundensis	8	100
Lactococcus raffinolactis	3	100
Pseudomonas putida	1	34	[68]
Enterobacter ludwigii	1	34
Citrobacter freundii	1	34
Kurthia pneumonia	1	34
Kurthia gibsonii	2	34
Lactococcus garvieae	1	34
Lactococcus lactis	2	34
Acinetobacter lwoffii	1	34
Enterococcus faecalis	1	34
Pseudomonas aeruginosa	1	34
Pseudomonas fluorescens	41	82	[69]
Acinetobacter guillouiae	4	697	[70]
Buttiauxella agrestis	5	697
Buttiauxella gaviniae	6	697
Buttiauxella noackiae	7	697
Citrobacter braakii	23	697
Citrobacter freundii	15	697
Escherichia coli	121	697
Enterobacter amnigenus	24	697
Enterobacter asburiae	43	697
Enterobacter cloacae	52	697
Enterobacter kobei	7	697
Enterobacter ludwigii	15	697
Ewingella americana	5	697
Hafnia alvei	52	697
Klebsiella oxytoca	70	697
Klebsiella pneumonia	16	697
Kluyvera intermedia	7	697
Pantoea agglomerans	9	697
Proteus mirabilis	2	697
Proteus vulgaris	2	697
Pseudomonas aeruginosa	2	697
Pseudomonas fluorescens	4	697
Pseudomonas lundensis	2	697
Pseudomonas nitroreducens	2	697
Rahnella aquatilis	9	697
Raoultella ornithinolytica	114	697
Raoultella terrigena	3	697
Serratia fonticolla	2	697
Serratia liquefaciens	48	697
Serratia marcescens	11	697
Stenotrophomonas maltophilia	2	697
Yersinia enterocolitica	12	697
Pseudomonas fragi	30	200	[71]
Pseudomonas gessardii	19	200
Pseudomonas lundensis	34	200
Acinetobacter johnsonii	11	200
Sphingobacterium multivorum	3	200
Pseudomonas fragi	20	200
Raoultella ornithinolytica	17	200
Corynebacterium casei	17	200
Lactococcus lactis	15	200
Aeromonas media	9	200
Kytococcus sedentarius	6	200
Citrobacter braakii	6	200
Enterococcus faecalis	1	200
Candida inconspicua	3	200
Bacillus muralis	38	188	[72]
Clostridium butyricum	6	188
Lactobacillus helveticus	1	188
Lactobacillus zeae	1	188
Paenibacillus amylolyticus	29	188
Staphylococcus equorum	1	188
Streptococcus infantis	1	188
Acinetobacter johnsonii	3	188
Acinetobacter iwofii	1	188
Azomonas insignis	1	188
Enhydrobacter aerosaccus	1	188
Hemophilus parainfluenza	1	188
Propionibacterium acnes	1	188
Pseudomonas fragi	2	188
Bacillus cereus	22	188
Pseudomonas fragi	3	188
Pseudomonas veronii	2	188
Sphingomonas echinoides	1	188
Anoxybacillus kestabolensis	1	188
Bacillus anthracis	1	188
Bacillus flexus	1	188
Bacillus horikoshii	1	188
Clostridium bowmanii	11	188
Clostridium butyricum	5	188
Clostridium neonatale	3	188
Clostridium pasteurianum	38	188
Clostridium tyrobutiricum	1	188
Lysinibacillus boronitolerans	6	188
Paenibacillus lentimorbus	2	188
Paenibacillus stellifer	1	188
Hemophilus parainfluenzae	1	188
Neisseria subflava	1	188
Bacillus licheniformis	67	100	[73]
Bacillus cereus	19	100
Bacillus thuringiensis	4	100
Bacillus subtilis	4	100
Bacillus pumilus	4	100
Paenibacillus glucanolyticus	2	100
Bacillus pumilus	9.52	100	[74]
Bacillus licheniformis	52.38	100
Terribacillus aidingensis	4.76	100
Bacillus sporothermodurans	4.76	100
Bacillus cereus	29	1000	[75]
Bacillus subtilis	191	1000
Bacillus licheniformis	323	1000
