INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

The ability of a pathogen to colonize and proliferate within a particular environmental niche in the host is essential for the initiation of an infection. Growth depends, in part, on the ability of a pathogen to scavenge essential nutrients including iron, which plays a crucial role in both the establishment of a pathogen and the progression of disease (21). Within the human host the majority of iron is found intracellularly in the form of hemoglobin, heme proteins, or ferritin (21). Following intracellular release, heme and hemoglobin are bound by serum proteins hemopexin and haptoglobin, respectively. Small quantities of extracellular iron are also complexed to iron-binding proteins, primarily transferrin, which is found in serum, and lactoferrin, which is present within mucosal surfaces.

To survive in the iron-limited environment of the host, microorganisms have developed diverse and elaborate systems to obtain this element, which is needed for growth. These include the production of low-molecular-weight iron-chelating compounds (siderophores) and iron-regulated outer membrane receptors that function to bind iron-containing compounds directly. A new class of heme- and hemoglobin-binding proteins (hemophores) recently described appears to function similarly to siderophores for the capture of heme (3, 9, 22, 29-31, 38). An extracellular heme- and hemoglobin-binding protein (Hbp) in a human pathogenic strain of Escherichia coli has also recently been described and has been proposed to function as a shuttle protein of a hemophore-dependent hemin acquisition system (43). In addition to binding hemoglobin, the E. coli Hbp can degrade hemoglobin and subsequently binds the released hemin.

Porphyromonas gingivalis, the etiological agent of adult periodontitis, requires iron for growth (5, 12, 18); however, the specific mechanisms utilized for the acquisition of iron are poorly understood. Hemoglobin bound to haptoglobin and hemin complexed to hemopexin can be used as iron sources by P. gingivalis (5), indicating that this bacterium has a mechanism for removing the hemin from these host iron-binding proteins. Several reports have described P. gingivalis iron-regulated outer membrane proteins, which could function to bind heme compounds directly (2, 6, 57, 58); however, conclusive evidence for the role of these proteins in heme binding and uptake has not been presented.

We recently identified a hemin and hemoglobin receptor in P. gingivalis strain A7436 (HmuR), which exhibits a high degree of homology to TonB-dependent outer membrane hemin and hemoglobin receptors from several different gram-negative bacteria (56). A P. gingivalis hmuR mutant exhibited diminished ability to bind hemoglobin and to grow with either this protein or hemin as the sole iron source. We also demonstrated that the recombinant HmuR protein bound hemoglobin, hemin, and several different metalloporphyrins in vitro (42, 56). Recombinant HmuR binds hemopexin and serum albumin complexed with hemin and haptoglobin complexed with hemoglobin, but with affinity lower than that for hemoglobin alone (42). Recent studies have described three P. gingivalis genes (hemR, tlr, and ihtA) which have homology to genes encoding TonB-dependent receptors (13, 26, 57). The role of the protein products of the hemR, tlr, and ihtA genes in binding and utilization of heme or iron from different serum proteins in P. gingivalis has not been delineated.

In addition to requiring outer membrane receptors, utilization of hemin from hemoglobin in P. gingivalis requires participation of the P. gingivalis cysteine proteases, collectively referred to as gingipains (11, 25, 47, 55). These bacterial proteases exhibit activity against a wide range of host proteins including immunoglobulins, extracellular matrix proteins, bactericidal proteins, collagen, fibronectin, fibrinogen, tumor necrosis factor, interleukin-8, and proteins involved in the complement, coagulation, and kallikrein/kinin cascades (20, 28, 33, 50). The arginine-specific gingipains (HRgpA and RgpB) are encoded by genes rgpA and rgpB, respectively, while a lysine-specific gingipain (Kgp) is encoded by gene kgp (1, 4, 10, 39, 40, 44, 45). The translated portions of the rgpA and kgp genes encode prepropeptide, catalytic, and hemagglutinin domains, and the initial polyprotein is subject to posttranslational processing by Kgp itself. RgpB is expressed as a prepropeptide missing the majority of the hemagglutinin domain but is otherwise closely related to the rgpA gene product (34, 36). Kgp, HRgpA, and RgpB can be found both associated with the outer membrane and in a soluble form.

