Subscriber access provided by UNIV OF MISSISSIPPI
Article
Converting transaldolase into aldolase through swapping of the multifunctional acid-base catalyst COMMON AND DIVERGENT CATALYTIC PRINCIPLES IN F6P ALDOLASE AND TRANSALDOLASE Viktor Sautner, Mascha Miriam Friedrich, Anja Lehwess-Litzmann, and Kai Tittmann Biochemistry, Just Accepted Manuscript • DOI: 10.1021/acs.biochem.5b00283 • Publication Date (Web): 01 Jul 2015 Downloaded from http://pubs.acs.org on July 8, 2015
Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Biochemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 34
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Biochemistry
3
Converting transaldolase into aldolase through swapping of the multifunctional acid-base catalyst COMMON AND DIVERGENT CATALYTIC PRINCIPLES IN F6P ALDOLASE AND
4
TRANSALDOLASE
1 2
5 Viktor Sautner, Mascha Miriam Friedrich, Anja Lehwess-Litzmann & Kai Tittmann * 6 7 8 9 10 11 12
Göttingen Center for Molecular Biosciences, Department of Molecular Enzymology, Georg-August University Göttingen, Germany, Ernst-Caspari-Haus, Justus-von-Liebig-Weg 11, D-37077 Göttingen, Germany Running title: Origins of reaction specificities in F6P aldolase and transaldolase * Correspondence to: Kai Tittmann,
13 E-mail:
[email protected], 14 phone: +49-551-3914430, 15 fax: +49-551-395749 16 17 18 19 20 21 22
Keywords: enzyme mechanism, enzyme catalysis, crystallography, kinetics, carbohydrate metabolism
23
ABBREVIATIONS
24
TAL, transaldolase; TacTAL, TAL from Thermoplasma acidophilum; EcTAL, TAL from Escherichia
25
coli; HsTAL, TAL from Homo sapiens; FSA, fructose-6-phosphate aldolase; F6P, D-fructose 6-
26
phosphate; G3P; D-glyceraldehyde 3-phosphate; E4P, D-erythrose 4-phosphate; S7P, D-sedoheptulose 7-
27
phosphate; DHA, dihydroxyacetone
28 29 30 31 32 33 34 35 36 37 38 1 ACS Paragon Plus Environment
Biochemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 2 of 34
39
ABSTRACT
40
Transaldolase (TAL) and fructose-6-phosphate aldolase (FSA) both belong to the class I aldolase
41
family, and share a high degree of structural similarity and sequence identity. The molecular
42
basis of the different reaction specificities (transferase versus aldolase) has remained enigmatic
43
so far. A notable difference between the active sites is the presence of either a TAL-specific Glu
44
(Gln in FSA) or a FSA-specific Tyr (Phe in TAL). Both residues seem to have analoguous
45
multifunctional catalytic roles but are positioned at different faces of the substrate locale. We
46
have engineered a TAL double variant (Glu-to-Gln and Phe-to-Tyr) with an active site
47
resembling that of FSA. This variant indeed exhibits aldolase activity as main activity with a
48
catalytic efficiency even larger than that of authentic FSA, while TAL activity is greatly
49
impaired. Structural analysis of this variant in complex with the dihydroxyacetone Schiff base
50
formed upon substrate cleavage identifies the introduced Tyr (genuine in FSA) to catalyze
51
protonation of the central carbanion-enamine as a key determinant of the aldolase reaction. Our
52
studies pinpoint that the Glu in TAL and the Tyr in FSA, although located at different positions
53
at the active site, similarly act as bona fide acid-base catalysts in numerous catalytic steps
54
including substrate binding, dehydration of the carbinolamine and substrate cleavage. We
55
propose that the different spatial positions of the multifunctional Glu in TAL and of the
56
corresponding multifunctional Tyr in FSA relative to the substrate locale are critically
57
controlling reaction specificity through either unfavorable (TAL) or favorable (FSA) geometry
58
of proton transfer onto the common carbanion-enamine intermediate. The presence of both
59
potential acid-base residues, Glu and Tyr, in the active site of TAL has deleterious effects on
60
substrate binding and cleavage, most likely resulting from a differently organized H-bonding
61
network. Large-scale motions of the protein associated with opening and closing of the active
62
site are observed that seem to bear relevance for catalysis as covalent intermediates are
63
exclusively observed in the ‘closed’ conformation of the active site. Pre-steady-state kinetics are
64
used to monitor catalytic processes and structural transitions, and to refine the kinetic framework
65
of TAL catalysis.
