A Novel Heme Transporter from the Energy Coupling Factor Family Is Vital for Group A Streptococcus Colonization and Infections

ECF systems are new transporters that take up various vitamins, cobalt, or nickel with a high affinity. Here, we establish the GAS SiaFGH proteins as a new ECF module that imports heme and demonstrate its importance in virulence. SiaFGH is the first heme ECF system described in bacteria. We identified homologous systems in the genomes of related pathogens from the Firmicutes phylum. Notably, GAS and other pathogens that use a SiaFGH-type importer rely on host hemoproteins for a source of iron during infection. Hence, recognizing the function of this noncanonical ABC transporter in heme acquisition and the critical role that it plays in disease has broad implications.

ABSTRACT

Group A streptococcus (GAS) produces millions of infections worldwide, including mild mucosal infections, postinfection sequelae, and life-threatening invasive diseases. During infection, GAS readily acquires nutritional iron from host heme and hemoproteins. Here, we identified a new heme importer, named SiaFGH, and investigated its role in GAS pathophysiology. The SiaFGH proteins belong to a group of transporters with an unknown ligand from the recently described family of energy coupling factors (ECFs). A siaFGH deletion mutant exhibited high streptonigrin resistance compared to the parental strain, suggesting that iron ions or an iron complex is the likely ligand. Iron uptake and inductively coupled plasma mass spectrometry (ICP-MS) studies showed that the loss of siaFGH did not impact GAS import of ferric or ferrous iron, but the mutant was impaired in using hemoglobin iron for growth. Analysis of cells growing on hemoglobin iron revealed a substantial decrease in the cellular heme content in the mutant compared to the complemented strain. The induction of the siaFGH genes in trans resulted in the induction of heme uptake. The siaFGH mutant exhibited a significant impairment in murine models of mucosal colonization and systemic infection. Together, the data show that SiaFGH is a new type of heme importer that is key for GAS use of host hemoproteins and that this system is imperative for bacterial colonization and invasive infection.

IMPORTANCE ECF systems are new transporters that take up various vitamins, cobalt, or nickel with a high affinity. Here, we establish the GAS SiaFGH proteins as a new ECF module that imports heme and demonstrate its importance in virulence. SiaFGH is the first heme ECF system described in bacteria. We identified homologous systems in the genomes of related pathogens from the Firmicutes phylum. Notably, GAS and other pathogens that use a SiaFGH-type importer rely on host hemoproteins for a source of iron during infection. Hence, recognizing the function of this noncanonical ABC transporter in heme acquisition and the critical role that it plays in disease has broad implications.

INTRODUCTION

Group A streptococcus (GAS) is a significant pathogen and a leading cause of human morbidity and mortality. Gram-positive GAS is the etiological agent of common infections such as pharyngitis and impetigo. These simple episodes can lead to immune-based complications in susceptible individuals, with severe outcomes such as acute rheumatic fever, rheumatic heart disease (RHD), and poststreptococcal glomerulonephritis. GAS also produces rare but life-threatening conditions, including bacteremia, necrotizing fasciitis, and streptococcal toxic shock syndrome (1, 2). Altogether, GAS is responsible for several million cases globally and claims approximately 500,000 lives annually (3). RHD and invasive infections are the primary culprits in GAS-related mortality (4, 5). The development of a vaccine against GAS has met serious challenges due to hypervariability and cross-reactions with host epitopes. Hence, the control of GAS infections continues to rely on conventional antibiotics. Still, resistance to macrolides and clindamycin, which are used as an alternative (6) or together with β-lactams (7), is on the rise.

The establishment of infection requires pathogens such as GAS to successfully compete with the host for iron (8, 9). GAS cannot obtain iron from host ferric proteins, transferrin, or lactoferrin (9). But this beta-hemolytic pathogen thrives on heme iron and can readily remove heme from hemoglobin and other host hemoproteins (9). The best-characterized heme uptake system in GAS is encoded by the 10-gene Sia operon (10). The Sia mechanism consists of two surface proteins, Shr and Shp, and the heme ABC transporter SiaABC. The remaining siaDEFGH genes are yet to be described. Shr binds to hemoglobin on the surface, extracts the heme, and delivers it through the cell wall to Shp, using a shared heme-binding module named NEAT (11, 12). A rapid and affinity-driven mechanism then mediates the transfer of heme from Shp to SiaA, the substrate-binding component of the SiaABC transporter, for import into the cytoplasm (13–15). Shr, the first protein in the Sia heme relay system, is important for virulence (16, 17) and serves as a protective antigen in both passive and active vaccination models (18). The bicistronic operon hupYZ also contributes to GAS growth on heme iron. HupY is a cell wall receptor that binds heme in vitro (19). A hupY deletion mutant is impaired in the use of serum or hemoglobin as a source of iron and exhibits lower iron content when cultivated in vitro. The mechanism by which HupY captures heme from the surface and delivers it into the cytoplasm remains unknown. HupZ is a cytoplasmic enzyme that binds and degrades heme in vitro (20, 21). The sia and hupYZ operons are both negatively regulated by iron availability via the metal-responsive regulator MtsR (22). Besides SiaABC, the only other iron complex transporter reported in GAS is the SiuADBG/FtsABCD system. The SiuADBG/FtsABCD proteins consist of an ABC transporter that was implicated in the uptake of ferrichrome (23) as well as heme (24) and thus may exhibit broad specificity for iron compounds. In summary, three surface proteins and two conventional ABC transporters function in heme capture and import in GAS. This redundancy highlights the essential nutritional role that heme iron plays in GAS.