Bacillus safensis	59	1000
Bacillus aerophilus	59	1000
Bacillus odorifer	44	1000
Bacillus aerophilus	8	106	[76]
Bacillus cereus	3	106
Bacillus clausii	1	106
Bacillus horneckiae	1	106
Bacillus licheniformis	34	106
Bacillus pumilus	36	106
Bacillus safensis	6	106
Bacillus subtilis	9	106
Bacillus weihenstephanensis	1	106
Paenibacillus amylolyticus	1	106
Paenibacillus cookii	2	106
Paenibacillus peoriae	3	106
Bacillus cereus	36	89	[77]
Pseudomonas fragi	9	21	[78]
Pseudomonas putida	2	21
Pseudomonas prosekii	1	21
Pseudomonas fluorescens	6	21
Pseudomonas psychrophila	3	21
Bacillus cereus	38	125	[79]
Bacillus cereus	31.82	100	[80]
Lactococcus lactis	5.09	100
Alkaliphilus oremlandii	4.17	100
Cronobacter sakazakii	3.76	100
Bacillus circulans	3.39	100
Acinetobacter johnsonii	2.41	100
Paenibacillus barcinonensis	1.9	100
Paenibacillus amylolyticus	1.53	100
Paenibacillus odorifer	1.08	100
Lactococcus lactis	35	141	[81]
Enterobacter kobei	22	141
Aerococcus urinaeequi	12	141
Acinetobacter lwoffii	9	141
Kurthia gibsonii	8	141
Serratia ureilytica	8	141
Bacillus licheniformis	7	141
Staphylococcus epidermidis	7	141
Macrococcus caseolyticus	6	141
Streptococcus parauberis	5	141
Enterococcus faecium	4	141
Acinetobacter johnsonii	3	141
Enterococcus thailandicus	2	141
Stenotrophomonas rhizophila	2	141
Yersinia enterocolitica	2	141
Acinetobacter bouvetii	1	141
Bacillus cereus	1	141
Bacillus pumilus	1	141
Enterococcus hermanniensis	1	141
Staphylococcus warneri	1	141
Bacillus licheniformis	45	132	[82]
Brachybacterium nesterenkovii	19	132
Enterococcus faecalis	12	132
Streptococcus thermophilus	11	132
Micrococcus aloeverae	6	132
Streptococcus thermophilus	6	132
Kocuria varians	5	132
Bacillus circulans	4	132
Paenibacillus cookii	4	132
Kocuria kristinae	3	132
Bacillus clausii	2	132
Bacillus invictae	2	132
Paenibacillus puldeungensis	2	132
Streptococcus gallolyticus	2	132
Macrococcus carouselicus	1	132
Paenibacillus wynnii	1	132
Bacillus licheniformis	22	40	[83]
Bacillus pumilus	2	40
Bacillus circulans	1	40
Pseudomonas putida	11	67	[84]
Pseudomonas fluorescens	21	67
Pseudomonas fragi	11	67
Pseudomonas moraviensis	1	67
Pseudomonas rhodesiae	1	67
Pseudomonas protegens	1	67
Pseudomonas lurida	1	67
Pseudomonas reactans	1	67
Pseudomonas chlororaphis	1	67
Pseudomonas poae	1	67
Pseudomonas chlororaphis	1	67
Pseudomonas koreensis	3	67
Pseudomonas graminis	2	67
Pseudomonas psychrophila	1	67
Pseudomonas veronii	1	67
Pseudomonas reinekei	1	67
Pseudomonas cedrina	1	67
Microbacterium resistens	1	30	[85]
Rhodococcus pyridinivorans	3	30
Bacillus cereus	7	30
Enterococcus faecalis	3	30
Enterococcus faecium	2	30
Cellulosimicrobium funkei	2	30
Brachybacterium nesterenkovii	1	30
Streptococcus macedonicus	1	30
Sphingobacterium arenae	1	30
Gordonia sputi	1	30
Pseudomonas fluorescens	92	148	[86]
Acinetobacter guillouiae	6	148
Acinetobacter haemolyticus	8	148
Comamonas thiooxydans	3	148
Enterobacter ludwigii	2	148
Hafnia paralvei	2	148
Janthinobacterium lividum	2	148
