molecular oral

microbiology

molecular oral microbiology

Identification of amino acid residues involved in

hemin binding in Porphyromonas gingivalis
hemagglutinin 2
Q.B. Yang*, F.Y. Yu*, L. Sun, Q.X. Zhang, M. Lin, X.Y. Geng, X.N. Sun, J.L. Li and Y. Liu
Beijing Institute for Dental Research, Beijing Stomatological Hospital and School of Stomatology, Capital Medical University, Beijing, China

Correspondence: Qiu Bo Yang, Beijing Institute for Dental Research, Beijing Stomatological Hospital and School of Stomatology, Capital
Medical University, Tian Tan Xi Li No. 4, Dong Cheng District, Beijing 100050, China Tel.: +86 10 5709 9307; fax: +86 10 5702 5154;
E-mail: Qiuboyang2003@163.com
*These authors contributed equally to this work.

Keywords: Porphyromonas gingivalis; hemagglutinin 2; heme; binding

Accepted 6 March 2015
DOI: 10.1111/omi.12097

SUMMARY

Porphyromonas gingivalis (P. gingivalis) is a
major etiological agent in the development and
progression of chronic periodontitis. It produces
cysteine proteases (gingipains), including a lysine-
specific gingipain and two arginine-specific gingi-
pains. Heme binding and uptake are fundamental
to the growth and virulence of P. gingivalis. The
recombinant hemagglutinin 2 domain (rHA2) of gin-
gipain binds hemin with high affinity. The aim of
the present work was to identify the key residues
involved in its hemin-binding activity. A functional
rHA2 was expressed and bound to hemin-agarose,
and then digested with endopeptidases. The pep-
tides bound to hemin-agarose were identified by
mass spectrometry and the amino acids were
assessed by mutation and peptide binding inhibi-
tion analysis. The DHYAVMISK sequence was iden-
tified in peptides derived from both Asp-N and Lys-
C endopeptidase digestions of rHA2. A monoclonal
antibody, mAb QB, was produced and its epitope
was associated with the DGFPGDHYAVMISK pep-
tide within the HA2 domain. Hemin was shown to
competitively inhibit the immunoreactivity of rHA2
or the peptide to mAb QB. The peptide DHYAVM-
ISK inhibited hemin-binding activity; although, this
inhibition was not seen when the peptide contained
the H1001E mutation (DEYAVMISK). Based on
these results, we propose that residue His1001 is
involved in the hemin-binding mechanism of the
P. gingivalis rHA2 and the peptide containing this
residue, DHYAVMISK, may be an inhibitor of hemin
binding.
INTRODUCTION
Anaerobic bacterial infections are believed to progres-
sively destroy periodontal tissues, thereby playing a
role in periodontal diseases. A number of studies have
suggested an important role for Porphyromonas gingi-
valis in the etiology of human periodontal disease
(Slots et al., 1986; Bodet et al., 2007). Several poten-
tial virulence factors, including hemin-binding activity,
have been identified for the organism (Holt et al., 1999;
Curtis et al., 2001; Dixon & Darveau, 2005).
P. gingivalis requires exogenous porphyrin and iron
for its growth and virulence (Bramanti & Holt, 1991;
Schifferle et al., 1996). In the inflamed periodontal
pocket, hemoglobin would be a ready source of
heme-associated porphyrin and iron. Porphyromonas
gingivalis possesses outer membrane proteins that
are used for the binding and transport of hemin (Bra-
manti & Holt, 1993; Genco et al., 1994) and binding
of hemoglobin (Amano et al., 1995; Fujimura et al.,

