Listeria and Erysipelothrix

Taxonomy

The genus Listeria consists of gram-positive, non-spore- forming, facultative anaerobic,

regular rod-shaped bacteria with a low G+C content of 36 to 42 mol%. While early

phylogenetic studies suggested a close relation between Listeria and the Lactobacillaceae,

comparisons of 16S rRNA gene sequences have shown that Listeriais most closely related

to Staphylococcus and Bacillus. Together with Brochothrix, Listeria is provisionally assigned

to the family “Listeriaceae” within the order Bacillales. Synthesis of menaquinones and major

amounts of branched-chain fatty acids confirms the taxonomic separation of Listeria from

the Lactobacillaceae (29, 51).

Until recently, the genus Listeria comprised six validated species including Listeria

monocytogenes as the type species in the genus, L. grayi, L. innocua, L. ivanovii, L.

seeligeri, and L. welshimeri. Within L. ivanovii two subspecies, L.

ivanovii subsp. ivanovii and L. ivanovii subsp. londoniensis, are differentiated (8). Recently, a

seventh Listeria species, L. marthii, has been described from the natural environment and is

most closely related to L. monocytogenes (38).

Based on the results of multilocus enzyme electrophoresis, DNA-DNA hybridization, and 16S

rRNA gene sequencing, the species of Listeria are divided into two closely related but distinct

lines of descent: (i) the L. monocytogenes group of species, including L. innocua, L. ivanovii,

L. marthii, L. monocytogenes, L. seeligeri, andL. welshimeri; and (ii) L. grayi (8, 16, 38). Of

the seven species within the genus Listeria, only L. monocytogenesand L. ivanovii are

pathogenic for humans and animals.

Description of the Agent

Members of the genus Listeria are gram-positive, facultative anaerobic, non-spore-forming,

nonbranching, regular, short (0.5 to 2 by 0.4 to 0.5 μm) rods that occur singly or in short

chains. Filaments of 6 to 20 μm in length may occur in older or rough cultures. Temperatureregulated

expression of flagellin results in a characteristic tumbling motility at 20 to 28°C by

means of one to six peritrichous flagella. At 37°C the organisms are much less motile.

Colonies are small (1 to 2 mm in diameter after 1 or 2 days of incubation at 37°C), smooth,

and blue-gray on nutrient agar when examined with obliquely transmitted light. Listeria spp.

show an exceptionally large growth temperature range from 0 to 50°C. The optimum growth

temperature is between 30 and 37°C, but at 4°C growth is also observed within a few days.

Catalase is typically produced, but catalase-negative strains causing disease in humans have

been described (13, 25). The oxidase test is negative. Acid is produced from D-glucose and

other sugars. The Voges-Proskauer and methyl red tests are positive. Esculin is hydrolyzed in

a few hours. Urea and gelatin are not hydrolyzed. Neither indole nor H2S is produced. The

cell wall contains a directly cross-linked peptidoglycan based on meso-diaminopimelic acid,

as well as lipoteichoic acid, but no mycolic acids. The two predominant cellular fatty acids are

Cai15:0 and Cai17:0(branched-chain type) (6).

Epidemiology and Transmission

The primary habitat of Listeria species is the environment, where they exhibit a saprophytic

lifestyle. L. monocytogenes has been isolated from various animals, like mammals, birds,

fish, and crustaceans. Infected animals can asymptomatically pass the organism or develop

clinical disease. Due to its widespread distribution, L. monocytogenes has many

opportunities to enter human food production, resulting in contamination of fresh and

processed poultry, meat, and vegetables; raw milk; cheese; smoked salmon; etc. (27).

Numbers of organisms exceeding 103 CFU/g were detected in food products (27). Infection of

humans ingesting colonized food is potentiated by the ability of the organism to multiply at

4°C. The intestinal tract of adults is consistently colonized with

nonpathogenic Listeria species and, to a lesser extent (1 to 5%), with pathogenic L.

monocytogenes (41). Cervicovaginal carriage in women has not been reported. Apart from

food-related infections, a few nosocomial outbreaks, mainly in neonatal wards, have been

described (28, 69). The number of sporadic cases of listeriosis in countries that report the

illness is typically in the range of 0.1 to 0.9 cases per 100,000 persons (37, 89). While the

number of cases and the mortality in the United States have decreased (incidence, 0.4 per

100,000 per year) in recent years (89), the incidence of sporadic listeriosis has increased in

several European countries, reaching numbers from 0.4 up to 1.0 per 100,000 per year

(12, 37).