Lelliottia amnigena	5	148
Limnohabitans planktonicus	3	148
Pantoea anthophila	2	148
Pseudomonas putida	4	148
Rahnella aquatilis	2	148
Stenotrophomonas rhizophila	11	148
Pseudomonas azotoformans	8	54	[87]
Pseudomonas aeruginosa	5	54
Pseudomonas koreensis	4	54
Pseudomonas gessardii	3	54
Pseudomonas lurida	3	54
Pseudomonas monteilii	3	54
Pseudomonas brenneri	2	54
Pseudomonas lini	2	54
Pseudomonas kilonensis	2	54
Pseudomonas mucidolens	2	54
Pseudomonas proteolytica	2	54
Pseudomonas veronii	2	54
Pseudomonas antarctica	1	54
Pseudomonas chlororaphis	1	54
Pseudomonas congelans	1	54
Pseudomonas ficuserectae	1	54
Pseudomonas fluorescens	1	54
Pseudomonas jessenii	1	54
Pseudomonas lundensis	1	54
Pseudomonas mediterranea	1	54
Pseudomonas migulae	1	54
Pseudomonas panacis	1	54
Pseudomonas plecoglossicida	1	54
Pseudomonas salomonii	1	54
Pseudomonas simiae	1	54
Pseudomonas taetrolens	1	54
Pseudomonas trivialis	1	54
Pseudomonas Vancouverensis	1	54
Pseudomonas fluorescens	3	19	[88]
Staphylococcus epidermidis	1	19
Lactococcus lactis	2	19
Moraxella osloensis	1	19
Kocuria rosea	1	19
Klebsiella pneumoniae	1	19
Staphylococcus warneri	1	19
Klebsiella pneumonia	1	19
Bacillus cereus	148	380	[90]
Pseudomonas fluorescens	105.6	300	[91]
Pseudomonas putida	127.2	300
Pseudomonas aeruginosa	13.3	300
Pseudomonas oleovorans	6.7	300
Pseudomonas anguilliseptica	2.2	300
Pseudomonas lundensis	24.41	100	[92]
Pseudomonas fragi	20.47	100
Pseudomonas fluorescens	5.51	100
Pseudomonas taetrolens	2.36	100
Pseudomonas synxantha	0.79	100
Pseudomonas rhodesiae	3.15	100
Pseudomonas protegens	0.79	100
Pseudomonas libanensis	0.79	100
Pseudomonas koreensis	2.36	100
Pseudomonas gessardii	2.36	100
Pseudomonas extremorientalis	1.57	100
Pseudomonas antarctica	0.79	100
Rahnella aquatillis	0.79	100
Hafnia alvei	0.79	100
Lactococcus raffinolactis	0.79	100
Carnobacterium maltaromaticum	0.79	100
Buttiauxella brennerae	0.79	100
Bacillus cereus	0.79	100
Acinetobacter guillouiae	0.79	100
Pseudomonas fluorescens	23.1	100	[93]
Bacillus subtilis	23	139	[94]
Bacillus pumilis	7	139
Bacillus amyloliquefaciens	23	139
Bacillus licheniformis	16	139
Bacillus coagulans	1	139
Bacillus badius	1	139
Bacillus cereus	51	139
Geobacillus toebii	3	139
Bacillus cereus	89	340	[96]
Lysinibacillus fusiformis	80	229	[97]
Macrococcus caseolyticus	22	229
Micrococcus aloeverae	22	229
Bacillus licheniformis	15	229
Lysinibacillus varians	15	229
Bacillus toyonensis	14	229
Aestuariimicrobium kwangyangense	12	229
Carnobacterium divergens	12	229
Microbacterium hominis	9	229
Microbacterium lacticum	8	229
Kocuria indica	6	229
Micrococcus yunnanensis	6	229
Bacillus invictae	3	229
Gordonia paraffinivorans	3	229
Kocuria indica	2	229
Bacillus cereus	52	81	[98]
Bacillus flexus	1	22	[100]
Bacillus thuringiensis	1	22
Bacillus licheniformis	5	22
Bacillus cereus	13	22
Streptococcus agalactiae	4	29	[99]
Geobacillus stearothermophilus	2	29
Bacillus megaterium	3	29
Bacillus subtilis	3	29
