Identification and significance of Weissella species infections

Weissella spp. are non-spore forming, catalase-negative, gram-positive coccobacilli. They are often misidentified by traditional and commercial phenotypic identification methods as Lactobacillus spp. or Lactobacillus-like organisms. Weissella spp. were previously grouped along with Lactobacillus spp., Leuconostoc spp., and Pediococcus spp. Utilization of more sensitive methods like DNA sequencing or Matrix-Assisted Laser Desorption/Ionization Time of Flight Mass Spectrometry (MALDI-TOF MS) has facilitated identification of Weissella as a unique genus. Nineteen species have been identified to date. W. confusa, W. cibaria, and W. viridescens are the only species isolated from humans. The true prevalence of Weissella spp. continues to be probably underestimated. Weissella spp. strains have been isolated from a wide range of habitats including raw milk, feces, fermented cereals, and vegetables. Weisella is believed to be a rare cause of usually nonfatal infections in humans, and is often considered a contaminant. However, in recent years, Weissella spp. have been implicated in bacteremia, abscesses, prosthetic joint infections, and infective endocarditis. Alterations of the gut flora from surgery or chemotherapy are believed to facilitate translocation of Weissella spp. due to disruption of the mucosal barrier, predisposing the host to infection with this organism. Implications of the isolation of Weissella spp. from blood must be interpreted in context of underlying risk factors. Weissella spp. are inherently resistant to vancomycin. Therefore, early consideraton of the pathogenic role of this bacteria and choice of alternate therapy is important to assure better outcomes.


INTRODUCTION
Weissella spp. are non-spore forming, hetero-fermentative, facultative anaerobic, gram-positive, catalase-negative, alpha hemolytic bacteria that appear as short rods or coccobacilli in pairs and chains (Collins et al., 1993;Olano et al., 2001;Björkroth et al., 2009). Based on their unusual Gram stain morphology and inherent resistance to vancomycin, Weissella spp. have been often confused with Lactobacillus spp. or Lactobacillus-like organisms. They are usually considered contaminants when recovered from clinical specimens and rarely identified to the species level due to their fastidious nature (Facklam et al., 1989;Facklam and Elliott, 1995;Kumar et al., 2011).

EPIDEMIOLOGY
Weissella spp. have a widespread distribution and have been isolated from a wide variety of habitats including raw milk, feces, saliva, breast milk, urine, fermented cereals, meat and meat products, sugar cane, carrot juice, banana leaves and vegetables (Kandler and Weiss, 1986;Björkroth et al., 2002;Fairfax et al., 2014;Fusco et al., 2015). They have also been recovered from feces of healthy individuals (Green et al., 1990;Walter et al., 2001) and are common inhabitants of the vaginal microbiota (Jin et al., 2007). Although Weissella are of rare occurrence in humans, several disease outbreaks involving W. ceti have been reported in cultured rainbow trout from geographically diverse locations (United States, China, Brazil;Liu et al., 2009;Figueiredo et al., 2012;Welch and Good, 2013;Costa et al., 2015). The true incidence of Weissella infection in humans is probably underestimated due to its misidentification.

DIAGNOSIS
Identification of Weissella at the genus and species level has been challenging. It is often misidentified as Lactobacillus-like or viridans streptococci and accurate identification is not possible by traditional or commercial phenotypic identification methods that include morphological analysis, growth characteristics and sugar fermentation profiles as these techniques have low taxonomic discrimination (Fusco et al., 2015). Commercial biochemical based identification kits namely API (Analytical Profile Index systems, bioMéreiux, France) and RapiID Strep panel (Remel, USA) are unable to identify Weissella species. The identification table of API 20 Strep does not include Weissella. API 50 CHL kit (version 5.1) has W. confusa and W. viridescens in its identification table; however, it cannot differentiate W. cibara from W. confusa (Kulwichit et al., 2007). RapID ™ STR System does include W. confusa in its database but only give a high probability result (Fairfax et al., 2014). Automated systems namely Vitek 2 (bioMéreiux, France), MicroScan (Beckman Coulter Inc. USA), and Phoenix Automated Microbiology System (BD Diagnostic Systems, USA), do not have Weissella in their database and as such cannot reliably identify Weissella species Fairfax et al., 2014;Fusco et al., 2015).
Molecular DNA sequencing involving 16S rRNA gene sequence analysis can accurately identify Weissella to the species level and remains the current gold standard. It has also emerged as powerful tool for identification of phenotypically atypical microorganisms (Petti et al., 2005) and has been successfully used to identify Weissella to the species level (Collins et al., 1993;Vasquez et al., 2015). Most of the cases of W. confusa reported were originally misidentified as Lactobacillus-like or viridians streptococci organisms by phenotypic methods. These were subsequently confirmed to be W. confusa using 16S rRNA gene sequence analysis. Amplified ribosomal DNA restriction analysis (ARDRA; Jang et al., 2002) and ribotyping (Björkroth et al., 2002) have also been used to correctly identify Weissella species. Molecular typing techniques for Weissella species include DNA finger printing and restriction of ribosomal DNA (Villani et al., 1997), numerical analysis of HindIII and EcoRI ribopatterns (Koort et al., 2006), repetitive element-PCR fingerprinting using (GTG) 5 -PCR (Bounaix et al., 2010), and fluorescent-amplified fragment length polymorphism (fAFLP; Fusco et al., 2011). MALDI-TOF MS is now being routinely used for the identification of bacterial organisms (Bizzini et al., 2010;Wieser et al., 2012) and can also accurately identify W. confusa Fairfax et al., 2014). Two MALDI-TOF MS systems have found increasing use in microbiology laboratories and both are sensitive for the identification of unusual and/or difficultto-identify microorganisms isolated from clinical specimens (McElvania TeKippe and Burnham, 2014). The Bruker Biotyper (Bruker Daltonics, Germany) software version 3.0 also includes W. confusa, W. halotolerans, W. minor, and W. viridescens, and VITEK MS (bioMérieux, France) database v2.0 has W. confusa and W. viridescens. The future versions of both these mass spectrometry systems are likely to incorporate other Weissella spp. which will facilitate their early identification and provide insight into the true prevalence of these infections.

