Foodborne Klebsiella pneumoniae: Virulence Potential, Antibiotic Resistance, and Risks to Food Safety

ABSTRACT

Gastrointestinal carriage of Klebsiella pneumoniae is a predisposing factor for liver abscess in several Asian countries. To determine whether hypervirulent K. pneumoniae in the gut may be transmitted through food, we screened a range of raw and ready-to-eat retail food by culture and recovered K. pneumoniae in 21% (147 of 698) of samples tested. Based on PCR, no K. pneumoniae isolates carried the rmpA gene linked to community-acquired pyogenic liver abscess, providing no evidence of a link between food and liver disease. However, phenotypic resistance to multiple antibiotic classes was seen through disk diffusion tests, and carriage of genetic elements (wcaG and capsule types K1, K2, and K54) associated with increased virulence (8%, 11 of 147) was observed by PCR. Multidrug-resistant isolates were from raw vegetables, chicken or pork liver, and a ready-to-eat poultry dish; one multidrug-resistant K. pneumoniae isolate from raw bean sprouts was resistant to a third-generation cephalosporin (ceftriaxone). Although K. pneumoniae may be present in food without causing harm, we found isolates belonging to the K1 capsular serotype coexisting with the wcaG gene, one also conferring multidrug resistance. K. pneumoniae that carry antibiotic resistance genes, regardless of pathogenicity, may increase the available genetic pool of resistance along the food chain. Hygienic food handling practices are necessary to lower risks of acquiring K. pneumoniae and other opportunistic pathogens. .

HIGHLIGHTS

One hypervirulent clone, sequence type 283, accounts for a large proportion of invasive Streptococcus agalactiae isolated from humans and diseased tilapia in Southeast Asia

Background

In 2015, Singapore had the first and only reported foodborne outbreak of invasive disease caused by the group B Streptococcus (GBS; Streptococcus agalactiae). Disease, predominantly septic arthritis and meningitis, was associated with sequence type (ST)283, acquired from eating raw farmed freshwater fish. Although GBS sepsis is well-described in neonates and older adults with co-morbidities, this outbreak affected non-pregnant and younger adults with fewer co-morbidities, suggesting greater virulence. Before 2015 ST283 had only been reported from twenty humans in Hong Kong and two in France, and from one fish in Thailand. We hypothesised that ST283 was causing region-wide infection in Southeast Asia.

Methodology/Principal findings

We performed a literature review, whole genome sequencing on 145 GBS isolates collected from six Southeast Asian countries, and phylogenetic analysis on 7,468 GBS sequences including 227 variants of ST283 from humans and animals. Although almost absent outside Asia, ST283 was found in all invasive Asian collections analysed, from 1995 to 2017. It accounted for 29/38 (76%) human isolates in Lao PDR, 102/139 (73%) in Thailand, 4/13 (31%) in Vietnam, and 167/739 (23%) in Singapore. ST283 and its variants were found in 62/62 (100%) tilapia from 14 outbreak sites in Malaysia and Vietnam, in seven fish species in Singapore markets, and a diseased frog in China.

Conclusions

GBS ST283 is widespread in Southeast Asia, where it accounts for a large proportion of bacteraemic GBS, and causes disease and economic loss in aquaculture. If human ST283 is fishborne, as in the Singapore outbreak, then GBS sepsis in Thailand and Lao PDR is predominantly a foodborne disease. However, whether transmission is from aquaculture to humans, or vice versa, or involves an unidentified reservoir remains unknown. Creation of cross-border collaborations in human and animal health are needed to complete the epidemiological picture.