© 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd 337
Molecular Oral Microbiology 30 (2015) 337–346
Hemin binding in P. gingivalis rHA2
1996; Simpson et al., 2004). Several putative TonB-
dependent outer membrane receptors, including Tlr
(TonB-linked receptor) (Slakeski et al., 2000), IhtA
(iron heme transport) (Dashper et al., 2000), HmuR
(hemin utilization receptor) (Simpson et al., 2000) and
HemR (hemin-regulated receptor) (Karunakaran
et al., 1997) have been described. The HBP35 pro-
tein has also been reported as an important hemin-
binding protein in P. gingivalis (Shibata et al., 2003;
Shoji et al., 2010). These genes (ihtA, tlr, hemR and
hmuR) exhibit homology to genes encoding bacterial
TonB-dependent receptors. Their products are
thought to act as outer membrane receptors in iron
and/or heme transport. Heme-starved P. gingivalis
cells express high- and low-affinity heme-binding
activity with dissociation constants (Kd values) of ~10–
10
and ~10–7 M, respectively (Tompkins et al., 1997).
The affinity of heme binding to Escherichia coli cells
expressing the TonB-dependent hemagglutinin recep-
tor HmuR of P. gingivalis is low (Kd of ~10–5 M).
Gingipains are cysteine proteases produced in large
quantities by P. gingivalis. The gingipains include argi-
nine-specific proteases [Arg-gingipain (Rgp) A and B]
and a lysine-specific protease [Lys-gingipain (Kgp)]. All
the proteins have catalytic domains with either Arg- or
Lys-specific proteolytic activities; in addition, RgpA and
Kgp have four C-terminal hemagglutinin/adhesin
domains, HA1 to HA4 (Curtis et al., 1999). The hemag-
glutinin 2 domain (HA2) is highly conserved and binds
heme with an apparent Kd of ~10–8 M. Gingipains are
thought to be involved in heme acquisition and delivery
to the organism (Pike et al., 1994). The HA2 of the gin-
gipains has been designated the hemoglobin receptor
(HbR) domain (Okamoto et al., 1998). The hemaggluti-
nin A (HagA) protein of P. gingivalis possesses four
repeats of the HA2 sequence in its structure (Shi et al.,
1999).
In cell surface extracts, polypeptides derived from
HA2-containing proteins were predominantly involved
in hemoglobin binding in P. gingivalis (Paramaesva-
ran et al., 2003). The in vitro porphyrin-binding prop-
erties of a recombinant HA2 domain (rHA2) were
investigated and found to be iron independent in
P. gingivalis (DeCarlo et al., 1999). In addition, the
rHA2 was found to bind hemin with high affinity and
provide porphyrin and ferric iron, which are required
for the growth and virulence of P. gingivalis (DeCarlo
et al., 1999). There was partial neutralization of
P. gingivalis hemoglobin binding by HA2-related
338
Q.B. Yang et al.
immunoglobulin G, which implicates HA2-containing
proteins as major contributors to cell surface hemo-
globin and heme binding (Paramaesvaran et al.,
2003). However, the amino acid residues of the HA2
involved in heme binding in P. gingivalis have not
been characterized; in addition, the exact molecular
mechanism by which this domain captures heme is
yet to be elucidated.
DeCarlo et al. (1999) have reported on the monoclo-
nal antibody (mAb) 5A1 and its epitope, which was
associated with the ALNPDNYLISKDVTG peptide
within the HA2. The mAb 5A1 does not interfere with
the binding of the HA2 to hemoglobin and the peptide,
ALNPDNYLISKDVTG, is not the porphyrin-binding site
for this domain (DeCarlo et al., 1999). An anti-HA2
mAb that has the ability to neutralize heme binding to
the HA2 has not been developed yet; although, the
partial neutralization of P. gingivalis hemoglobin bind-
ing by HA2-related immunoglobulin G suggests that
this is possible (Paramaesvaran et al., 2003).
In the present work, we successfully expressed
functional rHA2 and investigated its binding site, with
the aim of identifying the key residues involved in its
hemin-binding activity. We first used endopeptidase
digestion to identify the putative binding peptide for
hemin. The residues in this peptide were then
assessed by mutation and peptide-binding inhibition
analysis. Based on our findings, we propose that resi-
due His1001 is involved in the hemin-binding mecha-
nism of the P. gingivalis rHA2 and the DHYAVMISK
peptide containing His1001 may be an inhibitor of
hemin binding to this domain.
METHODS
Expression and purification of rHA2
To construct an expression plasmid, the target DNA
fragment was amplified by polymerase chain reaction
(PCR) from P. gingivalis ATCC 33277 genomic DNA,
using the primers HA2F (50 -CCG GAA TTC CGC
GCA GAC TTC ACG GA-30 ) and HA2R (50 -CCG CTC
GAG TTA GAA CTG AAT ATC ATC C-30 ), which con-
tain EcoRI and XhoI restriction sites, respectively.
The PCR-amplified fragment was digested with EcoRI
and XhoI and then inserted into the expression vector
pET28a (Novagen, Madison, WI) by conventional
cloning methods. Sequence analysis confirmed that
the sequence of the inserted PCR fragment in
© 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
Molecular Oral Microbiology 30 (2015) 337–346
Q.B. Yang et al. Hemin binding in P. gingivalis rHA2

pET28a was 100% homologous to the sequence of Hemin binding assay

P. gingivalis 33277 AP 009380.1 (GenBank).
The concentrations of the HA2, H1001D mutant and
The constructed plasmid (pET28a-HA2) was trans-
a trypsin inhibitor were determined by the Bradford
formed into E. coli strain BL21 (DE3) pLysS cells for
assay so that the same quantities of proteins were
protein production. Isopropylthio-b-D-galactoside was
used in the binding experiment. We randomly
used to induce protein expression at a final concen-
selected a trypsin inhibitor protein as a negative con-
tration of 0.5 mM. Escherichia coli strain BL21 (DE3)
trol. The binding capacities of HA2, H1001D and the
pLysS cells were resuspended in 20 ml of buffer A
trypsin inhibitor were tested over a range of concen-
(20 mM Tris–HCl pH 7.5, 500 mM NaCl), and dis-
trations in 20 mM Tris–HCl pH 7.5, 200 mM NaCl by
rupted by sonication. After centrifugation (15,570 g,
incubating with 10 ll of hemin-agarose (Sigma, St
40 min, 4°C; Beckman, Rotor ID JA25.50, Fullerton,
Louis, MO) at room temperature for 30 min. The mix-
CA), the supernatant was collected and applied to a
ture was then centrifuged at 3944 g at 4°C for 5 min,
5-ml Hitrap chelating column (Ni2+ charged, GE-
and the supernatant was discarded. The pelleted
Amersham Pharmacia Biotech, Uppsala, Sweden) for
ligand affinity gel was then washed three times with
the purification of recombinant histidine-tagged HA2.
phosphate-buffered saline (PBS, pH 7.4). After a final
Impurities were removed by washing with 5% buffer
PBS wash, bound proteins were eluted by the addi-
B (20 mM Tris–HCl pH 7.5, 500 mM NaCl, 500 mM
tion of 20 ll of sodium dodecyl sulfate–polyacryl-
imidazole) and the target protein was eluted with 30%
amide gel electrophoresis (SDS–PAGE) sample
buffer B. The protein was further purified using size
buffer (0.5 M Tris–HCl, pH 6.8, 2% SDS, 0.05% brom-
exclusion chromatography (HiloadSuperdex 75 XK16/
ophenol blue, 30% glycerol), boiled at 100°C for
60; GE-Amersham Pharmacia Biotech) with buffer C
10 min, and analyzed by SDS–PAGE (see below).
(20 mM Tris–HCl pH 7.5, 200 mM NaCl). The molecu-
lar weight of the purified rHA2 was 25 kDa.
Enzymatic digestion

