PCR for Detection and Identification of Abiotrophia spp.

Abstract

Members of the genus Abiotrophia, formerly known as nutritionally variant streptococci, are important pathogens causing septicemia and endocarditis. Cultivation and biochemical differentiation of Abiotrophia spp. are often difficult. Based on 16S rRNA sequences, two PCR assays for detection and identification of Abiotrophia spp. were developed. The first PCR assay was positive for all Abiotrophia spp. Subsequently performed restriction fragment length polymorphism analysis allowed the verification of the PCR amplicons and the differentiation of the three species. The second PCR assay was positive only for A. elegans, the most fastidious species of Abiotrophia. Both PCR assays were shown to be specific and sensitive and should facilitate the identification of Abiotrophia spp.

Nutritionally variant streptococci (NVS) were first described by Frenkel and Hirsch (6) as ungroupable viridans streptococci that grow as satellite colonies around other bacteria. NVS are fastidious microorganisms which require additional growth factors for multiplication in complex media. l-Cysteine or pyridoxal hydrochloride (vitamin B₆) has been shown to stimulate the growth of most NVS (4, 6). Like other viridans streptococci, NVS cause sepsis and bacteremia. They are responsible for 5% of all cases of streptococcal endocarditis, including most of the so-called blood culture-negative endocarditis cases, and have been isolated from a variety of other infectious diseases (2, 9). NVS are members of the normal flora of the human throat as well as of the urogenital and intestinal tracts (9, 11).

Taxonomic studies based on DNA-DNA hybridization and sequence analysis of 16S rRNA demonstrated that NVS are unique bacteria. They were transferred to the new genus Abiotrophia, with A. adiacens and A. defectiva as species (2, 7). Recently a third species, A. elegans, was identified (10).

The genus Abiotrophia presents as a heterogeneous group of bacteria. Cultivation and identification of the different species is often difficult and can result in misidentification of the pathogen. Many different enzyme profiles have been found in biochemical differentiation. Members of the genus Abiotrophia have shown at least three major biotypes (1, 3). Moreover, the growth factors l-cysteine and pyridoxal hydrochloride do not stimulate growth of all known Abiotrophia species, as demonstrated for A. elegans (10). Beighton et al. (1) have also suggested that Abiotrophia spp. may have different nutritional requirements.

To solve some of these problems, a molecular approach for identification and differentiation of Abiotrophia spp. would be of practical relevance. Ohara-Nemoto et al. (8) described the identification of A. adiacens and A. defectiva and differentiation from other streptococci by restriction fragment length polymorphism (RFLP) analysis of universal 16S rRNA PCR. Drawbacks of this method are that (i) it is not possible to detect and identify Abiotrophia spp. among bacteria in mixed cultures and (ii) atypical strains or unknown species cannot be identified. Moreover, the study of Ohara-Nemoto et al. did not consider A. elegans.

To improve detection and facilitate the differentiation of Abiotrophia spp., we developed two PCR assays based on the sequences of the 16S rRNAs. The first PCR assay was designed to detect all Abiotrophia spp. Subsequent RFLP analysis of the PCR products can be performed for verification and differentiation between the three species. The second PCR assay was designed to identify A. elegans, the most fastidious species of Abiotrophia.

MATERIALS AND METHODS

Bacterial strains.

The following strains were used in this study: A. elegans DSM 11693, CCUG 26024, CCUG 27554, and A-5980 (clinical isolate, Munich); A. adiacens ATCC 49175, CCUG 27809, CCUG 35130, CIP103895, CIP103898, CIP 103900, CIP 103901, and B-2057 (clinical isolate, Munich); A. defectiva ATCC 49176, CCUG 36937, and CIP 103895; Streptococcus pyogenes ATCC 19615; Streptococcus agalactiae DSM 2134; Streptococcus pneumoniae ATCC 6303; Streptococcus mutans ATCC 35668; Streptococcus equisimilis ATCC 35666; Streptococcus mitis NCTC 12261; Streptococcus sanguis ATCC 10556; Streptococcus bovis DSM 20065; Enterococcus faecalis ATCC 29212; Enterococcus faecium ATCC 35667; Peptostreptococcus anaerobicus ATCC 27337; Staphylococcus aureus ATCC 25923; Staphylococcus epidermidis ATCC 14990; Lactobacillus acidophilus DSM 20242; Lactobacillus vaginalis (clinical isolate, Munich); Leuconostoc sp. (clinical isolate, Munich), Bartonella henselae ATCC 49797; and 40 different blood culture isolates (20 gram-positive and 20 gram-negative microbes) from patients of the university hospital of Munich. Bacteria were grown on Schaedler agar base (Difco Laboratories GmbH, Augsburg, Germany) supplemented with 7% defibrinated sheep blood (Oxoid, Unipath Ltd., Basingstoke, England) at 37°C under aerobic or anaerobic conditions for 24 to 48 h. Clinical isolates used in this study were verified by 16S rRNA sequencing.

