Stabilizing and destabilizing effects of phenylalanine → F 5-phenylalanine mutations on the folding of a small protein

Article abstract of DOI:10.1021/ja0634573

We report a systematic evaluation of phenylalanine-to-pentafluorophenylalanine (Phe → F₅-Phe) mutants for the 35-residue chicken villin headpiece subdomain (c-VHP), the hydrophobic core of which features a cluster of three Phe side chains (residues 6, 10, and 17). Phe → F₅-Phe mutations are interesting because aryl-perfluoroaryl interactions of optimal geometry are intrinsically more favorable than aryl-aryl interactions and because perfluoroaryl units are more hydrophobic than are analogous aryl units. One mutant, Phe-10 → F₅-Phe, provides enhanced tertiary structural stability relative to the native sequence. The other six mutants analyzed caused a decrease in stability. Copyright

Full text of DOI:10.1021/ja0634573

Published on Web 11/22/2006
Stabilizing and Destabilizing Effects of Phenylalanine f F5-Phenylalanine
Mutations on the Folding of a Small Protein
Matthew G. Woll,† Erik B. Hadley,† Sandro Mecozzi,†,‡ and Samuel H. Gellman*,†
Department of Chemistry and School of Pharmacy, UniVersity of Wisconsin, Madison, Wisconsin 53706
Received May 17, 2006; E-mail: gellman@chem.wisc.edu
Strategies that stabilize protein folding patterns and/or protein
associations are useful for achieving biomolecular engineering goals.
Amino-acid residues bearing fluorinated side chains have been
explored for this purpose,1 and several groups have reported that
fluorination of saturated carbon atoms leads to enhanced tertiary
and/or quaternary structural stability.2 This stabilization has usually
been attributed to enhanced hydrophobicity of fluorocarbons relative
to hydrocarbons, and, where appropriate, to a preference for
fluorocarbon-fluorocarbon interaction relative to fluorocarbon-
hydrocarbon interaction. Little attention has been paid to the impact
of fluorinated aromatic side chains on protein structure.3 Phe f
F5-Phe mutation at buried sites could stabilize protein folding
patterns via two distinct mechanisms. First, the mutant could benefit
from the enhanced hydrophobicity of F5-Phe.4 Second, if the Phe
side chain contacts other aromatic side chains in the native state,
then aromatic-aromatic attraction (e.g., from quadrupolar interac-
tions) could be enhanced in the mutant; the optimal C6H6/C6F6
stacking energy is estimated to be -3.7 kcal/mol.5 On the other
hand, the modest increase in size resulting from H f F replacements
(molecular volumes: C6H6, 106 Å3; C6F5H, 141 Å3)6 could cause
destablization if steric repulsions develop. Because fluoroalkyl
substitutions have stabilized higher-order protein structure in many
cases,1,2 it is important to evaluate the effects of fluoroaromatic
substitutions.
Figure 1. (A) Ribbon diagram of folded cVHP,6d with helical backbone
Here we document the impact of Phe f F5-Phe mutations on
the conformational stability of the chicken villin headpiece sub-
domain (cVHP), a 35-residue module that adopts a discrete tertiary
structure.7 The nonpolar core features a cluster of three Phe side
chains, from residues 6, 10, and 17 (Figure 1a). Examination of all
Phe f F5-Phe mutants involving the three core residues has allowed
us to survey the effect of aryl f fluoroaryl replacement in several
specific contexts and thus acquire a general appreciation of such
modifications.
