Salmonella enterica Serovar Panama, an Understudied Serovar Responsible for Extraintestinal Salmonellosis Worldwide

In recent years nontyphoidal Salmonella has emerged as one of the pathogens most frequently isolated from the bloodstream in humans. Only a small group of Salmonella serovars cause this systemic infection, known as invasive nontyphoidal salmonellosis. Here, we present a focused minireview on Salmonella enterica serovar Panama, a serovar responsible for invasive salmonellosis worldwide. S.

have also been involved in the spread of S. Panama during hospital outbreaks in France and in other Western European countries during the 1960s and 1970s (19,20). Over this period, there was a 3-fold increase in S. Panama cases in the United Kingdom, which led to a doubling of the number of salmonellosis cases (18). Subsequently, between 1969 and 1984, S. Panama was one of the top five serovars responsible for invasive disease in the United Kingdom (21). It is thought that these isolates were exposed to high antibiotic selective pressure in humans or food animals and consequently became resistant to antibiotics via acquisition of many types of plasmids (22)(23)(24)(25)(26)(27)(28). Elsewhere in the European Union, S. Panama was reported among the top 10 most frequently isolated serovars during 2012, following 706 confirmed cases of S. Panama salmonellosis associated with outbreaks in Germany and Italy (29). Sporadic outbreaks of S. Panama salmonellosis also occurred in Switzerland (1972), Hungary (1979), Spain (1998), and the Netherlands (2008) (30)(31)(32)(33). S. Panama maintained its ranking in the top 20 serovars associated with salmonellosis in the European Union until 2017, when it was replaced by other serovars (Salmonella enterica serovar Brandenburg, Salmonella enterica serovar Kottbus, and Salmonella enterica serovar Coeln) (34).

CLINICAL PICTURE IN HUMANS
Although S. Panama can cause gastrointestinal infection in humans (9), the serovar is more widely known for its ability to cause invasive disease and to colonize extraintestinal sites. For most salmonellae, extraintestinal colonization refers to bloodstream infection (2). However, S. Panama can also invade specific body sites, causing atypical presentations, including throat infection, brain abscess, and Bartholin's abscess (35)(36)(37) (summarized in Fig. 1). These unexpected symptoms of S. Panama infection can impede diagnosis and delay treatment.
( Fig. 1). A common complication of neonatal S. Panama infection is the development of Salmonella meningitis (8,14,36,(38)(39)(40)(41)(42)(43)(44)(45), a lethal disease that has previously been linked to localized outbreaks in hospital maternity wards (8,31). For example, S. Panama was recovered from 138 babies, new mothers, and staff during an outbreak of salmonellosis in a neonatal nursery in Michigan in 1934 to 1944 that resulted in 18 fatalities due to Salmonella meningitis (8). Similar outbreaks have historically occurred in other countries, including Germany, where a hospital outbreak in a maternity unit caused prolonged contamination despite radical disinfection of the entire ward (46). S. Panama causes more cases of clinically invasive disease in humans than most Salmonella serovars. Historically, S. Panama infections have been 11 times more likely to cause invasive disease than those by other serovars in Martinique (10,11). In England, 7% of all S. Panama isolates were isolated from extraintestinal sites compared to 2% of Salmonella enterica serovar Typhimurium and 3% of S. Enteritidis isolates (21). In Taiwan, 70% of S. Panama isolates were isolated from invasive disease compared to 12% of S. Enteritidis isolates (47). In addition to these epidemiologically suggestive data, multivariate analysis has recently confirmed the association of S. Panama with clinically invasive infection (P Ͻ 0.001) as part of a retrospective study of Salmonella infections in children living in Guadeloupe (7). A gnotobiotic-mouse model has been described for S. Panama (48), which could help to elucidate the mechanisms behind the increased invasiveness.

