Characteristic Signatures of the lytA Gene Provide a Basis for Rapid and Reliable Diagnosis of Streptococcus pneumoniae Infections

Abstract

The nucleotide sequences of the lytA gene from 29 pneumococcal isolates of various serotypes and 22 additional streptococci of the mitis group (including two Streptococcus pseudopneumoniae strains) have been compared and found to correspond to 19 typical (927-bp-long) and 20 atypical (921-bp-long) alleles. All the Streptococcus pneumoniae strains harbored typical lytA alleles, whereas nonpneumococcal isolates belonging to the mitis group always carried atypical alleles. A sequence alignment showed that the main difference between typical and atypical lytA alleles resided in 102 nucleotide positions (including the 6 bp absent from atypical alleles). These nucleotides were perfectly conserved in all the typical alleles studied, and the corresponding nucleotides of the atypical alleles were also perfectly conserved. The presence in these signatures of distinctive restriction sites (namely, SnaBI, XmnI, and BsaAI) allowed the development of a simple, reliable, and fast method that combines PCR amplification of the lytA gene, digestion with BsaAI, and separation of the products by agarose gel electrophoresis. This assay allows the rapid and consistent identification of true S. pneumoniae strains and represents an improved diagnostic tool for the study of pneumococcal carriage.

Streptococcus pneumoniae (the pneumococcus) is an important human pathogen that is currently the main cause of acute otitis media, sinusitis, community-acquired pneumonia, and acute meningitis. Overall, pneumococcal infections cause substantial morbidity and mortality worldwide (16). The progressive emergence and rapid dissemination of antibiotic resistance in the pneumococcus require a rapid and accurate identification of S. pneumoniae to provide the appropriate antimicrobial therapy. Classically, differentiation of S. pneumoniae from other alpha-hemolytic (i.e., viridans group) streptococci depends on the optochin (Opt) susceptibility test, bile or deoxycholate (Doc) solubility, and immunological reaction with type-specific antisera (the capsular reaction or the Quellung test) (23). In most countries, clinical identification of S. pneumoniae still relies exclusively on the Opt susceptibility test, although Doc solubility is also currently inspected in many U.S. laboratories. These two tests are usually sufficient with samples obtained from sterile locations; but isolates in samples from nonsterile sites, such as pharyngeal swabs, frequently provide conflicting results for one (or both) of these tests (see reference 1 for a recent review). Besides, nonpneumococcal, Opt-susceptible (Opt^s) streptococci have also been reported (24). Isolates that are putatively identified as pneumococci on the basis of other analyses (4, 18, 33) but that give negative results in one or more of the classical assays are usually referred to as “atypical” pneumococci. This designation is puzzling, and we propose here the term “streptococci of the mitis group” (SMG) as a more appropriate name for these strains. This name is equivalent to “Smit group,” previously used to assemble the streptococcal species belonging to the Streptococcus mitis-Streptococcus oralis group (1). Recently, a new streptococcal species of the mitis group (Streptococcus pseudopneumoniae) closely related to S. pneumoniae has been described (1). Isolates of this species are reported to be resistant to optochin (Opt^r) when they are incubated under an atmosphere of increased CO₂ and nontypeable (NT), are not soluble in Doc, and give a positive reaction with the GenProbe Accuprobe Pneumococcus test. It has been stated that a clear differentiation between isolates of S. pneumoniae and S. pseudopneumoniae could be obtained only by DNA-DNA hybridization (1), a technique difficult to handle in clinical laboratories. Multilocus sequence typing has been recently proposed as an alternative tool for determination of whether an isolate of the SMG is or is not a pneumococcus (13). Unfortunately, for the same reasons previously mentioned for DNA-DNA hybridization, multilocus sequence typing cannot be implemented as a routine method in most clinical laboratories. As an alternative, PCR amplification of several genes usually considered specific for S. pneumoniae (namely, lytA, ply, and psaA) has been used to determine which SMG isolate corresponds to a true pneumococcus. In a recent comparative study, Messmer et al. reported that PCR amplification of an internal fragment of lytA was the most appropriate approach to the correct identification of S. pneumoniae (26). A similar conclusion had been reached in previous studies (17, 25).