Extraction of DNA.

One to 10 freshly grown bacterial colonies were resuspended in 500 μl of sterile H₂O. About 35 μl of acid-washed glass beads (<106 μm; Sigma-Aldrich Chemie GmbH, Deisenhofen, Germany) were added. Bacteria were disintegrated in a swing-mill (Retsch GmbH, Haan, Germany) at maximum speed (about 1,560 min⁻¹) for 7 min. Glass beads and bacterial debris were pelleted by centrifugation (12,000 × g, 3 min). Two hundred fifty microliters of the DNA-containing supernatant was stored at −20°C. To lyse Abiotrophia spp., the repeated freeze-and-thaw procedure also works well.

Primers.

Specific primers to identify the genus Abiotrophia (ABSP, 5′-TACCCTCGCGAGTTCGCTG-3′ [bp 1280 to 1268]) or the species A. elegans (ABEL-1, 5′-CGCATAGGTTCTTTAGTCGCAT-3′ [bp 168 to 189]; ABEL-2, 5′-GTTCTTTAGTCGCATGACTGAA-3′ [bp 175 to 196]; ABEL-3, 5′-AGAAGGAAAAGAGGCTTCGG-3′ [bp 196 to 215]) were selected by comparing the 16S rRNAs of the three known Abiotrophia spp. with the existing 16S rRNA database by using the ARB software package (Department of Microbiology, Technical University of Munich, Munich, Germany). The universal primer fD1 (bp 8 to 28) was used, together with primer ABSP, in Abiotrophia genus-specific PCR (AGE-PCR) as the forward primer (5). Universal PCR (U-PCR) was done with primers fD1 and 606r (bp 1391 to 1408). Oligonucleotides were synthesized by Roth (Karlsruhe, Germany).

PCR.

PCR buffer, Taq DNA polymerase, and the model 2400 DNA thermal cycler were obtained from Perkin-Elmer Cetus (Foster City, Calif.), and deoxynucleoside triphosphates were purchased from Pharmacia LKB (Uppsala, Sweden). For a 50-μl PCR mixture, 2.5 μl of template DNA was added. PCRs were performed for 30 cycles with a profile of 94°C for 10 s, 58°C for 30 s, and 72°C for 90 s. The amplification products were analyzed in a 1.5% ethidium bromide-stained agarose gel. The correct sizes of the amplicons are 1,084 bp for A. elegans-specific PCR (AEL-PCR), 1,272 bp for AGE-PCR, and 1,400 bp for U-PCR.

16S rRNA gene sequence analysis.

Amplification and direct sequencing of the gene encoding the 16S rRNA were done as described previously (10). Universal primers corresponding to the Escherichia coli rRNA gene from bp 8 to 28 and bp 1542 to 1522 were used for PCR amplification. The hypervariable regions V1 and V2 were sequenced with a primer corresponding to bp 361 to 341. For solid-phase DNA sequencing, one of the oligonucleotides was biotinylated at the 5′ end. Dynabeads were used for preparation of single-stranded DNA, as recommended by the manufacturer (DYNAL GmbH, Hamburg, Germany). Sequence data were compared with those in the EMBL/GenBank database. (HUSAR-DKFZ, Heidelberg, Germany).

RFLP analysis.

Ten microliters of AGE-PCR mix were digested with 5 U of either HaeIII or MspI (New England Biolabs, Beverly, Mass.) for 2 h at 37°C, and the samples were analyzed in a 2.1% ethidium bromide-stained agarose gel.

RESULTS AND DISCUSSION

PCR specific for the genus Abiotrophia.

