Nocardia veterana as a Pathogen in North American Patients

JOURNAL OF CLINICAL MICROBIOLOGY, June 2003, p. 2560–2568 0095-1137/03/$08.00⫹0 DOI: 10.1128/JCM.41.6.2560–2568.2003 Copyright © 2003, American Society for Microbiology. All Rights Reserved.

Vol. 41, No. 6

Nocardia veterana as a Pathogen in North American Patients Patricia S. Conville,1* June M. Brown,2 Arnold G. Steigerwalt,2 Judy W. Lee,1† Dorothy E. Byrer,3‡ Victoria L. Anderson,4 Susan E. Dorman,4§ Steven M. Holland,4 Barbara Cahill,5 Karen C. Carroll,6储 and Frank G. Witebsky1 Microbiology Service, Department of Laboratory Medicine, Warren G. Magnuson Clinical Center,1 and Laboratory of Host Defenses, National Institute of Allergy and Infectious Diseases,4 National Institutes of Health, U.S. Department of Health and Human Services, Bethesda, Maryland; Meningitis and Special Pathogens Branch, Division of Bacterial and Mycotic Diseases, National Center for Infectious Diseases, Centers for Disease Control and Prevention, U.S. Department of Health and Human Services, Atlanta, Georgia2; American Type Culture Collection, Manassas, Virginia3; and Division of Pulmonary and Critical Care Medicine, University of Utah School of Medicine,5 and Department of Pathology, University of Utah,6 Salt Lake City, Utah Received 27 January 2003/Returned for modification 14 February 2003/Accepted 24 February 2003

The molecular methodologies used in our laboratories have allowed us to define a group of Nocardia isolates from clinical samples which resemble the type strain of Nocardia veterana. Three patient isolates and the type strain of N. veterana gave identical and distinctive restriction fragment length polymorphisms (RFLPs) for an amplified portion of the 16S rRNA gene. These three isolates and the N. veterana type strain also gave identical RFLPs for an amplified portion of the 65-kDa heat shock protein gene, but this pattern was identical to that obtained for the Nocardia nova type strain. Sequence analysis of both a 1,359-bp region of the 16S rRNA gene and a 441-bp region of the heat shock protein gene of the patient isolates showed 100% identities with the same regions of the N. veterana type strain. DNA-DNA hybridization of the DNA of one of the patient isolates with the DNA of the N. veterana type strain showed a relative binding ratio of 82%, with 0% divergence, confirming that the isolate was N. veterana. Biochemical and susceptibility testing showed no significant differences among the patient isolates and the N. veterana type strain. Significantly, the results of antimicrobial susceptibility testing obtained for our isolates were similar to those obtained for N. nova, indicating that susceptibility testing alone cannot discriminate between these species. We present two case studies which show that N. veterana is a causative agent of pulmonary disease in immunocompromised patients residing in North America. We also describe difficulties encountered in using 16S rRNA gene sequences alone for discrimination of N. veterana from the related species Nocardia africana and N. nova because of the very high degree of 16S rRNA gene similarity among them.

Nocardia species have been implicated as the causes of cutaneous, ocular, pulmonary, and disseminated diseases (16) in both immunocompetent and immunocompromised human hosts. Over the past several years the spectrum of disease caused by Nocardia species has changed due to the increase in the number of immunocompromised patients. While cutaneous disease is still the most common presentation in immunocompetent individuals, such infections also occur in the immunocompromised host. However, pulmonary and disseminated infections are the more common presentations in immunocompromised individuals (16).

Molecular methodologies have been instrumental in the recognition or description of several new species of Nocardia which have been recognized as human pathogens (11, 23–25, 30–32). For several years we have been using such methodologies to identify Nocardia species isolated from clinical specimens. Our procedure involves PCR amplification of portions of both the 16S rRNA gene and the 65-kDa heat shock protein (HSP) gene and subsequent restriction endonuclease analysis (REA) of the PCR products (6, 20). Our experience has shown that in most cases, correct species assignment can be made when the identification obtained by REA of the 16S rRNA gene region is identical to that obtained by REA of the HSP gene region. However, when the identifications obtained from REAs of the two gene regions differ or when no identification can be obtained from REAs of one or both gene regions, subsequent sequencing of the 16S rRNA gene frequently reveals that the organism belongs to an unusual or undescribed species. We have isolated and studied a cluster of Nocardia isolates with similar 16S rRNA gene sequences which to various degrees resemble Nocardia africana, Nocardia nova, Nocardia veterana, and Nocardia vaccinii in their phenotypic and molecular features. Further evaluation of three of these isolates showed that they were inseparable from N. veterana by the

* Corresponding author: Mailing address: Microbiology Service, Department of Laboratory Medicine, 10 Center Dr., MSC 1508, National Institutes of Health, Bethesda, MD 20892-1508. Phone: (301) 496-4433. Fax: (301) 402-1886. E-mail: [email protected]. † Present address: Northwestern University, Feinberg School of Medicine, Chicago, Ill. ‡ Present address: Meningitis and Special Pathogens Branch, Division of Bacterial and Mycotic Diseases, National Center for Infectious Diseases, Centers for Disease Control and Prevention, U.S. Department of Health and Human Services, Atlanta, Ga. § Present address: Johns Hopkins Center for Tuberculosis Research, Johns Hopkins Medical Institutions, Baltimore, Md. 储 Present address: Johns Hopkins University School of Medicine, The Johns Hopkins Hospital, Baltimore, Md. 2560

