The galU gene expression in Streptococcus pneumoniae


Correspondence: Marta Mollerach, Cátedra de Microbiología, Facultad de Farmacia y Bioquímica, Universidad de Buenos Aires, Junín 956, 1113 Buenos Aires, Argentina. Tel. +54 11 49648285; fax: +54 11 9648274; e-mail:


The polysaccharide capsule of Streptococcus pneumoniae is the main virulence factor making the bacterium resistant to phagocytosis. The galU gene of S. pneumoniae encodes a UDP-glucose pyrophosphorylase absolutely required for capsule biosynthesis. In silico analyses indicated that the galU gene is co-transcribed with the gpdA gene, and four putative promoter regions located upstream of gpdA were predicted. One of them behaved as a functional promoter in a promoter reporter system. It is conceivable that the sequence responsible for initiating transcription of gpdA-galU operon is an extended −10 site TATGATA(T/G)AAT. Semi-quantitative real-time reverse transcription PCR experiments indicated that galU was expressed mainly in the exponential phase of growth.


Streptococcus pneumoniae is a leading human pathogen causing both mucosal (such as otitis media and pneumonia) and systemic diseases (including septicemia and meningitis). To date, 93 different pneumococcal capsular types have been described (Henrichsen, 1995; Park et al., 2007; Bratcher et al., 2010; Calix & Nahm, 2010). This remarkable phenotypic variability appears to be present also at the genetic level (Bentley et al., 2006). Early studies showed that uridine diphosphoglucose (UDP-Glc) is a key component in the biosynthetic pathway of pneumococcal capsular polysaccharides containing glucose, galactose, and/or UDP-glucuronic or UDP-galacturonic acids (Mills & Smith, 1965). At least one of these sugars is a component of every capsular polysaccharide of S. pneumoniae (Kamerling, 2000). The enzyme UTP-Glc-1-phosphate uridylyltransferase (UDP-Glc pyrophosphorylase; EC is encoded by the galU gene. This enzyme catalyzes the formation of UDP-Glc, which is the substrate for the synthesis of UDP-glucuronic acid. Also, UDP-Glc is also required for the interconversion of galactose and glucose by way of the Leloir pathway (Frey, 1996).

Previously, the galU gene was cloned and overexpressed, and the gene product was biochemically characterized (Mollerach et al., 1998; Bonofiglio et al., 2005). In addition, knockout galU mutants of type 1 and type 3 pneumococci are unable to synthesize a detectable capsular polysaccharide. Southern blot hybridization experiments using DNAs prepared from pneumococcal isolates belonging to different types showed that every strain tested contained a galU homologue (Mollerach et al., 1998). Thus, the UDP-Glc pyrophosphorylase, which is directly involved in the synthesis of the capsular polysaccharide in S. pneumoniae, might represent a suitable target in the search for inhibitors to control the biosynthesis of the main pneumococcal virulence factor. In this sense, it should be emphasized that eukaryotic UDP-Glc pyrophosphorylases appear to be completely unrelated to their bacterial counterparts, suggesting the possibility that putative inhibitors of the bacterial enzymes would not be harmful for the host.

The gpdA gene (also named gpsA) is located immediately upstream of the galU gene and is predicted to encode a glycerol-3-phosphate dehydrogenase [NAD(P)(+)], (EC The spr1901 gene, which is annotated as a possible transcriptional regulator, is located upstream of the gpdA gene and transcribed from the opposite strand. Nothing is known about the promoter region of galU and its differential expression at different growth phases. In this report, we identified a promoter-active DNA sequence from S. pneumoniae located upstream of gpdA that is involved in controlling the expression of galU through co-transcription with gpdA. These findings provide insight into the expression of GalU, an enzyme with a key role in virulence.

Materials and methods

Bacterial strains and growth conditions

Streptococcus pneumoniae 406 and M31 were grown in liquid C medium (Lacks & Hotchkiss, 1960) supplemented with 0.08% of each yeast extract (CY medium) and bovine serum albumin without shaking, or on reconstituted tryptose blood agar base plates (Difco Laboratories Inc., Detroit, MI) supplemented with 5% defibrinated sheep blood. Lincomycin (0.6 μg mL−1) was added when required. Streptococcus pneumoniae M31 (ΔlytA) is a nonencapsulated, serotype 2 (S2) mutant having a deletion of at least 5.5 kb containing the gene lytA that encodes the main pneumococcal autolysin (Sánchez-Puelles et al., 1986). Streptococcus pneumoniae 406 is a clinical isolate of serotype 3 (García et al., 1993).

