The surface protein Shr of Streptococcus pyogenes binds heme and transfers it to the streptococcal heme-binding protein Shp.(Research article)

Authors: Hui Zhu [1]; Mengyao Liu [1]; Benfang Lei (corresponding author) [1]

Background

Specific ATP-binding cassette (ABC) type transporters are involved in acquisition of essential iron in various forms in bacterial pathogens. The substrate-binding component of the ABC transporters specifically binds free Fe

3+ , ferric siderophore complex, or heme, which is transported across the cytoplasmic membrane by the permease component using the energy from the hydrolysis of ATP catalyzed by the ATPase component [1]. The human pathogen Streptococcus pyogenes produces three ABC transporters, FtsABCD [2], HtsABC or SiaABC [3, 4], and MtsABC [5], which acquire ferric ferrichrome, heme, and Fe3+ and Mn2+ , respectively. Expression of htsABC and mtsABC are negatively regulated by the same metalloregulator MtsR in response to levels of Fe only and Fe or Mn, respectively [6, 7].

Heme is abundant in mammalian hosts and a preferred iron source for bacterial pathogens [8, 9, 10, 11]. Besides heme-specific ABC transporters, cell surface heme-binding proteins are required for heme acquisition in Gram-positive pathogens. These proteins have been identified in

S. pyogenes [12], Staphylococcus aureus [13], and Streptococcus equi [14]. Heme is usually bound to host proteins at extremely high affinities [15, 16]. Furthermore, Gram-positive bacteria have thick cell walls, and heme-specific ABC transporters are most likely buried in the cell wall. Thus, heme acquisition machinery in Gram-positive bacteria may have to have evolved mechanisms to overcome these obstacles in heme acquisition, i.e. inability of host hemoproteins to reach the ABC transporters and the extremely high affinity of host proteins for heme. The cell surface heme-binding proteins are believed to have evolved to overcome these obstacles in heme acquisition in Gram-positive pathogens. However, how these proteins are involved in heme acquisition is largely unknown.S. pyogenes is capable of utilizing heme derived from human hemoproteins as a source of iron [9, 17]. The heme acquisition machinery in S. pyogenes is believed to consist of Shr, Shp, and HtsABC. Shp and HtsABC are the cell surface heme-binding protein and heme-specific ABC transporter, respectively [4, 12]. Shr, another cell surface protein, is proposed to be a receptor of host hemoproteins [3]; however, its function has not been firmly established. We have been studying the S. pyogenes heme acquisition machinery as a model system for understanding the heme acquisition process in Gram-positive pathogens. Shp rapidly and directly transfers its heme to HtsA, the lipoprotein component of HtsABC, in a concerted two-step process with one kinetic phase [18, 19]. The structure of the Shp heme-binding domain reveals that the Shp heme iron is ligated to Met66 and Met153 [20], and the Met axial ligands are both important for rapid heme transfer from Shp to HtsA [21]. However, the heme source of Shp is not known.

We have proposed that Shp functions to relay heme from host proteins or another

S. pyogenes heme-binding protein to HtsABC [18]. In this study, we found that the rates of ferric heme transfer from oxidized hemoglobin (methemoglobin) to heme-free Shp (apoShp) are similar to those for the dissociation of ferric heme from methemoglobin, suggesting that Shp cannot directly acquire heme from methemoglobin and that Shp may mainly acquire heme from another S. pyogenes protein. We hypothesize that this protein is Shr. To test this hypothesis, recombinant Shr was prepared and characterized. We found that Shr indeed binds heme and, more interestingly, efficiently transfers it to Shp but inefficiently to HtsA. These findings suggest the possibility that Shr is the heme source of Shp and that heme transfer from Shr to Shp is part of the heme acquisition process in S. pyogenes .

Results

Shp appears to indirectly acquire ferric heme from methemoglobin

Heme-binding form of Shp (holoShp) was formed in the incubation of apoShp with human methemoglobin [18]. It was not known whether the formation of holoShp was due to direct heme transfer from methemoglobin or indirect scavenge of ferric heme dissociated from methemoglobin. To examine this issue, the rate for formation of holoShp was compared with that of heme dissociation from methemoglobin. Limited methemoglobin (total 5.8 [mu]M heme) was incubated with 7 to 56 [mu]M apoShp at 25[degrees]C, and the formation of holoShp was monitored by absorbance change at 405 and 425 nm, which present the loss and gain of heme by methemoglobin and apoShp, respectively. The time courses of normalized absorbance change, [DELTA](A

425 -A405 ), fit a two exponential equation (Fig. 1A), but not a single exponential equation (Fig. 1B). Fitting of the data to a two exponential equation resulted in two first-order rate constants. The values of the rate constants did not change significantly with [apoShp] from 7 to 56 [mu]M, giving the mean values [+ -] standard deviation of 0.0065 [+ -] 0.0013 and 0.00041 [+ -] 0.00012 s-1 for the rate constants of the fast and slow phases, respectively. The biphasic kinetics and rate constants are similar to those in the dissociation of heme from methemoglobin using H64Y/V68F apomyoglobin as a heme scavenger [22], in which the fast and slow phases are heme dissociation from the [beta] and [alpha] subunits, respectively [22, 23]. To confirm these similarities, the dissociation of heme from methemoglobin using …