66
2 ACS Paragon Plus Environment
Page 3 of 34
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Biochemistry
67
Introduction
68
Transaldolase (TAL) and fructose-6-phosphate aldolase (FSA) both belong to the class I aldolase
69
family, which operate via Schiff base chemistry using an active center lysine to reversibly cleave
70
phosphoketose D-fructose-6-phosphate (F6P) in an aldol-type reaction 1-3. Transaldolase shuttles
71
a three-carbon dihydroxyacetone unit derived from F6P to aldose acceptor D-erythrose-4-
72
phosphate (E4P) affording ketose D-sedoheptulose-7-phosphate (S7P) and aldose D-
73
glyceraldehyde-3-phosphate (G3P) (eq 1) 1. FSA reversibly cleaves substrate F6P into G3P and
74
dihydroxyacetone (DHA) (eq 2)2. While F6P aldolase activity of FSA could be clearly
75
demonstrated in vitro, its physiological function is still unclear.
76
TAL:
F6P + E4P ⇔ S7P + G3P
(eq 1)
77
FSA:
F6P ⇔ G3P + DHA
(eq 2)
78
The catalytic cycles of both enzymes are highly analogous and proceed through a series of
79
identical covalent reaction intermediates including the dipolar and neutral carbinolamine, the
80
F6P Schiff base, and the carbanion-enamine with multiple critical proton transfers (Scheme 1) 3.
81
Key to the different reaction specificities of the two enzymes is the stability of the carbanion-
82
enamine intermediate formed upon cleavage of the common F6P Schiff base. While TAL retains
83
the covalent C3-lysine carbanion-enamine conjugate bound in stable form for eventual transfer to
84
aldose acceptor E4P (aldol addition), this intermediate undergoes facile hydrolysis in FSA
85
liberating DHA as product. Aldolytic cleavage requires stereospecific protonation at C3 of the
86
carbanion-enamine by a suitably placed acid-base catalyst followed by hydrolysis of the formed
87
DHA Schiff base (Scheme 1) 3. The different reaction specificities of the two enzymes is very
88
remarkable, because the overall structures as well as the active site architectures are highly
89
equivalent (Figure 1). FSA from E. coli has a similar overall structure (doughnut-shaped
90
homodecamer of (β/α)8 barrels, Z-score = 30, r.m.s.d. for Cα atoms = 1.5 Å for monomer), and
91
shows 31.1 % sequence identity and 60.9 % sequence similarity to TAL from T. acidophilum
92
(TacTAL), of which recently several structures were determined at different stages of the
93
catalytic cycle
94
substrate binding and catalysis are conserved in FSA (e.g.: Asp6, Asn28, Lys86, Arg135and
95
Thr110), some active site residues are replaced in FSA by either hydrophobic
96
(Ser130TAL→AlaFSA,
97
(Glu60TAL→GlnFSA, Phe132TAL→TyrFSA, Arg169TAL→LysFSA). The absence of Glu60 in the
98
active site of E. coli FSA (Gln59) along with the presence of Tyr131 (Phe132 in TacTAL) is of
4, 5
. While most of the active site residues of TacTAL with putative roles in
Asn108TAL→LeuFSA,
Ser58TAL→PheFSA)
or
homologous
residues
3 ACS Paragon Plus Environment
Biochemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 4 of 34
99
particular interest. Glu60 has been suggested to have a key catalytic role in multiple proton
100
transfers in TacTAL including dehydration of the carbinolamine and substrate cleavage at the
101
F6P Schiff base state by protonation of the 2-OH of the neutral carbinolamine and subsequent
102
deprotonation of 4-OH of the F6P Schiff base (see Scheme 1) 5. This mechanism seems to be
103
conserved in the different TAL subfamilies 6. The X-ray structures with bound substrate Schiff
104
bases indicated that the proton transfers between Glu60 and the intermediates are shuttled
105
through an adjacent catalytic water molecule (Figure 1). The absence of an equivalent Glu
106
residue in the active site of FSA is therefore surprising, since the acid-base chemistry required
107
for the aldol-cleavage of substrate F6P is also part of the reaction pathway of this enzyme. With