Backbone thioester exchange (BTE)8 was used to compare
conformational stabilities among all possible Phe/F5-Phe combina-
tions at positions 6, 10, and 17. BTE is implemented by replacing
a single backbone amide linkage in a polypeptide with a thioester
(i.e., a single R-amino-acid residue is replaced by an R-thio-acid
residue). Conformational stability can then be probed by monitoring
a thioester-thiol exchange process: equilibration between a full-
length molecule and peptide fragments that cannot form a complete
tertiary core.9 Mixing t-cVHP (Figure 2) with the small thiol
indicated in Figure 1c, for example, induces equilibration with a
thioester/thiol pair that corresponds to N-terminal and C-terminal
fragments of the protein (Figure 1b). The equilibrium constant
(KBTE) provides insight on the free energy of tertiary structure
formation. We have previously shown that BTE is very useful for
studying small protein mutagenesis,8 which frequently causes
incomplete folding; the conformational stability of partially folded
segments in red. The side chains of residues Phe-6, Phe-10, and Phe-17
are shown in space-filling format; all other side chains are omitted. Graphic
B shows the backbone thioester exchange (BTE) process for t-cVHP
(sequence in Figure 2). The thioester-thiol pair on the left is composed of
fragments that contain residues 1-10 (green) and 11-35 (red), while the
pair on the right contains full-length t-cVHP and a small thiol, the structure
of which is shown in graphic C.
structures can be challenging to quantify via traditional methods10
but is readily assessed with BTE.
To implement BTE, we replaced the amide bond between Phe-
10 and Gly-11 of cVHP with a thioester (Figure 2). Splitting the
backbone at this point separates the first R-helical segment from
the other two R-helices, which prevents tertiary folding. Preliminary
studies, however, revealed that the cVHP-derived thioester was
irreversibly depleted by acylation of a lysine side chain. This
problem was eliminated by replacing all five lysine residues in
cVHP with arginine. In order to evaluate the cumulative effects of
the five Lys f Arg mutations plus the backbone amide f thioester
replacement on cVHP conformational stability in the native state,
we used classical methods to compare the conformational stability
of three molecules: (1) cVHP, (2) the mutant with the five Lys f
Arg modifications but a purely amide backbone (m-cVHP), and
(3) the hybrid polypeptide containing the five side-chain modifica-
tions along with the thioester replacement in the backbone (t-cVHP)
(Figure 2a). These three have similar circular dichroism (CD)
signatures,6 and all three molecules displayed cooperative dena-
turation upon addition of guanidinium chloride. Extrapolation of
these data to zero GdmCl concentration11 provided estimates of
† Department of Chemistry.
‡ School of Pharmacy.
9
15932
J. AM. CHEM. SOC. 2006, 128, 15932-15933
10.1021/ja0634573 CCC: $33.50 © 2006 American Chemical Society

C O M M U N I C A T I O N S
by BTE faithfully mirrors that of the full-fledged polypeptides. In
both series the Phe-10 f F5-Phe mutant is most stably folded.12
Our systematic examination of all F5-Phe variants involving core
Phe residues in a small folded protein has revealed a specific
mutation that enhances tertiary structural stability. Six of the seven
F5-Phe-containing mutants, however, proved to be less stably folded
than the version containing the three native Phe residues. This trend
differs from that seen with fluoroalkyl side chain substitutions,
which have been stabilizing in several different systems.2 Our
observations suggest that it is difficult for the cVHP tertiary fold
to take advantage of stabilization mechanisms that become available
when F5-Phe is introduced. The destabilizing effects could reflect
steric repulsions that arise from the increase in side-chain volume
and/or from minor conformational rearrangements that might be
necessary for harnessing improved side-chain interactions. Identify-
ing the precise mechanisms of destabilization induced by each
mutation pattern may be difficult or impossible, as is also generally
the case with stabilizing mutations. Nevertheless, our results are
valuable in the context of protein engineering because they show
that introduction of fluoroaryl side chains can stabilize folded
conformations, but that stabilization is not a general effect of Phe
f F5-Phe mutation.
Figure 2. (A) Sequence comparison for cVHP, m-cVHP, and t-cVHP; (B)
correlation of ∆Gfold/Gdm determined for four polypeptides (cVHP and the
three single Phe f F5-Phe mutants at position 6, 10, or 17) and ∆Gfold/BTE
determined for an analogous set of four thioester-containing molecules (t-
cVHP and the three single Phe f F5-Phe mutants at position 6, 10, or 17).