Recent reports have documented the ability of the Kgp complex to bind hemoglobin, hemin, porphyrins, and metalloporphyrins (14, 17, 27, 37, 41, 42). Characterization of defined P. gingivalis mutants also supports a role for Kgp in hemin and hemoglobin binding and utilization (19, 41, 53, 55); W. Simpson and C. A. Genco, unpublished data). Furthermore, a recent report has documented that Kgp can cleave soluble hemoglobin (32). Collectively these studies indicate that Kgp may function to both bind and degrade hemoglobin, ultimately releasing heme, which could be utilized for growth. In this study we demonstrate that, in addition to its ability to bind and degrade hemoglobin, purified Kgp can degrade soluble haptoglobin, hemopexin, and transferrin present in normal human serum. Degradation of these serum proteins was observed during growth of P. gingivalis in a minimal medium supplemented with human serum as a sole source of heme, and cleavage of these proteins coincided with an increase in lysine- and arginine-specific cysteine proteinase activity in bacterial cultures.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Bacterial strains and growth conditions. P. gingivalis strain A7436 was used in these studies and was maintained on anaerobic blood agar plates (Remel, Lenexa, Kans.). All P. gingivalis cultures were incubated at 37°C in an anaerobic chamber (Coy Laboratory Products, Inc., Ann Arbor, Mich.) with 85% N₂, 5% H₂, and 10% CO₂ for 3 to 5 days. To examine the ability of P. gingivalis to degrade serum proteins, strain A7436 was first grown on anaerobe broth MIC (Difco, Detroit, Mich.) at 37°C for 24 h and then inoculated into a basal medium (BM; 10 g of Trypticase peptone, 0.2 g of tryptophan, 2.5 g of NaCl, 0.1 g of sodium sulfite, and 0.4 g of cysteine per liter). After another 24 h of cultivation under anaerobic conditions, the culture was inoculated into BM or BM supplemented with 10% normal human serum (Sigma, St. Louis, Mo.).

Purification of gingipains. Kgp, HRgpA, and RgpB were purified from P. gingivalis strain HG66 cultures as previously described (46, 49). Approximately 5 mg of each gingipain from 1 liter of bacterial culture was purified to homogeneity as determined by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), and the concentration of active gingipain in each batch was determined by active-site titration using specific inhibitors Z-Phe-Lys-2,4,6-trimethylbenzoyloxymethylketone and H-D-Phe-Phe-Arg-chloromethylketone (FFRck) (Bachem Biosciences Inc., King of Prussia, Pa.) for gingipain K and gingipains R, respectively (48).

Degradation of hemoglobin by purified gingipains. Samples of hemoglobin (0.5 mg/ml, 7.5 µM) were incubated at 37°C with gingipains at 50 to 200 nM concentrations in 200 mM HEPES-150 mM NaCl-1 mM CaCl₂, pH 7.6, supplemented with 10 mM cysteine. At specific times the reaction was stopped by the addition of FFRck (1.0 µM), and the mixture was subjected to SDS-PAGE analysis (52). SDS-PAGE gels were also subjected to laser densitometry, and hemoglobin degradation was monitored as the relative optical densities of the hemoglobin bands. The rate of hemoglobin degradation was also measured as a decrease of absorbance at 415 nm (the Soret band) during incubation of hemoglobin with gingipains. For these studies 7.5 µM hemoglobin and 100 to 200 nM gingipains were used. Degradation of hemoglobin by gingipains in human serum was also examined. For these studies hemoglobin (0.5 mg/ml) dissolved in 20 mM Tris-150 mM NaCl, pH 7.4, was added to human serum (0.5 mg/ml, final concentration) and the serum was incubated with Kgp (50 nM) in the presence of 10 mM cysteine at 37°C for 3 or 7 h. After SDS-PAGE, proteins were electrotransferred onto a nitrocellulose membrane and Western blot analysis was performed using antihemoglobin serum (see below).