108
the exception of the conserved Asp6, only Tyr131 could potentially act as an acid-base catalyst
109
in the active site of FSA. Since the structural studies of intermediates in transaldolase were
110
suggesting a role of Asp6 for substrate binding and correct intermediate positioning with no
111
direct role in proton transfers (it H-bonds to 3-OH and 5-OH of the substrate), we hypothesized
112
that Tyr131 in FSA is a key acid-base catalyst of aldol-cleavage akin to Glu60 in the reaction
113
mechanism of TAL 5. Tyr residues are already known to play crucial roles in proton-transfer
114
reactions in other class I aldolases, most prominently as in case of mammalian fructose-1,6-
115
bisphosphate (FBP) aldolase 7, 8. In e.g. rabbit muscle fructose-1,6-bisphosphate aldolase, the C-
116
terminal tyrosine (Tyr363), which is located on a highly flexible segment, acts in concert with
117
Lys146 and Glu187 as acid-base catalyst during the hydrolytic cleavage of fructose 1,6-
118
bisphosphate into dihydroxyacetone phosphate and G3P 9. The deletion of this Tyr by treatment
119
of the protein with carboxypeptidase resulted in a decrease of aldolase activity and in an increase
120
of transaldolase activity
121
multifunctional role akin to the conserved Glu in transaldolases and FBP aldolases from
122
eukaryotes 11, 12. Recent studies demonstrated that an exchange of the aforementioned active site
123
Phe in TAL of E. coli and of human origin (Phe132 in TacTAL) by a Tyr greatly stimulated
124
aldolase activity, while at the same time TAL activity was impaired
125
TAL activity in these variants was still 30-40-fold higher than aldolase activity. The exact role of
126
this Tyr in substrate binding and catalysis of the Phe→Tyr TAL variants and in FSA remained,
127
however, unclear. Modeling of the reaction intermediates trapped in TacTAL into the active site
128
of FSA suggested a potential role of the introduced Tyr as acid-base catalyst in multiple proton
129
transfers (dehydration, substrate cleavage, protonation of carbanion-enamine)
130
hypothesis were correct, the absence of an additional acid-base catalyst such as the TAL-specific
131
Glu would likely be catalytically advantageous as two neighboring acid-base catalysts (a Glu and
10
. In archaeal aldolases, Tyr residues were shown to possess a
13
. Nonetheless, residual
3
. If this
4 ACS Paragon Plus Environment
Page 5 of 34
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Biochemistry
132
a Tyr) might obscure the required order and directionality of the numerous proton transfers due
133
to a differently organized hydrogen-bonding network at the active site.
134
Here, we have examined the catalytic role of this Tyr residue in the FSA mechanism while using
135
a transaldolase-derived ‘pseudo-FSA’ enzyme, in which the TAL-specific key catalytic residues
136
Glu60 and Phe132 in TacTAL were both changed to Gln60 and Tyr132 in order to restore an
137
active site that closely resembles that of FSA. Our study underscores the role of this Tyr as a key
138
acid-base catalyst in the FSA reaction using TacTAL as model by measurements of steady-state
139
and transient kinetics, and high-resolution structural analysis of the central DHA Schiff-base
140
intermediate for the first time in this enzyme family. Remarkably, introduction of both mutations
141
affords an enzyme variant (Glu60Gln/Phe132Tyr) with even higher catalytic efficiency for F6P
142
aldolase activity than authentic FSA from E. coli and any TAL Phe-to-Tyr single variant tested
143
so far. In broader context, our studies delineate the governing principles of how the relative
144
orientation of the multifunctional general acid-base catalyst to the reactive intermediate
145
determines the different reaction specificities in aldolases and transaldolases.