Linear regression was used to determine the line of best fit.
Table 1. Comparison of ∆G fold/BTE for VHP-Derived Thioestersa
peptide
∆G fold
/BTE
t-cVHP
-1.9
-1.2
-2.5
-1.5
-1.7
-1.2
-1.2
-1.1
t-cVHP (Phe6 f f5-Phe)
t-cVHP (Phe10 f f5-Phe)
t-cVHP (Phe17 f f5-Phe)
t-cVHP (Phe6,10 f f5-Phe)
t-cVHP (Phe6,17 f f5-Phe)
t-cVHP (Phe10,17 f f5-Phe)
t-cVHP (Phe6,10,17 f f5-Phe
Acknowledgment. This research was supported by the NIH
Grant GM-61238 (S.H.G.). M.G.W. was supported in part by a
Fellowship from the Organic Division of the American Chemical
Society, sponsored by Eli Lilly and Company.
Supporting Information Available: Experimental details, CD
analysis, chemical denaturation data, and HPLC chromatograms. This
material is available free of charge via the Internet at http://pubs.acs.org.
a Values are reported in kcal/mol. We estimate the error in the BTE
measurements to be ca. (0.1 kcal/mol.
the free energies of folding (∆Gfold/Gdm). The five Lys f Arg
mutations cause a moderate diminution of tertiary structural stability
(∆Gfold/Gdm ) -3.3 kcal/mol for cVHP vs -2.5 kcal/mol for
m-cVHP). The amide f thioester change, however, causes very
little additional change in stability (∆Gfold/Gdm ) -2.2 kcal/mol for
t-cVHP), consistent with previous BTE results.8 These findings
suggest that insights gathered via BTE analysis of t-cVHP and
related molecules are relevant to the folding of analogues containing
pure amide backbones (i.e., proteins).
HPLC was used to measure KBTE for t-cVHP (Figure 1b). By
initiating thioester-thiol exchange from each side of the equation
shown in Figure 1b, we could show that equilibrium was attained
within 2.5 h (pH 7).6 Tertiary folding occurs in t-cVHP under the
reaction conditions, as verified by CD, but not in the other
thioester-thiol pair because residues 1-10 and residues 11-35
reside in different molecules. KBTE can be used to calculate the
free energy of tertiary folding (∆Gfold/BTE) in the full-length thioester,
as previously described.8 For t-cVHP, ∆Gfold/BTE ) -1.9 kcal/mol,
References
(1) (a) Marsh, E. N. G. Chem. Biol. 2004, 14, R153. (b) Yoder, N. C.; Kumar,
K. Chem. Soc. ReV. 2002, 31, 335. (c) Jackel, C.; Koksch, B. Eur. J.
Org. Chem. 2005, 4483.
(2) (a) Holmgren, S. K.; Taylor, K. M.; Bretscher, L. E.; Raines, R. T. Nature
1998, 392, 666. (b) Tang, Y.; Ghirlanda, G.; Vaidehi, N.; Kua, J.; Mainz,
D. T.; Goddard, W. A.; DeGrado, W. F.; Tirrell, D. A. Biochemistry 2001,
40, 2790. (c) Bilgicer, B.; Fichera, A.; Kumar, K. J. Am. Chem. Soc. 2001,
123, 4393. (d) Bilgicer, B.; Xing, X.; Kumar, K. J. Am. Chem. Soc. 2001,
123, 11815. (e) Horng, J.-C.; Raleigh, D. P. J. Am. Chem. Soc. 2003,
125, 9286. (f) Lee, K.-H.; Lee, H.-Y.; Slutsky, M. M.; Anderson, J. T.;
Marsh, E. N. G. Biochemistry 2004, 43, 16277. (g) Lee, K.-H.; Lee, H.-
Y.; Al-Hashimi, H. M.; Marsh, E. N. G. J. Am. Chem. Soc. 2006, 128,
337. (h) Ja¨ckel, C.; Salwiczek, M.; Koksch, B. Angew. Chem., Int. Ed.