Clinical Significance

The majority of cases of listeriosis occur in individuals who have an underlying condition that

leads to suppression of their cell-mediated immunity. However, infections in

immunocompetent individuals are increasingly reported. About one-half of the cases of

listeriosis occur in individuals older than 60 years and younger than 1 month. In adults, L.

monocytogenes causes primarily septicemia, meningitis, and encephalitis with a mortality

reaching up to 50%. Focal infections with Listeria spp. have been infrequently described and

include endocarditis, arthritis, osteomyelitis, intra-abdominal abscesses, endophthalmitis,

(sclero-)keratitis, peritonitis, and intravenous catheter and pleuropulmonary infections (19).

Among veterinarians and abattoir workers, primary cutaneous listeriosis with or without

bacteremia has been reported (53).

In pregnant women, L. monocytogenes often causes a mild, self-limited influenzalike illness.

Transient bacteremia can result in placentitis and/or amnionitis, and since Listeria is able to

cross the placenta (49), it can infect the fetus, causing abortion, stillbirth, or, most

commonly, preterm labor. In neonates, an early-onset form and a late-onset form of

listeriosis occur. The early form is presumably caused by intrauterine infection and manifests

as granulomatosis infantisepticum. The organism is widely disseminated in the body,

including the central nervous system. The source of the organism in the late-onset cases,

which manifest at a mean of 14 days after birth, is unclear and may comprise the mother ’s

genital tract or environmental sources.

The infectious dose and the incubation period for human listeriosis have not been firmly

established, and reported incubation periods vary from a few days to 2 to 3 months. Doses

of 105 CFU or greater have been reported to cause gastroenteritis in outbreak situations (2).

A dose-response model using rhesus monkeys as a surrogate for pregnant women recently

indicated that oral exposure to 107 CFU of L. monocytogenes results in about 50% stillbirths

(71). Thus, it may be much less than the extrapolated estimate of 1013 CFU from the

FDAU.S. Department of Agriculture-CDC risk assessment based on mouse data (32).

Most cases of Listeria gastroenteritis are linked to foodborne outbreaks. Typically, patients

with Listeriagastroenteritis have no known predisposing risk factors for listeriosis, illness

occurs about 24 h after ingestion of a food item that is contaminated with a large number of

bacteria (105 to 109 CFU/g or ml), and illness lasts about 2 days. Apart from gastroenteritis,

fever, headache, and pain in joints and muscles are frequently seen (59).

After ingestion of L. monocytogenes, pathogen and host factors as well as the number of

pathogens ingested determine whether invasive infection develops. Immunity to listeriosis is

effected primarily via the cell-mediated immune system. Penetration of the epithelial barrier

in the gut by L. monocytogenes is facilitated by its ability to escape from the host cell

vacuole, intracytoplasmic multiplication, movement via bacterially induced polymerization of

host cell actin, and spread to neighboring cells through pseudopodlike extensions of the host

cell membrane. Virulence genes are clustered on an 8.2-kb pathogenicity island and include

genes coding for internalin A and B and listeriolysin, a hemolysin (65). Interaction between

internalin and E-cadherin, a receptor of the trophoblast, facilitates the spread of the

organism to the fetus (49).

L. ivanovii is primarily a pathogen of ruminants. Systemic infections in human

immunodeficiency virus-infected and nonimmunosuppressed patients have, however, been

described (17, 72).

Collection, Transport, and Storage of Specimens

Suitable specimens for detection of listeriosis include blood and cerebrospinal fluid (CSF). In

neonates with suspicion of listeriosis, investigation of blood, CSF, amniotic fluid, respiratory

secretions, placental or cutaneous swabs, gastric aspirates, or meconium can facilitate

detection of the organism. For epidemiologic purposes or rare causes of gastroenteritis, stool

specimens are preferred to rectal swabs. In general, specimens for detection of Listeria do

not need special handling during collection.