Acinetobacter lwoffii	3	29
Bacillus cereus	69	239	[101]
Bacillus cereus	72	103	[102]
Bacillus subtilis	31	103
Pseudomonas fluorescens	88	116	[14]
Pseudomonas veronii	2	116
Pseudomonas psychrophila	1	116
Pseudomonas lundensis	8	116
Pseudomonas lactis	2	116
Pseudomonas azotoformans	3	116
Pseudomonas granadensis	4	116
Pseudomonas lurida	3	116
Pseudomonas rhizosphaerae	1	116
Pseudomonas extremorientalis	2	116
Pseudomonas rhodesiae	2	116
Bacillus cereus	12	35	[103]
Bacillus cereus	5	30
Bacillus toyonensis	66	107	[104]
Bacillus licheniformis	11	107
Lysinibacillus fusiformis	20	107
Bacillus pumilus	3	107
Bacillus flexus	3	107
Bacillus invictae	2	107
Bacillus megaterium	2	107
Pediococcus pentosaceus	1	36	[105]
Lactococcus lactis	2	36
Enterococcus faecium	4	36
Pediococcus acidilactici	1	36
Enterococcus faecalis	2	36
Levilactobacillus brevis	1	36
Staphylococcus aureus	1	36
Lactococcus garvieae	2	36
Bacillus licheniformis	2	36
Pediococcus acidilactici	2	36
Staphylococcus epidermidis	2	36
Streptococcus pyogenes	1	36
Lactiplantibacillus plantarum	1	36
Bacillus cereus	1	36
Staphylococcus aureus	1	36
Staphylococcus hyicus	2	36
Staphylococcus haemolyticus	1	36
Bacillus mycoides	1	36
Weizmannia coagulans	2	36
Microbacterium lacticum	2	36
Bacillus subtilis	1	36
Bacillus thuringiensis	1	36
Paenibacillus polymyxa	1	36
Leuconostoc mesenteroides	1	36
Bacillus cereus	52	131	[106]
Bacillus subtilis	26	131
Serratia marcescens	23	131
Bacillus subtilis	3	8	[107]
Bacillus velezensis	5	8
Pseudomonas fluorescens	19	50	[109]
Pseudomonas lurida	31	106	[110]
Pseudomonas gessardii	20	106
Pseudomonas psychrophile	16	106
Pseudomonas lundensis	6	106
Pseudomonas qingdaonensis	6	106
Microbacterium lacticum	1	31	[111]
Paenibacillus odorifer	1	31
Lactococcus raffinolactis	5	31
Pseudomonas libanensis	1	31
Bacillus licheniformis	1	31
Sphingobium yanoikuyae	1	31
Bacillus cereus	1	31
Staphylococcus aureus	1	31
Lactococcus lactis	4	31
Staphylococcus warneri	1	31
Klebsiella pneumoniae	1	31
Pseudomonas qingdaonensis	2	31
Microbacterium lacticum	1	31
Paenibacillus odorifer	1	31
Lactococcus raffinolactis	1	31
Pseudomonas libanensis	1	31
Bacillus licheniformis	3	31
Sphingobium yanoikuyae	1	31
Bacillus cereus	1	31
Staphylococcus aureus	1	31
Lactococcus lactis	1	31
Pseudomonas putida	2	208	[112]
Pseudomonas atacamensis	6	208
Pseudomonas fragi	21	208
Pseudomonas fluorescens	34	208
Pseudomonas lundensis	21	208
Pseudomonas shahriarae	11	208
Pseudomonas koreensis	7	208
Pseudomonas salmasensis	4	208
Pseudomonas canadensis	4	208
Pseudomonas veronii	4	208
Pseudomonas weihenstephanensis	3	208
Pseudomonas putida	1	208
Pseudomonas atacamensis	2	208
Pseudomonas fragi	1	208
Pseudomonas fluorescens	1	208
Pseudomonas lundensis	1	208
Pseudomonas shahriarae	2	208
Pseudomonas koreensis	1	208
Pseudomonas salmasensis	1	208
Pseudomonas canadensis	1	208
Pseudomonas veronii	2	208
Pseudomonas fluorescens	121	1430	[113]
Acinetobacter junii	73	1430
Pseudomonas brenneri	68	1430