Site-directed mutagenesis of P. gingivalisrHA2 The hemin-agarose-bound rHA2 was digested in

25 mM ammonium bicarbonate buffer pH 7.8 with Asp-
The H1001D mutation was introduced into the rHA2
N endopeptidase or Lys-C endopeptidase (Sigma) at
gene, by site-directed mutagenesis using two-step
37°C for 10 h (1 : 1000 enzyme : substrate weight
overlap extension PCR and the HA2 plasmid (pET28a-
ratio). The mixtures were then centrifuged at 3944 g at
HA2) as the template. In the first PCR step, two HA2
4°C for 5 min, and the supernatant was discarded. The
gene fragments were generated. The first fragment
pelleted ligand affinity gel was then washed three
was amplified with UMEcoR (50 -AG-
0 times with PBS. After a final PBS wash, rHA2 peptides
TGAATTCCGCGCAGACTTCACGGAAAC-3 ) and dhd
were eluted from hemin-agarose by boiling in PBS for
(50 -CCGCATAGTCATCCCCGG-30 ). The latter primer
Ultra Performance Liquid Chromatography (UPLC)
contained a mutation site. The second fragment was
(Waters Corporation, Milford, MA) and ThermoFisher
amplified with uhd (50 -CCGGGGATGACTATGCGG-30 )
LTQ Orbitrap XLMS analysis.
and DMXho (50 -TATCTCGAGTTAGAA
CTGAATATCATCCAAAAG-30 ). The amplicons were
then purified from a gel and used as templates in the MSAnalysis
next PCR step. In the next PCR step, UMEcoR (50 -AG-
Samples were analyzed by nanoUPLC (nanoAcquity;
TGAATTCCGCGCAGACTTCACGGAAAC-30 ) and
Waters) coupled to an LTQ-Orbitrap XL mass spec-
DMXho (50 -TATCTCGAGTTAGAACTGAATATCATC
trometer (Thermo-Finnigan, Bremen, Germany). LC
CAAAAG-30 ) were used to synthesize a full-length
separations were performed on a 180-lm 9 20-mm
mutated HA2 fragment. The mutant HA2 was identified
C18 column with a 5-lm particle size and 75-
by plasmid sequencing. The full-length mutated HA2
lm 9 150-mm C18 column with a 3.5-lm particle size
fragment was purified from a gel and cloned into the
(Waters). The protein was dissolved in 12 ll of l5%
pET28a+ vector using the EcoRI and XhoI restriction
aqueous acetonitrile containing 0.1% formic acid and
sites. The rHA2 containing H1001D was purified using
18 ll of this solution was injected into the liquid chro-
the method described above for rHA2.

