Abstract
Listeria monocytogenes, a ubiquitous bacterium in nature, can lead to human listeriosis through food consumption. Listeriosis is a rare, preventable, and treatable foodborne disease but can cause hospitalizations and fatalities. We reviewed the literature published in China to better understand the prevalence of L. monocytogenes in food products, incidence of human listeriosis, and characteristics of L. monocytogenes strains in China. The average prevalence of L. monocytogenes in Chinese food products in 28 provinces was 4.42%, with the highest prevalence of 8.91% in meat-poultry products, followed by aquatic animals, Chinese salad and salad, rice and flour products, and so on. Two hundred fifty-three invasive listeriosis cases were reported from 2011 to 2016 in 19 provinces, and the overall case-fatality rate was 25.7% with no deaths reported of pregnant women and children. L. monocytogenes strains were generally susceptible to most antibiotics, with ampicillin and penicillin G still effective in treatment. The predominant sequence types (STs) in food were ST9 and ST8, while clinically ST87 was most common ST in China. The national human listeriosis pilot surveillance started in 2013, and a total of 133 listeriosis cases have been collected until now. On the basis of the surveillance program, further research should be conducted to uncover the reason for the prevalence and pathogenic mechanism of the highly epidemiological hypervirulent ST87 strains in China.
Introduction
L
The national microbiological food safety surveillance network in China was established in 2010, including surveillance of L. monocytogenes in different food categories (Pei et al., 2015). However, listeriosis is not yet a notifiable disease in China, and the national foodborne disease surveillance system was established in 2011 (Li et al., 2017), which has currently developed into a comprehensive and composite program (Liu et al., 2017). Listeriosis sporadic case surveillance was included in the national surveillance plan in 2013 in several provinces. We reviewed relevant published articles to gain an understanding of the prevalence of L. monocytogenes in food products, the incidence of human listeriosis, and phenotypic and genetic characteristics of L. monocytogenes strains in China.
Isolation Methods of L. monocytogenes from Food and Human Samples
For food samples, the isolation of L. monocytogenes is generally conducted in accordance with China national food safety standards GB4789.30 (M.O.H.C, 2010; N.H.F.P.C., 2016), and according to the requirements of national food safety standard about pathogen limit in food (GB29921-2013), L. monocytogenes should not be detected in cooked meat products or ready-to-eat meat products (N.H.F.P.C., 2013). For qualitative analysis, food samples are homogenized and incubated in 225 mL of Listeria Enrichment Broth I culture (LB1) for 24 h. Then, LB1 enrichment culture is added to Listeria Enrichment Broth II culture (LB2) for a second 24 h enrichment. Next, Listeria Chromogenic Agar and PALCAM can be used together to isolate presumptive colonies, which can be purified on tryptic soy agar with yeast extract and identified via Gram staining and biochemical confirmation.
As for quantitative analysis, food samples are incubated in Buffered Peptone Water or Listeria Enrichment broth (LB) without supplementation to conduct the plate count method, or samples can be incubated in LB1 to conduct the most probable number (MPN) method. This national examination standard used in China is slightly different from other countries, for instance, the sample is enriched for 48 h at 30°C in LB only once according to the bacteriological analytical manual from the Food and Drug Administration (Hitchins, 2001) and the enrichment broth is Fraser broth according to the International Organization of Standards 11290 method (Donnelly, 2002).
For clinical specimens obtained from sterile sites, such as blood, cerebrospinal fluid (CSF), and amniotic fluid, L. monocytogenes can be recovered from chromogenic agar after enrichment in automatic blood culture for 7 d. With regard to nonsterile clinical specimens, such as stool samples, the isolation of L. monocytogenes requires LB1 and LB2 selective enrichment at 30°C, and then LB2 enrichment cultures are plated on selective agar plates (Schuchat et al., 1991; Zhang, 2010).
