Maltodextrin Utilization Plays a Key Role in the Ability of Group A Streptococcus To Colonize the Oropharynx

Bacterial strains and culture media.The genome of serotype M1 strain MGAS5005 has been fully sequenced (52). This strain has been extensively used for in vitro and animal models of GAS infection including transcriptome analysis during growth in human saliva (49, 51, 52, 61). The other bacterial strains used in this study are listed in Table 1. GAS was grown on Trypticase soy agar containing 5% bovine serum albumin (BSA; Becton Dickinson) or in Todd-Hewitt broth containing 0.2% (wt/vol) yeast extract (THY; Difco). Carbohydrates were added at a 1.0% (wt/vol) concentration to a carbohydrate-free preparation of a commercially available chemically defined medium (CDM; JR Biosciences) (62). For this article, we will use glucose medium to refer to the carbohydrate-free CDM with 1% glucose added and maltose medium to refer to the carbohydrate-free CDM with 1% maltose added. Spectinomycin (Sigma) was added to THY agar or broth at 150 μg/ml, and chloramphenicol (Sigma) was added at 4 μg/ml to spectinomycin THY agar or broth when appropriate. Blue-white screening was performed by adding 50 μg/ml of 5-bromo-4-chloro-3-indolyl phosphate (Sigma) to the chloramphenicol-spectinomycin THY agar.

Growth of GAS in human saliva.GAS was grown in human saliva as previously described (48). Saliva was collected on ice from healthy volunteers as previously described under a Baylor College of Medicine Institutional Review Board human subjects protocol (48). Pooled saliva collected from at least four donors was used to minimize the effects of donor variation on study results.

DNA sequence analysis.Chromosomal DNA was isolated using a phenol-chloroform extraction method as previously described (41). DNA sequencing primers were designed on the basis of the serotype M1 strain MGAS5005 genome (Table 2) (52). The malE open reading frame M5005_Spy_1058 corresponds to open reading frame SPy1294 in serotype M1 strain SF370 (12). Sequence data obtained from both DNA strands with an Applied Biosystems 3700 automated sequencer were assembled with Sequencher 4.5. Comparative sequence alignment of the inferred amino acid sequences was performed with CLUSTALW.

RNA isolation and transcript level analysis.Serotype M1 strain MGAS5005 was grown in various media to early- and late-logarithmic growth phases. RNA was isolated with a Fast Prep Blue Kit (Q/BioGene) and purified using an RNeasy kit (QIAGEN) as previously described (49). The concentration and quality of RNA were assessed with an Agilent 2100 Bioanalyzer and analysis of the A₂₆₀/A₂₈₀ ratio. cDNAs were created from the RNA using SuperscriptIII (Invitrogen) according to the manufacturer's instructions. TaqMan real-time quantitative reverse transcription-PCR (QRT-PCR) was performed in quadruplicate with an ABI Thermocycler 7700 (Applied Biosystems) using the change in cycle threshold for analysis, as previously described (49). TaqMan primers and probes for malE and the internal control gene proS are listed in Table 2.

Detection of MalE on the GAS cell surface.Surface localization of MalE was assessed with a FACSCaliber flow cytometer (BD Biosciences) using affinity-purified MalE-specific antibodies (Bethyl Laboratories) (28, 42). Antibody against streptococcal phospholipase A₂ (SlaA), a GAS protein not encoded by strain MGAS5005, was used as a control for nonspecific antibody binding. Strain MGAS5005 and the ΔmalE isogenic mutant strain (see below) were grown to exponential phase in either THY medium or human saliva, harvested by centrifugation, washed once in Dulbecco's phosphate-buffered saline (PBS), pH 7.2, and suspended in Dulbecco's PBS at 10⁸ CFU/ml. Anti-MalE or anti-SlaA antibody was added at a final concentration of 0.05 μg/100 μl to 100 μl of bacterial suspension and incubated for 30 min on ice. Samples were washed with Pharmingen stain buffer (BD Biosciences) and stained with phycoerythrin-conjugated donkey anti-rabbit immunoglobulin G (Jackson ImmunoResearch) (1:100 dilution) for 30 min on ice. The cells were washed again with stain buffer and analyzed via flow cytometry.