© 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd 339
Molecular Oral Microbiology 30 (2015) 337–346
Hemin binding in P. gingivalis rHA2
matography/mass spectrometry (LC-MS) system. The
protein was eluted with a 5% aqueous acetonitrile con-
taining 0.1% formic acid (solvent A) and 95% acetoni-
trile containing 0.1% formic acid (solvent B) over a 70-
min period at a flow rate of 500 nl min1 with a column
temperature of 35°C and directed into the nanospray
source with the following parameters: ion spray voltage
of 2000 V, survey full-scan MS spectra (m/z 400–
2000) were acquired at resolution R = 100,000. Pep-
tide ions were selected for LTQ-CID (Linear ion trap
quadrupole-Collision-induced dissociation). The full-
scan MS spectra of corresponding peptide pairs were
manually analyzed to consider all relevant charge
states and up to 10 Isotope Peaks. The obtained chro-
matograms were analyzed with BIOWORKSBROWSER
3.3.1 SP1 and the resulting mass lists were used for
database search using SEQUESTTM.
Development of the mAb QB
Polypeptides were synthesized based on the
sequence of the active peptide of HA2, peptide 1
(Table 1). The synthetic peptides were purified to 98%
and conjugated to bovine serum albumin. BALB/c mice
were immunized with conjugated peptides. Hybrido-
mas were generated by fusion with SP2/0 myelomas
and spleen cells, which were obtained from mice
immunized with the polypeptides. The hybridoma cell
lines were selected by enzyme-linked immunosorbent
assay (ELISA). The mAb titer was identified by ELISA.
Then, the hybridoma cells were distributed over 96-well
cell culture plates supplemented with culture medium
Table 1 Sequences and positions of the various peptides used in
this study
Peptide Sequence Position
Peptide 1 DGFPGDHYAVMISK 995–1008
Peptide 2 DHYAVMISK 1000–1008
Peptide 3 DEYAVMISK Same as peptide 2 with
H1001E mutant
Peptide 4 ALHPDHYLI 969–977 with mutants N971H
and N974H
Peptide 5 YYYAVNDGFPGDH
YAVMISKTGTNAGDF 989–1016
Peptide 6 DHYAVMISKTGTNAG 1000–1014
Peptide 7 YYYAVNDGFPGDHYA
VMISK 989–1008
Residues are numbered according to the P .gingivalis RGP-1
mature protein (Pavloff et al., 1995).
340
Q.B. Yang et al.
containing 80% Dulbecco’s modified Eagle’s medium,
20% fetal bovine serum and HAT (hypoxanthine, ami-
nopterin, thymidine), which was replaced with HT
(hypoxanthine, thymidine) 2 weeks after cell fusion. All
cells were incubated at 37°C with 5% CO2. The anti-
body titer and binding properties of the culture super-
natants were tested by the same methods as sera
7 days after fusion, and the cells that showed both high
affinity and specificity were subcloned by the limited
dilution method until an amonoclone was obtained that
could steadily secrete antibodies.
Competitive ELISA
The ELISAs were performed in polystyrene microtitre
wells. Proteins (1 lg ml1) and peptides (10 lg ml1)
in bicarbonate buffer (0.05 mol l1, pH 9.6) were
used to coat the surfaces. All wells were blocked and
washed in PBS with 0.05% Tween-20 (PBS/Tween).
The coated protein was subsequently incubated with
dilutions of hemin and mAb QB in PBS/Tween at a
concentration of 0.5 mg ml1 for 30 min. Secondary
goat anti-mouse antibodies conjugated with horserad-
ish peroxidase (Dako, Carpinteria, CA) were applied
(at a dilution of 1 : 5000) for 30 min, and then horse-
radish peroxidase activity was monitored by hydroly-
sis of the substrate at 450 nm.
SDS–PAGE
Proteins were analyzed by SDS–PAGE with the dis-
continuous buffer system of Laemmli (Laemmli,
1970). Gels, comprised of 5% (wt/vol) acrylamide
stacking and 15% (wt/vol) acrylamide separating
components with 0.8% bisacrylamide, were electro-
phoresed at 150 V constant voltage in a Bio-Rad
MiniProtean II (Bio-Rad Laboratories, Richmond, CA)
apparatus. The resultant gels were stained with Coo-
massie brilliant blue R-250. Molecular weights were
determined using known proteins as standards.
Inhibition of hemin binding
Hemin-agarose (10 ll) was pre-incubated with 10 ll
of peptide (peptide 2, peptide 3, or peptide 4)
(10 lg ll1) at room temperature for 30 min and
washed with PBS three times. Then, the binding of
peptide-treated hemin-agarose to rHA2 was exam-
ined (in the same way as the Hemin binding assay).
© 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
Molecular Oral Microbiology 30 (2015) 337–346
Q.B. Yang et al.
Peptide synthesis
Peptides were synthesized by Chiron Mimotopes with
terminal amines and carboxylic acids. The peptide
sequences are shown in Table 1. Residues are num-
bered according to the P. gingivalis RGP-1 mature
protein (Pavloff et al., 1995).
RESULTS
Identification of peptides potentially involved in
the hemin binding of P. gingivalis rHA2
The rHA2 was incubated with hemin-agarose and then
digested with Asp-N and Lys-C endopeptidase. The
hemin-agarose-bound peptides were then analyzed by
MS. The MS analysis of the Asp-N endopeptidase
digestion identified peptide DHYAVMISKTGTNAG
(peptide 6). Subsequent Lys-C endopeptidase diges-
tion generated peptide YAVNDGFPGDHYAVMISK
(peptide 7). The smaller peptide DHYAVMISK (peptide
2) was present in both peptide 6 and peptide 7. Hence,
peptide 2 was identified as being putatively involved in
the hemin binding of the P. gingivalis rHA2 (Fig. 1).
Hemin competitively inhibits the immunoreactivity
of rHA2 or peptide 1 to mAb QB
The DHYAVMISK peptide, potentially involved in the
hemin binding of P. gingivalis rHA2, was then evalu-
ated. We synthesized a peptide with a more optimal
length, named peptide 1 (DGFPGDHYAVMISK) that
contained DHYAVMISK (peptide 2) for this experiment.
An mAb, named QB, was isolated by immunization of
mice with the bovine serum albumin-conjugated pep-
tide 1. Hemin was used to compete with mAb QB to
bind the antigen, DGFPGDHYAVMISK (peptide 1),
resulting in a concentration-dependent reduction in the
immunoreactivity of rHA2 or peptide 1 to mAb QB, indi-
cating that it is likely that peptide 2 (DHYAVMISK) was
involved in hemin binding to the P. gingivalis rHA2
(Fig. 2).
Mutational analysis of the putative residues
involved in hemin binding of the P. gingivalis rHA2
The residues of peptide 2, identified as putatively
involved in hemin binding, were mutated and the
hemin binding capacity was tested by incubation with
© 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
Molecular Oral Microbiology 30 (2015) 337–346
Hemin binding in P. gingivalis rHA2
hemin-agarose. After incubation, unbound recombi-
nant protein was washed away with PBS. Bound pro-
teins were eluted by boiling and analyzed by SDS–
PAGE. Results showed that the HA2 containing a
mutation of His1001 reduced hemin-binding activity
compared with wild-type rHA2 (Fig. 3).
Assessment of peptide inhibition of the hemin-
binding activity of the P. gingivalis HA2
Peptide inhibition of the hemin-binding activity of the
P. gingivalis HA2 was assayed by pretreating hemin-
agarose with peptide 2 containing the hemin-binding
residue (His1001) (Fig. 4, lane 1), peptide 3 contain-
ing a mutation of the hemin-binding residue
(H1001E) (Fig. 4, lane 2), peptide 4 (amino acids
969–977 with the N971H and N974H mutations)
(Fig. 4, lane 3), and without peptide (Fig. 4, lane 4).
Trypsin inhibitor was used as a negative protein con-
trol in the binding experiment. After incubation,
unbound peptides were washed away with PBS.
Then, peptide-treated hemin-agarose was incubated
with rHA2, washed with PBS, and analyzed by SDS–
PAGE. Results showed that peptide 2, DHYAVMISK,
affected the inhibition of hemin-binding activity;
whereas peptide 3, DEYAVMISK, and peptide 4, AL-
HPDHYLI, did not (Fig. 4).
DISCUSSION
P. gingivalis growth must be controlled to prevent
periodontal pathologies and this might be achieved
by interfering with one or more pathways of heme
uptake. In this study, we identified the peptide,
DHYAVMISK, which can inhibit rHA2 hemin binding.
This peptide was identified by endopeptidase diges-
tion, and was evaluated by site-directed mutagenesis
and functional analysis. First, the P. gingivalis rHA2,
bound to hemin-agarose, was digested to identify the
peptides potentially involved in the hemin binding of
P. gingivalis rHA2. Asp-N endopeptidase digestion
identified the peptide, DHYAVMISKTGTNAG (peptide
6); whereas, Lys-C endopeptidase digestion gener-
ated the peptide, YAVNDGFPGDHYAVMISK (peptide
7). The sequence, DHYAVMISK (peptide 2), was
present in the peptides derived from both digestions
and, therefore, identified as being putatively involved
in the hemin binding of P. gingivalis rHA2. An mAb
(QB) was produced by immunization of mice with a
341
Hemin binding in P. gingivalis rHA2 Q.B. Yang et al.

