Preferential modification of CyaA-hemolysin by CyaC-acyltransferase through the catalytic Ser30-His33 dyad in esterolysis of palmitoyl-donor substrate devoid of acyl carrier proteins

Article abstract of DOI:10.1016/j.abb.2020.108615

We previously demonstrated that the ~130-kDa CyaA-hemolysin domain (CyaA-Hly) from Bordetella pertussis co-expressed with CyaC-acyltransferase in Escherichia coli was acylated at Lys⁹⁸³ and thus activated its hemolytic activity. Here, attempts

Full text of DOI:10.1016/j.abb.2020.108615

Archives of Biochemistry and Biophysics 694 (2020) 108615
Contents lists available at ScienceDirect
Archives of Biochemistry and Biophysics
journal homepage: www.elsevier.com/locate/yabbi
Preferential modification of CyaA-hemolysin by CyaC-acyltransferase
through the catalytic Ser30-His33 dyad in esterolysis of palmitoyl-donor
substrate devoid of acyl carrier proteins
,
Mattayaus Yentongchaia, Niramon Thamwiriyasatib **, Chompounoot Imtongc, Hui-Chun Lid,
,d,e,*
Chanan Angsuthanasombata
a Bacterial Toxin Research Innovation Cluster (BRIC), Institute of Molecular Biosciences, Mahidol University, Salaya Campus, Nakornpathom, 73170, Thailand
b Molecular Medicine Research Group (MMRG), Division of Medical Technology, Faculty of Allied Health Sciences, Burapha University, Chonburi, 20131, Thailand
c Division of Biology, Department of Science, Faculty of Science and Technology, Prince of Songkla University, Pattani, 94000, Thailand
d Department of Biochemistry, School of Medicine, Tzu Chi University, Hualian, 97004, Taiwan
e Laboratory of Synthetic Biophysics and Chemical Biology, Biophysics Institute for Research and Development (BIRD), Chiang Mai, 50230, Thailand
A R T I C L E I N F O
A B S T R A C T
Keywords:
We previously demonstrated that the ~130-kDa CyaA-hemolysin domain (CyaA-Hly) from Bordetella pertussis co-
expressed with CyaC-acyltransferase in Escherichia coli was acylated at Lys983 and thus activated its hemolytic
activity. Here, attempts were made to provide greater insights into such toxin activation via fatty-acyl modifi-
cation by CyaC-acyltransferase. Non-acylated CyaA-Hly (NA/CyaA-Hly) and CyaC were separately expressed in
E. coli and subsequently purified by FPLC to near homogeneity. When effects of acyl-chain length were
comparatively evaluated through CyaC-esterolysis using various p-nitrophenyl (pNP) derivatives, Michaelis-
Menten steady-state kinetic parameters (KM and kcat) of CyaC-acyltransferase revealed a marked preference
for myristoyl (C14:0) and palmitoyl (C16:0) substrates of which catalytic efficiencies (kcat/KM) were roughly the
same (~1.5 × 103 sꢀ 1mMꢀ 1). However, pNP-palmitate (pNPP) gave the highest hemolytic activity of NA/CyaA-
Hly after being acylated in vitro with a range of acyl-donor substrates. LC-MS/MS analysis confirmed such CyaC-
mediated palmitoylation of CyaA-Hly occurring at Lys983, denoting no requirement of an acyl carrier protein
(ACP). A homology-based CyaC structure inferred a role of a potential catalytic dyad of conserved Ser30 and His33
residues in substrate esterolysis. CyaC-ligand binding analysis via molecular docking corroborated high-affinity
binding of palmitate with its carboxyl group oriented toward such a dyad. Ala-substitutions of each residue
(S30A or H33A) caused a drastic decrease in kcat/KM of CyaC toward pNPP, and hence its catalytic malfunction
through palmitoylation-dependent activation of NA/CyaA-Hly. Altogether, our present data evidently provide
such preferential palmitoylation of CyaA-Hly by CyaC-acyltransferase through the enzyme Ser30-His33
nucleophile-activation dyad in esterolysis of palmitoyl-donor substrate, particularly devoid of a natural acyl-ACP
donor.
CyaC-acyltransferase
Esterolysis
A nucleophile activation dyad
Palmitoylation
p-Nitrophenyl derivatives
Toxin activation
1. Introduction
respiratory tract infection in humans has re-emerged globally as a
consequence of declining immunity, following vaccination and imper-
Bordetella pertussis, a Gram-negative pathogenic bacterium that
causes whooping cough (also known as ‘pertussis’), secretes the bi-
functional adenylate cyclase (AC)-hemolysin (Hly) toxin (CyaA) which
harbors both calmodulin-dependent AC and hemolytic/pore-forming
activities [1,2]. Whooping cough which is a highly contagious
fect vaccination of populations [3–5], hence requiring new approches
for effective treatment of such a re-emerging infectious disease. We
recently generated CyaA-specific humanized VH/VHH nanobodies that
would have promising applications in potential development of a novel
anti-pertussis agent [6].
* Corresponding author. Bacterial Toxin Research Innovation Cluster (BRIC), Institute of Molecular Biosciences, Mahidol University, Salaya Campus, Nakornpa-
thom, 73170, Thailand.
** Corresponding author.
E-mail addresses: niramon@go.buu.ac.th (N. Thamwiriyasati), chanan.ang@mahidol.ac.th (C. Angsuthanasombat).
https://doi.org/10.1016/j.abb.2020.108615
Received 10 June 2020; Received in revised form 4 September 2020; Accepted 29 September 2020
Available online 2 October 2020
0003-9861/© 2020 Elsevier Inc. All rights reserved.