TRANSMISSION VEHICLES
Wild reptiles are the natural reservoir for S. Panama in Latin America (12,(49)(50)(51)(52). A study focusing on the frequency and host distribution of Salmonella serovars in reptiles and amphibians captured in the Republic of Panama between 1965 and 1967 showed that 2.6% of 78 Salmonella isolates were serovar Panama (49). In a subsequent study (1966 to 1969), 6.8% of Salmonella organisms isolated from neotropical lizards in Panama were S. Panama (50). In the past decade, a high prevalence of Salmonella has been found in the largest lizards in South America (Tegu lizards), and 3% of the isolates were classified as S. Panama (51). In French Guiana, where S. Panama was the most frequently isolated human-associated serovar in 2011, the serovar was also isolated from wild reptiles (12). Reptiles are likely to be an important source for transmission of S. Panama in regions of the world where many lizards and other reptiles are present in and around households. A recent survey of Salmonella strains carried by African venomous snakes did not isolate S. Panama (53).
In addition to reptiles, S. Panama has also been isolated from other wildlife species and companion animals. A study on pouched wild birds found S. Panama in cloacal swabs of chestnut-capped blackbirds in Rio de Janeiro, Brazil (54). In regard to companion animals, S. Panama was isolated from a household dog in Taiwan (55). S. Panama contamination has been found in birds and fish tanks sampled from pet shops and households in Trinidad (56). Wildlife, therefore, represent a potential reservoir for S. Panama dissemination.
In Europe, S. Panama infection is primarily a foodborne disease, with the main transmission vehicles being pork-derived products, including cured meat, minced pork, and sausages (57). The transmission pathway for S. Panama begins in animal feed, from where it can enter porcine and poultry animal reservoirs and move into animal food products, eventually infecting humans (18).
At the animal level, S. Panama was found in 2.08% of 200 abattoir pigs sampled in Budapest, Hungary (58), and has been found in cattle and swine in Germany (59). Outside Europe, S. Panama has been identified in beef and dairy herds in Argentina (60) and is the second most common Salmonella serovar to be isolated from swine finishing herds in Brazil (61).
S. Panama is also recognized as a contaminant in food-processing facilities and retail establishments globally, including butcher shops (62), public markets (63), meat vans (64), and slaughterhouses (65). The process of manufacturing pork-derived products includes several steps designed to result in a microbiologically safe, shelf-stable prod-uct by tightly controlling physicochemical conditions, such as salt and nitrate concentrations, pH, water activity, and temperature (66). However, Salmonella viability throughout this curing process has been reported, including the presence of S. Panama in salami (67,68). In the Netherlands, S. Panama has additionally been implicated in the contamination of cattle-derived food products and was one of the three Salmonella serovars most frequently isolated from mincemeat over a 13-month period. Interestingly, mincemeat from slaughterhouses was more likely to contain Salmonella than mincemeat derived from slaughtering completed at butcher shops (69). Food-processing facilities themselves can play a role in the contamination of animal food products with S. Panama.
The impact of S. Panama entering the human food chain can be seen in an outbreak of salmonellosis that affected 300 people who had eaten contaminated roast pork in the United Kingdom in 1970. S. Panama was implicated as the etiological agent (18). S. Panama has also caused several foodborne outbreaks between 1990 and 1999 in Asturias, Spain, and isolates were collected from gastroenteritis and septicemia patients who had consumed contaminated fish puddings, cooked octopus, and cream cakes (32). Other studies have linked S. Panama infections to consumption of goat cheese, vegetables, beef, poultry, eggs, fruit juice, and shellfish (14,33,70).
In addition to the usual fecal-oral transmission route of Salmonella in humans, breast milk has also been suggested as a vector for S. Panama (71). A study demonstrated that S. Panama can infect the human mammary duct, can be shed for at least 2 weeks, and can remain stable during storage of breast milk at 4°C (71). Furthermore, it is possible that a case of meningitis in an exclusively breastfed 4-month-old patient was contracted from breast milk that was contaminated with an antimicrobial-susceptible S. Panama isolate (41).