Degradation of haptoglobin and haptoglobin-hemoglobin complexes by purified gingipains. Purified haptoglobin (1 mg/ml) was incubated with RgpB (90 nM), HRgpA (90 nM), or Kgp (30 nM) for various times and examined by SDS-PAGE analysis as described above. For some studies purified haptoglobin (1 mg/ml) was preincubated with hemoglobin (2 mg/ml) and then incubated with the gingipains.

Degradation of hemoglobin, haptoglobin, hemopexin, and transferrin in normal human serum by gingipains. Samples of human serum obtained from healthy volunteers (diluted fivefold) were incubated at 37°C with gingipains (RgpB, HRgpA, and Kgp) in the presence of 10 mM cysteine. Concentrations of gingipains for these studies ranged from 10 to 1,000 nM. At specific times, aliquots were withdrawn, the reaction was stopped by addition of FFRck (inhibits all three enzymes; 100 nM), and the mixture was subjected to SDS-PAGE analysis. Protein bands were visualized by Coomassie brilliant blue R-250 staining and by Western blot analysis (60) using antihemoglobin, antihaptoglobin, antihemopexin, and antitransferrin serum. Rabbit anti-human hemoglobin serum was purchased from DAKO (Glostrup, Denmark), goat anti-human haptoglobin serum and goat anti-human transferrin serum were purchased from Biomeda Co. (Foster City, Calif.), and rabbit anti-human hemopexin serum was purchased from Cortex Biochem (San Leadro, Calif.). Antihemoglobin serum was used at a 1:2,000 dilution, and antihaptoglobin, antihemopexin, and antitransferrin sera were used at 1:5,000 dilution. The appropriate secondary antibodies conjugated to alkaline phosphatase (Sigma) were added, and the blot was developed with a kit obtained from Bio-Rad according to manufacturer's protocol. For some studies hemoglobin (50 µg/ml to 10 mg/ml) or hemin (0.1 mg/ml) was added to serum samples and samples were incubated for 1 h at 37°C prior to the addition of gingipains.

Degradation of hemoglobin, haptoglobin, hemopexin, and transferrin in normal human serum during bacterial growth. P. gingivalis A7436 was cultivated in BM or BM supplemented with 10% normal human serum (Sigma) as previously described (19). At specific times, samples were withdrawn, gingipains were inhibited by addition of FFRck (100 nM final concentration), and after centrifugation (10,000 × g, 10 min) the resulting supernatant was subjected to SDS-PAGE analysis. Protein bands corresponding to hemoglobin, haptoglobin, hemopexin, and transferrin were visualized by Western blot analysis as described above.

Lysine- and arginine-specific activities. The amidolytic activities of P. gingivalis A7436 in whole cultures and in cell-free supernatant fractions were determined with either N-benzoyl-L-arginine-p-nitroanilide or Z-lysine-p-nitroanilide (Z-Lys-pNA). Samples were preincubated in 200 mM Tris-HCl-100 mM NaCl-5 mM CaCl₂ (pH 7.6), supplemented with 10 mM cysteine, for 5 min at 37°C and assayed for amidase activity with 0.5 mM substrate. The formation of p-nitroanilide was monitored spectrophotometrically at 405 nm.

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

Kgp can degrade hemoglobin. To investigate and compare the functions of P. gingivalis proteolytic enzymes in hemoglobin degradation and heme release, we incubated hemoglobin with active-site-titrated Kgp, HRgpA, or RgpB. The purity of the three enzymes used in this study was confirmed by SDS-PAGE analysis (Fig. 1). Following incubation of each purified gingipain with hemoglobin, we monitored the rate of hemoglobin degradation by SDS-PAGE analysis. Following a 3-h incubation of hemoglobin with Kgp we observed almost complete digestion of this protein (Fig. 2A, lane e). In contrast, at a 100 nM gingipain concentration hemoglobin was refractory to degradation by HRgpA and RgpB (Fig. 2A, lanes c and d). Degradation of the hemoglobin chains by Kgp occurred in a time-dependent manner and without accumulation of any discrete cleavage products (Fig. 2A, lanes h to n). Hemoglobin was completely degraded by Kgp following incubation for 360 min (Fig. 2A, lane n). For samples of hemoglobin incubated with Kgp, the decrease of the absorbance correlated with the disappearance of the globin chains as determined by laser densitometry analysis of SDS-PAGE gels (data not shown).