Figure 1 Identification of the amino acid sequences around the hemin-binding site of the Porphyromonas gingivalis haemagglutinin 2 domain
(HA2) by analysis of Asp-N- and Lys-C-derived peptides. The rHA2 was incubated with hemin-agarose and then digested with Asp-N and
Lys-C endopeptidase. The hemin-agarose-bound peptides were identified by mass spectrometry analysis. Residues are numbered according
to the P. gingivalis RGP-1 mature protein (Pavloff et al., 1995). (A) The Asp-N- and Lys-C-derived peptides. (B) Mass spectrometry fragmen-
tation spectrum of the Asp-N-derived peptide 6. (C) Mass spectrometry fragmentation spectrum of Lys-C- derived peptide 7.

342 © 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
Molecular Oral Microbiology 30 (2015) 337–346
Q.B. Yang et al.
A dently of iron, as hemin only differs from protoporphy-
B with the iron (DeCarlo et al., 1999).
Figure 2 Competitive inhibition of the immunoreactivity of recombi-
nant haemagglutinin 2 domain (rHA2) or peptide 1 to monoclonal
antibody (mAb) QB. Microtitre wells were coated overnight with
rHA2 (A) or peptide 1 (B). The mAb QB and dilutions of hemin were
then incubated with the coated proteins. Binding of a secondary
anti-mouse horseradish peroxidase-conjugated antibody to mAb QB
was monitored by hydrolysis of the substrate at 450 nm. These data
are representative of two separate experiments. Abs, absorbance.
bovine serum albumin-conjugated peptide (peptide 1,
DGFPGDHYAVMISK) that contained peptide 2 as
well as additional amino acids to achieve the opti-
mum length. We found hemin competed with mAb
QB to bind peptide 1, which resulted in a concentra-
tion-dependent reduction in the immunoreactivity of
rHA2 or peptide 1 to mAb QB. The recombinant HA2
containing the H1001D mutation was expressed in
E. coli. The H1001D mutation of HA2 expressed in
E. coli reduced hemin-binding activity. H1001E muta-
tion was introduced into the peptide by chemical syn-
thesis. The peptide inhibition test showed that peptide
2, DHYAVMISK, has an effect on the inhibition of
hemin-binding activity. Taken together, the data indi-
cate that H1001 of HA2 is involved in the binding of
heme to HA2.
Protoporphyrin IX and hemin did not differ in their
ability to bind rHA2. Also, protoporphyrin IX inhibited
binding of rHA2 to hemin. This indicated that the
sequestering of porphyrin by HA2 functioned indepen-
© 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
Molecular Oral Microbiology 30 (2015) 337–346
Hemin binding in P. gingivalis rHA2
rin IX by a ferric iron ion. The rHA2 specifically binds
three porphyrin compounds, hemin, protoporphyrin IX
and hematoporphyrin, in the region of the propionic
acid groups, as this binding was abolished by direc-
tionally attaching the carboxylic acids of the porphyrin
compounds to fixed amines. Additionally, the iron
chelator, 2,29-dipyridal, also inhibited the binding of
rHA2 to hemin; although, the Ki of the dipyridal was
200-fold higher than the Ki of protoporphyrin IX. This
indicates that rHA2 also had some weak interaction
Hemin affinity is regulated through a combination of
covalent, hydrophobic, electrostatic and steric effects
between the globins and bound hemin (Hargrove,
1996) . It is well established that histidines are com-
mon axial ligands for heme (Poulos, 1996; Chakrab-
orti, 2003). Both hemoproteins and outer membrane
receptors contain histidines that can bind heme
(Bracken et al., 1999; Paoli et al., 1999). The cova-
lent bond between the His(F8) residue and the fifth
coordination site of iron is the key force securing
heme to ferrous globins. Hydrophobic interactions
between the heme vinyls and the apolar environment
proximal and distal to bound hemin also contribute to
hemin retention. Hydrogen bonding between His64
(E7) and coordinated water also plays a role in
anchoring hemin to the globin.
The amino acid residues reported to be ligands for
the propionic acid groups of heme are Arg, Lys and
His. (Bonaventura et al., 1991; Whitaker, 1995; de-
Sanctis et al., 2004). The molecular and biochemical
mechanisms involved in the binding of His1001 in
P. gingivalis rHA2 to heme were not characterized in
this study. DeCarlo et al. suggested PPIX competes
with both heme and hemoglobin for HA2 binding and
heme recognition by HA2 may be solely porphyrin
mediated. It is likely that His1001 of HA2 interacted
with the heme propionate. However, the possibility of
interaction of His1001 of HA2 with the fifth coordination
site of iron is also considered (DeCarlo et al., 1999).
Porphyromonas gingivalis possesses outer mem-
brane proteins, which are used for the binding and
transport of hemin (Bramanti & Holt, 1993; Karunaka-
ran et al., 1997) and binding of hemoglobin (Amano
et al., 1995; Fujimura et al., 1996; Simpson et al.,
2004). Several putative TonB-dependent outer mem-
brane receptors, including Tlr, IhtA, HmuR and HemR,
have been described. Hemin, hemoglobin and serum
343
Hemin binding in P. gingivalis rHA2 Q.B. Yang et al.