2006, 45, 4198.
(3) Butterfield, S. M.; Patel, P. R.; Waters, M. L. J. Am. Chem. Soc. 2002,
124, 9751.
(4) A quantitative measure of the hydrophobicity for Phe and F5-Phe sidechains
was obtained by partitioning amino acid analogues in water/heptanol:
HΠPhe ) 1.71, ΠF -Phe ) 2.12. (see Supporting Information).
H
(5) (a) Dunitz, J. D. C5hemBioChem 2004, 5, 614. (b) Williams, J. R. Acc.
Chem. Res. 1993, 26, 593. (c) West, A. P.; Mecozzi, S.; Dougherty, D.
A. J. Phys. Org. Chem. 1997, 10, 347.
(6) See Supporting Information.
(7) (a) McKnight, C. J.; Matsudaira, P. T.; Kim, P. S. Nat. Struct. Biol. 1997,
4, 180. (b) Vardar, D.; Buckley, D. A.; Frank, B. S.; McKnight, C. J. J.
Mol. Biol. 1999, 294, 1299. (c) Chiu, T. K.; Kubelka, J.; Herbst-Irmer,
R.; Eaton, W. A.; Hofrichter, J.; Davies, D. R. Proc. Natl. Acad. Sci.
U.S.A. 2005, 102, 7517.
which is similar to ∆Gfold/Gdm
.
∆Gfold/BTE was determined for all variants of t-cVHP containing
one, two, or three Phe f F5-Phe replacements at residues 6, 10,
and/or 17 (Table 1). Each variant appeared to adopt a stable tertiary
fold (∆Gfold/BTE < 0), but only the Phe-10 f F5-Phe mutant,
displayed greater conformational stability than does t-cVHP. The
range of ∆Gfold/BTE values (1.4 kcal/mol) is significant because it
is comparable to the conformational stability of t-cVHP. We
compared our VHP thioester results with data from a more
traditional approach in an effort to validate the BTE method. Figure
2b shows a comparison of ∆Gfold/BTE for t-cVHP and its three single
Phe f F5-Phe mutants with ∆Gfold/Gdm for m-cVHP and its three
single Phe f F5-Phe mutants. The linear correlation shows that
the conformational stability trend indicated for the thioester series
(8) Woll, M. G.; Gellman, S. H. J. Am. Chem. Soc. 2004, 126, 11172.
(9) Thioester-thiol exchange is a form of dynamic covalent chemistry: (a)
Larrson, R.; Pei, Z.; Ramstro¨m, O. Angew. Chem., Int. Ed. 2004, 43, 3716.
(b) Leclaire, J.; Vial, L.; Otto, S.; Sanders, J. K. M. Chem. Commun.
2005, 1959. (c) Rowan, S. J.; Cantrill, S. J.; Cousins, G. R. L.; Sanders,
J. K. M.; Stoddart, J. F. Angew. Chem., Int. Ed. 2002, 41, 898.
(10) Garcia-Mira, M. M.; Sadqi, M.; Fischer, N.; Sanchez-Ruiz, J. M.; Munoz,
V. Science 2002, 298, 2191.
(11) Pace, C. N.; Shaw, K. L. Proteins: Struct. Funct., Genet. Suppl. 2000, 4,
1.
(12) The solution structure of a Phe-10 f F5-Phe mutant of c-VHP has been
determined: Cornilescu, G.; Hadley, E. B.; Woll, M. G.; Markley, J. L.;
Gellman, S. H.; Cornilescu, C. C. Protein Sci. In press.
JA0634573
9
J. AM. CHEM. SOC. VOL. 128, NO. 50, 2006 15933