The molecular peculiarities of the lytA gene encoding the major pneumococcal autolysin (LytA), an N-acetylmuramoyl-l-alanine amidase (NAM-amidase), have been studied in detail in our laboratory (21, 22). The LytA NAM-amidase is the enzyme responsible for the solubilization of pneumococci upon addition of 1% Doc (29). It should be underlined that all the pneumococci isolated from humans so far are solubilized by 1% Doc and, consequently, contain a fully active lytA gene. A partially deleted lytA gene has been found only in the case of pneumococci isolated from the respiratory tracts of horses (47), suggesting that there is a strong selective advantage in maintaining the function of the LytA NAM-amidase in human populations.

It is well known that some isolates of the SMG that are insoluble in 1% Doc also contain a lytA gene (31). Our group reported that the lytA alleles present in isolates of the SMG are “atypical” because they possess a characteristic 6-bp deletion at the lytA 3′ moiety (6, 31), which opens up the possibility of distinguishing between true pneumococci and other SMG. We also observed that most lytA-containing isolates of the SMG that showed a Doc-insoluble phenotype when they were tested with a Doc concentration of 1% (Doc⁻) were completely solubilized when 0.1% Doc or 1% Triton X-100 was used (31). The importance of this finding must be underlined, because most clinical laboratories will go no further with the identification if a streptococcal isolate is bile soluble. Besides, it has been found that inactivation of LytA-like NAM-amidases by 1% Doc is caused not only by the 2-amino-acid deletion characteristic of atypical LytA enzymes (31) (see above) but also by a variety of point mutations in the lytA gene (36). All these observations taken together strongly suggested that the presence of a lytA-like gene in many isolates of the SMG may result in misidentification if PCR amplification experiments with DNA prepared from these strains is performed.

A detailed comparison of the nucleotide sequences of the lytA alleles from S. pneumoniae isolates and from isolates of the SMG revealed that the pneumococcal (i.e., typical) lytA alleles have characteristic signatures that fully differentiate them from atypical alleles. Based on these data, we propose a rapid and easy PCR method for the identification of true pneumococcal isolates.

MATERIALS AND METHODS

Bacterial strains, plasmids, and growth conditions.

Two conjunctival, NT, Doc-soluble S. pneumoniae strains (strains ST344 and ST942) (2) were included in this study. In addition, three Opt^s, NT strains of the SMG (strains 578, 1504, and 3072) (Table 1) were also studied. The aforementioned strains were provided by the Spanish Pneumococcal Reference Laboratory (Majadahonda, Spain). S. pseudopneumoniae strains CCUG 49455^T (type strain) and CCUG 48465 were purchased from the Culture Collection, University of Göteborg (Göteborg, Sweden). The bacteria were grown without shaking in Todd-Hewitt broth supplemented with 0.5% yeast extract (THY) or in C medium (19) supplemented with 0.08% yeast extract (C+Y) at 37°C.

TABLE 1.