Escherchia coli C600 cells (thi-1, leuB, thr-1) were grown in Luria–Bertani growth medium (LB) at 37 °C or on LB solid agar supplemented when necessary with tetracycline (20 μg mL−1; Sambrook et al., 1989).

DNA techniques and plasmid construction

Restriction enzymes, T4 DNA ligase and the Klenow fragment of DNA polymerase were obtained commercially and used according to the recommendations of the suppliers. Chromosomal DNA from S. pneumoniae 406 was prepared as previously described (Fenoll et al., 1994; Wilson, 1997). PCR was performed using standard conditions with AmpliTaq DNA polymerase (PerkinElmer). The primers used are listed in Table 1. An SphI/BbuI restriction site was included in oligonucleotides pGalU1, pGalU3, pGalU5, and pGalU7 and an SmaI restriction site in pGalU2, pGalU4, pGalU6, and pGalU8. Four different DNA fragments (F1–F4), one overhanging the other and containing putative promoter regions, were amplified by PCR.

Table 1. Oligonucleotide primers used in this study
Primer nameNucleotide sequence (5′-3′)aPurpose on usePositionb
  1. a

    Bold-faced letters indicate nucleotides introduced to construct the appropriate restriction sites.

  2. b

    Position is given relative to the last nucleotide of spr1901 (see Fig. 1).


Plasmid pLSE4 is a promoter probe vector able to replicate in S. pneumoniae and E. coli that contains a promoterless lytA gene (Díaz & García, 1990). DNA fragments were ligated separately on pLSE4 previously digested with XbaI, filled with Klenow fragment and digested with BbuI.

Transformation procedures

Escherichia coli C600 was made competent and transformed with derivatives of pLSE4 as described elsewhere (Muñoz et al., 1997). The accuracy of the constructs was confirmed by nucleotide sequencing of the corresponding insert. Plasmid derivatives of pLSE4 containing DNA fragments (F1–F4) were transformed into S. pneumoniae M31.

Streptococcus pneumoniae was transformed with plasmid DNA by treating cells in C medium supplemented with 0.08% of bovine serum albumin with synthetic competence-stimulating pheromone (250 ng mL−1) at 37 °C for 10 min to induce competence (Moscoso & Claverys, 2004) followed by incubation at 30 °C during DNA uptake. Streptococcus pneumoniae clones obtained upon transformation with derivatives of pLSE4 were scored on CY agar plates containing lincomycin (0.6 μg mL−1) and catalase (250 units mL−1).

Assay for lytic activity

Crude sonicated extracts were obtained as previously described from mid-exponentially growing cultures for S. pneumoniae M31 derivatives (Ronda et al., 1987). Assays of cell wall lytic (N-acetylmuramoyl-l-alanine amidase; NAM-amidase) activity were performed according to standard procedures described elsewhere using [methyl-3H]choline-labeled pneumococcal cell walls as substrate (Höltje & Tomasz, 1976). One unit of NAM-amidase activity was defined as the amount of enzyme needed to catalyze the hydrolysis (solubilization) of 1 μg of cell wall material in 10 min at 37 °C.

RNA isolation, cDNA synthesis, and RT-PCR

Total RNA was extracted from S. pneumoniae cultures in CY medium using the RNeasy mini Kit (QIAGEN). Cells were harvested throughout the growth curve at 37 °C (A 550 nm of 0.12, 0.33, 0.67, and 0.66 that corresponds respectively to early, medium logarithmic, late logarithmic, and stationary growth phase) and stored in ice. Pellets were resuspended in 0.9% NaCl solution and stored at −80 °C. The concentration and the purity were estimated using an ND1000 Spectrophotometer (Nanodrop Technologies). Primers used for qRT-PCR are listed in Table 1.

cDNA was synthesized using SuperScript II Reverse Transcriptase (Invitrogen), according to the manufacturer's protocol. To ensure that the amplification observed in the PCRs was attributable to the cDNA template made from mRNA and not from contaminating genomic DNA, controls were carried out for each sample under the same conditions, except that transcriptase was not added to the reactions. Semi-quantitative real-time PCR (RT-PCR) experiments were performed using SYBR Green technology in an ABI Prism 7000 Sequence Detection System (Applied Biosystems). Each experiment was carried out in triplicate, so each relative gene expression reported for each point of the curve represents the average of three independent biologic replicates. Changes in sample gene expression were measured based on an external standard used as a calibrator (Wong & Medrano, 2005). Dunnet's test was used to determine whether the expression values of a given point were significantly different from other points of the curve.