Creation of ΔmalE isogenic mutant strain.The ΔmalE isogenic mutant strain was created from parental serotype M1 strain MGAS5005 using the PCR-mediated method of Kuwayama et al. (25). Primers were designed to amplify the 5′ and 3′ ends of the malE gene region along with nucleotide sequences that were complementary to the 5′ or 3′ portion of the spectinomycin (spc) resistance cassette (Table 2). A third set of primers with nucleotide sequences complementary to the 5′ and 3′ ends of the malE gene region was used to amplify the spc cassette from plasmid pSL60 (30). Fusion PCR was then used to link the 5′ and 3′ malE gene region PCR products to the spc cassette via the overlapping nucleotide sequence regions (9, 25). This resulted in a PCR product where the spc cassette, flanked by the 5′ and 3′ malE gene regions, was now inserted in the place of nearly the entire malE open reading frame. The gene disruption PCR construct was used to transform competent GAS cells with spectinomycin resistance used as the selection mechanism. The ΔmalE isogenic mutant strain was analyzed by Southern hybridization and DNA sequencing to confirm that the proper genetic construct was obtained (Fig. 1).

• Open in new tab
• Download powerpoint

FIG. 1.

Southern blot analysis of wild-type (MGAS5005) and ΔmalE isogenic mutant strains was performed following digestion of genomic DNA with XmnI. The spc cassette replacement of malE removes an XmnI-cut site, thereby increasing the fragment size by ∼2500 bps.

Complementation of ΔmalE isogenic mutant strain.For complementation purposes, we utilized plasmid pDC123 which includes the chloramphenicol acetyltransferase (cat) gene and the phoZ gene that allows for blue-white screening via alkaline phosphatase activity (27). The complete malE gene along with ∼250 bp upstream and ∼15 bp downstream was amplified from strain MGAS5005 chromosomal DNA using primers that introduced EcoR1-cut sites at the 5′ and 3′ ends (Table 2). The resulting PCR product was digested with EcoR1 (New England Biolabs) and cloned into the EcoR1 site of plasmid pDC123. The resulting plasmid, named pDCmalE, was used to transform competent ΔmalE cells. Chloramphenicol resistance and blue-white screening were used to choose transformed strains for further analysis. Presence of the desired pDCmalE construct was confirmed using PCR and DNA sequencing. pDC123 lacking the malE gene was used to transform strain MGAS5005 and the ΔmalE isogenic mutant strain as controls.

Carbohydrate binding assays.Strain MGAS5005, the ΔmalE isogenic mutant strain, and the compΔmalE (complemented mutant) strain were grown in either glucose or maltose medium to mid-exponential phase, collected by centrifugation, washed, and suspended to an optical density at 600 nm (OD₆₀₀) of 0.5 in 150 μl of CDM lacking carbohydrates. A total of 25 μl of 280 μM 1.85 MBq [¹⁴C]maltose or [¹⁴C]glucose (GE Healthcare) was added to obtain a final concentration of 40 μM radiolabeled carbohydrate. At various times 40-μl samples were removed and filtered on a 0.45-μm-pore-size nitrocellulose membrane (Millipore) which was then rinsed twice with CDM lacking carbohydrates. The radioactivity trapped on each filter was then determined by liquid scintillation counting (Beckman Model LS7500). Samples were taken every minute for the first 4 min during which carbohydrate binding rates were found to be linear. The amount of protein in the cultures was determined by Bradford assay (Bio-Rad). All experiments were performed in quadruplicate.