Figure 3 Hemin-binding capacity of proteins containing mutations of residues putatively involved in hemin binding. The hemin-binding capac-
ity of proteins containing mutations of residues identified as putatively involved in hemin binding (shown in Fig. 1) were tested by incubation
with hemin-agarose. M, molecular weight marker. After incubation of 10 nM (lane 1), 7.5 nM (lane 3), 5 nM (lane 5) and 2.5 nM (lane 7) of Por-
phyromonas gingivalis wild-type recombinant haemagglutinin 2 domain (rHA2); 10 nM (lane 2), 7.5 nM (lane 4), 5 nM (lane 6) and 2.5 nM (lane
8) of P. gingivalis rHA2H1001D mutant; and 25 nM (lane 9), 12.5 nM (lane 10) and 6.25 nM (lane 11) of trypsin inhibitor with 10 ll hemin,
unbound recombinant protein was washed away with phosphate-buffered saline. Bound proteins were eluted by boiling and analyzed by
sodium dodceyl sulfate–polyacrylamide gel electrophoresis. Results showed that the wild-type HA2 had hemin-binding activity; whereas the
protein containing a mutation of His1001 did not.

albumin utilization in P. gingivalis requires the partici-

Figure 4 Peptide inhibition of the hemin-binding activity of the Por-
phyromonas gingivalis haemagglutinin 2 domain (HA2). Peptide
inhibition of the hemin-binding activity of the P. gingivalis HA2 was
assayed by pretreating hemin-agarose with peptide 2 (lane 1), pep-
tide 3 (lane 2), peptide 4 (lane 3) and without peptide (lane 4) (pep-
tide sequences listed in Table 1). After incubation, unbound
peptides were washed away with phosphate-buffered saline (PBS).
Then, peptide-treated hemin-agarose was incubated with rHA2,
washed with PBS, and analyzed by sodium dodceyl sulfate–poly-
acrylamide gel electrophoresis. Results showed that peptide 2 inhib-
ited hemin-binding activity; whereas peptides 3 and 4 did not.
pation of cysteine proteases, including the lysine-spe-
cific gingipain Kgp and arginine-specific gingipain
RgpA, which contain a cysteine protease domain and
additional C-terminal hemagglutinin domains. Gingi-
pains have also been shown to degrade host iron-con-
taining and heme-containing proteins, including
hemoglobin, hemopexin, haptoglobin and transferrin
(Brochu et al., 2001; Sroka et al., 2001). Kgp and
RgpA may function in a catalytic capacity and as he-
mophore-like proteins to capture heme and deliver it to
an outer membrane receptor (Simpson et al., 2000).
The HA2-containing proteins contain the gingipains
Kgp and RgpA and the putative hemagglutinin protein
HagA (Nakayama et al., 1998; DeCarlo et al., 1999).
Heme-starved P. gingivalis cells express high- and
low-affinity heme-binding activity with dissociation con-
stants (Kds) of ~10–10 and ~10–7 M, respectively (Tomp-
kins et al., 1997). The HA2 domain is highly conserved
in HA2-containing proteins and is reported to bind
heme with an apparent Kd of ~10–8 M. The affinity of
heme binding to E. coli cells expressing the TonB-
dependent hemagglutinin receptor HmuR of P. gingi-
valis is low (Kd, ~10–5 M). In cell surface extracts, poly-
peptides derived from HA2-containing proteins were
predominantly involved in hemoglobin binding. Inhibi-