Clinical specimens for culture of L. monocytogenes should be processed as soon as possible

or stored and transported at room temperature or 4°C for up to 48 h. At 4°C even longer

storage times may be tolerated due to the specific cold resistance of the organism, but

multiplication of Listeria has to be regarded (15). Stool samples (1 g each) can be inoculated

into 100 ml of a selective University of Vermont or polymyxin-acriflavin-lithium chlorideceftazidime

esculinmannitol (PALCAM) enrichment broth and then shipped overnight at room

temperature. To avoid overgrowth of L. monocytogenes by contaminating microbiota during

longer periods of storage, nonsterile-site specimens should be stored at 4°C for 24 to 48 h or

frozen at −20°C.

Food samples should include a minimum of 100 g of a sample and should be collected

aseptically in sterile containers. Food packaged in original containers should always be

preferred. Samples should be shipped overnight frozen. Although L. monocytogenes is

relatively resistant to freezing, repeated freezing and thawing should be avoided.

Cultures of Listeria spp. should be frozen at −20 to −70°C for long-time storage. They can

be shipped on a non-glucose-containing agar slant and packaged and declared according to

the respective national and international requirements.

Because L. monocytogenes can infect the fetus, leading to stillbirths and abortions while

causing only mild symptoms in the mother, pregnant women should be particularly careful

when working in a laboratory whereL. monocytogenes is propagated or handled.

Direct Examination

Direct microscopy should be performed in CSF, positive blood cultures, and if available,

tissue samples. Detection of gram-positive, regular short rods in CSF or blood cultures

should lead to the suspicion of listeriosis. Nevertheless, L. monocytogenes may be confused

with members of the coryneform rods (especially in direct slides from positive blood

cultures), since the cells may be arranged in V forms or palisades. Commercial tests licensed

for antigen detection in clinical specimens other than nucleic acid-based tests are not

available.

Sensitive and specific in-house PCR assays have been described for detection of L.

monocytogenes in CSF, stool, or lung tissue (10, 34, 93) and may be particularly useful for

specimens from patients with prior antimicrobial therapy. Regarding commercial assays, the

Probelia Listeria monocytogenes assay (Bio-Rad, Hercules, CA) has been evaluated in clinical

stool specimens (41) while other commercial assays (e.g., LightCycler PCR, Roche

Diagnostics, Indianapolis, IN) have been validated only for food specimens.

Isolation Procedures

Clinical specimens from normally sterile sites should be plated onto tryptic soy agar

containing 5% sheep, horse, or rabbit blood. Plates should be incubated at 35 to 37°C under

room air with 5% CO2 for a minimum of 48 h. Blood samples should be inoculated into

conventional blood culture media. Clinical specimens obtained from nonsterile sites, like stool

samples, as well as food and environmental specimens should be plated onListeria spp.

selective agars. In addition, enrichment by inoculation into selective broth for Listeria spp.

(see above) should be done before plating.

Selective agars for culture of Listeria spp. include lithium chloride-phenylethanol-moxalactam

(LPM) (50), Oxford, modified Oxford, and PALCAM agars (34). On LPM agar, colonies have to

be examined under a stereomicroscope with Henry illumination (magnification, ×15 to ×25,

with oblique lighting directed to the microscope stage by a concave mirror positioned at a

45° angle to the incident light). Listeria colonies appear blue, and colonies of other bacteria

appear yellowish or orange. Oxford and PALCAM agars contain selective substances that

eliminate the need for examination under oblique lighting (84). On Oxford and modified

Oxford agars, Listeria colonies appear black due to esculin hydrolysis, are 1 to 3 mm in

diameter, and are surrounded by a black halo after 24 to 48 h of incubation at 35 to 37°C.

On PALCAM agar, Listeria colonies appear gray-green, are approximately 2 mm in diameter,

and have black sunken centers.

For the detection of Listeria spp. in food samples, enrichment methods have to be used. The

most widely used reference methods for food and environmental samples are the Food and

Drug Administration (FDA)Bacteriological and Analytical Manual (BAM) (83) and the U.S.