The energy coupling factor (ECF) transporter family is an unusual type of ABC importer that is widespread among Gram-positive bacteria (25). ECF transporters are comprised of a small membrane-embedded protein that provides substrate specificity (S component [EcfS]) and an ECF module consisting of a transmembrane protein (T component [EcfT]) and a pair of similar or identical cytosolic ATPases (A component [EcfA]) (26, 27). The suggested import mechanism of these noncanonical ABC transporters involves the toppling and repositioning of the S protein to facilitate substrate capture at the extracellular side and its release into the cytoplasm (25). Hence, the S component is the only ECF component suggested to interact with the ligand. Twenty-seven different S families have been identified, each for a separate micronutrient (e.g., various vitamins, amino acids, queuosine, or trace metals, not including iron or iron complexes) (26, 28). The ECF family is categorized into two groups. In group I, a single S component interacts with a specific ECF module, whereas in the group II system, multiple S components of unrelated sequences and various ligands can form a complex with the same ECF module (25, 26). Some S components, referred to as Solitary, do not have a recognized ECF module (29). The genetic organization suggests that the GAS siaFGH genes code for a group I ECF system in which all of the components are coexpressed and work exclusively together. The siaFGH genes are expressed in response to iron deprivation as part of the sia operon and along with the rest of the MtsR regulon. We hypothesize that the siaFGH genes encode a novel heme import system that is imperative for the establishment of GAS infection.

RESULTS

The siaFGH genes encode a putative ECF importer.

In silico analysis suggested that the SiaFGH proteins consist of the group I ECF family of transporters (25), with siaFGH respectively encoding the substrate-binding S component (EcfS), the transmembrane T component (EcfT), and the energy-coupling A component (EcfA) (Fig. 1A). Based on the substrate-binding component, SiaF, the SiaFGH transporter belongs to the HtsTUV ECF subgroup, whose ligands are unknown (26). SiaF is predicted to be an α-only protein with both the N and C termini in the cytosol (data not shown) (30). Six distinct helices make up the transmembrane domains, and an extended extracellular loop connects two of the helices in SiaF (31). The predicted SiaF ribbon structure (Fig. 1B) is highly similar to the overall fold and membrane topology shared by EcfS units, with one exception: the large extracellular loop that typically connects helices 1 and 2 in EcfS proteins joins helices 5 and 6 in SiaF. This difference is intriguing since this loop is proposed to serve as a lid over the binding site. The SiaFGH proteins are highly conserved in GAS isolates and the related bacteria Streptococcus dysgalactiae and S. equi, both of which encode the entire sia operon (Table 1).

FIG 1.

The siaFGH genes encode an ECF transporter. (A) Schematic representation of the SiaFGH ECF transporter. The SiaFGH proteins belong to the subgroups 3.A.1.31.1, 3.A.1.25.4, and 3.A.1.25.6, respectively, in the Transporter Classification (TC) database (http://www.tcdb.org) (54). The red oval indicates the substrate. (B) SiaF structural model (I-TASSER [53]) based on, which shares high similarity with the EcfS component in Lactobacillus brevis (PDB code 4RFS). The elongated extracellular loop between the fifth and sixth helices is in red.

TABLE 1.

In silico analysis suggests that the SiaFGH system is conserved in streptococcal strains

Organism	% identity to GAS SiaFGH protein (NCBI accession no.):
	SiaF (ACI61675)	SiaG (ACI61674)	SiaH (WP_136048563.1)
S. dysgalactiae	86.80 (WP_129556215.1)	66.37 (WP_111717518.1)	85.66 (WP_143928109.1)
S. equi	75 (WP_043040123.1)	64.6 (WP_165626339.1)	77.42 (WP_043026841.1)
S. canis	86.80 (WP_003045124.1)	74.34 (WP_164227913.1)	85.66 (WP_164405649.1)
S. castoreus	79.19 (WP_027969907.1)	68.37 (WP_027969908.1)	79.21 (WP_027969709.1)
S. phocae	73.10 (WP_054278834.1)	65.74 (WP_054278835.1)	78.85 (WP_037595860.1)
S. ictaluri	74.62 (WP_008086912.1)	59.11 (WP_008087003.1)	78.06 (WP_008087568.1)

Deletion of the siaFGH genes results in higher streptonigrin resistance and impaired use of hemoglobin iron.