Table A5.

Publication bias of the literature.

Analysis Type	Number of Studies (k)	Threshold (5k + 10)	Fail-Safe N Analysis	Egger’s Regression Test (p)
The overall prevalence	79	405	1664	0.00
Prevalence of Bacillus cereus	35	185	6659	0.13
Prevalence of Bacillus licheniformis	27	145	6450	0.34
Prevalence of Pseudomonas fluorescens	30	160	3361	0.37
Prevalence of Bacillus subtilis	21	115	4857	0.21
Prevalence of Bacillus pumilus	18	100	3877	0.30
Prevalence of Pseudomonas fragi	14	80	1733	0.05
Prevalence of Pseudomonas putida	14	80	1613	0.01
Prevalence of Lactococcus lactis	13	75	671	0.01
Prevalence of Pseudomonas lundensis	11	65	957	0.03
Prevalence of Pseudomonas aeruginosa	8	50	713	0.10
Prevalence of Enterococcus faecalis	8	50	162	0.01
Prevalence of Bacillus circulans	6	40	432	0.03
Prevalence of Enterococcus faecium	6	40	224	0.47
Prevalence of Pseudomonas gessardii	7	45	410	0.05
Prevalence of Pseudomonas psychrophila	6	40	218	0.03
Prevalence of Pseudomonas veronii	6	40	270	0.79