Department of Agriculture (USDA) method (82) in the United States and the International

Organization of Standards (ISO) 11290 method in Europe (34). All methods require

enrichment of the samples in a selective broth (Listeria enrichment broth, FDA BAM

formulation, or University of Vermont broth in the FDA BAM and USDA methods; and Fraser

broth in the ISO method) prior to plating onto selective agar and biochemical identification of

typical colonies (18). A detailed comparison of methods is given in reference 18.

New chromogenic media allow selective isolation of Listeria species (34, 62). Several media

identify L. monocytogenes by the production of a phosphatidylinositol-specific phospholipase

C. Media include ALOAgar (Biolife, Milan, Italy) (88), BCM L. monocytogenes (Biosynth,

Staad, Switzerland) (64), LIMONO-Ident-Agar (Heipha, Eppelheim, Germany), and BBL

CHROMagar (BD, Sparks, MD) (43). However, none of these agars differentiate between L.

monocytogenes and L. ivanovii. Specific detection of L. monocytogenes is facilitated on

RAPID L.mono agar (Bio-Rad, Hercules, CA) (4). Chromogenic media showed sensitivities

comparable to those of Oxford and PALCAM agar (4, 43, 62).

Identification

A simplified identification is based on the following tests: Gram staining, observation of

tumbling motility in a wet mount, and tests for a positive catalase reaction and esculin

hydrolysis. Acid production from D-glucose and positive Voges-Proskauer test are

confirmatory results.

Listeria spp. may be confused with other gram-positive bacteria due to similar morphologic

or biochemical characteristics. Streptococcus and Enterococcus spp. can be differentiated

from Listeria spp. on the basis of Gram stain morphology, motility, and catalase

reaction. Erysipelothrix spp. differ from Listeria spp. in motility, catalase reaction, and ability

to grow at 4°C (Erysipelothrix spp. do not grow at that temperature).Lactobacillus spp. are

usually nonmotile and catalase negative.

Identification of Listeria isolates to the species level is crucial, because all species can

contaminate foods but only L. monocytogenes is of public health concern. A scheme for

identification of Listeria species based on morphological and biochemical characteristics is

shown in (Table 1). Among these markers, hemolysis is essential for differentiating

between L. monocytogenes and the most frequently isolated nonpathogenic species, L.

innocua.

Typing Systems

Subtyping of L. monocytogenes is crucial for the workup of disease acquired from foodborne

agents. Based on somatic “O” and flagellar “H” antigens, 13 serovars of L.

monocytogenes are known (1/2a, 1/2b, 1/2c, 3a, 3b, 3c, 4a, 4ab, 4b/4bX, 4c, 4d, 4e, and

7). Since the vast majority of L. monocytogenes strains that cause sporadic infections or

outbreaks belong to the same serotypes, i.e., 1/2a, 1/2b, and 4b, and since serotyping

antigens are shared among L. monocytogenes, L. innocua, L. seeligeri, and L. welshimeri,

reliable discrimination below the level of serotype is necessary. Thus, serotyping is only

useful as a first-level discriminator or for the selection of further typing methods in

suspected outbreaks. Antisera are commercially available from Difco (Difco Laboratories/BD,

Sparks, MD) and Denka Seiken (Tokyo, Japan). A multiplex PCR has been described for

identification of the four major serovars of L. monocytogenes (1/2a, 1/2b, 1/2c, and 4b) and

has been validated by interlaboratory comparison (20, 21).

Pulsed-field gel electrophoresis (PFGE) is considered to be the standard typing method for L.

monocytogenes. Its discriminatory power and reproducibility of results have been confirmed

in a World Health Organization multicenter international typing study (11) as well as in a

large number of other studies. PFGE is particularly useful for subtyping of serovar 4b

isolates. Since interlaboratory comparison of results is difficult, considerable efforts have

been undertaken for standardization of the method.

In the mid 1990s, a network (PulseNet) of public health and food regulatory laboratories that

routinely subtype foodborne pathogenic bacteria in order to rapidly detect foodborne disease

outbreaks was established by the Centers for Disease Control and Prevention in the United

States.