M. Yentongchai et al.
Archives of Biochemistry and Biophysics 694 (2020) 108615
The ~180-kDa CyaA toxin preferentially binds to the
α
Mβ2 integrin
[22]. Complementary pairs of mutagenic primers were designed based
on the cloned cyaC gene sequence (see Supplementary Table S1). Mutant
plasmids were first identified by restriction endonuclease digestion and
subsequently verified by DNA sequencing.
(CD11b/CD18) of macrophages [7] that will enable entry of its AC
catalytic domain into the cell to induce apoptosis toward intracellular
accumulation of cyclic AMP [8]. Recently, this key virulence fac-
torꢀ CyaA has been shown to bind to an inactive form of an integrin
called complement receptor 3 (CR3) residing on the surface of many
immune cells, enabling the toxin to use the integrin without triggering
an immune response [9]. Nonetheless, CyaA also causes hemolysis of
sheep erythrocytes lacking the intergrin receptor [10], thus suggesting
an alternative mechanism of target cell binding. Moreover, the
~130-kDa truncated CyaA fragment, termed CyaA-Hly, retains hemo-
lytic activity independent of the N-terminal AC domain [11] and can
also induce ion-channel formation in receptor-free planar lipid bilayers
[12]. Particularly, positively-charged side-chains in the pore-lining helix
3 were demonstrated to be important for hemolytic activity and
ion-channel opening of CyaA-Hly [13].
2.2. Protein preparations of CyaC wild-type and its mutants
CyaC wild-type and its mutants were over-expressed in E. coli BL21
(DE3)pLysS under control of the IPTG-inducible T7 promoter as
described previously [23]. After harvested by centrifugation, E. coli cells
expressing each individual CyaC protein as inclusions were resuspended
in 20 mM Tris-HCl (pH 8.0) and disrupted by using a French Pressure
Cell (10,000 psi). Insoluble inclusions separated from crude lysate by
centrifugation (12,000×g, 20 min) were solubilized in 20 mM Tris-HCl
(pH 8.0) containing 8 M urea for 1 h. Unfolded CyaC proteins were
initially refolded in a low [urea]-refolding buffer (20 mM Tris-HCl, pH
8.0, 150 mM NaCl, 2 M urea) for 4 h and finally dialyzed twice against
the same buffer without urea for 4 h [22]. The refolded proteins were
then subjected to size-exclusion chromatography (SEC, Superdex™ 75,
GE Healthcare) equilibrated with 20 mM Tris-HCl (pH 8.0), 150 mM
NaCl, and the eluted factions were analyzed by SDS-PAGE (sodium
dodecyl sulfate-polyacrylamide gel electrophoresis).
CyaA which belongs to the RTX (Repeat-in-ToXin) cytolysin family is
initially produced as inactive proCyaA and converted intracellularly to
the mature active toxin via a post-translational acylation mediated by
the endogenous CyaC-acyltransferase [14]. An acyl group predomi-
nantly found to be attached to CyaA at Lys983 is a C16:0-hydrocarbon
chain (i.e. palmitoyl) [15,16], although the exact role in toxin function
of such conjugated palmitoyl is yet unclear. Nonetheless, it was initially
proposed that channel-forming activity of the ~180-kDa full-length
CyaA toxin was strongly dependent on toxin modification via palmi-
toylation, albeit not yet clear for the involvement of the attached pal-
mitoyl moiety in channel-pore formation [17]. The added palmitoyl
moiety was also suggested to increase membrane affinity of the
full-length CyaA toxin needed for efficient attachment to the target cell
membrane by serving as a membrane-associating mediator or a
receptor-toxin interaction determinant [18]. However, our recent
studies demonstrated that such toxin modification via Lys983-palmitoy-
lation was not essential for binding of the 130-kDa CyaA-Hly domain to
target erythrocyte membranes, but rather needed for stabilizing
toxin-induced ion-leakage pores [19]. More recently, we have shown
that the N-terminal hydrophobic region (Met482-Leu750) of CyaA-Hly is
conceivably required for not only membrane-pore formation but also
functional association with CyaC-acyltransferase, and hence effective
palmitoylation at Lys983 [20].
2.3. Expression and purification of soluble non-acylated CyaA-Hly
CyaA-Hly was expressed without CyaC-acyltransferase (termed non-
acylated CyaA-Hly, NA/CyaA-Hly) in E. coli BL21(DE3)pLysS under
control of the T7 promoter as described previously [23]. The cultured
cells were harvested by centrifugation, resuspended in 20 mM Tris-HCl
(pH 8.0) containing 5 mM CaCl2 and 1 mM PMSF (phenylmethylsulfonyl
fluoride) and disrupted in a French Pressure Cell. After centrifugation
(12,000×g, 20 min), the soluble NA/CyaA-Hly protein in the lysate su-
pernatant was sequentially purified by anion-exchange chromatography
(AEC, HiTrap Q HP, GE Healthcare) and SEC (Superose® 12, GE
Healthcare), and the purified protein was analyzed by SDS-PAGE.