An outbreak due to a bacterium called Streptococccus agalactiae in Singapore in 2015 was caused by a clone called ST283, and was associated with consumption of raw freshwater-fish. It was considered unique as it was the only reported foodborne outbreak of this bacterium. Our new data show that invasive ST283 disease is far from unique. ST283 has been causing disease in humans and farmed fish in SE Asian countries for decades. Reports of ST283 are almost absent outside Asia. We suspect that human ST283 is fishborne in other Asian countries, as it was in Singapore, but we haven’t looked at this yet. We don’t know where ST283 originally came from; it may have been transmitted from humans to fish, or come from another animal. More studies are needed to determine ST283’s geographical extent and burden of disease, as well as its origin, how it is transmitted, and what enables it to be so aggressive. We may then be able to interrupt transmission, to the benefit of fish, farmers, and the general public.

Microbial Quality and Safety of Sushi Prepared with Gloved or Bare Hands: Food Handlers' Impact on Retail Food Hygiene and Safety

Food handlers play an important role in retail food hygiene and safety. This study was conducted to better understand the impact of food handlers on the microbiological quality and safety of sushi and ingredients handled by gloved and bare hands. At retail premises, food handlers were asked to prepare a batch of sushi with raw fish followed by a batch of sushi with cooked ingredients. Food (sushi and ready-to-eat ingredients), hand, and glove samples were collected for analysis of overall microorganisms (standard plate counts) and targeted foodborne bacteria: Staphylococcus aureus, Bacillus cereus, Escherichia coli, Salmonella, and Listeria monocytogenes. Results suggested that cross-contamination was more prevalent at premises where bare hands were used to prepare sushi. When bare hands were used, significantly higher standard plate counts were obtained from samples of cooked rice (2.3 to 4.9 log CFU/g) and sushi (2.8 to 6.9 log CFU/g) and the prevalence of S. aureus in samples was higher on food (21.7%, 28 of 129 samples) and hands (30%, 18 of 60 samples) ( P < 0.05). Glove changing in combination with hand washing minimized cross-contamination during sushi preparation as indicated by the lower prevalence of S. aureus (0%, 0 of 28 samples) and total targeted foodborne bacteria (3.6%, 1 of 28 samples) on the gloves of food handlers who changed gloves and washed their hands compared with those handlers who did not don new gloves. Repeated use of dishcloths could be a cause of cross-contamination, and the prevalence of total targeted foodborne bacteria was significantly higher on hands dried with dishcloths (64.7%, 11 of 17 samples) than on hands dried with paper towels (12.5%, 1 of 8 samples) ( P < 0.05). The prevalences of B. cereus, L. monocytogenes, and Salmonella in the 356 food samples were 5.1% (18 samples), 0.8% (3 samples), and 0%, respectively. Improvements to hand washing, hand drying, and glove changing practices are needed to lower the occurrence of cross-contamination during sushi preparation.

Survival of an emerging foodborne pathogen: Group B Streptococcus (GBS) serotype III sequence type (ST) 283-under simulated partial cooking and gastric fluid conditions

Group B Streptococcus (GBS) was previously not known to be transmitted through food, but an outbreak investigation in Singapore in 2015 documented for the first time an association between GBS Type III Sequence Type 283 infection and consumption of raw fish dishes. As very little is known about the survival of GBS during heat treatment and the stomach transit, its survival under simulated conditions was studied, in comparison with that of Escherichia coli O157:H7 and Listeria monocytogenes. The mean D-values of four GBS strains ranging from 0.72 to 0.88 min in neutral pH tryptone soy broth at 56.4 °C and 0.44-1.43 min at pH 2.35 at 37 °C in simulated gastric fluid, were significantly lower (p < 0.05) than those of E. coli O157:H7 and L. monocytogenes. This study suggests possible factors other than acid or heat resistance of GBS to be instrumental to its pathogenicity.