The expression of group I ECFs is often regulated according to ligand requirements; hence, the repression of the siaFGH genes by iron and heme (22, 32) suggests that the SiaFGH substrate is related to iron metabolism. To test this hypothesis, we constructed a deletion mutant in the M49 NZ131 background and examined its impact on GAS physiology. The parental strain NZ131 and the isogenic ΔsiaFGH strain (ZE4950) grew equally well in Todd-Hewitt broth supplemented with yeast extract (THYB) (Fig. 2), but the mutant was more resistant to the antibiotic streptonigrin (50% lethal doses [LD₅₀s] of >3.5 μM for ZE4950 and 3 μM for NZ131). Iron potentiates bacterial killing by streptonigrin (33). Hence, the increased resistance suggests that the ΔsiaFGH strain has lower levels of cellular iron than the parental strain NZ131. We next tested if the function of the siaFGH genes is related to heme uptake, similarly to that of the neighboring shr, shp, and siaABC genes. We examined the wild-type and mutant strains cultivated in iron-depleted medium (THYB with 2,2′-dipyridyl [THYB-DP]) supplemented with hemoglobin. The deletion of the siaFGH genes impaired growth on low concentrations of hemoglobin (2.5 to 10 μM) (Fig. 3A). But the addition of hemoglobin at higher levels resulted in similar growth of both strains. These data support the hypothesis that SiaFGH is a high-affinity heme importer and are consistent with the redundancy of heme uptake mechanisms in GAS. The siaFG genes are integral membrane proteins and thus are difficult to clone under the control of a constitutive promoter. Thus, for complementation analysis, we cloned siaFGH under the control of the nisin-dependent promoter P_nisA (9). The growths of the isogenic mutant series in THYB with and without nisin were similar (see Fig. S1 in the supplemental material). The induction of siaFGH expression with nisin significantly improved the ability of the mutant harboring the complementation plasmid to grow on hemoglobin iron (i.e., in THYB-DP and nisin) compared with the mutant carrying an empty vector (Fig. 3B), demonstrating that siaFGH expression in trans complements the defect in the use of hemoglobin as an iron source. Complementation was observed only with 10 μM hemoglobin and higher, and the maximal growth of the complemented strain on 10 or 20 μM hemoglobin iron was slightly lower than that observed with the wild-type strain (0.537 versus 0.657 and 0.669 versus 0.75 for the complemented and wild-type strains, respectively), indicating that complementation is partial.

FIG 2.

The SiaFGH importer impacts intracellular iron levels. THYB containing 0 to 5 μM streptonigrin was inoculated with cultures grown overnight (starting OD₆₀₀ of 0.01). The growth (20 h) of NZ131 (wild type [WT]) and ZE4950 (ΔsiaFGH) at 37°C is shown. The data are from two independent experiments done in technical triplicates, with standard deviations (SD) shown. * indicates significance (P < 0.05 by Student’s t test, equal variance).

FIG 3.

The SiaFGH transporter impacts the use of hemoglobin iron. Cells from cultures grown overnight (starting OD₆₀₀ of 0.01) were grown in fresh THYB, THYB with 3 mM DP, or THYB-DP with hemoglobin. Nisin (0.5 μg/ml) was added to the cultures in panel B to induce siaFGH expression. Shown is the growth (20 h) of NZ131 (wild type) and ZE4950 (ΔsiaFGH) (A) or ZE4950/pNC111 (complement) and ZE4950/pMSP3535 (empty vector) (B). Data are derived from three experiments done in triplicates. * indicates significance (P < 0.05 by Student’s t test).

The SiaFGH proteins import heme but not iron ions.

To investigate if the SiaFGH system takes up heme, we studied heme accumulation in the mutant and complemented pair. GAS was grown overnight in THYB and 20 μM hemoglobin with or without nisin. The cellular heme levels in culture samples of equal cell densities were determined as previously described (34, 35). The complemented strain accumulated 90% more heme in response to nisin and about twice as much compared with the mutant (Fig. 4A). Therefore, the induction of siaFGH gene expression results in increased heme import. The complemented strain accumulated more heme even in the absence of nisin, indicating some basal level of P_nisA activity. To study heme uptake over time, we grew cells in THYB with hemoglobin and added nisin to the culture at the early logarithmic phase. The heme content in samples collected 0, 1, 2, and 3 h after induction was determined (Fig. 4B). While we observed a slow increase in heme levels over time in the mutant strain (harboring an empty vector), the addition of nisin to the complemented strain resulted in much faster heme uptake. Together, these observations establish that the SiaFGH system promotes heme uptake in GAS.

FIG 4.

The SiaFGH transporter promotes heme uptake. (A) Fresh THYB (HB) supplemented with 20 μM hemoglobin with or without nisin was inoculated with cultures of ZE4950 (ΔsiaFGH) harboring plasmid pNC111 (complement) or pMSP3535 (empty vector) grown overnight (starting OD₆₀₀ of 0.01). Cells were harvested, washed, and subjected to chloroform extraction. UV-visible spectra (250 to 700 nm) of the collected organic phases were recorded, and the concentrations of heme in the test samples were calculated as described previously (35). (B) Same as panel A except that nisin was added to the cultures at the early exponential phase, and the heme content was determined in samples collected 0, 1, 2, and 3 h after the addition of nisin.