2.4. CyaC-esterolytic activity assays toward p-nitrophenyl (pNP)
derivatives
All RTX toxins are conceivably believed to be activated by similar
mechanisms despite variations in the location, number and chain length
of the conjugated acyl group [21]. However, molecular basis for the
Esterolytic activity of CyaC-acyltransferase (15 μg/mL or ~0.7 μM)
was assayed toward a number of pNP derivatives, i.e., p-nitrophenyl
acetate (pNPA, C2:0), p-nitrophenyl carpyrate (pNPC, C8:0), p-nitro-
phenyl decanoate (pNPD, C10:0), p-nitrophenyl myristate (pNPM, C14:0),
p-nitrophenyl palmitate (pNPP, C16:0) and p-nitrophenyl stearate (pNPS,
C18:0) dissolved in isopropanol. The CyaC-catalyzed reaction was per-
formed in 50 mM Tris-HCl (pH 7.4) at 25 ◦C and its enzymatic rate was
substrate
specificity
and
catalytic
mechanism
of
the
CyaC-acyltransferase is not clearly defined. In our previous studies, we
demonstrated that the recombinant CyaC-acyltransferase, which was
effectively refolded after urea solubilization, was able to activate the
130-kDa non-acylated CyaA-Hly (NA/CyaA-Hly) fragment in the pres-
ence of E. coli lysate containing ACPs to become hemolytically active
[22]. In this study, to gain more critical insights into the molecular
mechanism of CyaC catalysis, we have successfully devised an in vitro
activation system for NA/CyaA-Hly toward pNP derivatives as mimics of
the natural acyl-ACP donor. Moreover, structural analysis of CyaC via
protein-ligand docking together with mutagenesis results revealed a
critical role of a potential catalytic dyad of conserved Ser30 and His33
residues in substrate esterolysis for toxin activation through preferential
determined via the formation of pNP product by measuring OD400 (ε =
11.6 mMꢀ 1cꢀ 1) [24] with SoftMax Pro spectrophotometer (0.7-cm
light-path). Steady-state kinetics were recorded by varying concentra-
tions of pNP substrates (10-500 μM) and the corresponding kinetic pa-
rameters (Vmax and KM) were determined by non-linear fitting of
untransformed data to the Michaelis-Menten (M-M) equation by using
GraphPad Prism 5 software. Catalytic constant (kcat) and catalytic effi-
ciency (kcat/KM) were calculated by using the active enzyme
concentration.
palmitoylation at Lys983
.
2.5. Assessment of hemolytic activity of activated CyaA-Hly
2. Materials and methods
In vitro activation of CyaA-Hly was set up by mixing purified NA/
2.1. Construction of CyaC mutants via site-directed mutagenesis
CyaA-Hly (10 μg) and purified CyaC (15 μg) together with varying
concentrations (10-500 μM) of pNP derivatives in 1 mL of TBS buffer
The pCyaC plasmid encoding the ~21-kDa CyaC-acyltransferase
[23] was used as a template for single-alanine substitutions at Ser30
and His33 performed by PCR-based directed mutagenesis using a high
fidelity Phusion DNA polymerase (Finnzymes, Finland), following the
Quick-Change Mutagenesis Kit (Stratagene) as previously described
(20 mM Tris-HCl, pH 7.4, 150 mM NaCl, 5 mM CaCl2) and incubated at
37 ◦C for 5 min. After the activation, 10-
μ
l sheep erythrocytes (108 cells/
mL) were added to each 1-mL mixture sample and incubated at 37 ◦C for
5 h. After centrifugation (10,000×g, 5 min), erythrocyte lysis was
determined by measuring OD540 of the supernatant. OD540 value
2

M. Yentongchai et al.
Archives of Biochemistry and Biophysics 694 (2020) 108615
Table 1
Kinetic parameters for CyaC esterolysis with different acyl chain-length
substratesa.
Substrate
Vmax (U/mg)b
kcat (sꢀ 1
)
KM (mM)
kcat/KM (sꢀ 1
mMꢀ 1
)
pNPA (C2:0
)
0.50 × 102
±
0.18 × 102
±
0.11 ±
0.01
1.7 × 102 ± 11
2.5 × 102 ± 4.8
3.0 × 102 ± 2.6
1.5 × 103 ± 86
1.5 × 103 ± 10
2.1 × 102 ± 5.3
1.0
0.2
pNPC (C8:0
)
4.5 × 102
9.0
±
1.6 × 102
3.3
±
0.64 ±
0.02
pNPD
3.3 × 102
4.9
±
1.1 × 102
1.7
±
0.38 ±
0.01
(C10:0
pNPM
(C14:0
pNPP
(C16:0
)
)
)
2.4 × 103 ± 96
1.5 × 103 ± 26
8.6 × 102 ± 35
8.4 × 102
33
±
±
±
0.55 ±
0.05
5.2 × 102
9.2
0.34 ±
0.01
pNPS (C18:0
)
3.0 × 102
12
1.4 ± 0.09
a
Results were averaged from three independent experiments. Both M-M and
Lineweaver-Burk plots of individual corresponding data are shown in Supple-
mentary Fig. S3.
Fig. 1. Left panel, SDS-PAGE (Coomassie brilliant blue-stained 12% gel) of NA/
CyaA-Hly preparation. Lane 1, crude extracts from E. coli cells expressing the
130-kDa NA/CyaA-Hly. Lanes 2 and 3, purified NA/CyaA-Hly obtained by AEC
and SEC, respectively. Right panel, SDS-PAGE (Coomassie brilliant blue-stained
12% gel) of CyaC preparation. Lane 1, insoluble fraction of crude extracts from
E. coli cells expressing the 21-kDa CyaC-acyltransferase. Lane 2, SEC-purified
CyaC after unfolding-refolding processes. M, molecular mass standards. (For
interpretation of the references to color in this figure legend, the reader is
referred to the Web version of this article.)
b
One unit of enzyme activity is defined as the amount of enzyme liberating 1
μ
mol of substrates in 1 min at 37 ◦C.
binding site of the CyaC modeled structure using SwissDock, an EADock
DSS-based web server (http://www.swissdock.ch) dedicated to the
docking of a small molecule on a target protein [27]. The resulting
docking models were analyzed by using UCSF Chimera and PyMOL.
corresponding to complete hemolysis was obtained by lysing the
erythrocytes with 0.1% Triton-X 100 while spontaneous hemolysis was
determined by incubating the erythrocyte suspensions in buffer alone.