Typical and atypical lytA alleles studied here

Strain	Serotypea	Optb	Docc	Allele no.d	Accession no.	Reference or source
Pneumococci
Rst7	R	S	+	1	M13812	10
R800	R	S	+	1	Z34303	28
R6	R	S	+	1	AE008540	15
TIGR4	4	S	+	2	AE007483	41
150	12	S	+	2	AF345845	Unpublished
147	29	S	+	3	AF345844	Unpublished
219	7F	S	+	4	AF345846	Unpublished
494	3	S	+	5	AJ243399	45
670	6B	S	+	6	AJ243400	45
PN107	1	S	+	7	AJ243401	45
CL2	1	S	+	7	AJ243402	45
INV104B	1	S	+	7	—e	Unpublished
PN58	19A	S	+	8	AJ243403	45
VA1	19	S	+	9	AJ243404	45
CL18	10	S	+	10	AJ243405	45
1012	35	S	+	11	AJ243406	45
PN15	12	S	+	12	AJ243407	45
234	23F	S	+	13	AJ243408	45
233	23F	S	+	13	AJ243409	45
472	3	S	+	14	AJ243410	45
860	NK	S	+	14	AJ243411	45
29044	14	S	+	14	AJ243412	45
PN8	23	S	+	15	AJ243413	45
7751	6	S	+	15	AJ243414	45
Spain23F-1	23F	S	+	15	—	Unpublished
8249	19A	S	+	16	AJ490328	32
949	23F	S	+	17	AJ490805	32
ST344f	NT	S	+	18	AM113493	2; this study
ST942	NT	S	+	19	AM113494	2; this study
SMG
101/87	NT	R	−	20	S43511	6
COL17	NT	R	−	21	AJ252190	46
COL16	NT	S	−	22	AJ252192	46
COL20	NT	S/R	−	23	AJ252194	46
COL26	NT	R	−	24	AJ252195	46
S. pseudopneumoniae
CCUG 49455T	NT	S/R	−	24	AM113495	1; this study
CCUG 48465	NT	S/R	−	24	AM113496	1; this study
COL27	NT	S/R	−	25	AJ252196	46
1508/92	NT	R	−	26	AJ419973	31
11923/1992	NT	S	−	27	AJ419974	31
8224/1994	NT	R	−	28	AJ419975	31
10546/1994	19	R	−	29	AJ419976	31
782/1996	NT	S	−	30	AJ419977	31
1230/1996	19	S	−	31	AJ419978	31
1283/1996	NT	S	−	32	AJ419979	31
1338/1996	NT	S	−	33	AJ419980	31
1078/1997	NT	R	−	34	AJ419981	31
1383/1997	19	S	−	35	AJ419982	31
1629/1997	19	S	−	36	AJ419983	31
578	NT	S	−	37	AM113497	This study
1504	NT	S	−	38	AM113500	This study
3072	NT	S	−	39	AM113504	This study

^aR, unencapsulated. The D39 parental strain of Rst7, R6, and R800 has a type 2 capsule. NK, not known; NT, nontypeable.

^bOpt, optochin susceptibility; S, susceptible; R, resistant; S/R, variable.

^cDoc, solubility in 1% Doc; +, soluble; −, insoluble.

^dThe alleles were ordered according their accession numbers.

^e—, preliminary sequence data (http://www.sanger.ac.uk/Projects/S_pneumoniae).

^fST, sequence type.

Determination of Doc solubility and autolytic phenotype.

Routinely, 0.5 ml of exponentially growing cultures of S. pneumoniae or SMG received 50 μl of 1 M potassium phosphate buffer (pH 8.0) and 50 μl of a 10% Doc solution in water. The mixtures were incubated for up to 15 min at 37°C. When the turbidity of the cell suspension decreased more than 50% from the initial value, the strain was designated Doc⁺. In some experiments, Doc was used at a final concentration of 0.1%. When Triton X-100 was used instead of Doc, the former detergent was used at a final concentration of 1%. Streptococcal isolates that autolyzed in the stationary phase of culture, that is, after overnight incubation in C+Y medium at 37°C, were designated Lyt⁺.

PCR amplification, cloning, and nucleotide sequencing.

S. pneumoniae chromosomal DNA was prepared as described previously (9). The DNA extraction procedure described by Ezaki et al. (8) was used for all other SMG isolates. For PCR amplification of the lytA gene, oligonucleotide primers LA5_Ext and LA3_Ext were used (31). PCR amplifications were carried out in 50-μl reaction mixtures that contained 0.5 units of Thermus thermophilus DNA polymerase (Biotools B&M Labs, Madrid, Spain), 250 μM deoxynucleoside triphosphates, 0.5 μM each oligonucleotide primer, and 0.5 μg of DNA template in standard buffer [75 mM Tris-HCl, pH 9.0, 2 mM MgCl₂, 50 mM KCl, 20 mM (NH₄)₂SO₄; (Biotools B&M Labs)]. The PCR conditions included an initial denaturation at 95°C for 5 min, followed by a 25-cycle amplification, with each cycle consisting of denaturation at 95°C for 30 s, annealing at 52°C for 30 s, and extension at 72°C for 1 min. The products amplified by PCR were purified with a High Pure PCR product purification kit (Roche). Electrophoresis in 1.5% agarose gels was carried out by standard methods (39). The DNA sequence was determined by the dideoxy chain-termination method (40) with an automated ABI Prism 3700 DNA sequencer (Applied Biosystems). Restriction enzymes were purchased from New England BioLabs and were used according to the recommendations of the supplier.

