JOurnal. ofr ILn.er,.r.3d. u.3- Re-:-earch
,,Oluiie 6, issue 4 - C:eceini 'er 2111:14
Crystal Structure of Mutant HCAII H64A Complexed with Bicarbonate:
Its Implication in Proton Shuttling Mechanism
Yashash Pathak, Zoe Fisher, Robbie Reutzel, and Robert McKenna
Carbonic anhydrases catalyze the reversible hydration reaction CO2 + H20 fl <--> HCO3- + H+2. In HCAII,
the residue His64 is primarily responsible as the proton acceptor/donor shuttle, which is the rate limiting step
in catalysis. The proton is shuttled through a hydrogen bonded solvent network. Crystals of HCAII H64A were
grown using the sitting drop method by vapor diffusion and soaked in 3 M NaHCO3 at 4oC. Diffraction data
were collected to 2.00 A resolution on an in-house R-AXIS IV++ image system. The crystal space group
was determined to be P21 with unit cell parameters a = 42.7, b = 41.6, c = 72.9 and _ = 104.60. Using the
known crystal structure of HCAII H64A as a molecular-replacement model, the bicarbonate crystal data was
phased. The structure was then refined (rigid body, anneal, b-factor, and minimize) and water-pick method
selected solvent in Foo-Fc electron density maps with R-factor of 0.17. Structural analysis of the active site
residues, and binding site of the HCO3- substrate, as well as the solvent network, has implications for the
mechanism of CA activity.
The carbonic anhydrases (CA's) are zinc containing enzymes that catalyze the reversible hydration reaction CO2
+ H20 fl <--> HCO3- + H+. This process has many physiological roles including respiration, acid/base
homeostasis, and gastric acidification (Dodgson et al., 1991).
The most intensely studied of the a-CA isozymes is CA II which is found in erythrocytes where it mediates
respiration by converting CO2 to HCO3- and through its interaction with hemoglobin to facilitate oxygen release to
the tissues. CA II represents a substantial amount of the protein mass of erythrocytes with 2mg/g
hemoglobin (Lindskog, 1997). There have been at least 14 human a-CA isozymes identified (Parkkila, 2000). CA II
is also the most efficient isozyme of the a-class with a catalytic turnover number 106s-1 (Hewett-Emmett &
Human Carbonic Anhydrases II (hCA) catalyses HCO3-/ CO2 reaction in a two-stage mechanism.
HCO3- + EZn H20 fl <--> CO2 + EZnOH- + H20 (Eq. 1)
EZnOH- + BH+ fl <--> EZnH20 + B (Eq. 2)
The zinc bound water (in eq. 2) is regenerated through a proton transfer mechanism. BH+ represents a proton
donor, which is an amino acid Histidine 64 (His64) within the wildtype enzyme.
This intramolecular proton transfer occurs between the zinc-bound H20 and residue His64, which accepts the H+.
The intermolecular proton transfer procedures with the imidazolium ion of His64 transferring its H+ to the
external buffer. Proton transfer between the zinc ion and His64 is not direct and involves the participation of
several hydrogen-bonded water molecules (Silverman & Lindskog 2000). The crystal structure of wildtype HCA II
at 2.0 A showed that there are two to three water molecules spanning the 7.5 A distance between the zinc-
bound hydroxyl and His64 (Eriksson et al., 1988).
The kinetic studies of mutant HCAII H64A have shown the reduction in proton-transfer activity (kcat) from 1 _ 106
s-1 in wild type to 1 _ 103 s-1 in the mutant (Tu et al., 1989). Therefore, the decrease in kcat of the mutant
enhances our chance to "see" bicarbonate binding sites. In this study, we will find the bicarbonate binding site
in H64A mutant to elucidate the mechanism of CA activity.
MATERIALS AND METHODS
The mutant H64A of HCAII was prepared using bacterial expression vectors that are used for the site-
specific mutagenesis and protein synthesis (Tanhauser et. al., 1992). The prepared mutant of HCA II was purified
by affinity chromatography (Khalifah et. al., 1977). Concentration of the enzyme was determined from
molar absorptivity at 280 nm and 5.4 _ 104 M-1 cm-1 (Duda et. al., 2001).
Crystals of CAII H64A were produced in limbro plates using the sitting drop vapor diffusion method
(McPherson, 1982). The protein concentration was 10 mg/ml in the solution of 50 mM Tris-HCI at pH 7.8 (Duda
et. al., 2001). The drops were prepared by mixing 5 pl of protein and 5 pl of precipitant solution. The drops
were equilibrated by vapor diffusion against 1 mL of precipitant solution which consisted of 2.4-2.65 M (NH4)2S04
in 50 mM Tris-HCI at pH 7.8. The crystal limbro plate was stored at 40C for two weeks.
Crystals of CAII H64A completed with bicarbonate were obtained by soaking them in 3 M NaHCO3 for one week.