Percentage of hemolysis was calculated by [(OD540 sample ꢀ OD540
negative control)/(OD540 of 100% hemolysis ꢀ OD540 negative con-
trol)] × 100. For comparison of acylation activities between CyaC wild-
3. Results and discussion
3.1. Substrate preference of CyaC-acyltransferase analyzed through its
esterolytic kinetics
At the start, the ~130-kDa non-acylated form of CyaA-Hly (NA/
CyaA-Hly) which was optimally expressed as a soluble protein was
sequentially purified by AEC and SEC to near homogeneity as analyzed
by SDS-PAGE (Fig. 1, left panel). Moreover as shown in Fig. 1 (right
panel), the recombinant CyaC-acyltransferase which was over-expressed
mostly as an insoluble inclusion was completely solubilized in 8M-urea
denaturing buffer. The soluble protein after refolding in a low [urea]-
refolding buffer was purified via SEC which indicated that the urea-
free refolded protein was recovered in a single peak containing a
>95%-pure band of the ~21-kDa CyaC protein as verified by SDS-PAGE
(see Fig. 1, right panel, lane 2).
type and its mutants, only pNPP with a fixed concentration of 500
was used as an acyl-donor substrate.
μM
2.6. Verification of acylated CyaA-Hly via LC-MS/MS analysis
To examine whether there is a presence of palmitate (C16:0) at Lys983
acylation site of CyaA-Hly mediated by CyaC or its mutants, such CyaA-
Hly protein bands resolved by SDS-PAGE were excised and eluted from
the gel prior to digestion with trypsin according to a standard protocol.
LC-MS/MS (liquid chromatography-tandem mass spectrometry, Fin-
nigan LTQ Linear Ion Trap Mass Spectrometer) was used to analyze the
purified trypsin-generated peptide fragments.
To gain more insights into toxin activation through CyaC catalysis,
the M-M kinetic parameters, i.e., the maximum reaction rate (Vmax) and
the Michaelis constant (KM), of the enzymatic esterolytic reaction to-
ward various pNP derivatives containing different even-type acyl-chain
lengths (short C2:0, medium C8:0-C10:0 and long C14:0-C18:0) were
comparatively determined. As can be seen from Table 1, the longest
chain-length (pNPS, C18:0) among the varying acyl groups resulted in the
highest KM value (~1.4 mM) for CyaC esterolysis, indicating the lowest
affinity of the enzyme toward this stearyl group. The Vmax and catalytic
constant (kcat) values of the enzyme appeared to be greatest for the
myristoyl (pNPM, C14:0) substrate, giving a maximum catalytic effi-
ciency (kcat/KM) of ~1.5 × 103 sꢀ 1mMꢀ 1. It should be noted that this
enzyme seemed to exhibit the highest-affinity binding (~KM value of
~0.1 mM) with the C2:0 chain-length acyl group (pNPA), albeit the
Vmax and kcat values toward this substrate were lowest (only ~50 U/mg
and ~18 sꢀ 1, respectively).
2.7. Homology-based modeling of CyaC structure
A plausible three-dimensional (3D) CyaC model was generated based
on the target-template alignment with the ApxC known structure (PDB-
4whn, the 172-amino acid toxin-activating acyltransferase from Acti-
nobacillus pleuropneumoniae) [25] via SWISS-MODEL homology
modeling. As analyzed via compositional score matrix adjustment of the
NCBI’s protein-query protein-database BLAST program, the pairwise
sequence alignment between CyaC and ApxC revealed ~39% identity
(60/155 residues) and 60% homology (93/155 residues). Coordinates
which are conserved between the target and the template were copied
from the template to the model. Insertions and deletions were
re-modeled using a fragment library. Side chains were then rebuilt and
the geometry of the resulting model was finally regularized by using a
force field. The global and per-residue model quality was assessed using
the QMEAN scoring function [26].
Minimal differences in kcat and KM values were observed between the
C
14:0 and C16:0 chain-length substrates, resulting in similar catalytic ef-
ficiencies (~1.5 × 103 sꢀ 1mMꢀ 1) with both substrates (see Table 1).
Overall, these kinetic parameters of such CyaC-esterolytic reaction to-
ward various acyl-chain length derivatives consistently revealed an
obvious preference for both pNPM (C14:0) and pNPP (C16:0) as substrates.
Our data are in agreement with other kinetic studies of HlyC-
2.8. CyaC-ligand docking
To determine an optimal ligand-binding mode of CyaC for a palmi-
tate, the ligand (free palmitic acid) was docked in silico into a potential
3

M. Yentongchai et al.
Archives of Biochemistry and Biophysics 694 (2020) 108615
○
Fig. 2. Hemolytic activities of CyaA-Hly acylated in vitro by CyaC with various pNP-derivetive substrates, pNPC, C8:0 (●), pNPD, C10:0 ( ), pNPM, C14:0 (▾), pNPP,
C
16:0 (Δ) and pNPS, C18:0 (■). Error bars indicate standard errors of the mean from three independent experiments.
acyltransferase from pathogenic E. coli producing
α
-hemolysin (HlyA)
donor substrates at low concentrations (10-100
μM), but the stearylated
that both myristoyl- and palmitoyl-ACPs are the most effective sub-
strates for HlyC catalysis [28].