View larger version (80K): [in this window] [in a new window]   FIG. 1.   Purity of gingipains. RgpB (lane 1), HRgpA (lane 2), Kgp (lane 3) were purified from P. gingivalis strain HG66. Samples were boiled in reducing sample treatment buffer and analyzed by SDS-PAGE. Molecular weights are indicated on the left.

View larger version (15K): [in this window] [in a new window]   FIG. 2.   Degradation of hemoglobin. (A) Samples of hemoglobin (0.5 mg/ml, 7.5 µM) were incubated with gingipains RgpB, HRgpA, and Kgp (100 nM) at 37°C in the presence of 10 mM cysteine. At specific times, aliquots were withdrawn and treated with FFRck (100 nM, final concentration) to terminate the reaction. Samples were boiled in reducing sample treatment buffer and analyzed by SDS-PAGE. Lane a, molecular mass standards; lanes b and g, hemoglobin control incubated without gingipains; lanes c to e, hemoglobin incubated for 3 h with HRgpA, RgpB, and Kgp, respectively; lane f, Kgp alone; lanes h to n, hemoglobin incubated with Kgp for 30, 60, 90, 120, 150, 180, and 360 min, respectively. (B) Samples of hemoglobin (7.5 µM) were incubated with Kgp (100 [solid circles] or 200 nM [open circles]) as described above, and the optical density at 415 nm (OD415nm) was recorded every 10 min. Triangles, control samples containing all reagents except Kgp.

1

1

1

Degradation of haptoglobin and the haptoglobin-hemoglobin complex by gingipains. Under physiological conditions, hemoglobin released from erythrocytes is tightly bound to haptoglobin with a dissociation constant greater than 10¹⁵ M (15, 21). Interestingly, the P. gingivalis hemoglobin receptor HmuR binds with very low affinity to haptoglobin complexed to hemoglobin and does not bind apo-haptoglobin (42). To utilize hemoglobin in vivo, P. gingivalis should thus also recognize and degrade haptoglobin and/or the hemoglobin-haptoglobin complex, releasing hemoglobin. Therefore, we examined the ability of gingipains to degrade haptoglobin. Three major phenotypic forms of human haptoglobin, designated Hp 1-1, Hp 2-2, and Hp 2-1 have been described; Hp 2-2 and 2-1 are polymerized forms of higher molecular mass (15). In vitro, Kgp (30 nM) efficiently cleaved the invariant heavy -chain and both light ¹- and ²-chains with purified Hp 1-2 (Fig. 3) and Hp 1-1 and Hp 2-2 (data not shown) phenotypes. This degradation was initially observed following a 15-min incubation period of haptoglobin with Kgp and was virtually complete by 60 min (Fig. 3). In contrast, HRgpA and RgpB (90 nM) degraded only the -chains of haptoglobin, as first observed following a 60-min incubation (Fig. 3).

View larger version (60K): [in this window] [in a new window]   FIG. 3.   Degradation of soluble haptoglobin by gingipains. Haptoglobin (1 mg/ml) was incubated with RgpB (90 nM; lanes c to e), HRgpA (90 nM; lanes f to h), and Kgp (30 nM; lanes i to k) for 15 (lanes c, f, and i), 60 (lanes d, g, and j), and 180 min (lanes e, h, and k) in 20 mM HEPES-150 mM NaCl-1 mM CaCl2-10 mM cysteine, pH 7.6. The reaction was terminated with FFRck (10 µM, final concentration); samples were boiled in reducing treatment buffer and analyzed by SDS-PAGE. Haptoglobin incubated without gingipains for 180 min was loaded in lane b. Lane a, molecular mass standards.

