Mycobacterium avium subsp. hominissuis Infection in 2 Pet Dogs, Germany

To the Editor: The genus Mycobacterium contains various obligate and opportunistic pathogens of animals, which may also be transmitted to humans and cause disease in, thus exhibiting a considerable zoonotic potential (1,2). During the past few decades, members of the Mycobacterium avium-intracellulare complex (MAIC) emerged as pathogens of human diseases, including lymphadenitis in children, pulmonary tuberculosis-like disease, and disseminated infections (occurring predominantly in immunocompromised persons, particularly AIDS patients) (1,2). Similarly, important animal diseases are caused by members of this group, e.g., avian tuberculosis and paratuberculosis in ruminants (1). MAIC includes M. intracellulare and 4 subspecies of M. avium, namely, M. avium subsp. avium, M. avium subsp. hominissuis, M. avium subsp. silvaticum, and M. avium subsp. paratuberculosis (3,4). Whereas members of the M. tuberculosis complex are transmitted by direct host contact, MAIC species are acquired predominantly from environmental sources, including soil, water, dust, and feed. Subclinical infections are common among birds (1,2). 
 
M. avium strains differ from M. intracellulare by containing the insertion sequence (IS) IS1245 (3) and are further discriminated by terms of IS901 (4). Avian isolates (M. avium subsp. avium) are usually positive for IS901 and represent the main pathogen of avian tuberculosis (5). In contrast, mammalian isolates are IS901-negative and have been designated as M. avium subsp. hominissuis because of their predominant hosts. This subspecies is only weakly virulent for birds but causes disease in animals and humans (5). 
 
Even though M. tuberculosis and M. bovis are the common etiologic agents of canine mycobacteriosis, dogs are reported to be relatively resistant to M. avium infection (6,7). Nonetheless, sporadic cases usually show nonspecific clinical signs, whereas necropsy consistently reveals granulomatous inflammation in numerous organs, including lymph nodes, intestine, spleen, liver, lung, bone marrow, and even spinal cord (7,8). The predominant involvement of the gastrointestinal tract indicates an oral route of infection (7,8), and simultaneously increases the risk for human infection by fecal spread of mycobacteria. 
 
Our report concerns 2 young dogs, a 3-year-old miniature schnauzer and a 1-year-old Yorkshire terrier, that lived in different geographic regions in Germany. Both had had therapy-resistant fever, lethargy, progressive weight loss, and generalized lymphadenomegaly for several weeks and were euthanized after a final phase of diarrhea. Necropsy findings, similar in both dogs, included generalized enlargement of lymph nodes with a whitish, granular to greasy cut surface, leading to intraabdominal adhesions by extensive involvement of mesenteric lymph nodes. In the terrier, the greater omentum and a part of the right apical lung lobe showed changes similar to those in the lymph nodes. Furthermore, numerous white 1-mm nodules were found in the spleen (both dogs), liver (schnauzer) and costal pleura (terrier). 
 
Histologic examination showed (pyo-)granulomatous inflammation of lymph nodes, tonsils, liver, spleen, and greater omentum. Additionally, pyogranulomatous pleuropneumonia was present in the terrier, and a granulomatous enteritis and pyelitis in the schnauzer. The granulomatous lesions frequently exhibited central necrosis surrounded by macrophages, epitheloid cells, and few neutrophils (Figure, panel A). However, multinucleated giant cells or mineralization was not observed. In both animals, Ziehl-Neelsen stain demonstrated large numbers of acid-fast bacilli within macrophages (Figure, panel B). Samples of lymph nodes and lung were processed for mycobacterial culture by using standard procedures (Lowenstein-Jensen, Stonebrink medium). Colonies emerging after 2-week incubation at 37°C were investigated by PCR targeting IS1245 and IS901 (3,4). In all samples, M. avium subsp. hominissuis was identified by growth characteristics as well as presence of an IS1245-specific and absence of an IS901-specific PCR product. Additionally, sequencing of hsp65 was conducted (9), which indicated M. avium subsp. hominissuis in both dogs (GenBank accession nos. {"type":"entrez-nucleotide","attrs":{"text":"EU488724","term_id":"169666617"}}EU488724 and {"type":"entrez-nucleotide","attrs":{"text":"EU488725","term_id":"169666619"}}EU488725). 
 