CyaA-Hly-induced hemolysis became drastically reduced when the
substrate concentrations were steadily increased from 200 to 500 μM
(Fig. 2). A possible explanation for this observation is that in such
high-concentration conditions (most likely above the critical micelle
3.2. Effects of acyl-chain lengths on CyaC-mediated CyaA-Hly activation
and hence toxin-induced hemolysis
concentration, CMC), pNPS with its long hydrophobic acyl-chain (C18:0
)
would be hardly soluble in the reaction buffer [20 mM Tris-HCl (pH 7.4),
150 mM NaCl and 5 mM CaCl2] and more likely to become aggregated,
thus decreasing a net number of free dissolved pNPS available for CyaC
catalysis. It was previously observed for BSA-induced hydrolysis of
pNP-acyl esters that the rate of esterolysis appeared to decrease with
increasing chain length of the acyl moiety at concentrations above CMC
Effects of such acyl-chain lengths on CyaC-acyltransferase activity
was also comparatively studied via its relative enzymatic ability to
modulate NA/CyaA-Hly in vitro to become hemolytically active against
sheep erythrocytes. As the results shown in Fig. 2, when NA/CyaA-Hly
was modulated by CyaC with pNPP substrate at ≥ 200 μM, the palmi-
(280 μM) [30].
toylated protein was revealed to be the most active (~90% hemolytic
activity at 500 M substrate) compared to four other acylated CyaA-Hly
It is worth mentioning that unlike our previous method for toxin
activation in vitro [22], we herein were able to establish an in vitro
modification system mediated by recombinant CyaC without the use of
an acyl-ACP donor from its E. coli-host cell lysate. As also noted that
CyaC-acyltransferase activity was measured indirectly through hemo-
lytic activity of CyaA-Hly after being modulated with an appropriate
acyl-donor substrate. Such observation at the highest hemolytic activity
through the toxin modification by pNPP might be dependent on the net
amount of palmitoylated toxins being formed and/or the physiological
and biochemical features of the attached palmitoyl moiety. Accordingly,
such results of CyaA-Hly activation via palmitoylation could suggest that
pNPP is the most preferential substrate of CyaC-acyltransferase and the
attached palmitoyl moiety (C16:0) could possess the most proper
acyl-chain length (~20 Å) for efficient membrane anchoring, and hence
toxin-induced hemolysis. Being in agreement with this notion, our
previous studies have shown that such toxin modification via
Lys983-palmitoylation was rather essential for stabilizing the
toxin-induced ion-leakage pores [19].
μ
proteins. For instance, CyaA-Hly modified with myristoyl group through
pNPM (another preferential substrate as shown above) appeared to
exhibit only a half-maximal hemolytic activity (~45%) observed for
palmitoylated CyaA-Hly. Nevertheless, it was previously shown that
myristoyl-ACP but not palmitoyl-ACP conferred the highest hemolytic
activity of the E. coli HlyA acylated in vitro by its activator (HlyC-acyl-
transferase) with a range of fatty acids (C12 to C18:1) [28]. It should be
noted that negative control reactions containing NA/CyaA-Hly incu-
bated with only CyaC exhibited merely negligible activity (<5%),
indicating the need of pNP substrates as mimics of the natural acyl-ACP
donor.
Noted that LC-MS/MS analysis of a trypsin-digested fragment
derived from palmitoylated CyaA-Hly (E972GVA TQTTAYGKC16:0R984
)
confirmed the C16:0 acylation of CyaA-Hly at Lys983 (see Supplementary
Fig. S1). In line with this, it has very recently been shown that the ~180-
kDa full-length CyaA toxin was exclusively activated in vivo to be
hemolytically active through predominant modification of its Lys983
residue by C16 fatty acyl chains [29]. It is also noticeable that the
stearylated-C18:0 CyaA-Hly protein was found to display hemolytic ac-
tivity similar to palmitoylated-C16:0 CyaA-Hly when modulated with
4

M. Yentongchai et al.
Archives of Biochemistry and Biophysics 694 (2020) 108615
Fig. 3. (A) The CyaC modeled structure of which the ribbon diagram is rainbow-colored from purple at the N terminus to red at the C terminus. Residue numbers in
the secondary structural elements are:
α
1 (Phe17 – Trp27),
α
2 (Pro31 – Arg34),
α
3 (Val38 – Gln50), β1 (Gln53 – Arg58), β2 (Val62 – Met72),
α
4 (Ala74 – Met82),
α
5 (Asn90
–
Asn93), β3 (Leu98 – Ile104),
α
6 (Asp109 – Arg121), β4 (Val126 – Arg133), β5 (Thr138 – Arg147) and
α
7 (Ala151 – Ser172). Ball-and-stick models denote a proposed Ser30
-
His33 catalytic dyad of CyaC. (B) Multiple sequence alignments of the deduced amino acid sequence of Bpt-CyaC, B. pertussis CyaC-acyltransferase (EF_592556)
against those of Apl-ApxC, the ApxC crystal structure (PDB-4whn) from Actinobacillus pleuropneumoniae [25] and four other RTX-C acyltransferases, i.e., Ecl-HlyC
(E. coli hemolysin-acyltransferase, AJ488511.1), Mhm-LktC (Mannheimia haemolytica leukotoxin-acyltransferase, P55121), Par-PaxC (Pasteurella aerogenes
PaxC-acyltransferase, U66588.1) and Apc-ApxIIC (A. porcitonsillarum ApxIIC-acyltransferase, Q5DI90) using ClustalX. Bpt-CyaC is used for denoting secondary
structures according to the modeled structure shown in A. Amino acids are shaded red, blue and green to denote degrees of homology (6/6), (5/6) and (4/6),
respectively. Proposed catalytic residues (e.g., Ser30 and His33 for CyaC) are indicated by *. (For interpretation of the references to color in this figure legend, the
reader is referred to the Web version of this article.)
Fig. 4. Protein-ligand interactions between
CyaC-acyltransferase and palmitic acid as
depicted by surface representation and vdW
spheres, respectively, showing that palmitic
acid is buried in a hydrophobic cavity of the
CyaC molecule. Upper inset, a zoom-in region
demonstrating the position of palmitic acid’s
carboxyl group placed closely to the pro-
posed Ser30-His33 catalytic dyad. Lower inset,
another zoom-in interaction region showing
a putative oxyanion hole generated by an
oxygen atom of ligand forming hydrogen
bonds with backbone amide (NH) groups of
Met32 and His33 from CyaC.