146 147
Materials and Methods
148
Mutagenesis, expression, purification and crystallization
149
Mutant strains producing TacTAL variants Phe132Tyr and Glu60Gln/Phe132Tyr were generated
150
by introduction of appropriate mutations using the QuikChange site-directed mutagenesis
151
protocol (Stratagene, La Jolla, CA). Expression and purification of TacTAL wt and of both
152
variants was carried out as described before
153
exhibits greatly increased aldolase activity, was crystallized according to the hanging-drop
154
vapor-diffusion method using a reservoir solution of 200 mM ammonium acetate pH 4.4, 10%
155
(w/v) PEG 6000 and 25% (v/v) glycerol. In order to obtain crystal structures with reaction
156
intermediates bound to the active site, TacTAL double variant Glu60Gln/Phe132Tyr (16 mg/mL
157
in 6 mM Tris-HCl/14 mM glycylglycine pH 7.5) was supplemented with 34 mM F6P prior to
158
crystallization; 3 µL of this solution were mixed with 3 µL of reservoir solution at room
159
temperature, and equilibrated against 250 µL of reservoir solution.
160
TacTAL crystallizes in two different space groups (P21 and C2221)
161
sufficiently large single crystals of TacTAL in the desired space group C2221, the crystallization
162
solution was micro-seeded with previously obtained TacTAL crystals of this space group, and
5, 14
. Double variant Glu60Gln/Phe132Tyr, which
14
. In order to grow
5 ACS Paragon Plus Environment
Biochemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 6 of 34
163
incubated at 30 °C for 10 min in a thermostatted incubator. Transaldolase crystals were then
164
further grown for 7 – 15 days at 20 °C.
165
X-ray data collection, processing and model building
166
Diffraction data of a single TacTAL crystal (Glu60Gln/Phe132Tyr variant) were collected using
167
synchrotron radiation at ESRF Grenoble (beamline ID23-1, wavelength 0.91 Å), France at
168
cryogenic temperatures at 100 K. Diffraction images were indexed, integrated, and scaled using
169
the XDS-package 15.
170
Model-building and refinement were performed using COOT and the PHENIX.REFINE
171
crystallographic package using the previously solved structure of TacTAL wild-type as a starting
172
model
173
MOLPROBITY-server 19 was used to verify the geometry of the final model, of which 97.93% are
174
in the favored region of the Ramachandran plot, 1.99% in the allowed region along with 0.09%
175
outliers. The representation of structures was performed with PyMOL (Schrödinger, LLC).
176
The structural alignment of monomers of TacTAL and FSA, and the calculation of Z-score and
177
r.m.s.d. were performed using the DALILITE-server 20 and T-COFFEE-server 21.
178
Database accession number
179
The final refined model and the corresponding structure factor amplitudes have been deposited in
180
the Research Collaboratory for Structural Biology (http://www.rcsb.org) under accession
181
numbers 4XZ9.
182
Steady-State Activity Assay
183
The steady-state kinetics for aldolase activity of TacTAL wt and variants (conversion of donor
184
F6P into G3P and DHA) were measured in a coupled spectrophotometric assay using the
185
auxiliary enzymes triosephosphate isomerase (TIM) and sn-glycerol-3-phosphate : NAD+ 2-
186
oxidoreductase (G3PDH) to detect formation of G3P formed by aldolytic cleavage of F6P 22. The
187
concomitant oxidation of NADH was monitored spectrophotometrically in a UV-Vis
188
spectrometer (V-650, Jasco GmbH, Gros-Umstadt) at 340 nm at 30 °C. The reaction mixture
189
contained 20 mM glycyl-glycine, pH 7.5, 3 U/mL TIM/G3PDH (8 mM (NH4)2SO4), 0.22 mM
190
NADH and varying concentrations of F6P (0.1 - 100 mM). The enzyme concentrations were 1
191
mg/mL (TacTAL wt), 0.17 mg/mL (TacTAL Phe132Tyr) and 0.035 mg/mL (TacTAL
192
Glu60Gln/Phe132Tyr), respectively.
16-18
. TLS grouping for each monomer was implemented as recently described 5. The
6 ACS Paragon Plus Environment
Page 7 of 34
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Biochemistry
193
Transaldolase activity was analyzed under the same conditions except that acceptor E4P was
194
added to the assay in concentrations up to 2 mM.
195
In order to determine the macroscopic kinetic constants (kcat, KM), the initial rates were plotted
196
against substrate concentration and analyzed using the Michaelis-Menten equation.