In order to improve quality and interlaboratory comparability of PFGE, standardized

laboratory protocols, including a 1-day protocol, were developed and rapid comparison of

PFGE patterns from different locations is possible via the Internet (39). According to the

standardized protocol, restriction endonucleases ApaI and AscI are used. In the last years,

PulseNet networks have been established in other continents as well, linking laboratories

from all over the world (http://www.pulsenetinternational.org) and facilitating rapid

detection and comparison of strains (35, 52). Recently, PulseNet surveillance provided

definitive linkages between ready-to-eat-meats and human cases in a large Canadian

outbreak of listeriosis, resulting in 22 deaths and 57 confirmed cases (http://www.phacaspc.

gc.ca/alert-alerte/listeria/listeria_2009-eng.ph).

In the recent past, faster and simpler molecular subtyping methods, like multilocus variablenumber

tandem-repeat analysis and multilocus sequence typing, have evolved, and their

application for subtyping of L. monocytogenes is supported by PulseNet. Both methods

showed a discriminatory power comparable to that of PFGE (54, 67, 73, 94). Recently, a

single-nucleotide-polymorphism-based multilocus genotyping assay that has a very high

discriminatory power for all lineages of L. monocytogenes has been developed (23, 91). In

addition, typing of L. monocytogenes isolates with a mixed-genome DNA microarray has

been established (9, 22) and compared to PFGE, ribotyping, and multilocus sequence typing.

Subtyping results were comparable to those obtained with PFGE (9).

Sequence-based molecular methods may further improve subtyping of L.

monocytogenes and allow easier data comparison via the Internet. They may further replace

typing methods with high discriminatory power but lacking interlaboratory standardization,

like random amplification of polymorphic DNA (RAPD) (31), amplified fragment length

polymorphism (31), multilocus single-strand conformation polymorphism (76), and

ribotyping (46, 75).

As a new technique, MALDI-TOF (MS) has been recently introduced for rapid typing of L.

monocytogenes. It allowed clear discrimination of all lineages and serotypes of L.

monocytogenes (3). Reproducibility, speed, and simplicity are major advantages of the

method.

Serologic Tests

Antibodies directed against listeriolysin-O have been detected in listeriosis patients by

blotting techniques with sensitivities from 50 to 96%, but the sensitivity was markedly lower

with complement fixation or O-agglutination tests (5, 63). A test based on the detection of

antibodies against recombinant truncated forms of listeriolysin O may be more specific (36).

Serologic tests cannot be recommended for the detection of past or acute listeriosis.

Antimicrobial Susceptibilities

Treatment with an aminopenicillin (ampicillin or amoxicillin) plus gentamicin is still regarded

as the most effective therapeutic regimen for listeriosis, and in vitro resistance to ampicillin

has not been described. Aminoglycosides exhibit a synergistic effect on penicillin and

aminopenicillins. Trimethoprim-sulfamethoxazole is recommended for patients who are

allergic to penicillin, and moxifloxacin may be a valuable alternative since it shows

bactericidal efficacy comparable to that of amoxicillin in vitro (40). L. monocytogenes is

intrinsically resistant to cephalosporins, fosfomycin, and fusidic acid, and even when in vitro

susceptibility may be determined, cephalosporins should not be used for therapy. Isolated

resistance against tetracycline has been noted (87) as well as multiresistance to

chloramphenicol, macrolides, and tetracyclines due to the presence of resistance plasmids

(42). Newer substances against gram-positive pathogens, like linezolid and daptomycin,

elicited high susceptibility in vitro (44, 66, 74). In addition, L. monocytogenes is generally

susceptible in vitro to erythromycin and vancomycin (81, 87). Antimicrobial susceptibility

testing should be performed in cases of suspected treatment failures, severe disease, and

patients with penicillin allergy. A CLSI guideline (M45-A2) for broth microdilution

antimicrobial susceptibility testing of L. monocytogenes including interpretive breakpoints for

penicillin, ampicillin, and trimethoprimsulfamethoxazole is available (14).