Sick pets as potential reservoirs of antibiotic-resistant bacteria in Singapore

An analysis of 186 diagnostic reports collected from a veterinary clinic in Singapore between 2014 to 2016 showed that sick companion animals can carry bacteria that are of significance to human health. Among the 186 specimens submitted, 82 showed polymicrobial growth (45%, 82/186) and in total, 359 bacteria were isolated. Of the 359 bacteria reported, 45% (162/359) were multi-drug resistant and 18% (66/359) were extended-spectrum-beta-lactamase species. Resistance to broad-spectrum antibiotics were also observed among individual species. Namely, methicillin-resistance among Staphylococcus pseudintermedius (63%, 32/51) and Staphylococcus aureus (50%, 4/8); fluoroquinolone-resistance among Escherichia coli (40%, 17/42) and carbapenem-resistance among Klebsiella pneumoniae (7%, 2/30) were noted. Our analysis suggests that sick pets may contribute to the pool of clinically relevant antibiotic-resistant bacteria and play a role in the spread of antibiotic resistance in Singapore. A more extensive study to better understand the extent of distribution and the factors affecting transmission of antibiotic-resistant bacteria to and from pets is necessary.

2015 Epidemic of Severe Streptococcus agalactiae Sequence Type 283 Infections in Singapore Associated With the Consumption of Raw Freshwater Fish: A Detailed Analysis of Clinical, Epidemiological, and Bacterial Sequencing Data

Background

Streptococcus agalactiae (group B Streptococcus [GBS]) has not been described as a foodborne pathogen. However, in 2015, a large outbreak of severe invasive sequence type (ST) 283 GBS infections in adults epidemiologically linked to the consumption of raw freshwater fish occurred in Singapore. We attempted to determine the scale of the outbreak, define the clinical spectrum of disease, and link the outbreak to contaminated fish.

Methods

Time-series analysis was performed on microbiology laboratory data. Food handlers and fishmongers were screened for enteric carriage of GBS. A retrospective cohort study was conducted to assess differences in demographic and clinical characteristics of patients with invasive ST283 and non-ST283 infections. Whole-genome sequencing was performed on human and fish ST283 isolates from Singapore, Thailand, and Hong Kong.

Results

The outbreak was estimated to have started in late January 2015. Within the study cohort of 408 patients, ST283 accounted for 35.8% of cases. Patients with ST283 infection were younger and had fewer comorbidities but were more likely to develop meningoencephalitis, septic arthritis, and spinal infection. Of 82 food handlers and fishmongers screened, none carried ST283. Culture of 43 fish samples yielded 13 ST283-positive samples. Phylogenomic analysis of 161 ST283 isolates from humans and fish revealed they formed a tight clade distinguished by 93 single-nucleotide polymorphisms.

Conclusions

ST283 is a zoonotic GBS clone associated with farmed freshwater fish, capable of causing severe disease in humans. It caused a large foodborne outbreak in Singapore and poses both a regional and potentially more widespread threat.

Microbial survey of ready-to-eat salad ingredients sold at retail reveals the occurrence and the persistence of Listeria monocytogenes Sequence Types 2 and 87 in pre-packed smoked salmon

BACKGROUND

As the preparation of salads involves extensive handling and the use of uncooked ingredients, they are particularly vulnerable to microbial contamination. This study aimed to determine the microbial safety and quality of pre-packed salads and salad bar ingredients sold in Singapore, so as to identify public health risks that could arise from consuming salads and to determine areas for improvement in the management of food safety.

RESULTS

The most frequently encountered organism in pre-packed salad samples was B. cereus, particularly in pasta salads (33.3%, 10/30). The most commonly detected organism in salad bar ingredients was L. monocytogenes, in particular seafood ingredients (44.1%, 15/34), largely due to contaminated smoked salmon. Further investigation showed that 21.6% (37/171) of the pre-packed smoked salmon sold in supermarkets contained L. monocytogenes. Significantly higher prevalence of L. monocytogenes and higher Standard Plate Count were detected in smoked salmon at salad bars compared to pre-packed smoked salmon in supermarkets, which suggested multiplication of the organism as the products move down the supply chain. Further molecular analysis revealed that L. monocytogenes Sequence Type (ST) 2 and ST87 were present in a particular brand of pre-packed salmon products over a 4-year period, implying a potential persistent contamination problem at the manufacturing level.