197
Pre-steady state kinetics
198
The pre-steady state of the donor-half reaction (cleavage of donor F6P) was analyzed by
199
stopped-flow kinetics using a coupled spectrophotometric assay as described before. The enzyme
200
solution (0.44 mg/mL TacTAL, 200 U/mL TIM/G3PDH, 0.44 mM NADH in 20 mM glycyl-
201
glycine, pH 7.5) and substrate solution (0.1-100 mM F6P in 20 mM glycylglycine, pH 7.5) were
202
rapidly mixed using a stopped-flow spectrometer (SX20, Applied Photophysics, UK) in a 1+1
203
mixing ratio at 30 °C. Progress curves were fitted with appropriate equations consisting of either
204
one or two single exponential terms (pre-steady state phase) and a linear term (steady-state
205
consumption of F6P) (vide infra).
206
The estimated rate constants kobs of the pre-steady-state phase were plotted versus the applied
207
F6P concentration, and data were fitted to either a hyperbolic equation (eq 3) for F6P
208
concentrations up to 12 mM or a hyperbolic equation taking into account substrate access
209
inhibition (eq 4).
210
obs =
211
obs =
max obs ∙[F6P]
S
(eq 3)
[F6P]
max obs
S [F6P]
(eq 4)
[F6P] I
212
where kobsmax denotes the rate constant at infinitely high F6P concentrations, KSapp the
213
equilibrium constant of the fast substrate binding (pre-)equilibrium, and KI the inhibition
214
equilibrium constant of a putative substrate access inhibition.
215 216
Results and Discussion
217
Steady-state kinetic analysis of aldolase and transaldolase activity
218
TacTAL wt and TacTAL variants Phe132Tyr and Glu60Gln/Phe132Tyr could be recombinantly
219
expressed in E. coli and purified to homogeneity. The expression yields of the two variants were 7 ACS Paragon Plus Environment
Biochemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 8 of 34
220
not as high as in case of the wt enzyme (~2 mg enzyme/g cell pellet), and amounted to 0.8 mg enzyme/g cell
221
pellet
222
We first assessed the steady-state kinetic properties of all three proteins for F6P aldolase activity
223
(F6P as sole substrate) and TAL activity (F6P and E4P as substrates) using a spectrophotometric
224
assay that detects liberated product G3P. TacTAL wt exhibits a very small, yet detectable F6P
225
aldolase activity that corresponds to a kcat of ~2.2·10-3 s-1 on a per active site basis (Table 1).
226
However, since very high concentrations of enzyme were required in the assay, true steady-state
227
(multiple turnover) conditions cannot be established at small F6P concentrations precluding a
228
reliable estimation of the KM F6P value. Notably, the aldolase activity of TacTAL wt is three to
229
four orders of magnitude smaller than its genuine TAL activity (kcat ~14 s-1). Single variant
230
Phe132Tyr exhibits a 102-fold stimulated F6P aldolase activity (kcat = 0.24 ± 0.01), and a KM
231
value for F6P (12.3 ± 1.5 mM) that is comparable to that of FSA (Table1, Figure 2). The
232
catalytic efficiency kcat/KM of TacTAL Phe132Tyr is approximately 5% of that of FSA. TAL
233
activity of this variant is greatly impaired, addition of acceptor E4P (tested up to 2 mM at
234
different F6P concentrations) to the F6P-containing assay did not lead to a measurable change of
235
the reaction rate suggesting that the TAL activity is smaller than (or equal to) FSA activity. This
236
translates into a reduction of TAL activity by a factor of at least ~60 in this variant. Introduction
237
of the second mutation in the Glu60Gln/Phe132Tyr double variant increases F6P aldolase
238
activity by an additional factor of three (kcat = 0.62 ± 0.01s-1), and at the same time decreased the
239
apparent Michaelis constant of F6P (KM = 1.5 ± 0.1 mM) compared to the Phe132Tyr single
240
variant (Table 1, Figure 2). Addition of acceptor E4P (tested up to 2 mM) to the assay containing
241
F6P resulted in a decrease of the reaction rate by up to 10% showcasing that i) TAL activity is
242
smaller than FSA activity and ii) that bound E4P slightly impairs protonation of the carbanion-
243
enamine and/or hydrolysis of the C3 Schiff base (see Scheme 1). Notably, the catalytic
244
efficiency of F6P aldolytic cleavage for TacTAL double variant Glu60Gln/Phe132Tyr (413 M-1
245
s-1) is the highest ever reported value, it is higher than these of authentic FSA and of Phe→Tyr
246
single variants from E. coli and human TAL (Table 1)
247
variants exhibited markedly elevated FSA activity, their residual TAL activity was still 30-40-
248
fold higher than FSA activity. In case of TacTAL double variant Glu60Gln/Phe132Tyr, FSA
249
activity is clearly higher than the residual TAL activity as indicated by the observed inhibitory
250
effect of E4P on F6P consumption.