NOCARDIA VETERANA IN NORTH AMERICAN PATIENTS

VOL. 41, 2003

phenotypic and molecular methods that we regularly use; DNA-DNA hybridization of the DNA of one of these isolates with the DNA of the N. veterana type strain confirmed the identification of these isolates as N. veterana. We describe here our clinical isolates of N. veterana and the methods used to recognize them and provide a brief clinical history of two immunocompromised patients infected with this pathogen. We conclude that N. veterana is a pathogen in patients residing in North America and may be more common than generally believed because of difficulties in discriminating this species from others, such as N. nova. We also discuss difficulties encountered in the identification of Nocardia isolates solely by comparison of 16S rRNA gene sequences. (Some of the data pertaining to case 1 have been included in a report summarizing the experience to date with Nocardia infections in chronic granulomatous disease [7].) CASE REPORTS Case patient 1 (isolate A). A 24-year-old white male with X-linked chronic granulomatous disease was admitted to the National Institutes of Health (NIH) Clinical Center, Bethesda, Md., in May 1995 after a routine computed tomographic scan of the chest showed a new left-sided pulmonary infiltrate. A fine-needle transthoracic biopsy of the lesion was performed. No organisms were seen on Gram and modified acid-fast staining of the specimen. Methenamine silver and acid-fast staining performed in a cytology laboratory was negative for organisms. Routine bacteriologic, fungal, and mycobacterial cultures all grew a Nocardia species. Because the patient was allergic to sulfa drugs, he received intravenous amikacin and ceftriaxone along with oral trimethoprim for 2 months. After 2 months the amikacin was stopped, oral minocycline was added, and intravenous ceftriaxone was continued. His pneumonia resolved completely, and after completing 3 months of ceftriaxone treatment, he was thereafter maintained on an oral regimen of amoxicillin-clavulanate, trimethoprim, and minocycline and continued to do well. Isolate A (case patient 1) has been deposited in the American Type Culture Collection as N. veterana and assigned the number ATCC BAA-509. Case patient 2 (isolate C). A 63-year-old man underwent a left-sided single-lung transplant for chronic obstructive lung disease at the University of Utah Medical Center, Salt Lake City, in April 2000. His postoperative course was complicated by cytomegalovirus pneumonia and recurrent acute rejection, bacterial pneumonia, and nasal herpes simplex virus infection. In addition, the patient received itraconazole and caspofungin for 12 weeks for an invasive Aspergillus fumigatus infection in his native lung. Seventeen months after transplantation, following several months of treatment with prednisone for bronchiolitis obliterans, the patient developed worsening dyspnea and a new nodule in his transplanted lung. At that time his immunosuppressive regimen included tacrolimus, azathioprine, and prednisone. A computed tomography-guided transthoracic fine-needle aspirate of the nodule was performed in August 2001. Histopathology revealed abundant necrotic debris and acute neutrophilic inflammation. Gram staining, methenamine silver staining, Wright staining, and acid-fast staining were negative for organisms. Staining with a modified acid-fast stain was not performed. Routine microbiologic cul-

2561

tures grew an isolate identified as most closely resembling N. nova. He was treated with trimethoprim-sulfamethoxazole, and the amount of prednisone that he was taking was reduced. The nodule had decreased in size 8 weeks into a planned 24-week course of therapy. After 16 weeks of therapy the trimethoprim-sulfamethoxazole was discontinued at the patient’s request because of side effects from the drug. A chest X ray obtained 4 weeks after antibiotic discontinuation showed no evidence of increased opacities or nodularity. Eight weeks after antibiotic discontinuation the patient elected to terminate immunosuppressive therapy; he died 1 week later. No autopsy was performed. MATERIALS AND METHODS Organisms. (i) Reference strains. The following reference strains were used for molecular, biochemical, and/or antimicrobial studies (GenBank accession numbers are given in parentheses): N. africana DSM 44491T and ATCC BAA280T (AY089701), N. nova ATCC 33726T (AY191250), N. vaccinii ATCC 11092T (AY191252), and N. veterana DSM 44445T (AY191253). (ii) Patient isolates. Three patient isolates were identified as N. veterana. Two of these isolates were from patients for whom clinical histories were available. One of these was from a patient being treated at the Warren G. Magnuson Clinical Center of NIH (isolate A from a lung biopsy specimen from case patient 1), and the other was from a patient being treated at the University of Utah Health Sciences Center (isolate C, also from a lung biopsy specimen from case patient 2). An additional isolate obtained from a bronchial wash (isolate B) had been referred to ARUP Laboratories, Salt Lake City, Utah. A fourth isolate was also preliminarily identified as N. veterana (isolate D); this isolate was obtained from a lung biopsy specimen from a patient being treated at the Clinical Center of NIH. (iii) GenBank sequences. For comparison of 16S rRNA gene sequences, sequences from the following organisms were used (GenBank accession numbers are given in parentheses; isolates were included only if unambiguous associations could be made between the type strain and the GenBank accession number): Nocardia abscessus ATCC 23824 (X84851), Nocardia asteroides ATCC 19247T (Z36934), N. asteroides drug pattern type II (AF163818), N. asteroides drug pattern type IV ATCC 49872 (AY191251), N. asteroides drug pattern type VI ATCC 14759 (AF162772), Nocardia beijingensis JCM 10666T (AF154129), Nocardia brasiliensis ATCC 19296T (Z36935), Nocardia brevicatena ATCC 15333T (X80600), Nocardia carnea DSM 43397T (Z36929), Nocardia crassostreae ATCC 700418T (Z37989), Nocardia cyriacigeorgica DSM 44484T (AF282889), Nocardia farcinica ATCC 3318T (Z36936), Nocardia flavorosea JCM 3332T (Z46754), Nocardia ignorata DSM 44496T (AY191254), Nocardia otitidiscaviarum ATCC 14629T (X80599), Nocardia paucivorans DSM 44386T (AF179865), Nocardia pseudobrasiliensis ATCC 51512T (X84857), Nocardia salmonicida JCM 4826T (Z46750), Nocardia seriolae ATCC 43993T (X80592), Nocardia transvalensis ATCC 6865T (X80598), Nocardia uniformis JCM 3224T (Z46752), Nocardia vinacea JCM 10988 (AB024312), and Rhodococcus equi ATCC 6939T (X80603). Phenotypic identification. Colonies were examined for aerial hyphae by using a dissecting microscope; microscopic morphology was determined by modified acid-fast staining (4). Biochemical tests were performed at the Actinomycete Reference Laboratory of the Centers for Disease Control and Prevention (CDC) by the methods of Berd (1). Growth requirements at 25, 35, and 45°C were determined on heart infusion agar for all isolates except N. vaccinii, for which growth requirements were determined on tryptone glucose yeast agar (Difco) to obtain better growth. Utilization of acetamide and citrate as the sole carbon and nitrogen sources was determined as described by Wallace et al. (25); arylsulfatase production tests were performed as described by Kent and Kubica (13). Esculin hydrolysis (Remel, Lenexa, Kans.) was performed on solid medium as described by Williams et al. (29). Susceptibility testing. Organisms were grown at 35°C in 25 ml of Middlebrook 7H9 broth (Difco, Sparks, Md.) with albumin-dextrose-catalase enrichment (Gibco, Carlsbad, Calif.) by using glass beads and continuous shaking (generally for 3 days) until good growth was achieved. The inoculum was standardized according to NCCLS guidelines (17). Ten microliters of a standardized inoculum was added to each well of a prepared microdilution plate (PML Microbiologicals, Wilsonville, Oreg.) containing antimicrobial agents diluted in 100 ␮l of cationsupplemented Mueller-Hinton broth per well. The concentrations of amoxicillinclavulanic acid tested differed from those recommended by the NCCLS in that

2562

CONVILLE ET AL.