To test if the SiaFGH system imports iron ions in addition to heme, we monitored radioactivity accumulation in cells grown in iron-depleted chemically defined medium (CDM) supplemented with ⁵⁵Fe in the ferrous (Fig. 5A) or the ferric (Fig. 5B) form. Interestingly, both strains incorporated radioactivity at a higher rate when supplied with reduced iron than with the oxidized form. The deletion of the siaFGH genes, however, did not affect the import of the metal in either state significantly. We also compared iron accumulation between a ΔsiaFGH strain carrying an empty vector and a mutant harboring the complementation plasmid. The mutant and complemented strains were grown in THYB supplemented with 80 μM iron and nisin, and we determined the iron content in cell samples collected after growth overnight using inductively coupled plasma mass spectrometry (ICP-MS) (Fig. 5C). Similar amounts of iron were found in both strains, indicating that the induction of the siaFGH genes did not result in enhanced uptake of free iron.

FIG 5.

Inactivation of the SiaFGH genes does not influence metal iron uptake. (A and B) Uptake of ferrous iron (A) or ferric iron (B) by NZ131 (wild type) and ZE4950 (ΔsiaFGH) grown in iron-depleted CDM supplemented with ⁵⁵Fe. The data are derived from two experiments done in duplicates; SD are shown (error bars). (C) Total iron content in the ZE4950/pNC111 (complement) and ZE4950/pMSP3535 (empty vector) strains grown overnight in THYB supplemented with 80 μM FeCl₃ and nisin, as determined by ICP-MS. The data are derived from two independent experiments.

SiaFGH contributes to GAS mucosal colonization as well as systemic infection.

We recently observed that the entire sia operon is induced during vaginal colonization in mice (19). We therefore used this infection model to test the role of the SiaFGH transporter in GAS colonization of the mucosa (Fig. 6). Estrus-synchronized mice were inoculated intravaginally with the M49 NZ131 (wild-type) or ZE4950 (ΔsiaFGH) strain, and the colonization rate was monitored for 5 days. Although there was an obvious trend of higher colonization rates in the mice inoculated with the wild-type strain immediately, we observed no statistical significance between the wild type and the mutant strain in the first day of the experiments. A significant colonization defect, however, was exhibited by the siaFGH mutant on the second day. The colonization rate remained much lower than that of the wild-type strain through the third day, but it was lost on day 5 for both the wild type and the mutant strain. To test if the SiaFGH system also impacts GAS invasive disease, a ΔsiaFGH mutant (ZE151) and a wild-type rescue strain (ZE152) were constructed in the virulent M1 MGAS5005 background and examined in a murine model of systemic infection (Fig. 7). Mice were infected intraperitoneally, and survival was recorded for 5 days. Infection by the wild-type rescue strain (ZE152) caused a rapid and lethal infection (resulting in a 20% survival rate on day 2). Mice infected with the mutant strain, however, exhibited slower disease with a lower mortality rate (50% final survival rate). Together, the data show that the siaFGH genes are imperative for mucosal colonization as well as systemic infection.

FIG 6.

Inactivation of siaFGH attenuates GAS vaginal colonization. NZ131 (wild type) and ZE4950 (ΔsiaFGH) were inoculated into the vaginal vault of mice, and colonization rates were determined. Significance was assessed on daily samples using a Mann-Whitney test. *, P < 0.05; **, P < 0.005; ns, not significant.

FIG 7.

Survival of CD-1 mice with systemic GAS infection. A Kaplan-Meier survival curve is shown for mice following intraperitoneal infection (1.4 × 10⁸ CFU) with ZE152 (wild-type rescue) (solid line) or ZE151 (ΔsiaFGH) (dashed line). The data are pooled from two independent experiments (P < 0.05 by a log rank test) (n = 30).

DISCUSSION

The 10-gene sia operon belongs to the core genome in GAS, S. dysgalactiae, and S. equi, which are chief human and animal pathogens. We and others have elucidated the structure-function of shr, shp, and siaABC and established their role in heme acquisition, adherence, and virulence in multiple serotypes (10–18, 36–38). However, the sia operon has five additional genes (siaDEFGH) that are yet to be described. The data in this study show that the last three genes, siaFGH, encode a novel ECF transporter that imports heme and promotes GAS colonization and infection. Notably, heme uptake in bacteria has so far been described only for conventional ABC transporters. In GAS, the two typical ABC systems that import heme are SiaABC and SiuADBG (which may also take up ferrichrome) (10, 23, 24). SiaFGH is the first heme ECF system to be described. The finding of a new heme transporter in the sia operon from the ECF system family expands the paradigm of heme uptake not only in GAS but also in the related bacteria S. dysgalactiae and S. equi and in bacteria overall.