Prevalence of L. monocytogenes in Food Products
We reviewed the relevant literature mostly published in Chinese and some in English about the prevalence of L. monocytogenes in different food products in China. We retrieved articles by directly searching two primary academic electronic databases, including the China National Knowledge Infrastructure (
Furthermore, food products sampled and detected were assigned to one of 14 food categories: meat-poultry products, rice and flour products, aquatic animals, vegetables, dairy, bean products, Chinese salad and salad, ready-to-eat food, infant food, bakery products, fresh juice, cold drinks, ice cream, and eggs. The average prevalence was 5.54% for L. monocytogenes by food categories, ranging from 0.00% to 8.91% (Table 2). L. monocytogenes strains were isolated from almost all food categories except infant food, bakery products, and eggs. The prevalence of L. monocytogenes in meat-poultry products was the highest at 8.91%, followed by aquatic animals (5.72%), Chinese salad and salad (3.65%), rice and flour products (2.97%), vegetables (2.84%), and dairy (1.82%). Meat poultry products, including pork, beef, mutton, and poultry, can also be grouped into raw meat, raw poultry, and cooked meat. The prevalence of L. monocytogenes in cooked meat (4.61%) was significantly lower than raw meat (11.68%) and raw poultry (11.59%). Aquatic animal products can be divided into frozen aquatic products and raw aquatic products. The prevalence of L. monocytogenes in frozen aquatic products (5.41%) was higher than raw aquatic products (2.90%).
The recovery rate of L. monocytogenes in most food categories from 2008 to 2016, shown in Table 2, was higher than a previous literature review from 2000 to 2007, perhaps due to enhanced detection ability and improved methods (Chen et al., 2009b). In other studies, the average prevalence of L. monocytogenes in retail raw foods in 14 provinces was 20.0%, with the MPN values ranging from 0.3 to 110 MPN/g; the edible mushroom, Flammulina velutipes, and quick-frozen meat were most frequently contaminated with L. monocytogenes, with a prevalence that reached 55.6% and 54.5% (Wu et al., 2015). Taken together, previous studies in China (Zhou and Jiao, 2006; Chen et al., 2015; Yang et al., 2016) showed that meat products, aquatic products, and Chinese salad are the main sources of L. monocytogenes contamination, while the risk of ready-to-eat food cannot be neglected.
Compared with other countries, about 2.3–25.4% of salad samples were positive for L. monocytogenes in Chile (Cordano and Jacquet, 2009), the United Kingdom (Little et al., 2007), and Malaysia (Ponniah et al., 2010), while the recovery rate in salad and vegetables was relatively lower in China. The prevalence of L. monocytogenes in food products ranged from 0.4% to 14% among different food categories and countries, seafood, and ready-to-eat food relatively high-risk food products (Norrung et al., 1999; Latorre et al., 2007; Jalali and Abedi, 2008; Lambertz et al., 2012; Martinez-Rios and Dalgaard, 2018).
Incidence of L. monocytogenes from Human Listeriosis
L. monocytogenes was first isolated by Murray in 1924 (Hof, 2003). A case of neonatal L. monocytogenes meningitis was first reported in China in the 1960s (Ding et al., 1964), and listeriosis was first recognized as a foodborne pathogen in the early 1980s (Schlech et al., 1983).
There was no national clinical listeriosis surveillance in China until 2013, but many Chinese medical journals have published listeriosis case reports. One previous systematic review was conducted to illustrate the clinical and epidemiologic characteristics of listeriosis in China through literature retrieval from 1964 to 2010, and the results showed that 52% of clinical cases were pregnancy-related patients; the most common clinical manifestation was septicemia (46%), followed by CNS infection (31%) and focal infection or gastroenteritis (23%), and the over-all case-fatality rate was 26% (Feng et al., 2013).