View larger version (19K): [in this window] [in a new window]   FIG. 4.   Degradation of haptoglobin and the haptoglobin-hemoglobin complex in normal human serum. Human serum (lanes a to d) and serum saturated with hemoglobin (10 mg/ml; 1 h at 37°C) (lanes e to h) were incubated with HRgpA (90 nM) (A), RgpB (90 nM) (B), and Kgp (30 nM) (C) in the presence of 10 mM cysteine for 3 (lanes b and f), 7 (lanes c and g), and 15 h (lanes d and h) at 37°C. Lanes a and e, control serum samples incubated overnight under the same conditions but without gingipains. Following SDS-PAGE, Western blot analysis was performed using antihaptoglobin serum. Molecular mass standards are shown on the left. Arrowheads, - and -chains of haptoglobin.

View larger version (24K): [in this window] [in a new window]   FIG. 5.   Degradation of hemoglobin in Tris-buffered saline and in normal human serum. Hemoglobin (0.5 mg/ml) dissolved in 20 mM Tris-150 mM NaCl, pH 7.4 (lanes a to c) or added to human serum (0.5 mg/ml, final concentration) (lanes d to f) was incubated with Kgp (50 nM) in the presence of 10 mM cysteine at 37°C for 3 (lanes b and e) or 7 h (lanes c and f). Lane a, hemoglobin in buffer; lane d, hemoglobin in serum without gingipains. After SDS-PAGE, proteins were electrotransferred onto a nitrocellulose membrane and Western blot analysis was performed using antihemoglobin serum. Molecular mass standards are shown on the left.

Hemopexin degradation by gingipains. Heme is avidly scavenged by the serum heme-binding proteins hemopexin and albumin (21). The affinity of hemopexin for hemin is higher than that of hemoglobin and several orders of magnitude higher than that of albumin (35). Since P. gingivalis can utilize hemin complexed to hemopexin as an iron source (5), it must be able to capture hemin from this host hemin-binding protein. We have established that the P. gingivalis hemoglobin receptor HmuR binds hemin and, with lower affinity, hemopexin complexed with hemin, but not apo-hemopexin (42). Because hemopexin has such a high affinity for hemin, we reasoned that the ability of P. gingivalis to acquire hemin from this protein could involve a mechanism in which the serum protein was degraded by proteinases from this organism. To determine if the gingipains could degrade hemopexin, normal human serum was incubated with Kgp, HRgpA, or RgpB and degradation was monitored over time using Western blot analysis with antihemopexin serum. When Kgp (30 nM) was added to the serum samples, we observed rapid cleavage of hemopexin as indicated by the disappearance of the intact 70-kDa polypeptide and the appearance of degradation products (23 and 47 kDa) immunoreactive with antihemopexin serum (Fig. 6C). Serum hemopexin was also efficiently cleaved by HRgpA and RgpB, yielding two main proteolytic products of 23 and 40 kDa (Fig. 6A and B). However, degradation of hemopexin by HRgpA or RgpB was achieved only at higher concentrations (90 nM) of these enzymes (Fig. 6A and B).

View larger version (26K): [in this window] [in a new window]   FIG. 6.   Hemopexin degradation in human serum by gingipains. Serum (lanes b to d) and serum saturated with hemin (0.1 mg/ml, 30 min, 37°C) (lanes f and g) was incubated with HRgpA (90 nM) (A), RgpB (90 nM) (B), and Kgp (30 nM) (C) at 37°C for 3 (lanes b and f), 7 (lanes c), and 15 h (lanes d and g). Lanes a and e, serum samples incubated without gingipains. After SDS-PAGE, proteins were electrotransferred onto a nitrocellulose membrane and Western blot analysis was performed using antihemopexin serum. Molecular mass standards are shown on the left.