Streptonigrin binding to metal iron leads to structural changes that facilitate a transition from a partially active form into a fully functioning state (39, 40). The inactivation of siaFGH resulted in decreased sensitivity to streptonigrin, suggesting that the ΔsiaFGH mutant accumulates lower levels of total iron when cultivated in normal THYB (Fig. 2). THYB is a beef heart infusion broth supplemented with iron/heme-rich yeast extract and contains both metal iron and heme. The observed changes in iron levels in the siaFGH mutant grown in THYB could therefore result from a reduction in the uptake of either free iron or heme. Growth experiments, however, suggested that the siaFGH mutant imports heme, since the mutant did not grow as well as the wild type in iron-depleted THYB (THYB-DP) supplemented with hemoglobin (≤10 μM), and expressing siaFGH in trans complemented the mutant phenotype (Fig. 3). The observation that the mutant phenotype was lost when the cells grew with 20 μM hemoglobin implies that, like other ECF systems, the SiaFGH system is a high-affinity importer whose role becomes redundant in the presence of a larger amount of hemoglobin. This redundancy is likely due to the activity of additional transporters, such as SiaABC and SiuADBG.

We used direct heme import assays to test the suggestion that heme serves as a ligand for the SiaFGH proteins (Fig. 4). The expression in trans of the siaFGH genes (from P_nisA) resulted in heme accumulation by the complemented strain when grown overnight with hemoglobin. Furthermore, the activation of the siaFGH genes in exponentially growing cells (in THYB with hemoglobin) triggered pronounced heme uptake in cells harboring the complementation plasmid. The noticeable amount of heme that was present in the siaFGH mutant, even in the absence of the complementation plasmid, likely resulted from the activity of other heme transporters. The high heme content found in the complemented strain (compared to the empty vector) in the absence of nisin suggests some basal activity of P_nisA. During iron uptake and accumulation, we did not observe any impact on the import of ferric or ferrous iron in the deletion mutant, nor did the induction of the siaFGH genes promote increased iron intake in cells grown on THYB supplemented with ferric iron (Fig. 5). Together, the data identify the SiaFGH importer as being specific for heme. We noted, however, that both wild-type and mutant cells imported ferrous iron readily compared to the oxidized ferric state. As acid-forming bacteria, GAS might create a reduced microenvironment during growth, and its import systems might be better adapted for the uptake of reduced iron.

The inactivation of siaFGH impaired GAS growth on hemoglobin as the sole source of iron (Fig. 3) and the pathogen’s ability to import heme from hemoglobin (Fig. 4). Hemoglobin is too large to diffuse through the cell wall and reach the membrane SiaFGH transporter. Gram-positive pathogens typically use surface receptors to capture heme from hemoglobin or other host sources on the bacterial surface and shuttle it through the cell wall to a membrane transporter (10, 41). In GAS, Shr, Shp, and SiaABC together create such a heme relay system (10–15). It seems possible that SiaF also interacts with one or more of the heme relay components encoded by the sia operon. SiaF may also receive heme from the recently described heme receptor HupY (19). Alternatively, small amounts of heme that are released by hemoglobin arrive at the membrane and reach SiaF. Additional studies are required to examine these possibilities and to describe the mechanism of heme capture by the SiaFGH transporter.

GAS is not able to obtain the iron that it needs for growth from the ferric glycoproteins, lactoferrin, or transferrin (9, 24). To survive and cause infection, this beta-hemolytic pathogen metabolizes the heme that it sequesters from hemoglobin and other host hemoproteins. Examination of the siaFGH mutant in a murine model of mucosal colonization revealed that the transporter is critical for GAS establishment in the vaginal tract in mice; the mutant colonized only one mouse (out of nine), compared to the eight animals that were colonized by the wild-type strain (day 3; P < 0.005) (Fig. 6). The strong activation of the MtsR regulon in this infection model (19) suggests that the transporter plays a key role in GAS adaptation to the vaginal mucosa and indicates that highly restricted iron availability confronts bacteria attempting to colonize this niche. Notably, the siaFGH mutant is also attenuated in a murine model of systemic GAS infection (Fig. 7). The attenuation of virulence by the mutant strain is less dramatic than the impact of the mutation in the model of colonization. This tempered reduction in virulence implies that iron availability might be less limiting during systemic growth (than at the mucosal surfaces).

This study identifies the SiaFGH system as a novel transporter that imports heme with a high affinity and as a new virulence factor in GAS. Homologs of this new heme transporter are found in the chromosomes of additional bacteria. For example, in addition to GAS, the related bacteria S. dysgalactiae and S. equi harbor the entire sia operon (including the siaFGH genes). We also identified homologs of siaFGH next to siaDE-like genes in Streptococcus gordonii and Eggerthella lenta or near iron-dependent repressors in group B streptococcus and Enterococcus faecalis. Finally, while this article was in preparation, a study describing a heme ECF system from Lactobacillus sakei was posted in bioRxiv (57). In silico analyses also suggest that ECF-type transporters mediate siderophore import. Zoe Heather et al. described a novel nonribosomal peptide synthetase (NRPS) system in S. equi that produces the siderophore equibactin (42). The eqbHIJ genes in this gene cluster, which are predicted to import an iron-equibactin complex, also belong to the ECF family (NCBI protein accession numbers CAW93973 for EqbH/EcfS, CAW93972 for EqbI/EcfT, and CAW93970 for EqbJ/EcfA). Therefore, transporters from the new ECF family may import other types of iron complexes in addition to heme.