Data analysis.

The multiple-sequence alignment of the lytA alleles was done with Clustal W (42) at the EMBL-European Bioinformatics Institute (http://www.ebi.ac.uk/clustalw).

Nucleotide sequence accession numbers.

The nucleotide sequences determined in this study have been deposited in the EMBL/GenBank/DDBJ databases under the accession numbers AM113493, AM113494, AM113495, AM113496, AM113497, AM113500, and AM113504.

RESULTS AND DISCUSSION

Autolysis and Doc phenotypes of isolates of the SMG.

SMG strains 578, 1504, and 3072 and the two strains of S. pseudopneumoniae exhibited a Lyt⁺ phenotype; that is, they autolyzed in the stationary phase of culture. Autolysis was more pronounced when the streptococci were grown in C+Y medium than in THY medium. In addition, all of these strains showed a Doc⁻ phenotype when they were treated with 1% Doc but lysed within 15 min when they were treated with 0.1% Doc or 1% Triton X-100. Also in this case, C+Y medium-grown cells lysed faster with the detergent than those incubated in THY medium (unpublished observations). The lytic behavior of the two S. pseudopneumoniae strains was of particular interest because they had been described as bile insoluble (1). Consequently, the final concentration of Doc used in a particular experiment should be clearly stated. These results taken together fully confirmed and extended the findings of previous reports from our laboratory on the lytic behavior of SMG isolates and strongly suggested that all those strains possessed a lytA-like gene encoding a Doc-sensitive, LytA-like NAM-amidase (6, 31).

PCR amplification of the lytA gene.

The strains analyzed here did contain a lytA-like gene, as demonstrated by PCR amplification, although only the NT pneumococcal strains ST344 and ST942 possessed a typical, full-length (927-bp) lytA allele. This was not unexpected because these strains were NT derivatives of true pneumococci (2). However, the isolates of the SMG (including the two S. pseudopneumoniae strains) contained atypical lytA alleles; that is, they had the distinctive 6-bp deletion at their 3′ ends. In particular, both strains of S. pseudopneumoniae harbored a lytA allele identical to that of the COL26 streptococcal isolate reported previously (46) (Table 1).

Sequence signatures in typical and atypical lytA alleles.

Although the EMBL database contains many partial lytA sequences from different isolates, to obtain reliable results from sequence comparisons, it was important to analyze only complete (or nearly complete) sequences with no indeterminate nucleotides. We carefully examined all the lytA sequences that were included in the EMBL database (last date accessed, 12 October 2005) or that belonged to current genomic sequencing projects, and only those sequences that comprised at least nucleotide positions 22 to 927 (∼95% of the gene length) were chosen for comparison (Table 1). Taking into account the strains sequenced here, the available lytA sequences corresponded to 29 pneumococci and 22 SMG (including the two strains of S. pseudopneumoniae). The lytA alleles (19 and 20 alleles from S. pneumoniae and isolates of the SMG, respectively) were aligned by using ClustalW (Fig. S1 in the supplemental material). Confirming previous observations (31), typical lytA alleles are 957 bp long, whereas atypical lytA alleles are always 951 bp long, as they have a 6-bp deletion (between positions 868 and 873 of any typical lytA allele). Unexpectedly, after a careful inspection of the alignment, it could be found that 102 nucleotide positions (including the 6 positions absent from atypical alleles) were perfectly conserved in all the typical alleles. Moreover, the nucleotides at these positions differed from those of the corresponding atypical alleles, which, in turn, were also completely conserved (Fig. 1A). In addition, the distribution of these signatures along the lytA gene suggested a mosaic-like organization, with the majority of nucleotide differences clustering at both lytA ends (Fig. 1B).

FIG. 1.