Mouse colonization experiments.All animal experiments were performed under a protocol approved by the Baylor College of Medicine Institutional Animal Care and Use Committee. Mouse throat colonization studies were conducted with adult (18 to 20 g) female outbred CD-1 Swiss mice (Harlan Sprague-Dawley, Inc.) as described previously (30). Strain MGAS5005, the ΔmalE isogenic mutant strain, and the compΔmalE strain were grown in THY medium and harvested at an OD₆₀₀ of ∼0.5. The cells were washed once and suspended in sterile, pyrogen-free PBS to a density of ∼1.5 × 10⁸ CFU in 100 μl. Both nares of each mouse were inoculated with 50 μl of the GAS suspension. This dose of MGAS5005 has been previously shown to result in the colonization of ∼75% of inoculated animals (30). The mouse throats were swabbed prior to inoculation and then daily thereafter. Throat swabs were plated onto BSA and grown overnight, and beta-hemolytic colonies were tested for the presence of GAS carbohydrate antigen via latex agglutination (BD Biosciences). Blood collected from dead animals by cardiac puncture was cultured on BSA. In addition, compΔmalE colonies recovered from colonized and dead mice were tested for loss of the complementary plasmid by culturing in THY broth with chloramphenicol.

Statistical analysis.Flow cytometry and RNA transcript level comparisons were performed using a Student's two-sided t test. Growth comparisons were done using analysis of variance. Radiolabeled carbohydrate binding rates were compared using linear regression. The chi-square test was used to assess statistical differences in throat colonization rates among the animal groups infected with strain MGAS5005, its ΔmalE isogenic mutant derivative, and the compΔmalE strain. Analysis of variance was used to test for differences in number of CFU recovered from animals infected with the three strains. Statistical significance was assigned at a two-sided P value of 0.05 using Bonferroni's adjustment for multiple comparisons when appropriate. Statistical calculations were performed with NCSS software, version 2004 (Kaysville, Utah).

Description of the malE gene region and homology of MalE among various GAS strains.MalE has been identified as a putative lipoprotein based on the presence and composition of its amino-terminal secretion signal sequence (28, 53). Analysis of the serotype M1 strain MGAS5005 genome (52) shows that downstream of malE are two genes, malF and malG, whose products are predicted to form a permease for solute transport (Fig. 2). Together with MalE, MalF and MalG are predicted to comprise two parts of an ATP binding cassette transport system, a common mechanism by which bacteria import substrates (18). The putative ATP-binding portion of the ATP binding cassette transport system is located outside of the malE genome region, an arrangement that is not unusual in gram-positive bacteria (39, 45). Upstream of malE and divergently transcribed are malR, which encodes a putative transcriptional regulator, and malP and malQ, which encode enzymes putatively involved in generating various forms of glucose from maltodextrins (Fig. 2). The identical gene arrangement is present in all 12 GAS strains sequenced to date (2-4, 12, 17, 34, 50).

• Open in new tab
• Download powerpoint

FIG. 2.

Schematic of the malE gene region in GAS strain MGAS5005. M5005_Spy# refers to the gene number in serotype M1 strain MGAS5005. ABC, ATP-binding cassette. The identical gene arrangement is present in all 12 GAS strains sequenced to date (2-4, 12, 17, 34, 50, 52).

MalE is the only protein encoded by this gene region that is putatively surface exposed and thus possibly under selective pressure from the host immune system. Some GAS surface-exposed proteins are highly variable among various strains, whereas others are well conserved (31, 42). We sequenced the malE gene from 28 strains comprising 22 different M serotypes (Table 1). The malE gene was present in all strains tested. There was an average of 3.2 amino acid replacements per 415 amino acid sites (Table 1). These data show that MalE is highly conserved among diverse GAS strains, indicating that information generated regarding MalE function in a particular GAS strain may be broadly applicable to diverse GAS strains.