CONCLUSIONS

Our findings highlighted a need to improve manufacturing and retail hygiene processes as well as to educate vulnerable populations to avoid consuming food prone to L. monocytogenes contamination.

MICROBIOLOGICAL SAFETY ASSESSMENT AND RISK MITIGATION OF INDIAN ROJAK (DEEP FRIED READYTO-EAT FOOD) IN SINGAPORE

We conducted a microbiological assessment of Indian Rojak, a popular deep fried food in Singapore to evaluate its overall microbial quality, assess the effectiveness of reheating and identify key food items that could contribute to the microbial load of the dish. In 2009, an outbreak of foodborne illness associated with this food led to 154 reported cases of acute gastroenteritis, 48 were hospitalized and 2 died. Vibrio parahaemolyticus was isolated from the patients. We evaluated 455 Indian Rojak ingredients from 35 stalls; no Salmonella spp, Vibrio cholerae/parahaemolyticus or Escherichia coli O157:H7 were recovered from the studied samples. The reheating by the food handlers significantly reduced the overall median Standard Plate Count (SPC) of food from 4.5 to 2.7 log colony forming units (CFU)/g (p<0.05). The cooked ingredients with the highest microbial loads were tofu and fish cake, with those purchased from wet markets having significantly higher bacterial loads than those purchased from supermarkets (p<0.05). The Rojak gravy had the lowest median bacterial load (1.9 log CFU/g). Raw, ready-to-eat vegetables, namely green chillis, cucumbers and onions had higher levels ranging from 5.9 to 6.1 log CFU/g. Contamination with E. coli, Staphylococcus aureus, and Bacillus cereus was seen with some of the ready-to-eat raw vegetables. Repeated education of food handlers with emphasis on good hygiene practices should be conducted to reduce the risk of foodborne illnesses.

Diarrheagenic pathogens in adults attending a hospital in Singapore

Background

Singapore’s diarrhoeal notification system is based on specific pathogens. Official data may thus be skewed towards notifiable diseases. Limited information is available on the profiles of aetiological agents responsible for acute gastroenteritis (AGE) cases, especially among the adult population. To understand the frequency and distribution of potential causative agents of diarrheal disease in Singapore, we screened adults’ stool samples collected from a large public hospital.

Methods

The stool samples were screened for 18 diarrheagenic pathogens using a combination of commercial multiplex polymerase chain reaction (PCR), in-house singleplex PCR and immunochromatographic assays. One hundred adult faecal samples that were collected from October 2013 to January 2014 for routine diagnostic purposes and submitted for culture at Tan Tock Seng Hospital, Singapore were used.

Results

Pathogens were detected in 32 % of the samples. The predominant organisms encountered were norovirus genogroup II (11 %), Aeromonas spp. (9 %) and Campylobacter spp. (5 %). One sample was positive for both verocytotoxigenic E. coli (VTEC) and E. coli O157:H7. Two other samples were positive for VTEC only, and one other sample was positive for E. coli O157:H7 only. Astrovirus, C. perfringens, Shigella spp. and toxigenic C. difficile were each detected in 2 % of the samples. Cryptosporidium parvum, Giardia lamblia, group A rotavirus, Salmonella spp. and Vibrio spp. were each detected in 1 % of the samples. No L. monocytogenes, Y. enterocolitica, enteric adenovirus, or norovirus genogroup I were detected.

Conclusion

Our preliminary findings suggest that pathogens causing non-notifiable diseases might have contributed considerably to the adult hospitalised AGE cases. However, as the samples were from an adult hospital, the data obtained may not be representative of the whole community. Thus, a larger study to collect clinical samples and risk exposure data from primary healthcare clinics and children hospital is planned for, to gain a more holistic perspective on the epidemiology of AGE in Singapore. A larger study may also offer valuable insights for improving the approach of microbiological surveillance of food, as well as strategizing inspection efforts along the food supply chain by public health authorities.