We reviewed human listeriosis cases reported in China from 2011 to 2016. There is no standard diagnosis and treatment guideline for listeriosis in China, considering the definition of the Centers for Disease Control and Prevention (CDC) (Silk et al., 2012) and the World Health Organization (WHO) (de Noordhout et al., 2014). We defined invasive listeriosis as isolation of L. monocytogenes strains from a normally sterile site or from products of conception. We counted a maternofetal infection as one case, including pregnant women and cases in newborn babies during the first 4 weeks of life. We also defined stillbirth and abortion as deaths. We included a total of 60 articles describing invasive listeriosis for study. All listeriosis cases were categorized into two main population groups: perinatal cases, including pregnant women, fetuses, and neonates (<28 d) and nonperinatal cases, including immunocompromised people, elderly people, children, and healthy people. Two hundred fifty-three invasive listeriosis cases were reported in 19 provinces from 2011 to 2016, in which 163 patients (64.4%) were perinatal cases and 90 patients (35.6%) were nonperinatal cases, including 64 immune-comprised patients, 18 healthy people, 4 elderly people, and 4 children (<14 years old). The most common clinical manifestation of nonperinatal cases was septicemia (62.2%), followed by CNS infection (36.7%) and pneumonia (1.1%). The overall case-fatality rate was 25.7%, while no deaths of pregnant women and children were reported. The rates of perinatal and nonperinatal cases were 26.4% and 24.4%, respectively.
All identified 253 listeriosis cases were sampled and diagnosed by clinical laboratory examination, and biological specimens included blood, CSF, uterine swabs, placenta swabs, nasopharyngeal swabs, sputum, marrow, amniotic fluid, lung fluid, and trachea secretion. A total of 63.6% of CNS infection cases were positive for L. monocytogenes in CSF samples, and other cases were confirmed by isolation of strains from blood, and followed by CSF biochemical and cytological analyses. A previous case study and literature review from 1964 to 1997 in England also revealed that only 31% of CNS infection cases were positive by Gram stain (Mylonakis et al., 1998); clinical features or characterization of CSF alone cannot determine the causative pathogen of bacterial meningitis, as it is hard to distinguish L. monocytogenes from other types of meningitis (Phan et al., 2017), so the most reliable diagnostic method is blood or CSF culture, but currently, CSF culture is still not the routine test utilized in most hospitals in China. L. monocytogenes meningitis was normally diagnosed in patients with bacterial meningitis in whom L. monocytogenes meningitis was isolated from CSF, blood culture, or both in China (Jiang et al., 2012), according to the diagnosis standard used in a previous descriptive prospective study (Amaya-Villar et al., 2010).
Most cases of listeriosis are sporadic; only one outbreak within a blood ward with three patients was reported in Beijing. The WHO (de Noordhout et al., 2014) and only a few countries (Kemmeren et al., 2006; Lake et al., 2010; Havelaar et al., 2012) have assessed the listeriosis burden. The CDC estimated that 1600 listeriosis cases occur each year and that about 260 die; the incidence per 100,000 population was 0.26 in the United States (Crim et al., 2014), but the burden and costs of specific foodborne disease are still undetermined to date, in China.
Distribution of Serotypes
L. monocytogenes is a highly heterogeneous species and can be divided into 13 serotypes, including 1/2a, 1/2b, 1/2c, 3a, 3b, 3c, 4a, 4ab, 4b, 4c, 4d, 4e, and 7 (Allerberger, 2003). L. monocytogenes strains can be divided into serotypes on the basis of somatic (O) and flagellar (H) (Gasanov et al., 2005). Although serotyping is a classic subtyping method with limited discrimination, serology is also a useful tool for epidemiological studies and tracking contamination. The national food safety risk surveillance showed that 226L. monocytogenes strains, isolated from ready-to-eat food from 2007 to 2009 in China were identified as serotypes 1/2a, 1/2b, 1/2c, and 4b, and the rate of human infection-related serotypes 1/2a, 1/2b, and 4b was relatively high at 87.61% (Ma et al., 2017). Other food surveillance data showed that 212 isolates from different food products in 12 provinces from 2000 to 2008 were identified into seven serotypes, and 1/2c, 1/2a, and 1/2b were the most frequent serotypes (Wang et al., 2012). Considering other previous studies carried out in provinces, we can conclude that 1/2a, 1/2b, 1/2c, and 4b are the dominant serotypes for food strains in China (Wang et al., 2012; Zhang et al., 2013; Yu and Jiang, 2014).