Sequence signatures of typical and atypical lytA alleles. (A) The nucleotide positions (taking 1 as the first nucleotide of the ATG initiation codon) should be read in vertical lines. Hyphens represent nucleotides absent from atypical alleles. (B) Mosaic-like distribution of the nucleotide differences between typical and atypical alleles. An open bar corresponds to the lytA sequences that have been compared (nucleotide positions 22 to 927). Hatched boxes indicate the location and number of conserved nucleotide signatures. The location of the 6-bp deletion characteristic of atypical lytA alleles is shown.

A method for the rapid identification of typical and atypical lytA alleles.

The presence of signatures that discriminate between typical and atypical lytA alleles has two important consequences: (i) lytA-specific primers previously used to amplify the lytA gene (or a fragment of it) by PCR might correspond only to typical sequences and would not be appropriate for use for amplification of any other allele, mainly in the case of atypical alleles. (ii) It would be possible to identify some restriction sites characteristic of typical (or atypical) alleles that may be useful for the development of a rapid technique that could be used to easily distinguish true pneumococcal isolates from other SMG.

The use of primers LA5_Ext and LA3_Ext (31), which amplify a 1,213-bp HindIII fragment containing the lytA gene (10), resulted in the amplification of the desired PCR product from any strain, either a pneumococcus or an isolate of the SMG, with the exception of SMG strain 101/87, which has a 1.5-kb deletion immediately after the termination codon of the lytA₁₀₁ gene (6). In contrast, most of the oligonucleotide primers that have been used to amplify the lytA gene in previous work are located in one of the highly divergent patches (Fig. 1 and Table 2). This finding strongly suggests that in most studies aimed at estimation of the presence of an important trait like lytA in clinical samples, a noticeable bias was introduced. In addition, the possible lytA-containing SMG could not be detected. In contrast, the results presented in this work, particularly the multiple-sequence alignment shown in Fig. S1 in the supplemental material, provide the rationale for the design of oligonucleotide primers for amplification by PCR of specifically either typical or atypical lytA alleles.

TABLE 2.

Oligonucleotides used in previous work for PCR amplification of the lytA gene

Name	Assigned no.	Sequence (5′→3′)a	Position in lytAb	Size (bp) of PCR product	Reference(s)
Up1	1	AGAATGAAGCGGATTATCACTGGCGGA	107-133 (+)	560 (Up1 + Down2)	38
Up2	2	AACGGTTGCATCATGCAGGTAGGACCT	172-198 (±)	456 (Up2 + Down1)
Down1	3	AAAGTAGTACCAAGTGCCATTGATTTT	627-601 (+)
Down2	4	CTTCCTCCAGCGGTCTGCAAGCATATA	666-640 (±)
SPI	5	ATGGAAATTAATGTGAGTA	1-19 (−)	247	14
SPII	6	AGGTCTCAGCATTCCA	247-232 (±)
Up	7	GGAGTAGAATATGGAAATTAATGT	−10-14 (−)	262	11
Down	8	GCTGCATAGGTCTCAGCATTCCAA	254-231 (±)
ALY1	9	TGAAGCGGATTATCACTGGC	111-130 (+)	273	1, 44
ALY2	10	GCTAAACTCCCTGTATCAAGCG	383-362 (+)
Forward	11	ACGCAATCTAGCAGATGAAGC	327-347 (−)	122	25
Reverse	12	TGTTTGGTTGGTTATTCGTGC	427-407 (+)
lytAup	13	GGAGTAGAATATGGAAATTGATGTGAGTAA	−10-20 (−)	968	46, 47
lytAdn	14	TTTATTTTACTGTAATCAAGCCATCTGGCTC	958-928 (−)
lytA101dn	15	CTACTTCATCGTAATCAAACCGTCAGGTTC	951-922c (+)
A750	16	GGCTACTGGTACGTACATTC	550-569 (+)	274 (A781 + A1055)	26, 27
A781	17	ATCCAAAAGACAAGTTTGAGA	581-601 (+)	395 (A750 + A1145)
A1055	18	CTGGATAAAGGCATTTGATAC	855-835 (+)
A1145	19	AATCAAGCCATCTGGCTCTA	946-926 (±)
Lyt A1	20	GTCGGCGTGCAACCATATAGGCAA	43-66 (−)	413	3
Lyt A2	21	GGATAAGGGTCAACGTGGTCTGAG	455-432 (±)
LA5_Ext	22	ggtctagAAGCTTTTTAGTCTGGGGTG	−200-181 (−)	1,213	31
LA3_Ext	23	ggggatccAAGCTTTTTCAAGACCTAATAATATG	1013-988 (−)

^aRestriction enzyme sequences introduced for cloning purposes are indicated in lowercase.