PREDISPOSING FACTORS AND CLINICAL MANIFESTATIONS
Most of the cases of W. confusa infections reported in humans have been from immunocompromised patients (Lahtinen et al., 2012;Fairfax et al., 2014;Medford et al., 2014). Malignancy has been the most common factor associated with immunocompromised and complicated medical status. Recent chemotherapy, organ transplant, burn, chronic alcoholism, longterm use of steroids, chronic renal insufficiency and diabetes seem to increase the chances of acquiring this infection (Flaherty et al., 2003;Salimnia et al., 2011; Table 1). Orthopedic procedures like joint replacements, arthroplasty, and postoperative osteomyelitis also put the patients at increased risk of bacterial infections. Prior exposure to vancomycin may results in the selection of Weissella, which is intrinsically resistant to this drug. Central line catheter insertion prior to any surgical procedure increases the risk of infection, although Weissella has not been recovered from catheter tips. Total parenteral nutrition has been suspected to be risk factor for the development of bacteremia involving Weissella species (Olano et al., 2001;Flaherty et al., 2003;Kumar et al., 2011;Lee et al., 2011;Vasquez et al., 2015). Weissella is a common inhabitant of the human gastrointestinal system. A compromise of the gastrointestinal mucosal barrier due to surgery is associated with increased risk of acquiring this infection and may be a probable route of entry of W. confusa resulting in bacteremia and endocarditis (Flaherty et al., 2003;Shin et al., 2007). Polymicrobial infection and subsequent antimicrobial measures resulting in the alteration of gut flora also favor the selection of Weissella in such patients (Kumar et al., 2011). Changes in the normal flora of the throat, gut, and vaginal tract and disruption of mucous integrity by invasive procedures, surgery, and/or antibiotics predispose the host to increased risk of Weissella infection. The possible risk factors for W. cibara and W. viridescens infection in humans remain unknown.

ANTIMICROBIAL SUSCEPTIBILITY TESTING
Weissella is known to be intrinsically resistant to vancomycin and has high minimum inhibitory concentration (MIC) of ≥256 µg/ml. Antimicrobial susceptibilities of W. confusa are not fully understood. There are no standard methods and interpretation criteria of antimicrobial susceptibilities established so far for Weissella spp. by the Clinical and Laboratory Standards Institute (CLSI). Susceptibility testing methods have included broth dilution, agar based methods (Bantar et al., 1991;Olano et al., 2001;Vay et al., 2007;Lee et al., 2011;Medford et al., 2014) and E-test (Svec et al., 2007; Table 2).

CONCLUSION
W. confusa is an opportunistic bacterial organism that warrants rapid and accurate identification to ensure appropriate therapy. It is difficult to distinguish Weissella species from other heterofermentative bacteria on the basis of phenotypic or biochemical tests alone. Accurate and rapid identification to the species level is feasible using 16S rRNA or MALDI-TOF MS techniques. Infections with Weissella spp. primarily occur in immunocompromised status and/or those with other underlying medical conditions. It is a common inhabitant of the gut flora, therefore, surgical procedures may translocate this organism and result in bacteremia, endocarditis, and abscess formation. The clinical significance of W. confusa remains unclear in the setting of polymicrobial infections, which comprise the largest proportion of total cases. The use of antimicrobial agents, especially vancomycin, may predispose patients to increased risk of infection with Weissella, which is intrinsically resistant to this drug (Lahtinen et al., 2012;Fairfax et al., 2014;Medford et al., 2014). When isolated in blood culture, Weissella may be confused with Lactobacillus-like or viridians streptococci and can be considered a probable contaminant (Ruoff, 2002;Fairfax and Salimnia, 2012;Fairfax et al., 2014). However, Weissella is not a part of the normal skin flora and should be considered significant when isolated from blood cultures (Petti et al., 2005).
Vancomycin is the empiric first-line therapy for bacteremia caused by gram-positive organisms. However, the use of vancomycin is likely to favor the growth of vancomycin-resistant organisms and may predispose these patients to subsequent infections with Weissella and other vancomycin resistant organisms (Fairfax and Salimnia, 2012). Early consideration of this organism in the differential diagnosis is important due to its inherent vancomycin resistance, which necessitates alternative therapy. Management of positive blood cultures especially from immunocompromised patients is a challenge both for the laboratory microbiologists as well as the clinicians. Penicillin, ampicillin, imipenem, clindamycin, erythromycin, moxifloxacin, doripenem, daptomycin, and tigecycline are all therapeutic agents for treating Weissella infections. The use of vancomycin, metronidazole, rifampin, teicoplanin, ceftazadime, and trimethoprim-sulphamethoxazole should be avoided if Weissella spp. is suspected. Antimicrobial susceptibility testing is vital to guide appropriate therapy in cases of severe infections.