To generate oligonucleotides specific for the genus Abiotrophia, we compared the 16S rRNA sequences of the three known Abiotrophia spp. with the existing 16S rRNA database by using the ARB software package. Between bp 1240 and bp 1280, the ARB program suggested probes which differentiate the genus Abiotrophia from other human-pathogenic bacteria. We chose one of them and denoted the probe ABSP. Using the universal forward primer fD1 (bp 8 to 28 [5]), we tested ABSP as a reverse primer in a PCR assay (AGE-PCR). AGE-PCR identified all tested Abiotrophia strains. The non-Abiotrophia microorganisms listed in Materials and Methods and 40 different blood culture isolates all gave negative PCR results (Table 1). AGE-PCR was applied to serial dilutions of A. elegans DSM 11693. Ten microorganisms were easily detected by this assay (data not shown).

TABLE 1.

Specificity of AEL-PCR and AGE-PCR

Bacterial strain(s)	AEL-PCR (1,084 bp)b	AGE-PCR (1,272 bp)b	U-PCRa (1,400 bp)b
A. elegans
DSM 11693	+	+	+
CCUG 26024	+	+	+
CCUG 27554	+	+	+
A-5980	+	+	+
A. adiacens
ATCC 49175	−	+	+
CCUG 27809	−	+	+
CCUG 35130	−	+	+
CIP 103897	−	+	+
CIP 103898	−	+	+
CIP 103900	−	+	+
CIP 103901	−	+	+
B-2057	−	+	+
A. defectiva
ATCC 49176	−	+	+
CCUG 36937	−	+	+
CIP 103895	−	+	+
17 Non-Abiotrophia micro-organismsc	−	−	+
20 Gram-negative blood culture isolatesc	−	−	+
20 Gram-positive blood culture isolatesc	−	−	+

^aU-PCR assays were performed with primers fD1 and 606r. 

^bSize of amplicon. 

^cSee list of bacterial strains in Materials and Methods. 

Ohara-Nemoto et al. (8) described the identification and differentiation of A. adiacens and A. defectiva by 16S rRNA gene PCR RFLP analysis with HaeIII or MspI. We tested the AGE-PCR amplicons of the A. elegans strains in comparison to the A. adiacens and A. defectiva strains by this method. As predicted by a computer search for restriction enzyme recognition sites of the Abiotrophia spp. 16S rRNA sequences, digestion of the AGE-PCR mixture with HaeIII showed a unique pattern profile for A. defectiva but did not differentiate between A. adiacens and A. elegans (data not shown). In contrast, MspI digestion resulted in characteristic band patterns for all three species when analyzed in a 2.1% agarose gel: A. adiacens (four strains), 407 and 538 bp; A. defectiva (three strains), 561 and 606 bp; A. elegans (four strains), 538 and 557 bp. However, in one case an A. adiacens strain (CCUG 35130) resembled A. elegans in the RFLP with MspI. The differentiation between A. adiacens and A. elegans is based on an additional MspI recognition site at bp 149 for A. adiacens. Sequencing of the 16S rRNA of strain CCUG 35130 showed a base pair substitution at bp 149 (T instead of C), resulting in the loss of an MspI recognition site.

These results demonstrated that AGE-PCR is a specific and sensitive assay for the detection of Abiotrophia spp. The PCR products can furthermore be verified by RFLP analysis using MspI. This allows differentiation between A. defectiva, A. elegans, and most A. adiacens strains. However, some strains of A. adiacens may be misidentified as A. elegans.

PCR specific for the species A. elegans.

Variable region 2 of the 16S rRNA showed the most significant differences between A. elegans and other bacteria. To generate an A. elegans-specific probe, we chose two primers within this region: ABEL-1 (bp 168 to 189) and ABEL-2 (bp 175 to 196) of A. elegans DSM 11693. Both primers were used in a PCR assay as forward primers with the Abiotrophia genus-specific reverse primer ABSP. To our surprise, the PCR assays were unspecific or insensitive. In addition to A. elegans, the PCR assay performed with ABEL-1 and ABSP also amplified A. adiacens. The PCR assay performed with ABEL-2 and ABSP did not recognize all A. elegans strains (negative amplification for CCUG 26024). Sequence analysis of variable region 2 of the 16S rRNA of strain CCUG 26024 and CCUG 27554 revealed a hypervariable region between bp 181 and 195 in the species A. elegans (Table 2). This heterogeneity is in good correlation with the failure of primer ABEL-2 to detect all A. elegans strains.