the clavulanate concentration increased twofold in each succeeding well beginning with 0.5/0.25 ␮g/ml (amoxicillin-clavulanic acid). The plates were incubated at 35°C for 72 h in ambient air; end points were determined according to NCCLS guidelines (17). The interpretive criteria used for all drugs except linezolid and vancomycin were those recommended by the NCCLS (17); no interpretive criteria for linezolid and vancomycin have yet been established for Nocardia species. Molecular analysis. (i) DNA extraction (for restriction fragment length polymorphism [RFLP] analysis and gene sequencing analyses). Organisms were grown on Sabouraud dextrose agar plates (Hardy Diagnostics, Santa Monica, Calif.); DNA was extracted as described previously (6). (ii) PCR. Amplification of a portion of the 16S rRNA gene was performed as described previously (6) or with Ready-To-Go PCR Beads (Amersham Pharmacia Biotech, Inc. Piscataway, N.J.) with the same primers at a concentration of 0.25 ␮M each. Amplification of a portion of the HSP gene was performed with the primers described by Telenti et al. (21) and the amplification conditions described by Steingrube et al. (20) or with Ready-To-Go PCR Beads with the same primers at a concentration of 0.3 ␮M each. (iii) REA. The PCR products from the amplification of the 16S rRNA gene were subjected to digestion with HinP1I and DpnII (New England Biolabs, Beverly, Mass.) (6). The PCR products from the amplification of a portion of the HSP gene were subjected to digestion with MspI and HinfI (New England Biolabs) (20). All digestions were performed under the conditions recommended by the manufacturer. The digestion reaction mixtures were electrophoresed (6), and the resulting RFLP patterns from all gene digestions were interpreted by using the Molecular Analyst software PC Fingerprinting Plus (Bio-Rad Laboratories, Hercules, Calif.) (6, 20). (iv) Determination of 16S rRNA gene sequence. The sequences of the first 999-bp region of the 16S rRNA gene were determined by using overlapping primers with tails containing M13 binding sites (previously described as reactions 1 and 2 [6]). The 16S rRNA gene sequence for N. asteroides drug pattern type IV was determined by using an alternative downstream primer for reaction 2 due to a lack of amplification with the previously described primers (6). The sequence of this primer was 5⬘-CAG-GAA-ACA-GCT-ATG-ACG-TAC-ACC-AGC-CAC-A AG-GGA-ACC-3⬘ (the M13 binding site sequence is indicated in boldface); the PCR conditions were identical to those used for the other Nocardia species. The sequence of an extended region of the 16S rRNA gene (reaction 3) corresponding to bases 910 to 1463 of the N. asteroides type strain (GenBank accession number Z3694) was determined by using an additional set of primers with tails containing M13 forward binding sites or M13 reverse binding sites. The primer sequences were 5⬘-GTA-AAA-CGA-CGG-CCA-GGA-TGC-AAC-GCG-AAG-A AC-CTT-ACC-T-3⬘ and 5⬘-CAG-GAA-ACA-GCT-ATG-ACT-ACG-GCT-ACC-T TG-TTA-CGA-CTT-CG-3⬘, respectively (the M13 binding site sequences are indicated in boldface). The PCR conditions were identical to those used for initial amplification of the 16S rRNA gene (6). The PCR amplification products from all regions were electrophoresed on 2% Tris acetate-EDTA gels. The bands were dissected, and the DNA was purified by using the GFX PCR DNA and Gel Band Purification kit (Amersham Pharmacia Biotech, Inc., Little Chalfont, England). Cycle sequencing of bases 1 to 999 was performed (6). Cycle sequencing of the extended region (bases 910 to 1463) was performed by using M13 (⫺20) forward and reverse primers (Invitrogen, Carlsbad, Calif.). All cycle sequencing reactions were performed with the ABI Prism BIG Dye Terminator Cycle Sequencing Ready Reaction kit (Perkin-Elmer Applied Biosystems, Foster City, Calif.) (6). Excess dye terminators were removed by ethanol-sodium acetate precipitation according to the guidelines of the manufacturer. Fluorescencebased sequence analysis of the extension products was performed with the ABI 310 Genetic Analyzer or the ABI 3100 Genetic Analyzer (Applied Biosystems/ Hitachi, Foster City, Calif.). (v) Determination of 65k-Da HSP gene sequence. The sequences of a 460-bp region of the HSP gene were determined by using the primers described by Telenti et al. (21), with the tails containing M13 binding sites added to the 5⬘ end. The primer sequences were 5⬘-GTA-AAA-CGA-CGG-CCA-GAC-CAA-CGA-TG G-TGT-GTC-CAT-3⬘ and 5⬘-CAG-GAA-ACA-GCT-ATG-ACC-TTG-TCG-AA C-CGC-ATA-CCC-T-3⬘ (M13 binding site sequences are indicated in boldface). The PCR conditions were identical to those used by Steingrube et al. (20), and template preparation and cycle sequencing were performed as described above for reaction 3. (vi) Sequence analysis and comparisons. Sequences were assembled by using SeqMan II software (DNASTAR, Inc., Madison, Wis.). Related sequences were identified by using the Basic Local Alignment Search Tool (BLAST; National Center for Biotechnology Information, NIH). Sequence lengths were adjusted to match the length of the shortest sequence, and all sequences were then aligned with Megalign software (DNASTAR, Inc.) by using the ClustalV algorithm.

J. CLIN. MICROBIOL. Percent similarity was determined and phylogenetic trees were constructed by using the Megalign software and sequences of similar lengths. (vii) DNA-DNA hybridization. Purified DNA of the type strains of N. africana, N. nova, N. vaccinii, and N. veterana and purified DNA of isolates D and A were prepared from lysed protoplasts as described previously (14). To improve the DNA yield, repeat extractions were performed with 20% sodium dodecyl sulfate by a procedure adapted from that of Loeffelholz and Scholl (15). N. veterana DNA was labeled with [32P]dCTP by using a nick translation kit (Gibco). The hybridization method for determination of the levels of DNA relatedness by absorbance to hydroxyapatite has been described previously (3). All reactions were performed in duplicate at 70°C. The relative binding ratio (RBR) ([percentage of heterologous DNA bound to hydroxyapatite/percentage of homologous DNA bound to hydroxyapatite] ⫻ 100) was calculated by the method of Brenner et al. (2). The percent divergence (calculated to the nearest 0.5%) was determined by assuming that each degree of heteroduplex instability, when compared to the melting temperature of the homologous duplex, was caused by 1% unpaired bases (2). Nucleotide sequence accession numbers. The nucleotide sequences which we determined and submitted to GenBank have been assigned the indicated accession numbers for the following organisms: N. africana DSM 44491, AY089701; N. asteroides drug pattern type IV ATCC 49872, AY191251; N. ignorata DSM 44496T, AY191254; N. nova ATCC 33726T, AY191250; N. vaccinii ATCC 11092T, AY191252; and N. veterana DSM 44445T, AY191253.