Genome analysis and multiple-sequence alignments

Sequences of S. pneumoniae genomes were retrieved from the NIH GenBank database ( Multiple-sequence alignments were performed using the ClustalW2 program (

Results and discussion

In silico analysis indicates that galU and gpdA genes are co-transcribed

The analysis of the 23 S. pneumoniae genomic complete sequences available at the NIH website ( revealed that galU and gpdA are adjacent and in the same orientation in the S. pneumoniae chromosome. Transcriptional terminator prediction was made using TransTermHP ( TransTermHP was run on seven complete S. pneumoniae genomes currently available at this site. The search process indicates that no terminator is present after gpdA gene although a rho-independent transcriptional terminator was found downstream of galU. In the case of S. pneumoniae R6, a predicted terminator was found with a confidence value of 70, which is regarded as high (Kingsford et al., 2007).

The gpdA and galU genes are located together and are transcribed from the same DNA strand in 61 different genomes belonging to the Firmicutes phylum. However, the galU gene and its flanking regions do not have the same organization in other bacterial species not closely related to S. pneumoniae (Varón et al., 1993; Dean & Goldberg, 2002; Silva et al., 2005).

Identification of a functional gpdA-galU promoter region

Promoter prediction on the 827-bp sequence upstream of the gpdA gene was carried out using the Neural Network Promoter Prediction program ( Four sequences were detected by this program as putative promoters with a score of at least 0.88 (Fig. 1).

Figure 1.

(a) Map of the 2.589-kb region containing the galU gene and putative promoters predicted by nnpp software (cut off > 0.88) (A–D). Numbers in brackets indicate the nucleotide positions corresponding to the sequence included in the GenBank database under accession number AE007317.1. Fragments F1–F4 were cloned in pLSE4 reporter plasmid generating pMMP1 to pMMP4 derivatives. (b) Putative promoter sequences (A–D). Putative starting point of each sequence is bold-faced.

To determine whether the proposed promoter sequences actually represent a gpdA-galU promoter, three DNA fragments, one overhanging the other (F1, F3, and F4) and containing the putative promoters, were PCR-amplified. A 1030-bp DNA fragment (F2) containing full-length gpdA gene was also amplified to explore the existence of a promoter region within this gene. After digestion with the appropriate restriction enzymes, the DNA fragments were ligated to the promoter probe vector pLSE4 previously treated with the same enzymes and used to transform competent cells of E. coli C600. The recombinant plasmids were transferred to pneumococcal M31 strain (ΔlytA). Lincomycin-resistant M31 transformants, harboring different recombinant plasmids designated pMMP1 to pMMP4, were obtained. Streptococcus pneumoniae M31 harboring pMMP1 lysed at the end of the exponential phase of growth (Fig. 2). Moreover, a detectable LytA amidase activity (7.4 U mg−1 of protein) was found in sonicated extracts prepared from M31 harboring pMMP1, indicating the existence of a functional promoter in the F1 fragment.

Figure 2.

(a) Functional characterization of gpdA-galU promoters. (a) Growth and lysis curves of Streptococcus pneumoniaeM31 (∆lytA) strain harboring plasmids pLSE4, pMMP1, pMMP2 pMMP3, or pMMP4. (b) Pneumococcal cell wall hydrolyzing activity of cell extracts obtained by sonication.

M31 cells containing pMMP2 also exhibited lysis at the end of exponential phase of growth although with a rate three times lower than that of the pMMP1 derivative. Moreover, LytA amidase activity was undetectable in sonicated extracts of this strain. By contrast, strains containing pMMP3, and pMMP4 and the promoterless vector (pLSE4), did not show any lysis (Fig. 2).