In summary, multiple bacteria, including dangerous pathogens, carry a SiaFGH-type heme transporter. Hence, describing the mechanism of this new import system and its role in bacterial pathophysiology may facilitate future methods for prevention and treatment with broad applicability.

MATERIALS AND METHODS

Strains, media, and growth conditions.

Escherichia coli cells were used for cloning and grown aerobically in Luria-Bertani (LB) medium at 37°C with shaking. GAS cells were grown statically at 37°C in Todd-Hewitt broth (Difco Laboratories) with 0.2% (wt/vol) yeast extract (THYB) or in chemically defined medium (CDM; SAFC Biosciences) (24). When necessary, 100 μg/ml ampicillin, 100 μg/ml spectinomycin, 70 or 300 μg/ml kanamycin (for E. coli and GAS, respectively), or 500 μg/ml erythromycin was added to the medium. All of the constructed chromosomal mutations are stable. Hence, antibiotics were used for strain construction and only when the strains were recovered from glycerol stocks.

Nucleic acid methods.

We used the PureLink genomic DNA minikit (Invitrogen) to extract chromosomal DNA and the Wizard Plus Minipreps DNA purification system (Promega) for plasmid DNA. DNA was amplified by PCR using the AccuTaq LA DNA polymerase (Sigma). DNA fragments were purified from agarose with the S.N.A.P. UV-free gel purification kit (Invitrogen). T4 DNA ligase (Roche) was used for ligation reactions. Restriction enzymes and DNA-modifying enzymes were purchased from NEB and used according to the manufacturer’s recommendations. Transformation and all molecular and genetic manipulations were performed according to the manufacturer’s instructions and standard protocols (43, 44).

Plasmid construction.

The plasmids and strains used in this study are listed in Table 2, and the primers are listed in Table 3. For plasmid pHNG7, a DNA fragment with the siaFGH genes and the ∼1-kb flanking region was amplified from the NZ131 chromosome with the primer pair ZE480/ZE481 and fused to pCR-XL-TOPO using the TOPO TA cloning kit (Invitrogen) (45). For plasmid pHNG10, a DNA fragment generated by inverse PCR from pHNG7 was cut with NheI, ligated into a PCR fragment encoding the aad9 gene (spectinomycin resistance), amplified from pJRS525 with the primer pair ZE408/ZE409, and digested with NheI. For plasmid pHNG12, a DNA fragment with the ΔsiaFGH::aad9 allele was amplified from pHNG10 with primer pair ZE451/ZE452, cut with ClaI, and ligated into pJRS700 (24, 32) linearized with ClaI. For plasmid pNC111, the siaFGH genes were amplified from the NZ131 chromosome using the primer pair ZE836/ZE837, digested with PstI, and ligated into pMSP3535 (46) cut with PstI. Plasmid-harboring GAS cells grown with antibiotics were used to prepare glycerol stocks. Since the plasmids were stably maintained in GAS cells grown overnight without antibiotics, we included antibiotics in the growth medium only when GAS cells were recovered from glycerol stocks but not during the experiments.

TABLE 2.

Strains and plasmids used in this study

Strain or plasmid	Characteristic(s)	Source and/or reference
Strains
Streptococcus pyogenes
NZ131	M49 serotype	Lab stock
ZE4950	NZ131 derivative with the ΔsiaFGH::aad9 mutation	This study
ZE4950/pNC111	ZE4950 with pNC111 expressing the siaFGH genes	This study
ZE4950/pMSP3535	ZE4950 with pMSP3535 (empty vector)	This study
MGAS5005	M1 serotype	Lab stock
ZE151	MGAS5005 derivative with the ΔsiaFGH::aad9 mutation	This study
ZE152	Wild-type rescue of MGAS151	This study
E. coli
JM109	endA1 glnV44 thi-1 relA1 gyrA96 recA1 mcrB+ Δ(lac-proAB) e14− [F′ traD36 proAB+ lacIq lacZΔM15] hsdR17(rK− mK+)	Lab stock
DH5α	F– endA1 glnV44 thi-1 recA1 relA1 gyrA96 deoR nupG purB20 ϕ80dlacZΔM15 Δ(lacZYA-argF)U169 hsdR17(rK− mK+) λ–	Lab stock
TOP10	F– endA1 recA1 mcrA Δ(mrr-hsdRMS-mcrBC) ϕ80lacZΔM15 ΔlacX74 araD139 galU galK Δ(ara-leu)7697 rpsL (Strr) nupG	Invitrogen
Plasmids
pCR-XL TOPO	TOPO cloning vector; Kanr	Invitrogen
pHNG7	pCR-XL TOPO derivative with the siaFGH region; Kanr	This study
pHNG12	pHNG12 derivative with the ΔsiaFGH::aad9 allele; Kanr	This study
pMSP3535	pAMβ1 (from pIL252), ColE1 replicon; Ermr nisRK PnisA	46
pNC111	pMSP3535 with the siaFGH genes under the control of PnisA	This study
pJRS525	Broad-host-range vector; Specr	Lab stock; 55
pJRS700	pVE6037 derivative; Kanr Tms	Lab stock; 56

TABLE 3.