RESULTS Biochemical results. All organisms were weakly acid fast (by modified acid-fast staining), and all isolates produced aerial hyphae. All isolates tested were negative for acid production from the oxidation of adonitol, L-arabinose, dulcitol, i-erythritol, lactose, melibiose, D-sorbitol, and sucrose; negative for hydrolysis of adenine, hypoxanthine, tyrosine, and xanthine; and negative for the utilization of acetamide. The results obtained for acid production from oxidation of i-myo-inositol and L-rhamnose for N. veterana differed from those obtained by Gu ¨rtler et al. (10) for the same organism. All other biochemical test results for reference strains and patient isolates are shown in Table 1. Susceptibility testing results. Patient isolates and the type strain of N. veterana showed antimicrobial susceptibility patterns essentially identical to those obtained for the type strains of N. nova and N. africana (Table 2). Slight variations in ampicillin MICs were seen. The results for isolate C showed the most variation from those for the type strains of N. africana, N. nova, and the other patient isolates, but except for the results for sulfamethoxazole, the results remained within the same interpretive range as those obtained for the type strains. REA results. The band sizes obtained by digestion of the PCR amplification products of portions of the 16S rRNA and HSP genes of the type strains of N. africana, N. nova, N. vaccinii, and N. veterana are shown in Table 3. Digestion of the 999-bp region of the 16S rRNA gene with the enzymes HinP1I and DpnII showed a unique set of RFLP patterns which were essentially identical for all four patient isolates and for the N. africana and N. veterana type strains (Fig. 1). Digestion of the portion of the HSP gene with the enzymes MspI and HinfI showed RFLP patterns for the four patient isolates and the N. africana and N. veterana type strains essentially identical to each other and to those obtained for the type strain of N. nova (Fig. 1). The set of RFLP patterns for the 16S rRNA gene of N. vaccinii was unique and resembled the patterns for N. asteroides drug pattern type IV or N. transvalensis by MspI and HinfI digestion of the HSP gene (Table 3) (20).

NOCARDIA VETERANA IN NORTH AMERICAN PATIENTS

VOL. 41, 2003

2563

TABLE 1. Biochemical assay results for N. veterana patient isolates compared with those for the type strains of N. africana, N. nova, N. vaccinii, and N. veterana Resulta Characteristic

N. africana ATCC BAA-280T

N. nova ATCC 33726T

N. vaccinii ATCC 11092

N. veterana DSM 44445T

Isolate A

Isolate B

Isolate C

Acid from oxidation of: Cellobiose D-Fructose D-Galactose D-Glucose Glycerol i-myo-Inositol Maltose D-Mannitol Mannose Raffinose L-Rhamnose Salicin Trehalose D-Xylose

⫺ ⫺ ⫺ ⫹ ⫹ ⫺ ⫹ ⫺ ⫺ ⫹ ⫺ ⫺ ⫹ ⫺

⫺ ⫺ ⫹ ⫹ ⫹ ⫺ ⫹ ⫺ ⫺ ⫺ ⫺ ⫺ ⫹ ⫺

⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫺ ⫹b ⫺ ⫹ ⫹

⫺ ⫺ ⫺ ⫹ ⫹ ⫺b ⫹ ⫺ ⫺ ⫺ ⫺b ⫹ ⫹ ⫺

⫺ ⫺ ⫺ ⫹ ⫹ ⫺ ⫹ ⫺ ⫺ ⫺ ⫺ ⫺ ⫹ ⫺

⫺ ⫺ ⫺ ⫹ ⫹ ⫺ ⫹ ⫺ ⫺ ⫺ ⫺ ⫹ ⫹ ⫺

⫺ ⫺ ⫺ ⫹ ⫾ ⫺ ⫹ ⫺ ⫺ ⫺ ⫺ ⫹ ⫺ ⫺

Growth on heart infusion agar at: 25°C 35°C 45°C

⫹ ⫹ ⫹

⫹ ⫹ ⫾

⫹c ⫹c ⫺c

⫹ ⫹ ⫹

⫹ ⫹ ⫹

⫹ ⫹ ⫹

⫹ ⫹ ⫹

Growth in lysozyme

⫹

⫹

⫹

⫹

⫹

⫹

⫹

Arylsulfatase production at: 3 days 14 days

⫹ (1⫹) ⫹ (2⫹)

⫹ (3⫹) ⫹ (3⫹)

⫺ ⫺

⫺ ⫺

⫺ ⫺

⫺ ⫺

⫺ ⫺

Hydrolysis of: Casein Esculin Esculin with bile

⫹ ⫺ ⫺

⫺ ⫺ ⫺

⫺ ⫺ ⫺

⫺ ⫹ ⫹

⫺ ⫹ ⫹

⫺ ⫹ ⫹

⫺ ⫹ ⫹

⫺

NG

⫺b

⫺

⫺

⫺

⫺

Utilization of citrate medium as sole carbon and nitrogen source

a Symbols and abbreviations: ⫺, negative; ⫹, positive; ⫾, weak reaction in 21 days; NG, no growth. See text for tests for which results for all organisms tested were negative. b Results differ from those obtained by Gu ¨rtler et al. (10) due to methodology differences (see text). c Tested on tryptone glucose yeast medium.

BLAST analysis of 16S rRNA sequences. Isolates A, B, and C showed the highest degree of similarity to N. veterana, N. africana, N. nova, and N. vaccinii by BLAST analysis. Alignment of 1,359-bp segments of the 16S rRNA genes of these

patient isolates with the 16S rRNA gene of the type strain of N. veterana showed 100% similarity; the similarities to the 16S rRNA genes of the type strains of N. africana, N. nova, and N. vaccinii were 99.0, 97.7, and 97.7%, respectively (Table 4).