Transferrin degradation. In addition to hemin- and heme-containing proteins, P. gingivalis can utilize nonhemin iron source transferrin for growth (5, 7, 24, 54, 59). However, specific receptors for transferrin in this organism have not been described. To determine if transferrin in normal human serum is susceptible to cleavage by gingipains, we incubated serum with HRgpA, RgpB, or Kgp and monitored the degradation over time. Incubation of human serum with each of the three enzymes resulted in limited proteolysis of transferrin, as detected by Western blot analysis using antitransferrin serum. We found that the patterns of cleavage differed considerably between the arginine-specific gingipains and Kgp. While the addition of HRgpA and RgpB to human serum resulted in only a slight truncation of the transferrin polypeptide chain (Fig. 7, lanes b to g), the addition of Kgp resulted in the cleavage of transferrin into two fragments (30 and 50 kDa) (Fig. 7, lanes h to j).

View larger version (78K): [in this window] [in a new window]   FIG. 7.   Degradation of transferrin in human serum by gingipains. Normal human serum was diluted fivefold and incubated with a 50 nM concentration of RgpB (lanes b to d), HRgpA (lanes e to g), and Kgp (lanes h to j) for 3 (lanes b, e, and h), 7 (lanes c, f, and g), and 15 h (lanes d, g, and j). After SDS-PAGE, proteins were electrotransferred onto a nitrocellulose membrane and Western blot analysis was performed using antitransferrin sera. Lane a, serum sample incubated without gingipain. Molecular mass standards are shown on the left.

P. gingivalis can degrade hemoglobin, haptoglobin, hemopexin, and transferrin in human serum during bacterial growth. To determine if P. gingivalis can cleave hemoglobin, haptoglobin, hemopexin, and transferrin in culture, we grew the organism in a minimal medium with human serum as the sole source of heme and iron. We monitored the degradation of serum proteins and the cell-associated and extracellular arginine- and lysine-specific activity during the growth period. We found that P. gingivalis strain A7436 exhibited a long lag phase in minimal media supplemented with normal human serum; this was followed by a rapid increase in growth (data not shown). The increase in bacterial growth correlated with a two- to threefold increase in both lysine- and arginine-specific proteinase activity (Table 1). The levels of lysine- and arginine-specific proteinase activity produced by P. gingivalis A7436 cultures were comparable to the levels of activity previously reported under similar growth conditions (19).

View larger version (47K): [in this window] [in a new window]   FIG. 8.   Degradation of serum proteins during growth of P. gingivalis. P. gingivalis strain A7436 was grown in minimal media supplemented with 10% normal human serum as a heme source. Samples were removed from cultures grown with serum at the indicated times, supernatants were separated by SDS-PAGE, and the Western blot was developed using antisera to hemoglobin (A), haptoglobin (B), hemopexin (C), and transferrin (D). Results are from one experiment and are representative of three separate experiments.

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

P. gingivalis can utilize several different iron sources for bacterial growth including hemin, hemoglobin, and transferrin (5, 7, 18, 24, 54, 59). Hemoglobin has been reported to be more effective in supporting the growth of P. gingivalis than hemin or transferrin (53, 54). However, under physiological conditions, free hemoglobin or hemin may not be available since these compounds are efficiently scavenged by haptoglobin and hemopexin, respectively (15, 21, 35). P. gingivalis can utilize hemoglobin bound to haptoglobin and hemin bound to hemopexin (5); however, the mechanism(s) by which hemin is released from these proteins has not been defined. One effective way to release iron or hemin from these proteins may involve their proteolytic degradation, and in this study we have explored this possibility. We have demonstrated that soluble gingipains can degrade the host heme iron-binding proteins, hemoglobin, haptoglobin, hemopexin, and transferrin, with Kgp being the most effective. This enzyme at nanomolar concentrations degraded free hemoglobin apparently due to significant catalytic efficiency against this substrate, as reflected by the high value of the k_cat/K_m ratio (3.2 × 10⁶ M¹ min¹). The kinetic value in this range indicates that this reaction can occur in vivo, especially in the absence of any endogenous inhibitors of Kgp (23, 51). In contrast to Kgp, gingipains R possessed weak hemoglobinase activity. Similarly, Kgp was more active than gingipains R in the degradation of haptoglobin, hemopexin, and transferrin in human serum. While arginine-specific gingipains HRgpA and RgpB also cleaved these serum proteins, the concentrations required for effective degradation were 5 to 10 times higher than that required by Kgp.