Evaluation, Interpretation, and Reporting of Results

The diagnosis of listeriosis can be made by isolation of L. monocytogenes from blood, CSF, or

specimens from other normally sterile sites. Species identification is necessary to

differentiate L. monocytogenes from nonpathogenic Listeria species. Especially for patients

with underlying immunosuppression and for individuals older than 60 years and younger

than 1 month, direct microscopic detection of gram-positive, regular, short rods in the

above-mentioned specimens should raise suspicion of listeriosis and should promptly be

communicated to the clinician in order to ensure eradication of L. monocytogenes by

antimicrobial therapy. While the presence of L. monocytogenes in specimens from normally

sterile sites indicates infection and should always be reported, detection of Listeria species in

stool samples likely represents colonization. Routine screening of stool samples

for Listeria remains unwarranted, although sporadic cases of L.

monocytogenesgastroenteritis have been reported (68).

Standard antimicrobial therapy of meningitis with cefotaxime or ceftriaxone is not active

against L. monocytogenes. Antimicrobial susceptibility testing should be performed in cases

of suspected treatment failures, severe disease, and patients with penicillin allergy. Cultures

from blood and CSF that were obtained after the initiation of antimicrobial therapy may be

negative. In these cases, detection of Listeria DNA may be useful. Commercial kits for PCRbased

detection of L. monocytogenes in CSF specimens are not yet available, but in-house

protocols and multiplex PCR formats are promising (10).

ERYSIPELOTHRIX Back to top

Taxonomy

The genus Erysipelothrix is taxonomically classified within the Erysipelotrichaceae, distinct

from the orderBacillales (86). The genus Erysipelothrix has three validly published species, E.

rhusiopathiae, E. tonsillarum, and the more recently described E. inopinata (51). Only E.

rhusiopathiae has been detected as a pathogen of humans. Based on peptidoglycan antigens

of the cell wall, several serovars can be distinguished in E. rhusiopathiae (serovars 1a, 1b,

2a, 2b, 3, 4, 5, 6, 8, 9, 11, 12, 15, 16, 17, 19, 21, and N) and E. tonsillarum(serovars 3, 7,

10, 14, 15, 16, 20, 22, and 23) (77). The vast majority of infections in humans are caused

by serovars 1 and 2.

Description of the Agent

Erysipelothrix organisms are facultatively anaerobic, non-spore-forming, non-acid-fast,

gram-positive bacteria that appear microscopically as short rods (0.2 to 0.5 μm by 0.8 to 2.5

μm) with rounded ends and occur singly, in short chains, or in long, nonbranching filaments

(60 μm or more in length). Some cells stain unevenly. They are nonmotile and grow in

complex media at a wide range of temperatures (5 to 42°C; optimum, 30 to 37°C) and at

alkaline pH (pH 6.7 to 9.2; optimum, pH 7.2 to 7.6). Like Listeria organisms, they can grow

in the presence of high concentrations of sodium chloride (up to

8.5%). Erysipelothrix organisms are catalase negative and oxidase negative, do not

hydrolyze esculin, and weakly ferment glucose without the production of gas. They are

methyl red and Voges-Proskauer negative and do not produce indole or hydrolyze urea but

distinctively produce H2S in triple sugar iron agar. Key fatty acids are C16:0 and C18:cis9 (6).

Epidemiology and Transmission

E. rhusiopathiae is distributed worldwide in nature and is remarkably stable under varying

environmental conditions. The organism is carried by a variety of animals, like mammals,

birds, and fish, in their digestive tract or tonsils but is most frequently associated with pigs.

Other domestic animals that are frequently infected include sheep, rabbits, cattle, and

turkeys. Infected animals, both sick and asymptomatic, pass the organism by urine and

feces, leading to contamination of water and soil.

Infection in animals is most likely acquired by ingestion of contaminated matter. Human

infection with E. rhusiopathiae is a zoonosis. Most cases are related to occupational

exposure, occurring most frequently among fish handlers, veterinarians, and butchers. The

disease is contracted through direct contact via skin abrasions, injuries, or animal bites (61).

E. tonsillarum has been recovered from tonsils of healthy pigs and cattle, water, and

seafood. E. inopinata has been isolated once from a vegetable-based peptone broth.