Three serotypes of L. monocytogenes, namely 4b, 1/2a, and1/2b, caused the majority of clinical cases (Gilot et al., 1996). The Italian surveillance network reported that the most frequent serotypes identified are 1/2a, 4b, and 1/2b in 251 clinical strains (Pontello et al., 2012). One previous study in China showed that 28 clinical strains were mainly divided into 1/2b, 1/2a, and 4b (Wang et al., 2015), and another study reported that 79% of strains were 4b, followed by 1/2a and 1/2c (Zhou et al., 2009). The latter two studies are not representative of the national prevalence of serotypes in China, because the limited number of strains and isolation regions may introduce bias.
Antibiotic Susceptibility and Treatment
L. monocytogenes strains are generally susceptible to most antibiotics, and the clinical and laboratory standard institutes select three agents for priority testing: penicillin, ampicillin, and trimethoprim-sulfamethoxazole (CLSI, 2014). A total of 1687 food strains from the National Food Safety Surveillance Network from 2003 to 2009 were analyzed using the E-test or broth microdilution susceptibility testing (Zhang et al., 2006; Yang et al., 2008; Zhao et al., 2012). At most, 15 antibiotics were examined. The average resistant rate was 6.82%, and tetracycline resistance was most prevalent, accounting for 5.69%, while antibiotics resistance also included doxycycline, erythromycin, ciprofloxacin, chloramphenicol, and streptomycin. All strains were susceptible to the three first choice antibiotics, although two strains were resistant to tetracycline, erythromycin, and chloramphenicol. The rates and profiles of antibiotic resistance were different among the provinces and food products.
Some studies showed high prevalence of antibiotic resistance in L. monocytogenes from food in China, up to about 40–70% (Yan et al., 2010; Yu and Jiang, 2014; Chen et al., 2015; Wu et al., 2015), partly owing to the inclusion of cefotaxime. The studies on the clinical strains revealed that all strains were susceptible to penicillin, ampicillin, and trimethoprim-sulfamethoxazole, and one previous study found that all 58 tested strains were resistant to tetracycline (Zhou et al., 2009; Wang and Chen, 2016). Other countries showed similar results as our national surveillance, such as 1–5% tetracycline resistance in the United Kingdom (Threlfall et al., 1998) and 2% resistance in France (Granier et al., 2011). Until now, L. monocytogenes strains in China were susceptible to most antibiotics commonly used in veterinary and human listeriosis treatment. Considering that after the first resistant strains isolated from a patient were found in 1988 (Poyart-Salmeron et al., 1990), more antibiotic-resistant strains have been reported, while continued surveillance of emerging antimicrobial resistance of L. monocytogenes is still important for diagnosis and treatment.
Third-generation cephalosporins are generally ineffective against L. monocytogenes and should not be used therapeutically (Allerberger and Dierich, 1992). The recommended treatment for L. monocytogenes meningitis is high-dose intravenous ampicillin or penicillin G. Alternatively, a combination of trimethoprim and sulfamethoxazole can be given. Fluoroquinolones, aminoglycosides, and rifampicin are reported to have some antibacterial effects on L. monocytogenes (Chau et al., 2010; Arslan et al., 2015). Treatment should last for at least 2 weeks in adults and should be extended to 3 to 4 weeks for neonatal listeriosis and up to 6 weeks for immunosuppressed patients (Braden, 2003; Roed et al., 2012). Delay treatment existed because cephalosporins were mainly empirically used in the treatment of CNS infections in China. Taken together, isolation of L. monocytogenes and antibiotic susceptibility tests are vital for treatment of listeriosis in China.