Our results fully confirm a recent study by Lewis et al. (32) on hemoglobin degradation by Kgp. Our study extends that report by providing kinetic data on hemoglobin degradation by Kgp as well as documenting the proteolysis of other heme iron-binding proteins by gingipains. Interestingly, in our study we did not observe extensive degradation of hemoglobin in human serum by soluble Kgp at the concentrations that we tested. Likewise the -chain of haptoglobin was partially protected from degradation by Kgp in a haptoglobin-hemoglobin complex. However, we observed the time-dependent degradation of all of the serum proteins examined in culture media supplemented with human serum during P. gingivalis growth, which was correlated with an increase in arginine- and lysine-specific proteinase activities. The observation that host heme and iron serum proteins are degraded by P. gingivalis cultures is in agreement with previous studies in which a suspension of P. gingivalis cells were reported to degrade serum proteins (8, 16). Collectively, these observations may result from differences in the enzymatic activity of soluble versus membrane-associated gingipains. It was previously reported that the activities of the gingipains are dependent on the form in which they are purified (34). This is not an unexpected finding considering that, when gingipains are embedded in the membrane, the exposure of the catalytic and hemagglutinin portions would be expected to differ from that for the protein in its soluble form. Alternatively, it is also possible that unidentified P. gingivalis proteinases may be utilized for the cleavage of hemoglobin bound to haptoglobin in serum during bacterial growth.

The results presented in this study and in an accompanying paper (42), together with several other reports (7, 14, 17, 27, 32, 37, 41), have established that Kgp can bind and degrade hemoglobin and other serum heme and iron proteins. Characterization of P. gingivalis kgp mutants further supports a role for Kgp in hemoglobin binding and utilization in P. gingivalis (19, 41, 55). Furthermore, we have recently established that hemoglobin utilization in P. gingivalis A7436 requires both outer membrane receptor HmuR and Kgp (42, 55, 56; Simpson and Genco, unpublished data). We found that a P. gingivalis hmuR kgp mutant did not grow with either hemin or hemoglobin as the sole iron source (Simpson and Genco, unpublished data). We have also shown that HmuR can interact with Kgp (42). Based on these results, we propose that Kgp may facilitate hemoglobin binding and utilization in P. gingivalis by functioning as a bacterial heme-scavenging protein or hemoglobinase.

In E. coli extracellular hemoglobin-binding protein Hbp has been proposed to function as a bacterial hemophore or hemoglobinase (43). Similar to Kgp, E. coli Hbp appears to function as a hemoglobin-degrading proteinase and as a hemin-binding protein. Hbp, similar to P. gingivalis Kgp, is synthesized as a polyprotein precursor that is processed following export to the cell surface. However, it is not known if the E. coli hemoglobinase can degrade hemoglobin when it is complexed to haptoglobin or when it is present in human serum. We know, however, that any Kgp activity in human serum would be fully functional because of the absence of any Kgp-specific inhibitor that might restrict degradation of heme-containing proteins in serum (23). In contrast, gingipain R activity can be, at least partially, limited by ₂-macroglobulin in human serum (23, 51).

The environmental conditions in the periodontal pocket during P. gingivalis infection are not precisely known. However, inflammation and tissue damage can result in an altered pattern of nutrients; when gingival crevicular fluid flow is increased and bleeding occurs, the availability of heme-containing proteins may increase, resulting in the enrichment of P. gingivalis. The ability to acquire hemin from hemoglobin released from erythrocytes appears to be of particular importance for the survival of P. gingivalis in vivo. In this study we have determined that lysine-specific cysteine proteinase Kgp can efficiently degrade soluble hemoglobin, as well as haptoglobin and hemopexin in serum. The proteinase-hemagglutinin complexes of Kgp may thus be important in the uptake of heme, via binding (42) and subsequent degradation of heme-containing proteins. The ability of P. gingivalis to efficiently bind and degrade host heme-containing proteins represents an additional mechanism adapted by this pathogen to ensure its successful growth and proliferation. Whether both soluble and cell-bound Kgps function in hemoglobin binding and degradation remains to be determined.