344 © 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
Molecular Oral Microbiology 30 (2015) 337–346
Q.B. Yang et al.
tors or mAbs, which eliminate the binding of HA2 to
hemoglobin or heme, may contribute to growth inhibi-
tion of the bacterium.
Further characterization of the binding between
rHA2 and heme should contribute to the design of
inhibitors of heme or hemoglobin binding. Heme acqui-
sition is considered fundamental to the growth of
P. gingivalis and intervention with peptide 2
(DHYAVMISK) or a related peptide to disrupt path-
ways for heme uptake may allow the control or preven-
tion of periodontal disease. Although the present
peptide is directed at P. gingivalis infection in the oral
cavity, such as during periodontal disease, its use
may extend to other diseases resulting from infection
by P. gingivalis or a related microorganism involving
the acquisition of iron, heme or porphyrin. Such micro-
organisms need to acquire iron, heme or porphyrin as
they do not possess a biosynthetic pathway for por-
phyrins and include Salmonella sp., Serratia sp., Yer-
sinia sp., Klebsiella sp., Vibrio sp., Pseudomonas sp.,
E. coli, Haemophilus sp. and Bordetella sp.
ACKNOWLEDGEMENTS
This work was supported by grants from the National
Natural Science Foundation of China (Grant
30572037, 81470753) and the Beijing Natural Sci-
ence Foundation (Grant 7062028, 7122078).
REFERENCES
Amano, A., Kuboniwa, M., Kataoka, K. et al. (1995) Bind-
ing of hemoglobin by Porphyromonas gingivalis. FEMS
Microbiol Lett 134: 63–67.
Bodet, C., Chandad, F. and Grenier, D. (2007) [Patho-
genic potential of Porphyromonas gingivalis, Treponema
denticola and Tannerella forsythia, the red bacterial
complex associated with periodontitis]. Pathol Biol
(Paris) 55: 154–162.
Bonaventura, C., Cashon, R., Bonaventura, J., Perutz, M.,
Fermi, G. and Shih, D.T. (1991) Involvement of the dis-
tal histidine in the low affinity exhibited by Hb Chico
(Lys beta 66 – Thr) and its isolated beta chains. J Biol
Chem 266: 23033–23040.
Bracken, C.S., Baer, M.T., Abdur-Rashid, A., Helms, W.
and Stojiljkovic, I. (1999) Use of heme-protein com-
plexes by the Yersinia enterocolitica HemR receptor:
histidine residues are essential for receptor function. J
Bacteriol 181: 6063–6072.
© 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
Molecular Oral Microbiology 30 (2015) 337–346
Hemin binding in P. gingivalis rHA2
Bramanti, T.E. and Holt, S.C. (1991) Roles of porphyrins
and host iron transport proteins in regulation of growth
of Porphyromonas gingivalis W50. J Bacteriol 173:
7330–7339.
Bramanti, T.E. and Holt, S.C. (1993) Hemin uptake in Por-
phyromonas gingivalis: Omp26 is a hemin-binding sur-
face protein. J Bacteriol 175: 7413–7420.
Brochu, V., Grenier, D., Nakayama, K. and Mayrand, D.
(2001) Acquisition of iron from human transferrin by Por-
phyromonas gingivalis: a role for Arg- and Lys-gingipain
activities. Oral Microbiol Immunol 16: 79–87.
Chakraborti, A.S. (2003) Interaction of porphyrins with
heme proteins – a brief review. Mol Cell Biochem 253:
49–54.
Curtis, M.A., Kuramitsu, H.K., Lantz, M. et al. (1999) Molec-
ular genetics and nomenclature of proteases of Porphyro-
monas gingivalis. J Periodontal Res 34: 464–472.
Curtis, M.A., Aduse-Opoku, J. and Rangarajan, M. (2001)
Cysteine proteases of Porphyromonas gingivalis. Crit
Rev Oral Biol Med 12: 192–216.
Dashper, S.G., Hendtlass, A., Slakeski, N. et al. (2000)
Characterization of a novel outer membrane hemin-bind-
ing protein of Porphyromonas gingivalis. J Bacteriol
182: 6456–6462.
DeCarlo, A.A., Paramaesvaran, M., Yun, P.L., Collyer, C.
and Hunter, N. (1999) Porphyrin-mediated binding to
hemoglobin by the HA2 domain of cysteine proteinases
(gingipains) and hemagglutinins from the periodontal
pathogen Porphyromonas gingivalis. J Bacteriol 181:
3784–3791.
Dixon, D.R. and Darveau, R.P. (2005) Lipopolysaccharide
heterogeneity: innate host responses to bacterial modifi-
cation of lipid a structure. J Dent Res 84: 584–595.
Fujimura, S., Shibata, Y., Hirai, K. and Nakamura, T.
(1996) Binding of hemoglobin to the envelope of Por-
phyromonas gingivalis and isolation of the hemoglobin-
binding protein. Infect Immun 64: 2339–2342.
Genco, C.A., Odusanya, B.M. and Brown, G. (1994) Bind-
ing and accumulation of hemin in Porphyromonas gingi-
valis are induced by hemin. Infect Immun 62: 2885–2892.
Hargrove, M.S. (1996) The Structural Determinants of
Myoglobin Stability. Rice University, Houston, TX. Avail-
able online at: http://hdl.handle.net/1911/16910.
Holt, S.C., Kesavalu, L., Walker, S. and Genco, C.A.
(1999) Virulence factors of Porphyromonas gingivalis.