for Phe132Tyr or even 0.3 mg enzyme/g cell pellet in case of Glu60Gln/Phe132Tyr.
13
. Although these previously studied
251 252
Pre-steady-state kinetic analysis of F6P conversion
8 ACS Paragon Plus Environment
Page 9 of 34
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Biochemistry
253
In order to gather more insights into the kinetics of TacTAL-catalyzed F6P cleavage, we
254
conducted stopped-flow experiments, in which TacTAL was mixed with substrate F6P. In a
255
minimal reaction scheme (Scheme 2), aldolytic cleavage of F6P into products G3P and DHA can
256
be subdivided into two half-reactions: i) F6P binding and cleavage (all steps until liberation of
257
G3P), and ii) protonation and hydrolysis of the C3 carbanion-enamine with concomitant release
258
of DHA. Since midway formed product G3P is detected in the spectrophotometric assay, the
259
potential occurrence or absence of a pre-steady-state phase is diagnostic for which of the two
260
half-reactions is rate-determining for overall catalysis akin to the classic kinetic analysis of
261
protease-catalyzed turnover of p-nitrophenyl-ester substrates 23, 24.
262
When TacTAL wt was mixed with substrate F6P in the stopped-flow instrument, an initial pre-
263
steady-state burst phase was observed in the first 100 ms followed by a linear decrease in
264
absorbance in the steady state (Figure 3). This observation indicates that the second half-reaction
265
(protonation of the carbanion-enamine, hydrolysis of DHA Schiff base and DHA release) is rate-
266
determining for the overall reaction, whereas the first half-reaction that is binding and cleavage
267
of F6P proceeds considerably faster. The fitted first-order rate constants kobs of the burst phase
268
display a hyperbolic dependence on the substrate concentration up to 10 mM with a KSapp of 0.49
269
± 0.06 mM, at higher F6P concentrations, however, rate constants are getting smaller. The reason
270
for the slowed substrate binding/cleavage at high F6P concentrations is unclear, this observation
271
could reflect substrate access inhibition or, alternatively, a negative cooperativity between the
272
different monomers (TacTAL forms homopentamers assembling into decamers) as suggested by
273
previous structural studies
274
intermonomer negative cooperativity is the finding that the maximal amplitude of the burst phase
275
corresponds to turnover of only 60% of the active sites indicating that at max only 3 out of 5
276
monomers of the pentamer are acting on F6P. Alternatively, a switch between an induced fit and
277
a conformational selection mechanism at higher F6P concentrations might account for the
278
observed dependence of kobs from the F6P concentration 25. The maximal rate constant kobs of the
279
burst phase amounts to ~50-55 s-1 (at cF6P 5-15 mM), which is greater then the turnover number
280
for the overall TAL reaction (kcat ~14 s-1). This pinpoints that all reaction steps until formation
281
of the carbanion-enamine are not rate-determining for the TAL reaction. This in turn suggests
282
that aldol addition of the carbanion-enamine with acceptor E4P and subsequent liberation of
283
product S7P must be (at least partially) rate-determining for the overall TAL reaction. As
284
expected, off-pathway protonation of the carbanion-enamine and subsequent DHA liberation
285
proceed extremely slow in TacTAL wild-type, and are rate-determining for conversion of F6P
286
into G3P and DHA (k’ = 2.25·10-3 s-1).