ANTIMICROBIAL RESISTANCE
Burden of antimicrobial resistance in S. Panama. Antimicrobial resistance (AMR) is an important public health concern (72). There are conflicting reports in the literature relating to the AMR status of the S. Panama serovar, with studies in Italy and Brazil reporting low levels of antibiotic resistance (41,73). They are supported by further reports from Martinique, where 91% of S. Panama isolates were susceptible to betalactams (11), and Guadeloupe, where all Salmonella serovars demonstrated high overall susceptibility to antibiotics (7). In contrast, other studies have seen higher levels of resistance in S. Panama, particularly against tetracycline (e.g., 67%) and chloramphenicol (e.g., 67%) since the 1980s (24,47,59,(74)(75)(76). Antibiotic stewardship promises to be an effective tool for decreasing antimicrobial resistance in the S. Panama serovar. For example, following a ban on tetracycline use in the pork industry in the Netherlands, S. Panama tetracycline resistance dropped from 90% to 1% (24).
In Asia, S. Panama has been associated with high levels of AMR since 1980, when 58% of the S. Panama isolates from Tokyo were resistant to at least one antibiotic agent (77). This figure appears to be on the rise. By the turn of the millennium, 83% of domestic and imported S. Panama isolates from cases in Tokyo were multidrug resistant. Similarly, in Taiwan, the serovar also exhibited resistance to multiple antibiotics, including cotrimoxazole (67%), ampicillin (56%), streptomycin (56%), kanamycin (56%), and gentamicin (45%) (74). The high proportion of S. Panama isolates that show AMR should be considered by clinicians working in Asia and by health care practitioners globally when treating Asiantravel-associated salmonellosis cases caused by S. Panama.
Genomic markers and trends in antimicrobial resistance. A large proportion of S. Panama antimicrobial resistance has been associated with plasmid carriage (P ϭ 0.012), class 1 integron presence, and transmissible drug resistance (R) factors (22,47,74,78). Resistance to tetracycline, for example, has often been mediated by the R factor R1 in S. Panama (26). Such R factors have been implicated in the transfer of multiple antimicrobial resistance genes, usually simultaneously, between S. Panama strains and other bacteria. However, an isolate from an epidemic of S. Panama infection in Paris showed unusual patterns of transferable resistance, which may extend to other strains in the S. Panama serovar. The isolate was able to transfer genes involved in antimicrobial resistance singly or in pairs, rather than as one antibiotic resistance cassette. The proposed mechanism involved the simultaneous transfer of several discrete genetic elements that were able to coexist stably and to replicate noncompetitively in S. Panama. The authors suggested that frequent cotransfer of genetic elements may be propagated by conjugative-transfer machinery (27).

INVASIVE DISEASE-GENOMIC INFERENCES IN S. PANAMA
Evolutionary history and virulence. The study of evolutionary history may explain why S. Panama is associated with invasive disease. The majority of salmonellae that cause disease in humans belong to S. enterica subsp. enterica, which is further divided into two main clades, A and B, and a number of smaller clades (79). Phylogenetically, S. Panama is in clade B, which is associated with increased levels of clinically invasive disease (53,80,81). Another review of the population structure within S. enterica found that S. Panama is in lineage 3 (equivalent to the above-mentioned clade B) (82).
The evolutionary history of S. Panama was studied by Selander et al. (83), who used multilocus enzyme electrophoresis to assess the relationships among Salmonella serovars that cause invasive disease. It was proposed that S. Panama evolved from the same ancestors that gave rise to Salmonella enterica serovar Paratyphi, Salmonella enterica serovar Sendai (which causes enteric fever), and Salmonella enterica serovar Miami (83). In the current era of genomically informed epidemiological analysis, phylogenetic methods can be used to understand the evolutionary history of Salmonella. However, no large-scale phylogenetic study has yet been conducted on S. Panama, and only one complete S. Panama genome sequence (from strain ATCC 7378; GenBank accession no. CP012346) is available (84). As part of the current review, virulence genes were identified in the complete genome of S. Panama strain ATCC 7378 using the program ABRicate v0.8.10 (https://github.com/tseemann/abricate) against a virulence factor database (85) with default parameters. In total, 131 virulence-associated genes were identified. The analysis confirmed the presence of typical Salmonella virulence determinants, including type III secretion systems, type III effector proteins, fimbriae, and flagella. Of interest, S. Panama was also found to carry the cytolethal distending toxin B gene (cdtB), which is characteristic of S. enterica clade B and the highly invasive Salmonella enterica serovar Typhi (53,80,81). A more detailed, epidemiologically representative analysis is required to further elucidate the uniqueness of the S. Panama serovar.
Accessory genome and virulence. Generally, plasmids play a key role in systemic Salmonella infection, but little is known about the plasmid complement of the S. Panama serovar. In the small number of available studies, it is reported that S. Panama, including the above-mentioned S. Panama ATCC 7378, does not commonly carry the large plasmids that have previously been associated with virulence in other Salmonella serovars (41). Rather, S. Panama strains carry a heterogeneous population of plasmids (86). Prophages can also make significant contributions to Salmonella virulence (87,88), but only one study has reported the presence of prophages in S. Panama (84). The Salmonella RE-2010 prophage was identified in the genome of S. Panama ATCC 7378. The prophage (also known as ElPhiS) has also been found in S. Enteritidis, where it has been associated with specific phylogenetic clusters (89,90). The importance of S. Panama for public health globally necessitates that a concerted comparative genomic analysis be conducted in the future.