Clinical Significance

E. rhusiopathiae has been recognized for more than 100 years as the agent of swine

erysipelas, an acute or chronic disease. In humans, E. rhusiopathiae causes erysipeloid, a

localized cellulitis developing within 2 to 7 days around the inoculation site. The infected area

is swollen, and the mostly painful lesion consists of a well-defined, slightly elevated,

violaceous zone which spreads peripherally as discoloration of the central area fades.

Vesicles may be present, but suppuration does not occur. Regional lymphangitis is present in

one-third of patients, and low-grade fever and arthralgias occur in about 10% of patients.

Healing of erysipeloids usually takes 2 to 4 weeks and sometimes months, and relapses are

frequently seen. Dissemination of the organism can occur and manifests in most of the cases

as endocarditis with a poor prognosis (61). Uncommon manifestations of infection with E.

rhusiopathiae include peritonitis, endophthalmitis, osteomyelitis, intracranial abscesses, and

prosthetic joint arthritis (26, 80).

Progress has been made in the understanding of E. rhusiopathiae pathogenesis, although

data are still scarce. E. rhusiopathiae has a capsule consisting of polysaccharide antigen that

confers increased resistance to phagocytosis. Neuraminidase plays a significant role in

bacterial attachment and subsequent invasion into host cells. The 69-kDa surface antigens

SpaA, SpaB, and SpaC appear to be the major protective antigens ofE. rhusiopathiae, and

recombinant SpaA and SpaC elicit a protective immune response in pigs and mice, making

them potential candidates for a new vaccine against erysipelas (70, 79).

Collection, Transport, and Storage of Specimens

Biopsy specimens from erysipeloid lesions are the best source of E. rhusiopathiae. Care

should be taken to cleanse and disinfect the skin before sampling. The organisms typically

are located deep in the subcutaneous layer of the leading edge of the lesion; hence, a biopsy

of the entire thickness of the dermis at the periphery of the lesion should be taken for Gram

staining and culture. Swabs from the surface of the skin are not useful. In disseminated

disease, the organism can be cultured in standard blood cultures or from aspirates of the

respective infected location. For transport and storage of specimens standard procedures

should be applied.

Direct Examination

Direct microscopy should be performed in aspirates, biopsy specimens, and positive blood

cultures. Gram stain morphology of E. rhusiopathiae includes short rods and very long

filaments and thus is not distinctive. However, the presence of long, slender, gram-positive

rods in tissue from an individual with a known exposure is suggestive of erysipeloid. It has to

be noted that the organism may appear gram negative in stains from cultures (see below).

PCR assays for specific detection of E. rhusiopathiae in animal tissue as well as for

discrimination of E. rhusiopathiae from E. tonsillarum have been described (78, 92), but their

application to human samples has not been evaluated yet.

Isolation Procedures

Tissue or biopsy specimens should be processed as described in chapter 16 and plated onto

blood agar or chocolate blood agar, placed in tryptic soy, Schaedler, or thioglycolate broth,

and incubated at 35 to 37°C aerobically or in 5% CO2 for 7 days. Special pretreatment of

samples is not necessary, but inoculation of an enrichment broth significantly increases the

detection rate. Blood from patients with septicemia or endocarditis can be inoculated into

commercial blood culture systems. E. rhusiopathiae colonies generally develop in 1 to 3

days, appearing as pinpoints (<0.1 to 0.5 mm in diameter) on blood agar plates after 24 h of

incubation; at 48 h, two distinct colony types can be observed. The smaller, smooth colonies

are 0.3 to 1.5 mm in diameter, transparent, convex, and circular with entire edges. Larger,

rough colonies are flatter and more opaque and have a matte surface and an irregular,

fimbriated edge. While a temperature of 37°C favors rough colonies, smooth colonies are

favored at 30°C. A zone of greenish discoloration frequently develops underneath the

colonies on blood agar plates after 2 days of incubation (48).

Identification

Cells stain gram positive, but especially those from rough colonies can decolorize and appear

gram negative, sometimes with a beaded morphology. Cells from smooth colonies appear as

rods or coccobacilli, sometimes in short chains. Cells from rough colonies appear as long

filaments, often more than 60 μm in length.