5
. In support of this mechanistic proposition invoking an
9 ACS Paragon Plus Environment
Biochemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 10 of 34
287
In TacTAL single variant Phe132Tyr, no burst phase can be observed in the stopped-flow kinetic
288
analysis of F6P conversion (Figure 3). As it could be established that a burst is not occuring in
289
the dead time of the stopped-flow experiment (t ~1.5 ms), this observation indicates that the
290
single site Phe132→Tyr substitution in TacTAL greatly impairs one or more step(s) of the first
291
half-reaction comprising all steps until formation of the carbanion-enamine and concomitant
292
G3P release. Owing to this kinetic behaviour, the first half-reaction that is binding and cleavage
293
of F6P has become rate determining in the variant (kobs ≤ 0.24 s-1), and thus proceeds at least
294
200-fold slower than in the wild-type enzyme (kobsmax = 55 s-1). This drastic effect implies that
295
the Phe→Tyr single site mutation has a large impact on the active site structure and/or reactivity.
296
TacTAL double variant Glu60Gln/Phe132Tyr exhibits a pre-steady-state phase that is similar to
297
that of TacTAL wt as a pre-steady-state burst phase preceding a linear decrease in the steady
298
state can be observed (Figure 3). However, in addition, a lag phase can be monitored in the very
299
first 50 ms of the reaction. Notably, the duration of this lag phase is not dependent on the applied
300
substrate concentration suggesting that this kinetic phase reflects an intrinsic process of the
301
enzyme. There is precedent in the literature for substrate-independent lag phases as e.g. in case
302
of glucanase 26. Since the previous structural analysis of TacTAL showed the enzyme to adopt an
303
open (presumably inactive) and closed (presumably active) conformation with an equilibrium
304
position close to unity in the resting state 5, one might speculate that the observed lag phase
305
corresponds to the equilibrium shift between the two states upon substrate binding according to a
306
conformational selection mechanism (Scheme 2) 27, 28. The fitted first-order rate constants of the
307
burst phase show a similar dependence on the F6P concentration as described above for the wild-
308
type enzyme with a kobsmax of ~50 s-1 (e.g. at 10 mM F6P kobs = 46.9 ± 1.4 s-1, see Figure 3). The
309
preceding lag phase can be reasonably fitted at F6P concentrations > 5 mM with a single
310
exponential function and amounts to 50-70 s-1 for the individual progress curves (e.g.
311
kobslag = 67.2 ± 3.7 s-1 at 10 mM F6P, see Figure 3). Admittedly, at lower F6P concentrations, kobs
312
of the lag phase cannot be reliably estimated because of the very small amplitude of this phase
313
such that a precise value cannot be determined with this type of experiments. In sum,
314
introduction of the second mutation (Glu60→Gln) in the TacTAL double variant
315
Glu60Gln/Phe132Tyr restored the active site in a way that F6P binding and cleavage are as
316
efficient as in the wild-type enzyme, whereas the single variant Phe132Tyr with both potential
317
acid-base catalysts at the active site (Glu60, Tyr132) is greatly impaired in that regard. The 2nd
318
half-reaction of aldolytic F6P cleavage (see Scheme 2) in TacTAL Glu60Gln/Phe132Tyr is rate
319
determining for the overall reaction (kcat = 0.62 s-1, see Table 1) pinpointing that either
320
protonation of the carbanion-enamine, hydrolysis of the C3 (DHA) Schiff base and/or product 10 ACS Paragon Plus Environment
Page 11 of 34
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Biochemistry
321
release are the slowest steps of catalysis. Interestingly, we can observe kinetic processes (initial
322
lag phase) with this variant that are independent of the applied substrate concentration tempting
323
us to assign these processes as intrinsic protein dynamics/conformational changes of reversible
324
opening⇔closing and substrate binding according to a conformational selection mechanism as
325
previously suggested 5. We did not observe this lag phase in case of the wild-type TAL enzyme
326
suggesting that these processes proceed considerably faster and cannot be resolved by stopped-
327
flow experiments (dead time 1.5 ms). Although F6P is efficiently bound and cleaved in the
328
double variant (kobs ~50 s-1, see Figure 3), TAL activity is greatly impaired as evidenced by the
329
inhibitory effect of E4P onto F6P consumption (kcat