Data Collection and Processing
X-ray diffraction data were collected using R-Axis IV++ image plate system with Osmic mirrors and a
Rigaku generator operating at 50 kV and 100 mA as seen in Figure 1. A 1200 sweep of data was collected from
a single crystal with a crystal-to-detector distance 150 mm and 10 oscillation angle with exposure time of 2
min/frame. The data set was indexed using Denzo and scaled with Scalepak to generate data for phasing
(Otwinowski et. al., 1997).
Figure 1. The diffraction pattern of the crystal of CAII H64A completed with bicarbonate.
A model of CAII H64A was refined using the software package CNS (Brunger et. al., 1998) and a model was
built using the program O version 7 (Jones et. al., 1991). The known structure of HCAII H64A at 1.0 A
resolution (pdb code, 1MOO) was used as the starting phase model for the data set (Duda et. al., 2002). All
water molecules and zinc were removed from the starting model to remove any model phase bias. After one cycle
of rigid body, anneal, and minimize refinement, the Fo-Fc maps showed the position of zinc and bicarbonate
ions, which were assigned to the model. This model was annealed and refined by heating to 3000 K and
gradually cooling, followed by temperature factor refinement with CNS. The solvent molecules were found using
the automatic water-picking program. Comparison of structure of HCAII H64A with and without bicarbonate was
done by superimposing both structures and analyzing the solvent network in the active site in the PyMOL
program (DeLano 2002).
RESULTS AND DISCUSSIONS
In the crystal limbro plate, each well had crystal growth; however, the lower precipitant condition (2.4 -2.5 M
(NH4)2SO4) had bigger and better diffracting crystals. The dimensions of the crystals were 0.3mm _ 0.3mm
0.7mm as shown in Figure 2.
Figure 2. Crystal of HCAII H64A with bicarbonate
Data Collection and Processing
A 1200 sweep of data was collected from a single crystal with a crystal-to-detector distance 150 mm
and 1 ooscillation angle with exposure time of 2 min/frame. The crystal space group was P21 with
unit cell parameters a = 42.7, b = 41.6, c = 72.9, and _ = 104.60 ; data completion was 75% with Rsym
= 0.16 as shown in Table 1.
[insert table 1]
Data Collection and Processing
Exposure time / frame
# of images
Rwork / Rfree
Final Rwork / Rfree
Prior to assignment of zinc and bicarbonate ions to the model, Rwork and Rfree were 0.27 and
0.28, respectively. After simulated annealing in the presence of ions, the Rwork and Rfree were 0.22
respectively (refinement R-factors have been recorded in Table 1). Several rounds of
refinement processes involved energy minimization, water pick and molecular modeling. At this
stage 138 water molecules had been incorporated. The final refined structure had the R-factor of 0.17.
Structure and the Active Site
Human Carbonic anhydrase II has 261 residues with ten-stranded twisted _-sheet, which is flanked
by seven _-helices. The catalytic active site is characterized by a conical cleft that is 15 A deep. The
zinc ion is tetrahedrally coordinated by three histidines (94, 96 and 119) and a water/
hydroxide molecule as seen in Figure 3-a (Duda et al., 2001).
Ilis 94 WHis 94
Zn | I IZis 96
Sicor I Ii-q 9
AlAla 64 O
Figure 3. Drawing of the active site residues, bicarbonate and water network using the PyMOL
program (a) The structure of HCAII H64A without the bicarbonate (b) HCAII H64A completed
with bicarbonate (this paper) (c) Superimposing both structures a-b. Zn is the Zinc ion in the active
site coordinated by three histides (94, 96, 119). Wat are the waters showing the solvent network in
the active site (Wat 82,118,22,36 and 137). Watl37 seen in (b) is absent in (a).
The distance between zinc and the waters network and between bicarbonate and the waters network
and are recorded in Table 2 and Table 3. It is interesting to note that bicarbonate binds bivalent to
the zinc, disrupting its normal tetrahedral configuration in the active site.
Active site geometry around Zinc ion
(distances in A)
Solvent Network (distances in A)
Wat82- HC03-300 (03)
Wat36 - Ala64 (0)
It has been proposed (Cox et al., 2000) that the zinc-bound hydroxyl ion hydrogen bonds to the
hydroxyl side chain of Thrl99 in wild type (not in Figure 3). The structure completed with bicarbonate
in H64A mutant shows HCO3- (01) atom hydrogen binds with Thrl99, which is 2.53 A away.
The bicarbonate binding supports that Thrl99 serves to stabilize the transition state of reaction
and serves to destabilize the bicarbonate ion product (Christianson & Cox 1999). In addition, HCO3-
(02) hydrogen bonds to the amide backbone of the Thrl99, supporting the proposed function of
amide group of Thrl99 in substrate binding and in polarization of bound CO2 (Huang et al., 2002).
The maximum decrease in Kcat/Km relative to the wild type for mutants maintaining the deep water
is about 100-fold, whereas without deep-water display, the minimum decrease is about 1000-fold
(Huang et al. 2002). In this structure, the deep water is absent and replaced by HCO3- (03),
which hydrogen binds to the N atom of Thrl99 instead. This 03 atom of the bicarbonate is thought to
be (-OH) group of the bicarbonate ion. Also the Watl37 is moved in the bicarbonate structure
compared to the CAII H64A structure without bicarbonate as seen in Figure 3.