TABLE 2. Susceptibility testing results for N. veterana patient isolates compared with those for a reference strain of N. africana and type strains of N. nova, N. vaccinii, and N. veteranaa Drug

Amikacin Amox-clav Ampicillin Ceftriaxone Ciprofloxacin Clarithromycin Imipenem Linezolidc Minocycline Smx Tmp-Smx Vancomycinc a b c

Breakpoint for resistance (␮g/ml)b

ⱖ16 ⱖ32/16 ⱖ32 ⱖ64 ⱖ4 ⱖ8 ⱖ6 ⱖ8 ⱖ64 ⱖ4/76

MIC (␮g/ml) N. africana ATCC BAA-280T

N. nova ATCC 33726T

N. vaccinii ATCC 11092T

N. veterana DSM 44445T

Isolate A

Isolate B

Isolate C

ⱕ0.25 32/16 4 8 8 ⱕ0.25 ⱕ0.25 0.25 2 32 0.5/9.5 32

ⱕ0.25 32/16 1.0 8 8 ⱕ0.25 ⱕ0.25 1 2 128 1/19 ⬎32

NG NG NG NG NG NG NG NG NG NG NG NG

ⱕ0.25 16/8 2 4 8 ⱕ0.25 ⱕ0.25 2 2 64 1/19 ⬎32

ⱕ0.25 16/8 2 8 4 ⱕ0.25 ⱕ0.25 2 2 64 1/19 ⬎32

ⱕ0.25 32/16 4 8 8 ⱕ0.25 ⱕ0.25 2 2 64 1/19 ⬎32

ⱕ0.25 32/16 1.0 2 8 2 ⱕ0.25 1 2 32 1/19 ⬎32

Abbreviations: NG, no growth; Amox-clav, amoxicillin-clavulanate; Smx, sulfamethoxazole; Tmp-Smx, trimethoprim-sulfamethoxazole. The MIC breakpoints used are those of the NCCLS (17). No established breakpoints exist.

2564

CONVILLE ET AL.

J. CLIN. MICROBIOL.

TABLE 3. Band sizes obtained by digestion of the PCR amplification products of portions of the 16S rRNA and HSP genes for the type strains of N. africana, N. veterana, N. nova, and N. vaccinii Gene

Enzyme

16S rRNA HSPb a b

Band(s) size (bp)a N. africana, N. veterana

N. nova

N. vaccinii

HinP1I DpnII

420, 225, 175, 125, 55 455, 250, 200, 95

420, 350, 225 640, 200, 95, 60

410, 225, 175, 125, 55 370, 250, 195, 90, 70

MspI HinfI

130, 110, 75 440

130, 110, 75 440

260, 180 250, 185

Rounded to the nearest 5 bp. The method of Steingrube et al. (20) was used.

Alignment of the same region of the 16S rRNA gene from isolate D showed that it was 99.6% similar to that of the type strain of N. veterana and 99.1, 98.0, and 98.0% similar to those of N. africana, N. nova, and N. vaccinii, respectively. HSP sequence analysis. Comparison of the HSP gene sequences of isolates A, B, and C showed that they were 100% similar to the HSP gene sequence of the type strain of N. veterana; the sequence similarities to the HSP gene sequences of the type strains of N. africana, N. nova, and N. vaccinii were 99.4, 99.1, and 93.0%, respectively (Table 4). Alignment of the same region of the HSP gene from isolate D showed that it was 99.8% similar to that of the type strain of N. veterana and 99.1, 98.9, and 92.8% similar to those of N. africana, N. nova, and N. vaccinii, respectively.

DNA-DNA hybridization. The RBR for isolate A to the N. veterana type strain was 82%, with 0% divergence. The RBRs for isolate A to the type strains of N. vaccinii, N. nova, and N. africana were 16, 33, and 39%, respectively, with divergences ranging from 8.7 to 11.2% (Table 5). Phylogenetic relationships. The phylogenetic relationships of patient isolates of N. veterana, the type strain of N. veterana, and most other described Nocardia species for which sequence information is available are shown in Fig. 2. All patient isolates and the type strain of N. veterana form a distinct cluster separate from the other species described. An incidental finding from our study was that the 16S rRNA gene sequence that we obtained for N. asteroides drug pattern type VI (ATCC 14759) and submitted to GenBank (accession number AF162772) dif-

FIG. 1. Bio-Rad-derived RFLP patterns for isolate A and type strains of Nocardia species. Bands smaller than 60 bp are not shown. Gel 1, HinP1I digests of 16S rRNA gene, 999-bp region; gel 2, DpnII digests of 16S rRNA gene, 999-bp region; gel 3, MspI digests of HSP gene, 439-bp region; gel 4, HinfI digests of HSP gene, 439-bp region. Lanes A, base pair marker; lanes B, isolate A; lanes C, N. veterana DSM 44445T; lanes D, N. africana DSM 44491T; lanes E, N. nova ATCC 33726T; lanes F, N. vaccinii, ATCC 11092T.

NOCARDIA VETERANA IN NORTH AMERICAN PATIENTS

VOL. 41, 2003

2565

TABLE 4. Percent identities of 16S rRNA and HSP gene sequences of patient isolates and type strains of N. veterana, N. africana, N. nova, and N. vacciniia % Similarity Isolate or taxon

Isolate A Isolate B Isolate C Isolate D N. veterana N. africana N. nova a

16S rRNA gene (1,359-bp segment)

HSP gene (441-bp segment)

N. veterana

N. africana

N. nova

N. vaccinii

N. veterana

N. africana

N. nova

N. vaccinii

100 100 100 99.6

99.0 99.0 99.0 99.1 99.0

97.7 97.7 97.7 98.0 97.7 98.4

97.7 97.7 97.7 98.0 97.7 98.5 98.1

100 100 100 99.8

99.4 99.4 99.4 99.1 99.4

99.1 99.1 99.1 98.9 99.1 99.4

93.0 93.0 93.0 92.8 93.0 92.3 92.1

Ambiguous bases, insertions, and deletions were considered base differences.

fered by the absence of 1 base compared to the sequence for N. cyriacigeorgica (DSM 44484) in GenBank (accession number AF282889). The 16S rRNA sequence that we subsequently determined for the N. cyriacigeorgica type strain was identical to that which we had obtained for the N. asteroides drug pattern type VI strain. DISCUSSION N. veterana was initially isolated from a bronchial lavage specimen from an Australian patient and described, but the isolate was not considered the cause of that patient’s pulmonary disease (10). Isolation of N. veterana from a mycetoma in Japan was recently reported (12); the ribosomal DNA sequence of this isolate showed 99.8% similarity to that of the type strain of N. veterana. An additional report mentions the isolation of N. veterana from a pulmonary specimen in Germany (28); the 999-bp region of the 16S rRNA gene of this isolate showed only 99.1% similarity to that of the type strain of N. veterana. The identification of neither of these isolates was confirmed by DNA-DNA hybridization with the type strain of N. veterana. We report here on the isolation of N. veterana from clinical specimens of three North American patients with pulmonary disease, two of which were considered to be the cause of significant disease in these immunocompromised patients. The use of molecular methods has allowed us to identify these three clinical isolates of N. veterana and has also highlighted the difficulties encountered in the differentiation of N. veterana from the closely related species N. nova and N. africana. While REA of the 16S rRNA of patient isolates and reference strains has been useful for the differentiation of N. veterana from N. nova and N. vaccinii, neither 16S rRNA nor HSP RFLP assays enabled discrimination between N. veterana and N. africana (Table 3; Fig. 1). REA of the HSP gene alone was also not useful in the differentiation of N. africana and N. veterana from N. nova. BLAST analysis of the 16S rRNA gene sequences indicated that our patient isolates had the closest similarity to N. africana, N. nova, N. vaccinii, and N. veterana. Subsequent 16S rRNA and HSP gene alignments showed that both gene regions of our patient isolates A, B, and C were identical to those of N. veterana. The sequencing results for these 16S rRNA and HSP gene sequences of the three patient isolates also showed that they had close similarity to the type