Primers used in this study

Primer	Sequence (5′–3′)	Restriction site
ZE389	CCTATTTGTACAGCAATATTGTCTGCAGG	NheI
ZE393	AAAACTGCAGGCGCTTGCTTATACTCTG	NheI
ZE408 (Spec)	TTTGCTAGCGGTCGATTTTCGTTCGTGAATACATG
ZE409 (Spec)	GGGGCTAGCCGAAAGTCTATGCAAGGGTTTATTG
ZE451	GGGAAATCGATACAGCAATATTGTCTGCAGG	ClaI
ZE452	CAAATGCGCATCGATTTCAAAAGCTG	ClaI
ZE480 (siaFGH)	ATGATAACAGGCGCATTTGC
ZE481 (siaFGH)	CACTACTTAGAAGTTCTTCATCATGTG
ZE691 (siaF)	CACCAAACAGTTAACAATAAAAGATATT
ZE692 (siaF)	TCATGATAACAATCCTGATT
ZE836 (siaFGH)	AAAGGATCCTTCCCTAAAAGAGGTG	BamHI
ZE837 (siaFGH)	CCCTGCAGCTATTATACAAGAGTTCC	PstI
SpecFw	GTGAGGAGGATATATTTGAATACATACGAA
SpecRev	GTCCATTCAATATTCTCTCCAAGATAACTA

Construction of ΔsiaFGH mutant and complemented strains.

GAS mutants were constructed by homologous recombination using a temperature-sensitive shuttle vector and established protocols. For ZE4950 (ΔsiaFGH::aad9 in NZ131), mutant construction involved the following steps: (i) introduction and propagation of pHNG12 in NZ131 under permissive conditions (i.e., growth at 30°C with kanamycin), (ii) selection for GAS clones with chromosomally integrated pHNG12 on spectinomycin at 37°C, and (iii) a screen for clones with a second homologous recombination using passages in broth at 30°C followed by plating on spectinomycin at 37°C. Mutants were identified by replica plating. The mutations in clones with the spectinomycin-resistant and kanamycin-sensitive phenotype were confirmed by PCR analysis.

Plasmids pNC111 (complementation) and pMSP3535 (vector) were introduced into the ΔsiaFGH mutant (ZE4950) for complementation studies.

For ZE151 (ΔsiaFGH::aad9 in MGAS5005) and ZE152 (wild-type rescue), plasmid pHNG12 was introduced into MGAS5005, and the mutant was selected as described above for ZE4950. To isolate the wild-type rescue strain, GAS was plated at 37°C without antibiotics following the passages at 30°C (the third step described above), and cells that were sensitive to both kanamycin and spectinomycin were identified by replica plating. The genotypes of the mutant and wild-type rescue strains were confirmed by PCR analysis. All of the constructed chromosomal mutations are stable. Hence, antibiotics were used for strain construction and only when the strains were recovered from glycerol stocks.

Growth assays for streptonigrin sensitivity and use of hemoglobin iron.

For streptonigrin sensitivity assays, streptonigrin (Sigma) stock solutions (2 mg/ml) were prepared in chloroform and methanol at a 1:1 (vol/vol) ratio and stored at −20°C. In each experiment, fresh THYB containing 0 to 5 μM streptonigrin was inoculated with cultures grown overnight (starting optical density at 600 nm [OD₆₀₀] of 0.01). The culture optical density was determined following a 20-h incubation at 37°C. For assays of GAS use of hemoglobin iron, a new 1 M 2,2′-dipyridyl (DP; Acros Organics) stock solution prepared in absolute ethanol and 1 mM human hemoglobin (Sigma) prepared in saline (and filter sterilized) were used in all experiments. In each experiment, fresh THYB containing 3 mM DP and hemoglobin (0 to 20 μM) was inoculated with GAS cultures grown overnight (starting OD₆₀₀ of 0.01). The culture optical density was determined following a 20-h incubation at 37°C.

Iron uptake assays.