Pathogenesis and Genetic Heterogeneity
L. monocytogenes can be divided into at least four evolutionary lineages (I, II, III, and IV) (Orsi et al., 2011) and four PCR serogroups (Doumith et al., 2004). Other common genetic analyses can include pulsed-field gel electrophoresis (Halpin et al., 2010), multilocus sequence typing (Ragon et al., 2008), virulence gene detection (Vazquez-Boland et al., 2001), multivirulence locus sequence typing (Filipello et al., 2017), and whole genome sequencing (WGS) (Kwong et al., 2016). Many pathogenicity studies have been applied in an effort to describe the pathogenic mechanism of L. monocytogenes in China, including detection of virulence genes, mouse virulence assays, cell assays, and genomic analysis (Zhou and Jiao, 2006; Chen et al., 2009a; Wang et al., 2012, 2015; Zhang et al., 2016).
A literature review showed that the most predominant sequence types (STs) were ST9 and ST8 of food isolates in China, and lineage II contained the largest number of STs and strains, which displayed a divergent structure between subgroups through phylogenetic analysis. When it comes to human listeriosis strains, ST87, ST3, and ST7 were the main STs, which mostly were allocated into lineage I. Previous researches showed that lineage I, including serotypes 1/2b, 3b, 4b, 4d, and 4e, has been linked to human listeriosis outbreaks and sporadic cases; lineage II, including serotypes 1/2a, 1/2c, 3a, and 3c, has been isolated from humans and animals but predominantly from food products; and lineage III, including 4a and 4c, has been proved to have a strong association with animals (Wiedmann et al., 1997; Orsi et al., 2011; Lomonaco et al., 2015). Outbreaks and sporadic cases have frequently and recently been related to lineage II (Camargo et al., 2017). It is important to declare that ST87, the dominant ST in clinical listeriosis in China (Huang et al., 2015), was seldom linked to human infection cases in other countries (Ragon et al., 2008; Knabel et al., 2012; Jensen et al., 2016; Maury et al., 2016), with only two outbreaks associated with ST87 strains in Spain (Perez-Trallero et al., 2014).
Most virulence genes are highly prevalent in L. monocytogenes strains from food in China (Yan et al., 2016; Du et al., 2017). So far, no clear correlation between origin or strain characteristics and virulence has been observed, so all strains of L. monocytogenes should be considered to be potentially capable of causing human disease (Allerberger, 2003). Recent research using WGS showed that a novel L. monocytogenes genomic island, denoted as LGI-2, was identified, and strains in China contained extensive diversification (Zhang et al., 2016). Another study conducted in France revealed that the cellobiose family phosphotransferase system was specific to certain hypervirulent complex clones (Maury et al., 2016). Now WGS on L. monocytogenes isolates is enhancing listeriosis surveillance, outbreak detection, and pathogenesis (Jackson et al., 2016). Research on the phylogenetic distribution of hyper virulence factors should be launched in China to discover the evolutional laws and divergent characteristics of L. monocytogenes in China.
Conclusions
This review provides only the general situation on the epidemiology of L. monocytogenes in China. L. monocytogenes occurs naturally in the environment, in food processing, and will inevitably contaminate the food we eat. We are facing severe challenges in this epidemiological investigation, including long incubation periods, difficulty in investigation of food exposure, and compliance of cases. The high prevalence of L. monocytogenes from food in China will definitely lead to more serious consequences. There is an urgent need for the government to enhance human listeriosis surveillance, with implementation of control measures in food processing facilities and health education to susceptible populations, especially pregnant women, to decrease the incidence of listeriosis. In the meantime, research should be carried out to explore the mechanism of the highly epidemiological hyper virulent STs that only exist in China.
Footnotes
Acknowledgment
This work was supported by the Key Research and Development of Food Safety of the Ministry of Science and Technology of China (Grant No. SQ2017YFC1601503).
Compliance with Ethics Guidelines
This article is a review article and does not involve a research protocol requiring approval by the relevant institutional review board or ethics committee.
Disclosure Statement
No competing financial interests exist.