3.3. Ser30 and His33 could potentially serve as a catalytic dyad in CyaC-
esterolysis of palmitoyl-donor substrate
Ramachandran plot indicated that the CyaC modeled structure could
remain in sterically favorable main-chain conformations (see Supple-
mentary Fig. S2). As illustrated in Fig. 3A, the modeled structure com-
prises a single domain with a β-sheet core of five strands (β1, β2, β3, β4,
Since no known structure of CyaC-acyltransferase has been eluci-
dated yet, attempts were first made to construct a plausible 3D-modeled
CyaC structure based on the ApxC crystal structure [25]. The resulting
and β5) connected by seven
α-helices (
α
1,
α2,
α
3,
α4
α
5,
α6 and
α7) to
form a two-layer /β sandwich similar to a classic fold of
α
α/β hydrolase
5

M. Yentongchai et al.
Archives of Biochemistry and Biophysics 694 (2020) 108615
binding to the cleavage site on ligand molecule as both functional groups
are only 3-4 Å from the ligand’s carboxyl group (see Fig. 4, upper inset).
Moreover, two adjacent backbone amide (NH) groups of Met32 and His33
could potentially form hydrogen bonds with a closet oxygen atom of the
substrate, thus generating an oxyanion hole which would probably
stabilize the tetrahedral intermediate formed during catalysis (see Fig. 4,
lower inset). This hole which is commonly described in many bacterial
enzyme structures (e.g., esterase and acyltranferase) [34,35] could allow
for positioning of such a substrate, which would suffer from steric hin-
drance if it could not occupy the hole.
Table 2
Kinetic parameters of CyaC wild-type and mutants with pNPP and corresponding
effects on the hemolytic activity of CyaA-Hlya.
Enzyme
Wt
Vmax (U/
kcat
KM (mM)
kcat/KM
Hemolytic
mg)
(sꢀ 1
)
(sꢀ 1mMꢀ 1
)
activity (%)
1.3 ×
4.7 ×
0.34 ±
1.4 × 103
±
88 ± 6.5
3.8 ± 4.7
4.5 ± 3.2
103 ± 39
102
14
±
0.05
11
S30A
H33A
45 ± 0.3
16 ± 0.1
23 ± 0.3
7.0 ×
2.2 × 102
±
±
10ꢀ 2
±
7.0
0.001
68 ± 0.9
8.0 ×
2.8 × 102
To test our notion from the docking data, two CyaC mutants (i.e.,
S30A and H33A) of which Ser30 and His33 were replaced individually
with Ala were initially determined for their esterolytic activity against
pNPP substrate in comparison with the wild-type (Wt) enzyme. As can
be seen in Table 2, both Ala-substitutions critically affected the CyaC
esterolytic activity in all kinetic aspects. Among the kinetic parameters
obtained, Vmax appeared to be mostly affected by both mutations as it
decreased by about 20–30 fold in each corresponding mutant compared
to that of Wt. The S30A and H33A mutations also decreased KM values,
suggesting an increase in binding affinity for the pNPP substrate, how-
ever, the greatest impact of the two mutations is the 20-30 fold reduction
in turnover (see Table 2). Consequently, their kcat/KM values were found
greatly reduced, thus clearly indicating an adverse influence of these
mutations on catalytic efficiency of CyaC and hence an importance of
Ser30 and His33 that would serve as a potential catalytic dyad in CyaC-
esterolytic catalysis.
10ꢀ 2
±
4.0
0.003
a
Results were averaged from at least three independent experiments per-
formed in duplicate. Both M-M and Lineweaver-Burk plots of individual corre-
sponding data are shown in Supplementary Fig. S4.
family [31], with a potential catalytic dyad of Ser30 and His33 which are
highly conserved among the RTX-acyltransferase family (see Fig. 3B). In
analogy to the catalytic Cys-His dyad of several cysteine proteases [32,
33], CyaC-His33 would serve as a general base catalyst, thus increasing
the nucleophilicity of Ser30
.
When in silico docking was performed to gain greater insights into the
architecture of CyaC interacting with its palmitoyl-donor substrate, the
results illustrated in Fig. 4 revealed a high potential affinity of the CyaC
molecule in binding to the docked palmitate ligand whose C16:0-chain
was found in a properly enclosed region within a relatively hydrophobic
cavity of the modeled structure. In addition, the palmitate’s carboxyl
group appeared to be oriented outward, directly interacting with the
two highly conserved Ser30 and His33 residues. As can be also inferred
from the docked model that both conserved side-chains which possess
functional groups for acyl-transfer catalysis (i.e., hydroxyl group of Ser30
and unprotonated imidazole of His33) were highly potent to get close in
Experiments were further conducted to determine an effect of such
Ala-substitutions on CyaC-acylation activity toward the resulting he-
molytic activity of activated CyaA-Hly. Consistent with observations
from the CyaC-esterolysis assay, NA/CyaA-Hly proteins after being
mediated by each CyaC mutant displayed an almost entire loss of cor-
responding hemolytic activity (see Table 2). Such a drastic decrease in
CyaA-Hly-induced hemolysis was likely due to incapability of each CyaC
Fig. 5. A putative catalytic mechanism involving roles of both Ser30 and His33 residues that could potentially act as a nucleophile-activation dyad in CyaC-
acyltransferase esterolysis of acyl-donor substrate, e.g. pNPP, as well as toxin activation through preferential palmitoylation at Lys983 of NA/Cya-HlyA.
6

M. Yentongchai et al.
Archives of Biochemistry and Biophysics 694 (2020) 108615
lining helix 3 of the Bordetella pertussis CyaA-hemolysin to hemolytic cctivity and
ion-channel opening, Toxins 9 (2017) 109–121.
mutant to palmitoylate NA/CyaA-Hly as verified by LC-MS/MS analysis
revealing no palmitoyl moiety being conjugated at Lys983 of CyaA-Hly.