Sequence analysis of the region upstream gpdA-galU

The region located immediately upstream of gpdA is highly conserved in pneumococcal genomes (Fig. 3a) and was searched for the presence of promoter-like sequences. A consensus, −10 extended promoter (TATGATATAAT; Sabelnikov et al., 1995) was identified in the six pneumococcal strains, and its single-nucleotide variant TATGATAgAAT was found in the rest of the 23 genome sequences analyzed. Changes in this nucleotide are not critical for the binding of the σ subunit of the RNA polymerase (Sevostyanova et al., 2007). A Shine- and Dalgarno-like sequence (SD) (AGGAGG) is located five nucleotides upstream of the gpdA start codon (AUG). As expected, the polycistronic transcript has a putative transcription start site within a purine (A) nucleotide located seven nucleotides downstream of the −10 extended promoter element. The putative promoter sequence consists of an extended −10 element without a −35 site and is situated 30 nucleotides upstream of the ATG initiation codon of the gpdA gene.

Figure 3.

(a) Multiple-sequence alignment of gpdA-galU promoter region from Streptococcus pneumoniae complete genome sequences deposited in databases. (b) Promoter regions of different pneumococcal genes are shown. sulA, murB,pcrA, comCDE, and fcsR genes have proven −10 extended elements; fcsK, yefM-yoefB, relB2, tts genes present −10 and −35 elements; ung and spxB exhibit −10 extended and −35 elements. Extended −10 sites, canonical −10 and −35 boxes are underlined. Conserved nucleotides are bold-faced. The consensus promoter sequence described by Sabelnikov is given: TnTGnTATAAT. The numbers of nucleotides that are not showed in yefM-yoefB and spxB genes are indicated in brackets. The sequences of the genes were taken from references cited in the text.

Regarding pneumococcal promoters, we compared similar extended −10 elements of sulA (dihydropteroate synthase), murB (UDP-N-acetylenolpyruvoylglucosamine reductase), pcrA (ATP-dependent DNA helicase), comCDE operon, and fcsR (regulator of fucose operon) with the gpdA-galU promoter described herein (Chan et al., 2003; Ware et al., 2005; Ruiz-Masó et al., 2006 and Martin et al., 2010). These sequences match the consensus extended −10 region previously described by Sabelnikov et al. (1995) (TnTGnTATAAT). The nucleotides in the canonical −10 hexamer TATAAT are conserved in murB, pcrA, comCDE, fcsR, and gpdA-galU (six strains). Moreover, −10 extended promoter element of comCDE showed an alteration of the T-TG extension. Also, −10 and −35 promoter elements found in fcsK (fuculose kinase), yefM-yoeB (toxin–antitoxin), relB2 (antitoxin), and tts (βglucosyltransferase) genes were compared, and most of them showed −10 canonical element (TATAAT) and the −35 box (TTGACA) with minor differences (Llull et al., 2001; Chan et al., 2003, 2011; Nieto et al., 2006). On the other hand, ung (DNA-uracil glycosylase) and spxB (pyruvate oxidase) exhibit −10 extended promoter element and a −35 box (Ramos-Montañez et al., 2008; Ruiz-Cruz et al., 2010), (Fig. 3b).

Expression of the galU gene is higher during exponential phase

Semi-quantitative real-time reverse transcription PCR (RT-PCR) was used to compare relative transcriptional abundance of galU transcripts at different points of the growth curve of S. pneumoniae R6. The results showed that galU-specific transcript levels in the exponential phase were 14-fold higher than during the other phases (Dunnet's test, < 0.05; Table 2), in agreement with that previously suggested for metabolic genes involved in biosynthetic processes whose expression is probably down-regulated in the stationary phase (Navarro Llorens et al., 2010). However, our results contrast with those carried out in Bacillus subtilis (Varón et al., 1993) and Lactobacillus casei (Wu et al., 2009) where the expression of the corresponding galU genes increased in the stationary phase.

Table 2. Relative expression of galU gene of Streptococcus pneumoniae 406 during different growth stages
Growth stageEarly exponentialMid-exponentialLate exponentialStationary phase
  1. qRT-PCR analysis was used to determine galU transcript abundance at different time points during growth. The transcription data are the means ± standard deviations of three independent experiments.

Relative galU expression2.4 ± 0.262.1 ± 0.322.6 ± 0.560.17 ± 0.06


This research was partly supported with grants from Universidad de Buenos Aires, Agencia Nacional de Promoción Científica y Tecnológica, ANPCYT, and Dirección General de Investigación Científica y Técnica (SAF2009-10824). Centro de Investigación Biomédica en Red de Enfermedades Respiratorias (CIBERES) is an initiative of the ISCIII. L.B. and M.M. are members of ‘Carrera del Investigador’, CONICET, Argentina. We thank E. Cano for skillful technical assistance.