Periodontol 2000 20: 168–238.
Karunakaran, T., Madden, T. and Kuramitsu, H. (1997)
Isolation and characterization of a hemin-regulated
gene, hemR, from Porphyromonas gingivalis. J Bacteriol
179: 1898–1908.
345
Hemin binding in P. gingivalis rHA2
Laemmli, U.K. (1970) Cleavage of structural proteins
during the assembly of the head of bacteriophage T4.
Nature 227:680–685.
Nakayama, K., Ratnayake, D.B., Tsukuba, T., Kadowaki,
T., Yamamoto, K. and Fujimura, S. (1998) Haemoglobin
receptor protein is intragenically encoded by the cyste-
ine proteinase-encoding genes and the haemagglutinin-
encoding gene of Porphyromonas gingivalis. Mol Micro-
biol 27: 51–61.
Okamoto, K., Nakayama, K., Kadowaki, T., Abe, N., Rat-
nayake, D.B. and Yamamoto, K. (1998) Involvement of
a lysine-specific cysteine proteinase in hemoglobin
adsorption and heme accumulation by Porphyromonas
gingivalis. J Biol Chem 273: 21225–21231.
Paoli, M., Anderson, B.F., Baker, H.M., Morgan, W.T.,
Smith, A. and Baker, E.N. (1999) Crystal structure of
hemopexin reveals a novel high-affinity heme site
formed between two beta-propeller domains. Nat Struct
Biol 6: 926–931.
Paramaesvaran, M., Nguyen, K.A., Caldon, E. et al.
(2003) Porphyrin-mediated cell surface heme capture
from hemoglobin by Porphyromonas gingivalis. J Bacte-
riol 185: 2528–2537.
Pavloff, N., Potempa, J., Pike, R.N. et al. (1995) Molecular
cloning and structural characterization of the Arg-gingi-
pain proteinase of Porphyromonas gingivalis. Biosynthe-
sis as a proteinase-adhesin polyprotein. J Biol Chem
270: 1007–1010.
Pike, R., McGraw, W., Potempa, J. and Travis, J. (1994)
Lysine- and arginine-specific proteinases from Porphyro-
monas gingivalis. Isolation, characterization, and evi-
dence for the existence of complexes with
hemagglutinins. J Biol Chem 269: 406–411.
Poulos, T.L. (1996) Ligands and electrons and haem pro-
teins. Nat Struct Biol 3: 401–403.
deSanctis, D., Dewilde, S., Pesce, A. et al. (2004) Crystal
structure of cytoglobin: the fourth globin type discovered
in man displays heme hexa-coordination. J Mol Biol
336: 917–927.
Schifferle, R.E., Shostad, S.A., Bayers-Thering, M.T.,
Dyer, D.W. and Neiders, M.E. (1996) Effect of protopor-
phyrin IX limitation on Porphyromonas gingivalis. J En-
dod 22: 352–355.
Shi, Y., Ratnayake, D.B., Okamoto, K., Abe, N., Yamam-
oto, K. and Nakayama, K. (1999) Genetic analyses of
346
Q.B. Yang et al.
proteolysis, hemoglobin binding, and hemagglutination
of Porphyromonas gingivalis. Construction of mutants
with a combination of rgpA, rgpB, kgp, and hagA. J Biol
Chem 274: 17955–17960.
Shibata, Y., Hiratsuka, K., Hayakawa, M. et al. (2003) A
35-kDa co-aggregation factor is a hemin binding protein
in Porphyromonas gingivalis. Biochem Biophys Res
Commun 300: 351–356.
Shoji, M., Shibata, Y., Shiroza, T. et al. (2010) Character-
ization of hemin-binding protein 35 (HBP35) in Por-
phyromonas gingivalis: its cellular distribution,
thioredoxin activity and role in heme utilization. BMC
Microbiol 25: 152.
Simpson, W., Olczak, T. and Genco, C.A. (2000) Charac-
terization and expression of HmuR, a TonB-dependent
hemoglobin receptor of Porphyromonas gingivalis. J
Bacteriol 182: 5737–5748.
Simpson, W., Olczak, T. and Genco, C.A. (2004) Lysine-
specific gingipain K and heme/hemoglobin receptor
HmuR are involved in heme utilization in Porphyromon-
as gingivalis. Acta Biochim Pol 51: 253–262.
Slakeski, N., Dashper, S.G., Cook, P., Poon, C., Moore,
C. and Reynolds, E.C. (2000) A Porphyromonas gingi-
valis genetic locus encoding a heme transport system.
Oral Microbiol Immunol 15: 388–392.
Slots, J., Bragd, L., Wikstrom, M. and Dahlen, G. (1986)
The occurrence of Actinobacillus actinomycetemcomi-
tans, Bacteroides gingivalis and Bacteroides intermedius
in destructive periodontal disease in adults. J Clin Peri-
odontol 13: 570–577.
Sroka, A., Sztukowska, M., Potempa, J., Travis, J. and
Genco, C.A. (2001) Degradation of host heme proteins
by lysine- and arginine-specific cysteine proteinases
(gingipains) of Porphyromonas gingivalis. J Bacteriol
183: 5609–5616.
Tompkins, G.R., Wood, D.P. and Birchmeier, K.R. (1997)
Detection and comparison of specific hemin binding by
Porphyromonas gingivalis and Prevotella intermedia. J
Bacteriol 179: 620–626.
Whitaker, T.L. (1995) Residues controlling the function
and stability of the CD corner in myoglobin and hemo-
globin. PhD dissertation, William Marsh Rice University,
Houston, TX. Available online at: http://hdl.handle.net/
1911/16913.
© 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
Molecular Oral Microbiology 30 (2015) 337–346