These results suggested that Ser30 and His33 would play a crucial role
in CyaC-directed NA/CyaA-Hly palmitoylation in vitro and hence toxin-
induced hemolysis. Taken together, all these independent assays
consistently signified that both Ser30 and His33 residues could poten-
tially act as a nucleophile-activation dyad in CyaC-acyltransferase
esterolysis of acyl-donor substrates as well as NA/Cya-HlyA acylation
through CyaC-catalyzed preferential palmitoylation as its putative cat-
alytic mechanism is illustrated in Fig. 5. Unlike other previous studies
for HlyC-directed pro-HlyA acylation [36,37], our present findings
evidently revealed that a natural acyl-ACP donor appeared not to be a
strict requirement for such CyaC-mediated NA/Cya-HlyA acylation in
vitro, possibly implicating an alternative post-translational modification
of CyaA-Hly underlining RTX-toxin maturation in vivo.
[14] G.D. Westrop, E.K. Hormozi, N.A. Da Costa, R. Parton, J.G. Coote, Bordetella
pertussis adenylate cyclase toxin: proCyaA and CyaC proteins synthesised
separately in Escherichia coli produce active toxin in vitro, Gene 180 (1996) 91–99.
[15] M. Hackett, L. Guo, J. Shabanowitz, D.F. Hunt, E.L. Hewlett, Internal lysine
palmitoylation in adenylate cyclase toxin from Bordetella pertussis, Science 266
(1994) 433–435.
[16] T. Basar, V. Havlicek, S. Bezouskova, M. Hackett, P. Sebo, Acylation of lysine 983 is
sufficient for toxin activity of Bordetella pertussis adenylate cyclase. Substitutions of
alanine 140 modulate acylation site selectivity of the toxin acyltransferase CyaC,
J. Biol. Chem. 276 (2001) 348–354.
[17] R. Benz, E. Maier, D. Ladant, A. Ullmann, P. Sebo, Adenylate cyclase toxin (CyaA)
of Bordetella pertussis; Evidence for the formation of small ion-permeable channels
and comparison with HlyA of Escherichia coli, J. Biol. Chem. 269 (1994), 27231-
27139.
[18] J. Masin, M. Basler, O. Knapp, M. El-Azami-El-Idrissi, E. Maier, I. Konopasek,
R. Benz, C. Leclerc, P. Sebo, Acylation of lysine 860 allows tight binding and
cytotoxicity of Bordetella adenylate cyclase on CD11b-expressing cells,
Biochemistry 44 (2005) 12759–12766.
[19] K. Meetum, C. Imtong, G. Katzenmeier, C. Angsuthanasombat, Acylation of the
Bordetella pertussis CyaA-hemolysin: functional implications for efficient membrane
insertion and pore formation, Biochim. Biophys. Acta Biomembr. 1859 (2017)
312–318.
Acknowledgment
This work has been generously supported by grants from Thailand
Science Research and Innovation (MRG-59-8-0040), Burapha University
and Mahidol University.
[20] V. Raksanoh, P. Prangkio, C. Imtong, N. Thamwiriyasati, K. Suvarnapunya,
L. Shank, C. Angsuthanasombat, Structural requirement of the hydrophobic region
of the Bordetella pertussis CyaA-hemolysin for functional association with CyaC-
acyltransferase in toxin acylation, Biochem. Biophys. Res. Commun. 499 (2018)
862–867.
Appendix A. Supplementary data
[21] P. Stanley, V. Koronakis, C. Hughes, Acylation of Escherichia coli hemolysin: a
unique protein lipidation mechanism underlying toxin function, Microbiol. Mol.
Biol. Rev. 62 (1998) 309–333.
Supplementary data to this article can be found online at https://doi.
org/10.1016/j.abb.2020.108615.
[22] N. Thamwiriyasati, B. Powthongchin, J. Kittiworakarn, G. Katzenmeier,
C. Angsuthanasombat, Esterase activity of Bordetella pertussis CyaC-acyltransferase
against synthetic substrates: implications for catalytic mechanism in vivo, FEMS
Microbiol. Lett. 304 (2010) 183–190.
References
[23] B. Powthongchin, C. Angsuthanasombat, High level of soluble expression in
Escherichia coli and characterisation of the CyaA pore-forming fragment from a
Bordetella pertussis Thai clinical isolate, Arch. Microbiol. 189 (2008) 169–174.
[24] Q. Wang, J.C. Allen, H.E. Swaisgood, Binding of vitamin D and cholesterol to beta-
lactoglobulin, J. Dairy Sci. 80 (1997) 1054–1059.
[1] N.H. Carbonetti, G.V. Artamonova, C. Andreasen, N. Bushar, Pertussis toxin and
adenylate cyclase toxin provide a one-two punch for establishment of Bordetella
pertussis infection of the respiratory tract, Infect. Immun. 73 (2005) 2698–2703.
[2] P. Glaser, H. Sakamoto, J. Bellalou, A. Ullmann, A. Danchin, Secretion of
cyclolysin, the calmodulin-sensitive adenylate cyclase-haemolysin bifunctional
protein of Bordetella pertussis, EMBO J. 7 (1988) 3997–4004.
[25] N.P. Greene, A. Crow, C. Hughes, V. Koronakis, Structure of a bacterial toxin-
activating acyltransferase, Proc. Natl. Acad. Sci. U.S.A. 112 (2015) E3058–3066.
[26] P. Benkert, S.C. Tosatto, D. Schomburg, QMEAN: a comprehensive scoring function
for model quality assessment, Proteins 71 (2008) 261–277.
[3] M.J. Bart, S.R. Harris, A. Advani, Y. Arakawa, D. Bottero, V. Bouchez, P.