^bPosition +1 is the first nucleotide of the ATG initiation codon of the lytA_R6 allele (957 bp) (GenBank accession number M13812). +, ±, and −, the primer is totally located in, partially located in, or located outside of one of the mosaic regions shown in Fig. 1, respectively.

^cPositions correspond to those of the lytA₁₀₁ allele (GenBank accession number S43511), which correspond to positions 957 to 928 of the lytA_R6 allele.

Sequencing of the 1,213-bp HindIII fragment containing the lytA gene would reveal conclusively whether the strain whose lytA gene was amplified is a typical pneumococcus or another, more or less related streptococcal isolate. However, the determination of the DNA sequence is unfeasible in clinical laboratories that handle large numbers of samples. Fortunately, the sequence alignment shown in Fig. S1 in the supplemental material suggested an achievable method of differentiation that could be used routinely in clinical laboratories. We have observed that typical lytA alleles possess an SnaBI restriction site (cleavage position between nucleotides 561 and 562) that is absent from all atypical alleles (Fig. 2A). On the contrary, atypical (but not typical) alleles have an XmnI sequence (cleavage position between nucleotides 290 and 291) (Fig. 2A). PCR amplification followed by digestion with SnaBI or XmnI and separation of the products by electrophoresis in agarose gels rendered DNA fragments of the predicted sizes (data not shown). Besides, since the SnaBI site (TAC↓GTA) is also cleaved by the restriction endonuclease BsaAI (PyAC↓GTPu), we looked for BsaAI sites in atypical alleles and found that all of them possess only one BsaAI cleavage site (CAC↓GTA) at positions 160 to 165 of the lytA gene that is not present in typical alleles (Fig. 2A; see Fig. S1 in the supplemental material). Four additional partial lytA sequences, either typical (GenBank accession numbers AJ240675 and AY204888) or atypical (GenBank accession numbers AJ252191 and AJ252193), that had been not previously considered because of their insufficient length (GenBank accession number AY204888) or because of a partial deletion, as in the case of a pneumococcus from horses (see above) (GenBank accession number AJ240675), or that contained indeterminate nucleotides (GenBank accession numbers AJ252191 and AJ252193) were examined for their restriction profiles, with the corresponding predicted results (data not shown). Figure 2B shows that BsaAI digestion of the purified PCR products obtained with primers LA5_Ext and LA3_Ext allowed the rapid and reliable identification of the lytA allele. When the alleles were amplified by PCR and restricted with BsaAI, typical or atypical alleles produced DNA fragments of 452 and 761 bp and fragments of 362 and 851 bp, respectively. Moreover, even the nonpurified PCR products could also be subjected to BsaAI digestion without any significant loss of resolution, although this was less efficient (Fig. 2C). It is noteworthy that this method accurately discriminated the S. pneumoniae isolates (even NT pneumococcal strains ST344 and ST942) from any other isolate of the SMG, either NT Opt^r Doc⁻ (strain 8224), encapsulated Opt^r Doc⁻ (strain 10546/1994), encapsulated Opt^s Doc⁻ (strain 1230/1996), or NT Opt^s Doc⁻ (strains 578 and 3072).

FIG. 2.