E. rhusiopathiae is catalase negative; it also tests negative for nitrate, urease, esculin,

gelatin, xylose, mannose, maltose, and sucrose but positive for glucose, lactose, and H2S.

The extent of H2S production is influenced by the culture medium, and the strongest reaction

is found on triple sugar iron agar. Vitek2 and Phoenix automated systems, as well as the API

system (API Coryne, API ID 32 Strep), identify E. rhusiopathiaereliably. E. tonsillarum differs

biochemically from E. rhusiopathiae by being sucrose positive.

Human-pathogenic genera that have morphological and physiological characteristics in

common withErysipelothrix include mainly Lactobacillus and Listeria (24). They are regular

nonpigmented, non-spore-forming, gram-positive rods. A major discriminatory characteristic

is that E. rhusiopathiae produces H2S in triple sugar iron, whereas species of the other

genera do not. Exceptions include some Bacillus strains, but they are easily differentiated

from E. rhusiopathiae by cellular morphology, spore formation, and catalase

reaction. Listeriaspecies are catalase positive, motile, esculin positive, and not alphahemolytic.

Corynebacteria and streptococci also can be confused with E. rhusiopathiae, but

careful examination of cell morphology should facilitate the distinction. An additional trait

highly characteristic of E. rhusiopathiae is its “pipe cleaner” pattern of growth in gelatin stab

cultures incubated at 22°C (48).

Typing Systems

Serotyping schemes are available for routine use in clinical laboratories but are of limited

value since most clinical isolates belong to serovar 1 or 2. RAPD and ribotyping methods

have proved useful for epidemiological analysis of Erysipelothrix strains (1, 58). PFGE using

SmaI was superior to RAPD and ribotyping in discriminating E. rhusiopathiae isolates (58).

Recently, nucleotide sequence analysis of a hypervariable region in the spaA gene has been

introduced allowing discrimination of certain serovars of E. rhusiopathiae (55, 79).

Serologic Tests

Serologic tests for detection of antibodies to E. rhusiopathiae in humans are not available.

Vaccines for active immunization of animals are available, and protective antibodies can be

measured by enzyme immunoassay (45).

Antimicrobial Susceptibilities

Penicillin or ampicillin is the treatment of choice for both localized and systemic infections.

Broad-spectrum cephalosporins or fluoroquinolones are suitable alternatives, since no

resistance has been described yet. E. rhusiopathiae is also usually in vitro susceptible to

clindamycin, erythromycin, daptomycin, imipenem, and tetracycline (30, 60). Of note, E.

rhusiopathiae is intrinsically resistant to vancomycin and usually also to aminoglycosides and

sulfonamides. Although antimicrobial susceptibility testing of isolates is not routinely

required, testing of erythromycin and clindamycin, or further substances, may be warranted

for patients with penicillin allergy. A CLSI guideline (M45-A2) for broth microdilution

antimicrobial susceptibility testing ofErysipelothrix including interpretative breakpoints for

penicillin, ampicillin, cefepime, cefotaxime, ceftriaxone, imipenem, meropenem,

erythromycin, ciprofloxacin, gatifloxacin, levofloxacin, and clindamycin has been published

(14).

Evaluation, Interpretation, and Reporting of Results

Since human infection is rare and clinical knowledge about the disease is scarce, diagnosis of

erysipeloid is usually made accidentally by culture of E. rhusiopathiae from tissue biopsy

specimens or blood. If there is no clinical suspicion, identification of E. rhusiopathiae in the

clinical laboratory may be challenging. Detection of gram-positive and gram-variable rods,

including decolorized, beaded cells and the presence of coccobacilli and very long filaments in

direct microscopy of the specimens, gives a hint to this organism. A major discriminatory

biochemical characteristic of E. rhusiopathiae is the production of H2S.

Detection of E. rhusiopathiae in clinical samples should always be reported. Occurrence of

this species in wound or tissue specimen indicates erysipeloid rather than contamination.

Species identification is essential in order to ensure adequate antimicrobial therapy. While

penicillin and ampicillin are generally active and recommended as first-line therapy, intrinsic

resistance to vancomycin has to be noted.