strain of N. africana (99.0 and 99.4%, respectively). Phylogenetic analysis based on the 16S rRNA sequences (Fig. 2) verified the close relationship of our patient isolates to N. veterana, N. africana, N. nova, and N. vaccinii, which form a separate branch on the phylogenetic tree. By using the accepted interpretive criteria (an RBR greater than 70% and less than 6% divergence is considered indicative of conspecificity [2, 27]), the DNA-DNA hybridization results confirmed that isolate A is N. veterana, with 82% relatedness and 0% divergence. The 16S rRNA and HSP gene sequences of isolate D showed 99.6 and 99.8% similarities to those of N. veterana, respectively. DNA-DNA hybridization results showed that isolate D is related to N. veterana, with an RBR of only 35% and 8.7% divergence, which is in contrast to the very close relationship of isolate A to N. veterana; detailed characterization of isolate D is in progress. Given the labor-intensiveness of the DNA-DNA hybridization procedure, hybridization studies were not performed with isolates B and C. However, we concluded that they are also isolates of N. veterana, given the 100% similarities of the 16S rRNA and HSP gene sequences of these two isolates with those of both isolate A and the N. veterana type strain. Biochemical analysis performed at CDC showed that isolates A, B, and C have phenotypic characteristics essentially identical to those of the type strain of N. veterana, with some variation in results for acid production from oxidation of salicin and trehalose. Also, the results obtained for the N. veterana type strain for utilization of i-myo-inositol and L-rhamnose differed from those obtained by Gu ¨rtler et al. (10) and may have resulted from differences in testing methodologies (assimilation detected photometrically [10] versus acid production TABLE 5. DNA relatedness among type strains of Nocardia spp. and patient isolates Source of unlabeled DNA

N. veterana DSM 44445T labeled DNA RBR (% D)a

N. veterana DSM 44445T ................................................ 100 Isolate A ........................................................................... 82 (0.0) Isolate D ........................................................................... 35 (8.5) N. africana DSM 44491T................................................. 39 (9.0) N. nova ATCC 33726T .................................................... 33 (9.0) N. vaccinii ATCC 11092T ............................................... 16 (11.0) a

See the text for the definition of RBR. % D, percent divergence (see text).

2566

CONVILLE ET AL.

J. CLIN. MICROBIOL.

FIG. 2. Phylogenetic tree of type strains of Nocardia species and patient isolates prepared by using the ClustalV algorithm of Megalign software (DNAstar, Inc.).

from oxidation of carbohydrates determined with bromcresol purple as the indicator [the CDC method]). Our patient isolates and the N. veterana type strain had identical results for these two carbohydrates by the CDC method. On the basis of our results, critical biochemical tests for the differentiation of isolates of N. veterana from N. africana, N. nova, and N. vaccinii appear to be 3- and 14-day arylsulfatase production and esculin hydrolysis; casein hydrolysis and acid production from oxidation of raffinose and D-galactose may also be useful in discriminating members of these species. N. vaccinii showed numerous biochemical differences from N. africana, N. nova, and N. veterana and is easily discriminated from these species (Table 1). Antimicrobial susceptibility testing showed no significant differences between the patient isolates and the type strain of N. veterana. The clarithromycin susceptibility results for isolate C showed a greater than 1 twofold dilution difference from the results for the N. veterana type strain, but the interpretive category was the same as that for the type strain (susceptible). Some phenotypic differences exist between the results obtained for the type strain of N. nova in this study and those

previously reported for that strain (24). These include differences in 3-day arylsulfatase production, acid production from oxidation of galactose and trehalose, and variations in ceftriaxone and sulfamethoxazole susceptibility results. These discrepancies may be due, in part, to methodologic variations and differences in interpretations, especially for susceptibilities. The effects of the use of different subcultures of the type strain in the two studies are unknown. These discrepancies emphasize the need for standardization of all aspects of the phenotypic methodologies used to identify Nocardia species. It is also important that some of the previously published results attributed to N. nova may in fact have included results for N. africana and N. veterana, as these two species would generally have been indistinguishable from N. nova by most phenotypic procedures used in the past. The identification of members of the genus Nocardia to the species level can be problematic even when different methodologies are applied simultaneously. We have found REA of both the 16S rRNA and the HSP genes to be essential for the correct identification of many Nocardia species and also for the

VOL. 41, 2003

NOCARDIA VETERANA IN NORTH AMERICAN PATIENTS

recognition of new or unusual species (6). The RFLP assay results obtained with the 16S rRNA and HSP genes of the four patient isolates discussed here did not lead to a definitive identification but prompted a broader investigation of the molecular makeup of these organisms. It was only after the results of these additional studies (sequence analysis of nearly the entire 16S rRNA gene and DNA-DNA hybridization) that we were able to establish that three of these isolates were indeed N. veterana. Sequence analysis of the 16S rRNA gene has become an increasingly important tool in establishing the species identifications of members of this genus. For some species, differences in 16S rRNA gene sequences are obvious (5, 18). However, caution must be used in identifying Nocardia isolates by 16S rRNA gene sequencing alone. The 16S rRNA gene sequence of isolate D in this study showed 99.6% similarity to that of the N. veterana type strain (6 base differences in a 1,359-bp region), yet DNA-DNA hybridization studies clearly showed that it did not belong to the species N. veterana. Yassin et al. (30) noted a similar identity of the 16S rRNA gene sequence of N. paucivorans to that of N. brevicatena (99.6%), with only 61.9% relatedness by DNA-DNA hybridization. It is unclear at present the extent to which 16S rRNA gene sequences need to differ between two isolates for each to be considered a separate species. Stackebrandt and Goebel (19) note that the ability of 16S sequencing to discriminate between closely related species may be limited. This limitation evidently applies to Nocardia species similar to N. nova. Our results show that two different species may share 99.6% of a 1,359-bp region of the 16S rRNA gene. On the basis of all of our results, we have concluded that our isolates B and C are in fact isolates of N. veterana. Fox et al. (8) comment that if strains have identical 16S rRNA gene sequences, they most likely belong to the same species. We think that the additional 100% similarity of a portion of a second gene (HSP) and the similar biochemical profiles and susceptibility testing results give more credence to our conclusion; however, we realize that by present criteria, only DNA-DNA hybridization can irrefutably confirm this identification. The use of sequences in public databases to make species determinations must also be done with care. Turenne et al. (22) noted numerous discrepancies in the sequence data for mycobacteria in these databases, and at least one discrepancy also exists for Nocardia spp. (6). Furthermore, many sequences in GenBank have ambiguous bases, and the use of these sequences may result in unreliable alignments. Pitfalls inherent in the uncritical use of public sequence databases prompted us to perform our own sequence analysis of some of the type strains most important for the comparisons used in this study. Phenotypic markers can be helpful in discriminating some species or species groups of Nocardia. However, complete biochemical characterization of isolates may be impossible because of the lack of commercial availability of many of the tests described in the literature for Nocardia identification. Laboratories that isolate Nocardia species infrequently and those that are unable to prepare their own biochemical tests are limited by the small battery of tests available for use. The use of such a limited battery of tests may have contributed to the heterogeneity of isolates previously included in the species (later, species complex) N. asteroides. It may at least partly explain