malE transcript level is increased in maltose medium compared to glucose medium and nutrient-rich medium.Many bacteria employ carbon catabolite repression, a phenomenon where the use of sugars other than glucose is minimized when glucose is available (19). By adding specific sugars to carbohydrate-free CDM, we tested the hypothesis that growth in maltose medium would lead to increased transcript levels of malE compared to growth in THY or glucose medium. As assessed by real-time QRT-PCR, during the mid-exponential phase of growth the transcript level of malE in the maltose medium showed a 35-fold increase compared with the glucose medium and 95-fold increase compared with the THY medium (P < 0.001 for both) (Fig. 3). At entry into stationary phase, the malE transcript level was fourfold and threefold higher in the maltose medium compared to the glucose and THY media, respectively (P = 0.024 and P = 0.008). These data show that a significant increase in malE transcript level occurred during growth in a medium containing maltose, suggesting that malE transcription is repressed in a nutrient-rich environment but induced in the presence of maltose.

• Open in new tab
• Download powerpoint

FIG. 3.

Transcript levels of malE are increased during growth in maltose medium and human saliva. Strain MGAS5005 was grown to mid-exponential or early stationary phase in either standard laboratory medium (THY medium), CDM with 1% glucose (wt/vol) or 1% maltose, or human saliva. TaqMan real-time QRT-PCR was performed using probe and primers specific for malE (Table 1) for cells at mid-exponential and early-stationary growth phase. The transcript level of malE was normalized to proS, a gene expressed constitutively throughout the GAS cell cycle and whose transcript levels are not significantly different during growth in THY medium compared with saliva (7, 49). The mean level ± standard deviation of quadruplicate measurements is presented.

malE transcript level is increased during growth in human saliva compared to growth in a nutrient-rich medium.Previous expression microarray analysis found that malE transcript levels were high during growth of GAS in human saliva, especially in the exponential growth phase (49). Thus, we tested the hypothesis that the interaction of GAS with saliva leads to increased malE transcript levels compared to growth in a nutrient-rich medium. Similar to the situation during growth in maltose, the malE transcript level in saliva at mid-exponential phase was 45-fold higher than growth in glucose medium and 120-fold higher than growth in THY medium (P < 0.001 for each) (Fig. 3). Upon entry into stationary phase in saliva, the transcript levels of malE decreased but remained 13-fold and 10-fold higher than growth in glucose or THY medium, respectively (P < 0.001 for each). Taken together, we conclude that growth of GAS human saliva led to a significant increase in the malE transcript level, suggesting that MalE is important for the successful interaction of GAS with human saliva.

Growth in human saliva increased the amount of MalE on the GAS cell surface.We used flow cytometry to test two hypotheses. First, we predicted that MalE would be expressed on the GAS cell surface. Second, we expected that the increased transcript levels of malE during growth in human saliva would result in increased MalE on the GAS cell surface compared to growth in THY medium. We used anti-streptococcal phospholipase A₂ (SlaA) antibody as the control antibody because SlaA is not produced by strain MGAS5005 (4, 52). The flow cytometry data showed no significant increase above the control level in the amount of bound anti-MalE when the ΔmalE isogenic mutant strain was examined, suggesting that the anti-MalE antibody is specific for MalE (Fig. 4A). When wild-type strain MGAS5005 was studied, there was an increase in binding above the control level for anti-MalE during the mid-exponential growth phase in THY medium (Fig. 4B). Moreover, there was an increase in bound anti-MalE at mid-exponential phase during growth in saliva compared to growth in THY medium (Fig. 4C) (P < 0.001). These data are consistent with the in silico analysis predicting that MalE is present on the GAS cell surface and therefore accessible to antibody. Also, the increased transcript level of malE observed during growth in saliva appeared to result in increased MalE levels on the GAS cell surface.

• Open in new tab
• Download powerpoint

FIG. 4.