Iron uptake assays were performed as previously described, with a small modification (24). Fresh CDM prepared without iron (and thus containing only trace iron levels) was inoculated with GAS cells, and the culture was allowed to grow to the mid-exponential phase (∼35 Klett units) at 37°C. A total of 1.3 μM ⁵⁵FeCl₃ (PerkinElmer) (specific activity, 18.55 mCi/mg; concentration, 38.80 mCi/ml) prepared in 1 mM sodium ascorbate (ferrous uptake) or 0.0075 M HCl (ferric uptake) was added to a 1.4-ml culture. Samples (200 μl) were drawn every 30 min and washed twice with 500 μl of CDM with 2 mM DP. Radioactivity (counts per minute) was measured for 5 min using a ³H standard with a Beckman LS6500 scintillation counter. The sample optical density was also determined using a Beckman DU730 UV-visible (UV-Vis) spectrophotometer. ⁵⁵Fe incorporation was standardized by the cell quantity by dividing the cpm in the cell pellet by the culture OD₆₀₀.

Inductively coupled plasma mass spectrometry analysis.

For inductively coupled plasma mass spectrometry (ICP-MS) analysis, fresh THYB containing 0.5 μg/ml nisin and 80 μM FeCl₃ (Fisher Scientific) was inoculated with GAS from a culture grown overnight (starting OD₆₀₀ of 0.01) and incubated at 37°C for 20 h. Culture samples (5 ml; OD₆₀₀ of 0.8) were washed three times with phosphate-buffered saline (PBS) prior to collection. The cell pellet was digested and analyzed (Center for Applied Isotope Studies, University of Georgia, Athens, GA) as described previously (47, 48).

Determination of cellular heme content and accumulation.

Fresh THYB containing 20 μM hemoglobin with or without 0.5 μg/ml nisin was inoculated with GAS cultures grown overnight (initial OD₆₀₀ of 0.01) and incubated at 37°C for 20 h. Culture samples (standardized according to the cell density) were collected, washed five times with PBS, resuspended in 2 ml of dimethyl sulfoxide (DMSO), and subjected to sonication (20% amplitude for 30 s). The amount of cellular heme was determined using acidified chloroform extraction as described previously (34, 35). Briefly, 2 ml of 50 mM glycine buffer (pH 2.0), 0.1 ml of 4 N HCl (pH 2.0), 0.2 ml of 5 M NaCl (pH 2.0), and 2 ml of chloroform were added to the experimental samples and standard heme solutions and mixed vigorously by vortexing. The reaction mixtures were incubated at room temperature for 1 min prior to centrifugation (20,000 × g for 20 min at 4°C). The absorbances of the organic phase at 388, 450, and 330 nm were fed into the correction equation Ac = 2 × A₃₈₈ − (A₄₅₀ + A₃₃₀). The heme content was estimated from the plot of Ac corrected for standard heme concentrations. For heme accumulation assays, cells were grown in THYB supplemented with 20 μM hemoglobin. Nisin (0.5 μg/ml) was added to the culture at the early logarithmic phase (20 to 30 Klett units) to induce siaFGH expression. Samples (standardized according to the cell density) were collected at the 0-, 1-, 2-, and 3-h time points, and the cellular heme content was determined as described above.

Mouse model of vaginal colonization.

Eight- to ten-week-old female outbred CD-1 mice were acclimated and randomly distributed into experimental groups. Experiments were performed as described previously (19, 49–52). Briefly, mice were estrus synchronized by an intraperitoneal administration of 0.5 mg of β-estradiol valerate (Acros Organics) suspended in 100 μl of filter-sterilized sesame oil (Sigma) 24 h prior to inoculation. Strains NZ131 and ZE4950 were grown to an OD₆₀₀ of 0.3 to 0.5 in THYB and concentrated to 10⁹ CFU/ml in PBS. Mice were intravaginally inoculated with 10⁷ CFU of the GAS culture in 10 μl of PBS. On days 1, 2, 3, and 5 postinoculation, the vaginal lumen was gently washed with 50 μl PBS. The bacteria in the resulting vaginal lavage fluid were enumerated by viable counts using CHROMagar StrepB selective plates.

Mouse model of systemic GAS infection.

CD-1 mice (weight, 20 to 22 g; Charles River Laboratories) were acclimated and randomly distributed into experimental groups. Cultures of ZE151 (ΔsiaFGH::aad9 in MGAS5005) and ZE152 (wild-type rescue) grown in THYB were harvested at the mid-logarithmic phase (OD₆₀₀ of 0.7), washed, and resuspended in saline. Mice were infected intraperitoneally with 0.1 ml of the cell suspension at 1.4 × 10⁸ CFU. The animals were observed 4 times per day after challenge, and mice exhibiting signs of severe distress were euthanized and counted as dead.

In silico methods.

The SiaF ribbon structure was predicted using I-TASSER (https://zhanglab.ccmb.med.umich.edu/I-TASSER/) (53) and visualized using PyMOL molecular visualization software. The FASTA sequence of each protein was retrieved from the NCBI database and then analyzed with the BLASTP server (https://blast.ncbi.nlm.nih.gov/) for the comparative analysis.

Ethics statement.

All mouse experimentation was conducted according to a protocol approved by the Institutional Animal Care and Use Committees of Georgia State University and Binghamton University.