 
 
Figure 
 
A) Mesenteric lymph node of Yorkshire Terrier shows diffuse granulomatous lymphadenitis with extensive infiltration of macrophages, foci of pyogranulomatous inflammation (arrowhead), and focal necrosis (asterisk). Hematoxylin and eosin stain; scale bar ... 
 
 
 
Despite improved therapeutic approaches, MAIC infection represents a frequent bacterial complication in persons with AIDS. However, several studies showed a very low incidence of M. avium subsp. avium infections in humans. Thus, most of these HIV-related infections are attributed to M. avium subsp. hominissuis (2,5). Unfortunately, the subspecies of M. avium was not identified in most canine cases reported in the literature (7,8). Nonetheless, different serotypes of M. avium, corresponding to either M. avium subsp. avium or M. avium subsp. hominissuis, have been identified sporadically (6,10). The source and route of infection were unclear in all reports including ours, albeit repeatedly observed enteritis strongly suggested an oral mode of infection. A common environmental or wildlife reservoir represents the most probable source of M. avium infection for both humans and animals. However, there is also evidence of direct transmission (1–3). Therefore, M. avium subsp. hominissuis infection in dogs may comprise a considerable zoonotic potential, particularly if pet dogs with close contact to the owner are affected and if prolonged nonspecific clinical signs and intestinal involvement occur, as demonstrated here.


Mycobacterium avium subsp. hominissuis Infection in 2 Pet Dogs, Germany
To the Editor: The genus Mycobacterium contains various obligate and opportunistic pathogens of animals, which may also be transmitted to humans and cause disease in, thus exhibiting a considerable zoonotic potential (1,2). During the past few decades, members of the Mycobacterium avium-intracellulare complex (MAIC) emerged as pathogens of human diseases, including lymphadenitis in children, pulmonary tuberculosis-like disease, and disseminated infections (occurring predominantly in immunocompromised persons, particularly AIDS patients) (1,2). Similarly, important animal diseases are caused by members of this group, e.g., avian tuberculosis and paratuberculosis in ruminants (1). MAIC (3,4). Whereas members of the M. tuberculosis complex are transmitted by direct host contact, MAIC species are acquired predominantly from environmental sources, including soil, water, dust, and feed. Subclinical infections are common among birds (1,2).
M. avium strains differ from M. intracellulare by containing the insertion sequence (IS) IS1245 (3) and are further discriminated by terms of IS901 (4). Avian isolates (M. avium subsp. avium) are usually positive for IS901 and represent the main pathogen of avian tuberculosis (5). In contrast, mammalian isolates are IS901-negative and have been designated as M. avium subsp. hominissuis because of their predominant hosts. This subspecies is only weakly virulent for birds but causes disease in animals and humans (5).