Table A6.

Characteristics of lipases from spoilage bacteria.

Host Strain	GenBank	Conservative Domain	Length /a.a	Weight /kDa	PI	Net Negative Charge	Net Positive Charge	Molar Absorptivity	Fat Coefficient	Average Hydrophilicity	Instability Index
Bacillus cereus	WCT66113.1	Abhydrolase super family	527	58.48	4.94	84	57	0.809	93.83	−0.306	23.42
	ETT75475.1	EstA super family	413	46.09	6.95	38	37	1.984	79.08	−0.355	26.43
	ETT75504.1	MenH super family/Abhydrolase super family	277	32.78	5.36	45	33	2.314	86.53	−0.481	40.24
	RXG07039.1	SGNH_hydrolase super family	293	32.97	8.77	35	39	1.213	94.4	−0.273	26.95
	ALZ63309.1	CusA super family/Aes super family	402	44.97	6.38	40	37	1.786	100.87	−0.018	37.22
	EOP28808.1	MenH super family	363	41.94	7.73	44	45	0.905	87.22	−0.287	34.83
	EOP56421.1	EstA super family/Abhydrolase super family	534	58.91	8.12	54	56	1.568	79.94	−0.321	31.69
Bacillus licheniformis	WKU31742.1	CVT17 super family	479	53.95	5.85	71	63	1.2	76.97	−0.741	22.19
	TWO06542.1	EstA super family	487	53.67	8.98	48	54	1.292	77.97	−0.351	25.25
	ARW52538.1	SGNH_hydrolase super family	212	24.12	5.54	33	28	0.851	90.14	−0.378	34.4
	VEH78556.1	SGNH_hydrolase super family	250	28.47	4.97	44	34	0.862	87.72	−0.416	35.75
Bacillus subtilis	CUB56573.1	Abhydrolase super family/CVT17 super family	416	47.24	5.82	66	57	1.118	69.11	−0.786	33.48
	WP_251187522.1	CVT17 super family	487	55.15	5.63	73	57	0.794	84.5	−0.597	30.83
	CON51271.1	DAP2 super family	257	29.74	9.68	26	36	1.291	73.54	−0.506	52.57
	KAF2426737.1	Lip2 super family	276	31.23	8.97	22	29	0.871	72.5	−0.226	40.41
Bacillus pumilus	OUZ06619.1	CVT17 superfamily	496	55.82	8.47	65	68	0.864	82.96	−0.55	40.05
Pseudomonas fluorescens	POH41489.1	Abhydrolase super family	716	81.58	7.23	83	82	1.086	78.39	−0.468	50.58
	VEF10173.1	COG4757 super family	329	36.59	7.15	36	36	2.17	82.19	−0.29	42.06
	ROO16718.1	Lip2 super family	776	86.82	6.05	93	76	1.2	79.96	−0.4	47.75
Pseudomonas fragi	ARQ76605.1	Abhydrolase super family	676	75.62	6.6	74	68	1.006	82.69	−0.281	48.5
	VFB20546.1	Abhydrolase super family/HemolysinCabind super family/COG2931 super family	616	64.93	4.65	71	36	1.244	82.86	−0.17	26.01
Pseudomonas putida	ANC02924.1	Abhydrolase super family	328	35.87	5.06	43	26	1.454	92.29	−0.1	56.56
	SUD66897.1	PldB super family	338	37.99	8.82	28	31	1.922	98.17	−0.063	54.12
	BAW21301.1	COG4782 super family	329	36.56	8.61	40	44	1.43	83.95	−0.48	30.04
Pseudomonas lundensis	AOZ11414.1	Axe1 super family/Abhydrolase super family	320	35.04	4.92	43	25	1.413	92.72	−0.027	46.82
Enterococcus faecalis	BDQ58792.1	Aes super family	337	37.85	8.69	30	34	1.356	85.64	−0.167	34.34
	GEJ62098.1	CwlO1 super family/CHAP super family	482	50.9	5.02	63	51	0.882	71.37	−0.53	40.43
	VTT41658.1	CVT17 super family	465	52.53	5.21	63	50	1.261	79.91	−0.535	28.71
Enterococcus faecium	OAQ37576.1	Abhydrolase super family	315	35.58	4.89	49	34	1.275	86.41	−0.357	40.67
	GJG91484.1	Aes super family	337	38.33	6.26	42	40	1.238	79.88	−0.383	30.72
	XLE04444.1	CVT17 super family	418	46.97	6.96	56	56	1.083	85.93	−0.47	23.8
Lactococcus lactis	WP_109991484.1	Abhydrolase super family	415	45.63	6.11	49	45	1.323	77.54	−0.462	23.53
	KRO23992.1	FrmB super family	258	29.67	5.34	32	25	2.151	79.77	−0.359	34.57
	AIS02893.1	KxYKxGKxW_sig superfamily/Abhydrolase super family	561	58.5	4.1	38	14	1.048	92.39	0.082	37.18
	PFG86633.1	SGNH_hydrolase super family	428	46.69	9.22	39	46	1.443	86.8	−0.264	23.34
	SPS11769.1	Aes super family/DAP2 super family	257	28.84	5.79	32	28	1.769	94.94	−0.342	26.13
	BFW28476.1	SGNH_hydrolase super family	224	25.81	4.85	36	22	0.733	82.72	−0.515	34.14

Figure A1.

Three-dimensional structural models of lipases from major dairy spoilage bacteria. (A) Homology model of lipase from Bacillus. (B) Homology model of lipase from Pseudomonas. (C) Homology model of lipases from other bacteria. The secondary structure elements are color-coded: cyan for α-helices, red for β-sheets, and purple for loops.

Author Contributions

Formal analysis, Y.L.; Methodology, Y.L.; Visualization, Y.L. and Q.W.; Data curation, Q.W.; Conceptualization, J.W., P.K.M., D.W. and X.W.; Validation, J.W., P.K.M., D.W. and X.W.; Writing—original draft, Y.L., Writing—review & editing, Q.W., J.W., P.K.M., D.W. and X.W. All authors have read and agreed to the published version of the manuscript.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

The original contributions presented in the study are included in the article. Further inquiries can be directed to the corresponding authors.

Conflicts of Interest

Author Jinling Wang, Xu Wang and Dongmei Wang were employed by the company Inner Mongolia Yili Industrial Group Co., Ltd. The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Funding Statement

This research was funded by National Center of Technology Innovation for Dairy, grant number 2024-JSGG-009.