TABLE 2.

16S rRNA (bp 165 to 224) of various A. elegans strains, A. adiacens, and A. defectiva and the relative position of the A. elegans-specific primer ABEL-3

Strain or primer	16S rRNA or primer sequence (bp 165–224)a
Primer ABEL-3	5′-AGAAGGAAAAGAGGCTTCGG-3′
A. elegans DSM 11693	AACCGCATAGGTTCTTTAGTCGCATGACTGAAGAAGGAAAAGAGGCTTCGGCTTCTGCTG
A. elegans CCUG 26024	AACCGCATAGGTTCTTGAGCTGCATGGCTCGAGAAGGAAAAGAGGCTTCGGCTTCTGCTG
A. elegans CCUG 27554	AACCGCATAGGTTCTTGAGTCGCATGACCCGAGAAGGAAAAGAGGCTTCGGCTTCTGCTG
A. elegans A-5980	AACCGCATAGGTTCTTTAGTCGCATGACTGGAGAAGGAAAAGAGGCTTCGGCTTCTGCTG
A. adiacens ATCC 49175	AACCGCATAGGTCTTCGAACCGCATGGTTTGAAGAGGAAAAGAGGCGCAAGCTTCTGCTG
A. defectiva ATCC 49176	TACCGCATAGGACATGGAATCACATGATTCAGTGAGGAAAGGTGGCGCAAGCTATCGCTG

^aSequence variations between the A. elegans strains are in boldface type. Sequence variations between A. adiacens and A. defectiva are underlined. Sequence conserved in A. elegans but different from that of the other two Abiotrophia spp. is double underlined. 

Neighboring the hypervariable region, bp 211 to 214 of the 16S rRNA were conserved in all sequenced A. elegans strains and were totally different from A. adiacens and A. defectiva (Table 2). We designed a probe covering this region (ABEL-3). A probe search with the ARB software package demonstrated that ABEL-3 is present only in A. elegans. We used probe ABEL-3 as a forward primer with the Abiotrophia genus-specific primer ABSP in a PCR assay (AEL-PCR). AEL-PCR recognized all four A. elegans strains and none of the other strains tested in this study (Table 1). As expected, verification of the AEL-PCR products with MspI digestion resulted in two bands, of 361 and 538 bp. With serial dilutions of A. elegans DSM 11693, AEL-PCR was shown to be sensitive, detecting as few as 10 bacteria (data not shown). Thus, AEL-PCR is a specific and sensitive PCR assay for detection of A. elegans.

Identification of a formerly unidentified bacterium as A. elegans.

We wondered whether A. elegans strains were present in our collection of blood culture isolates. We recultivated four unidentified gram-positive cocci isolated in recent years and analyzed the strains with AGE- and AEL-PCR. One of these strains (A-5980) was positive in both PCR assays. Sequencing of variable regions 1 and 2 of the 16S rRNA showed that A-5980 was most closely related to A. elegans (99%), followed by A. adiacens (92%). Strain A-5980 was isolated from a child with fever receiving immunosuppressive therapy by using the blood culture system BBL SEPTI-CHEK (Becton Dickinson GmbH, Heidelberg, Germany).

Summary.

We have developed two specific and sensitive PCR assays for the detection and identification of Abiotrophia spp. Primers were deduced from variable regions located in the 16S rRNA genes. In general, mutations in primer target sites (e.g., single mutations, gene deletions, and species-specific differences) can result in false-negative PCR results. This risk is low if highly conserved genes (such as the 16S rRNA) are used as PCR targets. However, PCR assays using the 16S rRNA gene as a target bear other risk factors. The differences between genera and species are sometimes low, and one or two mismatches (especially in the 5′ half of a primer) do not necessarily exclude amplification. For example, in the present study the A. elegans-specific primer ABEL-1 amplified A. adiacens despite five mismatches. Therefore, PCR assays based on 16S rRNA sequences have to be tested against related bacteria, and positive PCR results have to be verified. Both were performed with the new PCR assays presented herein. Abiotrophia is a heterogeneous genus of clinically important pathogens. For detection, identification, and subsequent characterization of further isolates, these PCR assays could be of great value. Identification and characterization is necessary, especially for A. elegans. Up to now, very few isolates of this species are known. Probably, A. elegans will be found frequently among unidentified Gram-positive cocci isolated from patients.