2567

why N. veterana isolates have not been more frequently recognized in the past. Indeed, one isolate from this study, isolate C, was initially identified as N. nova by the referring laboratory and was not recognized as N. veterana until molecular methodologies were applied. An additional problem with phenotypic testing is the difficulty of comparing results from different laboratories to each other as a consequence of different methods of biochemical testing; the lack of interlaboratory reproducibility of the tests can be a source of confusion and misidentification. Inconsistencies due to inoculum standardization or to various levels of experience with testing procedures and interpretation can also be the source of apparent phenotypic variations within and among laboratories. Our results show that the 3- and 14-day arylsulfatase production tests and esculin hydrolysis tests are potentially useful for the identification of N. veterana; further evaluation with more isolates by different laboratories is necessary to verify the usefulness and reproducibilities of these tests for achieving identifications. However, at present it seems probable that as the number of recognized Nocardia species continues to increase, the ability to identify isolates accurately by biochemical testing alone will further diminish. Antimicrobial susceptibility testing has also been suggested as a potential method for determination of the species of Nocardia isolates (9, 24, 26). The data presented here suggest that susceptibility data cannot be used as sole criterion for species identification, at least for isolates in the N. nova, N. veterana, and N. africana group. Due to the lack of other definitive identification methods, DNA-DNA hybridization may be the only procedure available for the unequivocal identification of some Nocardia species. This test, however, is labor-intensive and not widely available; performance of the test requires a significant amount of technical expertise to achieve reliable results. Because of the difficulties in the identification of N. veterana as detailed here, we conclude that this species is more common as a human pathogen than has previously been reported. N. veterana isolates have likely been identified as members of the “N. asteroides complex” or perhaps N. nova if only phenotypic characteristics have been assessed. Accurate identification of this species requires RFLP analysis of at least the 16S rRNA gene to recognize that the organism is a member of the N. africana-N. nova-N. veterana group. Subsequent determination of arylsulfatase production and esculin hydrolysis activity may aid in species identification if studies of more isolates substantiate that these biochemicals are able to reliably differentiate among species. Otherwise, sequencing of at least the 16S rRNA gene may be necessary to establish 100% identity with the type strain. Laboratories without molecular biology assay capabilities may consider providing a preliminary identification of “N. nova complex,” which would include the related species N. veterana, N. africana, and N. nova. However, because of the apparently similar antibiograms of these species, the lack of a definitive species determination may not be of immediate therapeutic relevance and at present may be important primarily to ascertain whether N. veterana has a particular propensity to infect certain hosts and to define more clearly its role in clinical disease.

2568

CONVILLE ET AL. ACKNOWLEDGMENTS

We thank Yvonne R. Shea, Caroline Dorworth Fukuda, and the technologists of the Mycology Section, Department of Laboratory Medicine, Microbiology Service, Clinical Center, NIH, for technical assistance. We thank Joann Cloud, ARUP Institute for Clinical and Experimental Pathology, for supplying isolates for this study. We thank Jessica Birkhead, Meningitis and Special Pathogens Branch, Division of Bacterial and Mycotic Diseases, National Center for Infectious Diseases, CDC, for performing biochemical studies and Patrick R. Murray, Department of Laboratory Medicine, Microbiology Service, Clinical Center, NIH, for critically reviewing the manuscript. REFERENCES 1. Berd, D. 1973. Laboratory identification of clinically important aerobic actinomycetes. Appl. Microbiol. 25:665–681. 2. Brenner, D. J., F. W. Hickman-Brenner, J. V. Lee, A. G. Steigerwalt, G. R. Fanning, D. G. Hollis, J. J. Farmer III, R. E. Weaver, S. W. Joseph, and R. Seidler. 1983. Vibrio furnissii (formerly aerogenic biogroup of Vibrio fluvialis), a new species isolated from human feces and the environment. J. Clin. Microbiol. 18:816–824. 3. Brenner, D. J., A. C. McWhorter, J. K. Leete Knutson, and A. G. Steigerwalt. 1982. Escherichia vulneris: a new species of Enterobacteriaceae associated with human wounds. J. Clin. Microbiol. 15:1133–1140. 4. Chapin, K. C., and P. R. Murray. 1999. Stains, p. 1674–1686. In P. R. Murray, E. J. Baron, M. A. Pfaller, F. C. Tenover, and R. H. Yolken (ed.), Manual of clinical microbiology, 7th ed. ASM Press, Washington, D.C. 5. Chun, J., and M. Goodfellow. 1995. A phylogenetic analysis of the genus Nocardia with 16S rRNA gene sequences. Int. J. Syst. Evol. Microbiol. 45:240–245. 6. Conville, P. S., S. H. Fischer, C. P. Cartwright, and F. G. Witebsky. 2000. Identification of Nocardia species by restriction endonuclease analysis of an amplified portion of the 16S rRNA gene. J. Clin. Microbiol. 38:158–164. 7. Dorman, S. E., S. V. Guide, P. S. Conville, E. S. DeCarlo, H. L. Malech, J. I. Gallin, F. G. Witebsky, and S. M. Holland. 2002. Nocardia infection in chronic granulomatous disease. Clin. Infect. Dis. 35:390–394. 8. Fox, G., J. D. Wisotzkey, and P. Jurtshuk, Jr. 1992. How close is close: 16S rRNA sequence identity may not be sufficient to guarantee species identity. Int. J. Syst. Bacteriol. 42:166–170. 9. Goodfellow, M., and V. A. Orchard. 1974. Antibiotic sensitivity of some nocardioform bacteria and its value as a criterion for taxonomy. J. Gen. Microbiol. 83:375–387. 10. Gu ¨rtler, V., R. Smith, B. C. Mayall, G. Po ¨tter-Reinemann, E. Stackebrandt, and R. M. Kroppenstedt. 2001. Nocardia veterana sp. nov., isolated from human bronchial lavage. Int. J. Syst. Evol. Microbiol. 51:933–936. 11. Hamid, M. E., L. Maldonado, G. S. Sharaf Eldin, M. F. Mohamed, N. S. Saeed, and M. Goodfellow. 2001. Nocardia africana sp. nov., a new pathogen isolated from patients with pulmonary infections. J. Clin. Microbiol. 39:625– 630. 12. Kano, R., Y. Hattori, N. Murakami, N. Mine, M. Kashima, R. Kroppenstedt, M. Mizoguchi, and A. Hasegawa. 2002. The first isolation of Nocardia veterana from a human mycetoma. Microbiol. Immunol. 46:409–412. 13. Kent, P. T., and G. P. Kubica. 1985. Public health mycobacteriology: a guide for the level III laboratory. Centers for Disease Control, U.S. Department of Health and Human Services, Atlanta, Ga. 14. Lasker, B. A., J. M. Brown, and M. M. McNeil. 1992. Identification and epidemiological typing of clinical and environmental isolates of the genus Rhodococcus with use of a digoxigenin-labeled rDNA gene probe. Clin. Infect. Dis. 15:223–233. 15. Loeffelholz, M. J., and D. R. Scholl. 1989. Method for improved extraction of DNA from Nocardia asteroides. J. Clin. Microbiol. 27:1880–1881.