As in Figure 3, Wat82 hydrogen binds to Watl18, which in turn hydrogen binds to Wat 36. This
Wat36 donates a hydrogen bond to 0 atom of main chain of Ala64. However, this solvent network is
weak likely due to the inability of Ala64 to act as proton shuttle, which explains the reduced
activity (Kcat) of HCAII H64A.
We report here the structure of HCAII H64A completed with bicarbonate at 2.00 A resolution.
The bicarbonate binding site found supports the proposed mechanism of CA as seen in Figure 4.
A , B C
F ....-ti--- -. - o f- hu-man a an ydas .- he
b d. e in H i te a nd 2
" ":EHO HH,
ï¿½--)j .... H HK. ...
C <4430 "L9 H h "110' - '- TU 1Wj
Wo H H
El H '.64 H T
Figure 4. Reaction mechanism of human carbonic anhydrase II. Dashed lines indicate hydrogen
bonds. Reaction intermediates correspond to equations 1 and 2 below. Zinc ion coordinated by
three histidine ligands is represented by a sphere with three extended lines. (A) EZnOH (B) EZn(OH)
CO2 (C) EZnHCO3- (D) EZnHC03- (E) EZnH20 (F) H+EZnOH-. Figure adapted from Lindskog &
The bivalent binding of bicarbonate in the structure is supported by step (D) in Figure 4; however,
the H20 molecule is absent in the bicarbonate structure. Ala64 is unable to be a proton acceptor/
donor shuttle, which may explain the decrease in activity (kcat) seen in the H64A mutant of CAIZ.
1. Briinger, A. T., Adams, P. D., Clore, G. M., DeLano, W. L., Gros, P., Grosse-Kunstleve, R. W., Jiang, J.
S., Kuszewski, J., Nilges, M., Pannu, N. S., Read, R. J., Rice, L. M., Simonson, T. & Warren, G. L.
(1998). Acta Cryst. D54, 905-921.
2. Christianson, D. W. & Cox, J. D. (1999) Annu. Rev. Biochem. 68, 33-57.
3. Cox, J. D., Hunt, J. A., Compher, K. M., Fierke, C. A. & Christianson, D. W. (2000) Biochemistry 39,
4. DeLano, W. L. The PyMOL Molecular Graphics System (2002) DeLano Scientific, San Carlos, CA,
5. Dodgson, S. J., Tashian, R. E., Gros, G., and Carter, N. D. (1991) "The Carbonic Anhydrases"
Plenum Press, New York.
6. Duda D., Govindasamy L., McKenna M., Tu, C., Silverman, D., and McKenna, R. (2002) Acta Cryst. 59,
7. Duda D.,Tu, C., Qian, M., Laipis, P., Mckenna, M., Silverman, D., and Mckenna, R. (2001) Biochemistry
8. Eriksson, A. E., Jones, T. A., and Liljas, A. (1988). Proteins Struct. Funct. Genet. 4, 274-282.
9. Hewett-Emmett, D., and Tashian, R. E. (1996) Mol. Phylo-genet. Evol. 5, 50-77.
10. Huang, S., Sjoblom, B., Sauer-Eriksson, A. E., Jonsson, B.-H. (2002) Biochemistry 41, 7628.
11. Jones, T.A., Zou, J.Y., Cowan, S.W. and Kjeldgaard, M. (1991). Improved methods for building
protein models in electron density maps and the location of errors in these models. Acta Cryst. A47,
12. Khalifah, R. G., Strader, D. J., Bryant, S. H., and Gibson, S. M. (1977) Biochemistry 16, 2241-2247.
13. Lindskog, S. (1997). Pharmacol. Ther. 74, 1-20.
14. McPherson, A. (1982) Preparation and analysis of protein crystals, Wiley and Sons, New York.
15. Otwinowski, Z., and W. Minor. "Processing of X-ray Diffraction Data Collected in Oscillation
Mode", Methods in Enzymology, Vol. 276: Macromolecular Crystallography, part A, 307-326, 1997, C.
W. Carter, Jr. & R. M. Sweet, Eds., Academic Press.
16. Parkkila, S. (2000). The Carbonic Anhydrases: New Horizons, edited by W. R. Chegwidden, N. D. Carter
& Y. H. Edwards, pp. 79-94. Berlin: Birkhauser Verlag.
17. Silverman, D. N., and Lindskog, S. (2000). The Carbonic Anhydrases; New Horizons, Birkhauser
18. Tanhauser, S. M., Jewell, D. A., Tu, C. K., Silverman, D. N., Laipis, P. J. (1992) Gene 117, 113-117.
19. Tu, C. K., Silverman, D. N., Forsman, C., Jonsson, B. H. & Lindskog, S. (1989) Biochemistry 28,
Back to the Journal of Undergraduate Research
College of Liberal Arts and Sciences I University Scholars Program I University of Florida I
ï¿½ University of Florida, Gainesville, FL 32611; (352) 846-2032.