PERSPECTIVES
S. Panama is a globally relevant pathogen that has consistently been reported as one of the most frequently isolated Salmonella serovars over the past 70 years. The proportion of clinical cases caused by S. Panama is particularly high in French territories in the Americas, where it is associated with invasion of extraintestinal sites, particularly in infants. Reptiles act as natural reservoirs for Salmonella in these regions, and it has been speculated that the large numbers of reptiles found in and around homes in tropical regions of America lead to high levels of S. Panama transmission to humans. The serovar was also introduced into Europe, where it spread through the pork industry and caused hospital outbreaks in the 1960s and 1970s. S. Panama continues to contribute to the global disease burden caused by salmonellae.
It is important to highlight the unusual clinical presentation of S. Panama in different patient populations to avoid delays in patient treatment. Clinicians and researchers should remain aware of the potential for increasing levels of antimicrobial resistance in the serovar, as has been described in Asia. Unraveling the molecular epidemiology and evolutionary history of S. Panama is the obvious next step in understanding more about this rarely studied serovar that continues to cause invasive salmonellosis worldwide. with a first-class degree in tropical disease biology, she became fascinated with bacterial genomic applications in public health. For the past 3 years, she has focused on understanding the molecular epidemiology of nontyphoidal Salmonella that causes bloodstream infection in Africa and the French Caribbean. She has made a significant contribution to global Salmonella sequencing efforts, working as part of the 10,000 Salmonella Genomes project. Her research has received numerous awards, including the NOVA for outstanding early contributions in biology and the John Lennon Memorial Scholarship in recognition of global health research. As an early career researcher, she has contributed a first author research paper focused on Salmonella diversity in venomous snakes. Her ambition is to pursue a career in the surveillance of bacterial pathogens during global epidemics.

ACKNOWLEDGMENTS
Blanca M. Perez-Sepulveda, Ph.D., is a molecular microbiologist working on invasive nontyphoidal Salmonella (iNTS). After completing an M.Sc. (Res) in biochemistry at the University of Chile, she moved to the United Kingdom, where she obtained a Ph.D. at the University of Warwick studying the molecular mechanisms of phage resistance. She moved to the University of Liverpool in 2016 to join Jay Hinton's laboratory as a postdoctoral research associate. She currently focuses on understanding the virulence determinants of novel invasive Salmonella Enteritidis clades identified in sub-Saharan African regions, using a combination of phenotypic characterization, comparative genomics, and transcriptomics. In collaboration with the Earlham Institute, she has been leading the 10,000 Salmonella Genomes project, a worldwide collaborative effort to understand the transmission and virulence of iNTS. Her interests lie in understanding how bacteria survive in the environment. Her focus has been to determine the environmental reservoirs and transmission of Salmonella and its phages by studying molecular mechanisms of virulence and phage resistance.