K. Cassiday, C.-S. Chiang, T. Dalby, N.K. Fry, et al., Global population structure and
evolution of Bordetella pertussis and their relationship with vaccination, mBio 5
(2014), e01074.
[27] A. Grosdidier, V. Zoete, O. Michielin, SwissDock, a protein-small molecule docking
web service based on EADock DSS, Nucleic Acids Res. 39 (2011) W270–W277.
[28] J.P. Issartel, V. Koronakis, C. Hughes, Activation of Escherichia coli prohaemolysin
to the mature toxin by acyl carrier protein-dependent fatty acylation, Nature 351
(1991) 759–761.
[4] J.M. Warfel, L.I. Zimmerman, T.J. Merkel, Acellular pertussis vaccines protect
against disease but fail to prevent infection and transmission in a nonhuman
primate model, Proc. Natl. Acad. Sci. U.S.A. 111 (2014) 787–792.
[5] A. Lim, J.K. Ng, C. Locht, S. Alonso, Protective role of adenylate cyclase in the
context of a live pertussis vaccine candidate, Microb. Infect. 16 (2014) 51–60.
[6] A. Malik, C. Imtong, N. Sookrung, G. Katzenmeier, W. Chaicumpa,
C. Angsuthanasombat, Structural characterization of humanized nanobodies with
neutralizing activity against the Bordetella pertussis CyaA-hemolysin: implications
for a potential epitope of toxin-protective antigen, Toxins 8 (2016) 99–111.
[7] P. Guermonprez, N. Khelef, E. Blouin, P. Rieu, P. Ricciardi-Castagnoli, N. Guiso,
D. Ladant, C. Leclerc, The adenylate cyclase toxin of Bordetella pertussis binds to
target cells via the alpha(M)beta(2) integrin (CD11b/CD18), J. Exp. Med. 193
(2001) 1035–1044.
[29] A. Osickova, H. Khaliq, J. Masin, D. Jurnecka, A. Sukova, R. Fiser, J. Holubova,
O. Stanek, P. Sebo, R. Osicka, Acyltransferase-mediated selection of the length of
the fatty acyl chain and of the acylation site governs activation of bacterial RTX
toxins, J. Biol. Chem. 295 (2020). https://www.jbc.org/cgi/doi/10.1074/jbc.RA1
20.014122.
[30] H. Østdal, H.J. Andersen, Non-enzymic protein induced hydrolysis of p-nitrophenyl
acyl esters in relation to lipase/esterase assays, Food Chem. 55 (1996) 55–61.
[31] M. Holmquist, Alpha/beta-hydrolase fold enzymes: structures, functions and
mechanisms, Curr. Protein Pept. Sci. 1 (2000) 209–235.
¨
[32] A. Paasche, A. Zipper, S. Schafer, J. Ziebuhr, T. Schirmeister, B. Engels, Evidence
for substrate binding-induced zwitterion formation in the catalytic Cys-His dyad of
the SARS-CoV main protease, Biochemistry 53 (2014) 5930–5946.
[33] S. Saraswat, M. Chaudhary, D. Sehgal, Hepatitis E virus cysteine protease has
papain-like properties validated by in silico modeling and cell-free inhibition
assays, Front Cell Infect Microbiol 9 (2020), https://doi.org/10.3389/
fcimb.2019.00478.
[8] P. Gueirard, A. Druilhe, M. Pretolani, N. Guiso, Role of adenylate cyclase-
hemolysin in alveolar macrophage apoptosis during Bordetella pertussis infection in
vivo, Infect. Immun. 66 (1998) 1718–1725.
[9] R. Osicka, A. Osickova, S. Hasan, L. Bumba, J. Cerny, P. Sebo, Bordetella adenylate
cyclase toxin is a unique ligand of the integrin complement receptor 3, Elife 4
(2015), e10766.
[34] L.-C. Lee, Y.-L. Lee, R.-J. Leu, J.-F. Shaw, Functional role of catalytic triad and
oxyanion hole-forming residues on enzyme activity of Escherichia coli thioesterase
I/protease I/phospholipase L1, Biochem. J. 397 (2006) 69–76.
[10] O. Knapp, E. Maier, G. Polleichtner, J. Masin, P. Sebo, R. Benz, Channel formation
in model membranes by the adenylate cyclase toxin of Bordetella pertussis: effect of
calcium, Biochemistry 42 (2003) 8077–8084.
¨
[35] A. Rottig, A. Steinbüchel, Acyltransferases in bacteria, Microbiol. Mol. Biol. Rev. 77
[11] B. Powthongchin, C. Angsuthanasombat, Effects on haemolytic activity of single
proline substitutions in the Bordetella pertussis CyaA pore-forming fragment, Arch.
Microbiol. 191 (2009) 1–9.
(2013) 277–321.
[36] M.S. Trent, L.M. Worsham, M.L. Ernst-Fonberg, The biochemistry of hemolysin
toxin activation: characterization of HlyC, an internal protein acyltransferase,
Biochemistry 37 (1998) 4644–4652.
[12] C. Kurehong, C. Kanchanawarin, B. Powthongchin, G. Katzenmeier,
C. Angsuthanasombat, Membrane-pore forming characteristics of the Bordetella
pertussis CyaA-hemolysin domain, Toxins 7 (2015) 1486–1496.
[37] P. Stanley, V. Koronakis, C. Hughes, Acylation of Escherichia coli hemolysin: a
unique protein lipidation mechanism underlying toxin function, Microbiol. Mol.
Biol. Rev. 62 (1998) 309–333.
[13] C. Kurehong, C. Kanchanawarin, B. Powthongchin, P. Prangkio, G. Katzenmeier,
C. Angsuthanasombat, Functional contributions of positive charges in the pore-
7