Analysis of MalE cell surface expression on GAS using flow cytometry. The ΔmalE isogenic mutant strain (A) and strain MGAS5005 (B and C) were grown to mid-exponential phase in THY medium (A and B) or saliva (C), washed with PBS, and treated with MalE-specific polyclonal rabbit antibody (gray shading) or control antibody (black line). Phycoerythrin-coated donkey anti-rabbit immunoglobulin G was used as secondary antibody, and the cells were analyzed by flow cytometry.

The ΔmalE isogenic mutant strain had decreased growth in maltose medium compared to wild-type.To directly test the role played by MalE in various GAS activities, we created a nonpolar ΔmalE isogenic mutant strain from parental M1 strain MGAS5005 and genetically complemented the mutant strain to create strain compΔmalE (see Material and Methods). We tested the hypothesis that MalE is important for the utilization of maltose but not glucose by comparing the growth patterns of wild-type GAS, the ΔmalE isogenic mutant strain, and the compΔmalE strain in various media. There was no growth for any strain in the carbohydrate-free CDM when no exogenous carbohydrate was added (data not shown). We observed no differences when the three strains were grown in THY medium or during growth in glucose medium (Fig. 5A, 5B) (P = 0.791 for THY; P = 0.972 for glucose medium). However, the ΔmalE isogenic mutant strain was significantly less able to grow in maltose medium compared to the wild-type and complemented strains (Fig. 5C) (P = 0.001). There was no significant difference in growth between the wild-type and the compΔmalE strain during growth in maltose medium (P = 0.611). These data show that MalE is needed for GAS to optimally grow in maltose medium.

• Open in new tab
• Download powerpoint

FIG. 5.

Growth curves of wild-type, ΔmalE, and compΔmalE strains in various media. OD₆₀₀ readings were taken at indicated times for growth in nutrient-rich medium (THY) and CDM with either 1% glucose or 1% maltose. The number of CFU was used to analyze growth in human saliva as previously described (48). Strains are represented as indicated on the figure. Growth media were THY broth (A), CDM with 1% glucose (B), CDM with 1% maltose (C), and human saliva (D). Data graphed are mean values ± standard deviation for five independent experiments.

The ΔmalE isogenic mutant strain grew less in human saliva compared to wild type.A key finding of our earlier GAS-saliva investigations was that diverse GAS strains were readily able to grow in human saliva and, contrary to growth in laboratory media, persisted at maximal CFU counts for up to 30 days (48). The observation that the malE transcript level increased 100-fold during growth in saliva compared to growth in THY medium suggested that, similar to its role in a maltose medium, MalE might be critical to the ability of GAS to grow in human saliva. We tested this hypothesis by comparing the growth of strain MGAS5005, the ΔmalE isogenic mutant strain, and the compΔmalE strain in saliva pooled from healthy donors. As previously described, GAS saliva growth experiments are performed using CFU analysis rather than density readings because saliva can aggregate GAS and thereby interfere with OD readings (8, 48). Whereas both the wild-type and complemented mutant strains grew ∼2.5 log₁₀ CFU/ml from baseline and persisted at a density of ∼1 × 10⁷ CFU/ml, the ΔmalE isogenic mutant strain only grew ∼1.0 log₁₀ CFU/ml and persisted at ∼5.5 × 10⁵ CFU/ml (Fig. 5D) (P value for growth = 0.002; P value for persistence = 0.007). The addition of glucose at 100 mg/dl to the saliva growth medium restored the growth of the ΔmalE isogenic mutant strain to wild-type levels (data not shown). Therefore, we conclude that MalE is key to the optimal growth and persistence of GAS in human saliva.