LETTERS
Even though M. tuberculosis and M. bovis are the common etiologic agents of canine mycobacteriosis, dogs are reported to be relatively resistant to M. avium infection (6,7). Nonetheless, sporadic cases usually show nonspecifi c clinical signs, whereas necropsy consistently reveals granulomatous infl ammation in numerous organs, including lymph nodes, intestine, spleen, liver, lung, bone marrow, and even spinal cord (7,8). The predominant involvement of the gastrointestinal tract indicates an oral route of infection (7,8), and simultaneously increases the risk for human infection by fecal spread of mycobacteria.
Our report concerns 2 young dogs, a 3-year-old miniature schnauzer and a 1-year-old Yorkshire terrier, that lived in different geographic regions in Germany. Both had had therapy-resistant fever, lethargy, progressive weight loss, and generalized lymphadenomegaly for several weeks and were euthanized after a fi nal phase of diarrhea. Necropsy fi ndings, similar in both dogs, included generalized enlargement of lymph nodes with a whitish, granular to greasy cut surface, leading to intraabdominal adhesions by extensive involvement of mesenteric lymph nodes. In the terrier, the greater omentum and a part of the right apical lung lobe showed changes similar to those in the lymph nodes. Furthermore, numerous white 1-mm nodules were found in the spleen (both dogs), liver (schnauzer) and costal pleura (terrier).
Histologic examination showed (pyo-)granulomatous infl ammation of lymph nodes, tonsils, liver, spleen, and greater omentum. Additionally, pyogranulomatous pleuropneumonia was present in the terrier, and a granulomatous enteritis and pyelitis in the schnauzer. The granulomatous lesions frequently exhibited central necrosis surrounded by macrophages, epitheloid cells, and few neutrophils ( Figure,  panel A). However, multinucleated giant cells or mineralization was not observed. In both animals, Ziehl-Neelsen stain demonstrated large numbers of acid-fast bacilli within macrophages ( Figure, panel B). Samples of lymph nodes and lung were processed for mycobacterial culture by using standard procedures (Löwenstein-Jensen, Stonebrink medium). Colonies emerging after 2-week incubation at 37°C were investigated by PCR targeting IS1245 and IS901 (3,4). In all samples, M. avium subsp. hominissuis was identifi ed by growth characteristics as well as presence of an IS1245-specifi c and absence of an IS901-specifi c PCR product. Additionally, sequencing of hsp65 was conducted (9), which indicated M. avium subsp. hominissuis in both dogs (GenBank accession nos. EU488724 and EU488725).
Despite improved therapeutic approaches, MAIC infection represents a frequent bacterial complication in persons with AIDS. However, several studies showed a very low incidence of M. avium subsp. avium infections in humans. Thus, most of these HIV-related infections are attributed to M. avium subsp. hominissuis (2,5). Unfortunately, the subspecies of M. avium was not identifi ed in most canine cases reported in the literature (7,8). Nonetheless, different serotypes of M. avium, corresponding to either M. avium subsp. avium or M. avium subsp. hominissuis, have been identifi ed sporadically (6,10). The source and route of infection were unclear in all reports including ours, albeit repeatedly observed enteritis strongly suggested an oral mode of infection. A common environmental or wildlife reservoir represents the most probable source of M. avium infection for both humans and animals. However, there is also evidence of direct transmission (1-3). Therefore, M. avium subsp. hominissuis infection in dogs may comprise a considerable zoonotic potential, particularly if pet dogs with close contact to the owner are affected and if prolonged nonspecifi c clinical signs and intestinal involvement occur, as demonstrated here.

Serogroup Y Meningococcal Disease, Colombia
To the Editor: Neisseria meningitidis is the etiologic agent of outbreaks, epidemics, and sporadic cases of meningitis or meningococcemia. Such infections have high illness and death rates, especially in children <5 years of age and adolescents. N. meningitidis serogroups A, B, C, Y, and W135 cause most meningococcal disease worldwide (1).
In Colombia, public health notifi cation is required for all cases of invasive meningococcal disease. This reporting system is supported by a laboratory-based surveillance network for acute bacterial meningitis that has been coordinated by the Microbiology Group at the Instituto Nacional de Salud since 1994 (2,3). Clinical laboratories in Colombia submit isolates with associated information including geographic origin, specimen source, age, sex, and clinical diagnosis of the patient. Identifi cation is confi rmed by traditional phenotypic methods (4). Isolates are serogrouped by agglutination using commercial antisera (Difco, Detroit, MI, USA, and Becton Dickinson, Franklin Lakes, NJ, USA) and subtyped by dot blot with monoclonal antibodies (RIVM, Bilthoven, the Netherlands; and Institute Adolfo Lutz [IAL], São Paulo, Brazil) (5). Antimicrobial drug susceptibility testing for penicillin and rifampin is performed by the agar dilution, according to Clinical and Laboratory Standards Institute methods (6); for the breakpoints, we used those recommended by the Mesa Española de Normalización de la Sensibilidad y Resistencia a los Antimicrobianos (MENSURA) group (7). The reference laboratory participates in an external quality assurance program coordinated by the Pan American Health Organization ( Serogroup distribution was 338 (77.9%) group B, 42 (9.7%) group C, 40 (9.2%) group Y, and 2 (0.5%) group W135; 12 isolates were nongroupable. There was little annual variation for groups B and C, but there was an unexpected increase in serogroup Y (Figure), from 0% in 1994 to 50% in 2006. When the period 1994-2002 was compared with 2003-2006, this change was signifi cant, increasing from 2.2% to 29.5% (p<0.001).