J. CLIN. MICROBIOL. 16. McNeil, M. M., and J. M. Brown. 1994. The medically important aerobic actinomycetes: epidemiology and microbiology. Clin. Microbiol. Rev. 7:357– 417. 17. National Committee for Clinical Laboratory Standards. 2000. Susceptibility testing of mycobacteria, Nocardia and other aerobic actinomycetes; tentative standard, 2nd ed. NCCLS document M24-T2. National Committee for Clinical Laboratory Standards, Wayne, Pa. 18. Ruimy, R., P. Riegel, A. Carlotti, P. Boiron, G. Bernardin, H. Monteil, R. J. Wallace, Jr., and R. Christen. 1996. Nocardia pseudobrasiliensis sp. nov., a new species of Nocardia which groups bacterial strains previously identified as Nocardia brasiliensis and associated with invasive diseases. Int. J. Syst. Bacteriol. 46:259–264. 19. Stackebrandt, E., and B. M. Goebel. 1994. Taxonomic note: a place for DNA-DNA reassociation and 16S rRNA sequence analysis in the present species definition in bacteriology. Int. J. Syst. Bacteriol. 44:846–849. 20. Steingrube, V. A., R. W. Wilson, B. A. Brown, K. C. Jost, Jr., Z. Blacklock, J. L. Gibson, and R. J. Wallace, Jr. 1997. Rapid identification of clinically significant species and taxa of aerobic actinomycetes, including Actinomadura, Gordona, Nocardia, Rhodococcus, Streptomyces, and Tsukamurella isolates, by DNA amplification and restriction endonuclease analysis. J. Clin. Microbiol. 35:817–822. 21. Telenti, A., F. Marchesi, M. Balz, F. Bally, E. C. Bo¨ttger, and T. Bodmer. 1993. Rapid identification of mycobacteria to the species level by polymerase chain reaction and restriction enzyme analysis. J. Clin. Microbiol. 31:175– 178. 22. Turenne, C. Y., L. Tschetter, J. Wolfe, and A. Kabani. 2001. Necessity of quality-controlled 16S rRNA gene sequence databases: identifying nontuberculous Mycobacterium species. J. Clin. Microbiol. 39:3637–3648. 23. Wallace, R. J., Jr., B. A. Brown, Z. Blacklock, R. F. Ulrich, K. Jost, J. M. Brown, M. M. McNeil, G. O. Onyi, V. A. Steingrube, and J. L. Gibson. 1995. New Nocardia taxon among isolates of Nocardia brasiliensis associated with invasive disease. J. Clin. Microbiol. 33:1528–1533. 24. Wallace, R. J., Jr., B. A. Brown, M. Tsukamura, J. M. Brown, and G. O. Onyi. 1991. Clinical and laboratory features of Nocardia nova. J. Clin. Microbiol. 29:2407–2411. 25. Wallace, R. J., Jr., M. Tsukamura, B. A. Brown, J. Brown, V. A. Steingrube, Y. Zhang, and D. R. Nash. 1990. Cefotaxime-resistant Nocardia asteroides strains are isolates of the controversial species Nocardia farcinica. J. Clin. Microbiol. 28:2726–2732. 26. Wallace, R. J., Jr., K. Wiss, R. Curvey, P. H. Vance, and J. Steadham. 1983. Differences among Nocardia spp. in susceptibility to aminoglycosides and ␤-lactam antibiotics and their potential use in taxonomy. Antimicrob. Agents Chemother. 23:19–21. 27. Wayne, L. G., D. J. Brenner, R. R. Colwell, P. A. D. Grimont, O. Kandler, M. I. Krichevshy, L. H. Moore, W. E. C. Moore, R. G. E. Murray, E. Stackebrandt, M. P. Starr, and H. G. Tru ¨per. 1987. Report of the Ad Hoc Committee on Reconciliation of Approaches to Bacterial Systematics. Int. J. Syst. Bacteriol. 37:463–464. 28. Wellinghausen, N., T. Pietzcker, W. Kern, A. Essig, and R. Marre. 2002. Expanded spectrum of Nocardia species causing clinical nocardiosis detected by molecular methods. Int. J. Med. Microbiol. 292:277–282. 29. Williams, S. T., M. Goodfellow, G. Alderson, E. M. H. Wellington, P. H. A. Sneath, and M. J. Sackin. 1983. Numerical classification of Streptomyces and related genera. J. Gen. Microbiol. 129:1743–1813. 30. Yassin, A. F., F. A. Rainey, J. Burghardt, H. Brzezinka, M. Mauch, and K. P. Schaal. 2000. Nocardia paucivorans sp. nov. Int. J. Syst. Evol. Microbiol. 50:803–809. 31. Yassin, A. F., F. A. Rainey, U. Mendrock, H. Brzezinka, and K. P. Schaal. 2000. Nocardia abscessus sp. nov. Int. J. Syst. Evol. Microbiol. 50:1487–1493. 32. Yassin, A. F., F. A. Rainey, and U. Steiner. 2001. Nocardia cyriacigeorgica sp. nov. Int. J. Syst. Evol. Microbiol. 51:1419–1423.