The ΔmalE isogenic mutant strain was deficient in maltose, but not glucose, binding.In earlier portions of this work, we found that MalE was expressed on the GAS cell surface and that a ΔmalE isogenic mutant strain grew poorly in maltose medium. To test the hypothesis that MalE mediates maltose transport, we studied GAS binding of ¹⁴C-radiolabeled carbohydrates. We found no difference among the ability of the wild-type, ΔmalE, or compΔmalE strains to bind glucose (P = 0.541) (Fig. 6A). In contrast, in comparison to the wild-type and the complemented ΔmalE strain, the ΔmalE mutant strain showed a 95% decrease in maltose binding (P < 0.001) (Fig. 6B). These results show that the presence of MalE is required for maximal binding of maltose by GAS but not for the binding of glucose. These findings are in concordance with the earlier data demonstrating that MalE is needed for the optimum growth of GAS in maltose medium but not in glucose medium.

• Open in new tab
• Download powerpoint

FIG. 6.

Radiolabeled carbohydrate binding by GAS. Serotype M1 strain MGAS5005, ΔmalE isogenic mutant strain, and compΔmalE strain were grown in CDM with either 1% glucose (A) or 1% maltose (B) to mid-exponential phase. Cells were harvested by centrifugation, washed with carbohydrate-free CDM, and suspended to an OD₆₀₀ of 0.5. [¹⁴C]glucose (A) or [¹⁴C]maltose (B) was added for a final concentration of 40 μM. Samples were removed every minute for 4 min and passed through a 0.45-μm-pore-size filter. Filters were washed twice with carbohydrate-free CDM, and retained radioactivity was determined using a liquid scintillation counter. Shown are mean values from four independent experiments ± standard deviations.

The ΔmalE isogenic mutant strain was unable to efficiently colonize the mouse oropharynx.Although mice are not normally susceptible to pharyngeal infection by GAS, temporary colonization of the mouse oropharynx occurs following nasopharyngeal inoculation (30, 35, 43). We used adult outbred CD-1 mice to test the hypothesis that MalE is needed for optimum GAS colonization of the oropharynx. Each group of 30 mice was challenged intranasally with 1.5 × 10⁸ CFU GAS in a total of 100 μl (50 μl per nostril). The oropharynx of each mouse was then swabbed daily using a fine-tip, sterile cotton applicator (Fisher), which was then used to inoculate a BSA plate. The plates were incubated for 24 h and examined for beta-hemolytic colonies, which were tested for the presence of GAS carbohydrate antigen via latex agglutination (BD Biosciences). At each time point tested, significantly more mice were colonized with the MGAS5005 and the compΔmalE strains compared to the ΔmalE isogenic mutant strain (Fig. 7A). Moreover, at each time point tested, the number of GAS CFU was significantly higher in the animals infected with the wild-type and complemented strains compared to the isogenic mutant strain (Fig. 7B). These data support our hypothesis that MalE is critical in the ability of GAS to successfully colonize the oropharynx.

• Open in new tab
• Download powerpoint

FIG. 7.

Colonization and CFU recovery rates among mice infected with GAS. Adult outbred CD-1 mice (30 per group) were inoculated with ∼1.5 × 10⁸ CFU of either serotype M1 strain MGAS5005, the ΔmalE isogenic mutant strain, or the compΔmalE strain. Mice oropharynxes were swabbed daily onto BSA. Plates were incubated for 24 h, and beta-hemolytic colonies were counted and tested for GAS carbohydrate antigen using latex agglutination. (A) Percentage of mice with GAS isolated by day. (B) Number of CFU isolated by day. Data graphed are mean values ± standard deviation. P values refer to either a chi-square (A) or analysis of variance (B) test among the three groups at indicated times. Subsequent testing for differences between wild-type and ΔmalE strains remained significant at P < 0.05.

As we were unable to maintain chloramphenicol selection on the compΔmalE strain during infection, we assayed for loss of the complementary plasmid. A total of 43 of 50 colonies recovered from animals infected with the complemented ΔmalE strain remained chloramphenicol resistant. Therefore, even in the absence of chloramphenicol pressure, the majority of compΔmalE strains maintained the pDCmalE plasmid, providing further evidence of the critical nature of MalE during oropharyngeal colonization.