Differentiation between typical and atypical lytA alleles. (A) Schematic representation of the DNA fragments produced by digestion of the 1,213-bp fragment containing the lytA gene with SnaBI (solid arrow), XmnI (open triangle), or BsaAI (solid triangles). (B) Separation of BsaAI-digested fragments by agarose gel electrophoresis. Lanes 1 to 12, digests obtained from strains 949, ST344, ST942, R6, TIGR4, 8249, 8224, 10546/1994, 1230/1996, 578, 3072, and S. pseudopneumoniae CCUG 49455^T, respectively. The fragments amplified by PCR were purified before digestion with the restriction enzyme. As size standards, a mixture of BstEII-digested bacteriophage λ DNA and HaeIII-digested φX174 DNA was used. The sizes of the standards are indicated on the left. (C) After PCR amplification, the nonpurified mixtures received NaCl (final concentration, 100 mM) (lanes 1 and 2). The samples received BsaAI (2 units), and after incubation at 37°C for 1 h, the mixtures were subjected to agarose gel electrophoresis. Lanes 1 and 4, strain R6; lanes 2 and 3, strain 3072. The size standards (lane S) were HaeIII-digested φX174 DNA.

Many natural pneumococcal isolates harbor one or more prophages or prophage remnants (34). This was deduced by Southern hybridization with a lytA_R6-specific probe and DNA samples prepared from 791 S. pneumoniae strains. Moreover, the lytA-like gene from three related temperate pneumococcal prophages has been studied in detail and reported to be of the same length as typical lytA alleles (957 bp) (30, 32, 35). Interestingly, this was also the case for two S. mitis prophages that were recently isolated (36). In contrast, the ejl gene from the S. mitis bacteriophage EJ-1 exhibited the 6-bp deletion characteristic of atypical lytA alleles (5, 37). Fortunately, the DNA regions flanking the bacteriophage lytA-like genes completely differ from those surrounding any bacterial lytA gene from either S. pneumoniae or SMG. In agreement with these data, PCRs with both the different phage DNAs mentioned above and oligonucleotide primers LA5_Ext and LA3_Ext did not render any amplified product. Furthermore, only the bacterial lytA gene was amplified when DNA prepared from the corresponding lysogenic strains was used (unpublished data).

Molecular diagnostics for S. pneumoniae has used PCR-based detection of lytA from clinical samples. This method has real advantages in terms of speed and sensitivity over the classical culture method (11, 14, 27, 38). This is particularly true when typically sterile fluids, such as blood, cerebrospinal fluid, or pleural fluid, are used. However, until now it has been considered that the use of PCR amplification is more problematic with samples from nonsterile sites, from which the pneumococcus-related organisms (such as the SMG studied here) isolated may harbor a lytA-like gene (7). However, the method described here identified unambiguously molecular signatures that allow the rapid and accurate discrimination between typical lytA alleles (characteristic of pneumococci) and atypical lytA alleles (present in nonpneumococcal SMG isolates). In addition, the analysis of the lytA sequences provided here is a must in the development of even more rapid diagnostic techniques for S. pneumoniae (and SGM) infections, such as fluorogenic 5′ nuclease PCR (TaqMan assay), which enables amplification and detection to be carried out at the same time in a closed-tube system (20). Furthermore, the discovery of the polymorphisms that differentiate between typical and atypical lytA alleles also allows the easy implementation of very specific assays, e.g., real-time quantitative PCR combined with the mismatch amplification mutation assay (12).

Currently, antibiotic-resistant S. pneumoniae strains and other members of the mitis group constitute a major health problem, since they are the etiological agents of several community-acquired infections. Obviously, the phenotypic variations induced by LytA-like NAM-amidases reflect alterations in the lytic behavior of SMG isolates that should influence the pathogenic properties of these strains compared with those of true pneumococci. Interestingly, it has been documented that pneumococcal strains that show alterations in their lytic systems appear to contribute to higher morbidity and mortality during infection by playing a role in shaping the course of pneumococcal meningitis (43).

In summary, this study has analyzed the lytA alleles from 29 S. pneumoniae strains and 22 nonpneumococcal isolates of the mitis group. Although experience with a larger number of clinical isolates would provide additional support to our conclusions, the present study provides a strong experimental basis not only to resolve the diagnostic problem of the reliable identification of S. pneumoniae strains, which may drive specific antibacterial therapy without delay, but also to open a new way to reveal alterations in the LytA-like NAM-amidases. This, in turn, should contribute to a better understanding of the importance of lytA variations in the virulence of different streptococci of the mitis group.