Download De Novo Design of an Enzyme

PERSPECTIVES
Crab nebula (a “plerion”) and hence would
show evidence for continuing energy input.
Now, Bietenholz et al. (1) have provided
the direct evidence.
In a superb long-term radio mapping
study of SN 1986J, using the full spatial
resolution of intercontinental radio astronomy linking 20 telescopes through the Very
Long Baseline Array (VLBA), Bietenholz
et al. have imaged the appearance of a
compact energetic source at the heart of the
expanding, cooling remnant of the supernova. This provides the first direct evidence
for a young neutron star (or perhaps black
hole) associated with a supernova. It identifies our youngest known compact object
for continuing study, and promises to start
to fill the many lacunae in our understanding of the behavior of matter at extreme
densities.
Refererences and Notes
1. M. F. Bietenholz, N. Bartel, M. P. Rupen, Science 304,
1947 (2004); published online 10 June 2004
(10.1126/science.1099460).
2. M. Livio, A. Reiss, Astrophys. J. 594, L93 (2003).
3. S. D. Van Dyk, Science 302, 1161 (2003).
M. Tegmark, Science 296, 1427 (2002).
R. P. Kirshner, Science 300, 1914 (2003).
C. Seife, Science 303, 1271 (2004).
M. Heger et al., Astrophys. J. 591, 288 (2003).
M. Limongi, A. Chieffi, Astrophys. J. 592, 404 (2003).
L. Wang, J. Wheeler, Astrophys. J. 504, L87 (1998).
S. J. Smartt et al., Science 303, 499 (2004).
T. Matheson, Nature 427, 109 (2004).
J. R. Maund et al., Nature 427, 129 (2004).
Special section on Pulsars, Science 304 (23 April 2004).
Supernovae are named by the year of discovery, with
a letter sequence denoting the discovery order. Thus,
SN 1986J was the 15th supernova discovered in 1986.
It probably exploded 3 years before discovery.
15. R. Chevalier, Nature 329, 611 (1987).
16. K. Weiler, N. Panagia, R. Sramek, Astrophys. J. 364, 611
(1990).
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
B I O C H E M I S T RY
RBP into a high-affinity receptor for nonnatural ligands as diverse as serotonin and
trinitrotoluene (5). Now, in going beyond
binding and on to catalysis, they have again
Reinhard Sterner and Franz X. Schmid
used RBP as the scaffold. The authors followed a three-step procedure.
First, they defined geometries
nzymes catalyze biological
in which the three catalytically
reactions under mild conessential residues glutamate,
ditions with high specificihistidine, and lysine are in a faties and rate enhancements of
vorable orientation with respect
up to 1017-fold (1). This enorto the enediol intermediate.
mous catalytic power is the
Then they used a combinatorial
product of natural evolution,
E
search algorithm to find amino
and biochemists have tried for
H
H
acid positions within the RBPmore than a century to underK
K
E
binding region that satisfy these
stand the underlying chemical
geometrical constraints (6).
principles. The salient test of
Finally, they used their recepour understanding of enzyme
tor-design algorithm to opticatalysis would be the design of
mize the potential active site
an enzyme from scratch. On
for the binding of the enediol
page 1967 of this issue, Dwyer,
intermediate (5).
Looger, and Hellinga (2) deIn this way 14 RBP variants
scribe an important step toward Activity by design. Comparison of native TIM with NovoTim1.2 (RBP
achieving this goal. Using com- containing a modeled TIM-like active site). (Left) X-ray structure of hu- were designed, produced, and
puter-based rational protein de- man TIM with bound transition-state analog 2-phosphoglycolate assayed for TIM activity. Resign, they turned the catalytical- [Protein Data Bank (PDB) entry code 1hti]. (Right) Structure of RBP from markably, seven of these conly inert ribose-binding protein Escherichia coli (PDB entry code 2dri) with modeled TIM active site and verted GAP into DHAP at a rate
(RBP) into an enzyme that is enediol intermediate. Ribbon diagrams are shown with α helices in red above background.
The most active RBP variant
highly active as a triose phos- and β strands in yellow. The catalytically essential residues glutamate (E),
phate isomerase (TIM).
histidine (H), and lysine (K) are in blue. Residues introduced into RBP for contained 10 amino acid substitutions, in addition to the catTIM is the prototype of the optimizing the binding of the enediol are in green.
alytic residues glutamate, histilarge family of (βα)8-barrel or
TIM-barrel enzymes (3). It is active in gly- mediates the transfer of a proton from the dine, and lysine. This variant, which was
colysis, catalyzing the interconversion be- C1 to the C2 oxygen within the forming much more labile than native RBP, was
tween the ketose dihyroxyacetone phos- cis-enediol intermediate, which in turn is stabilized—again with computational dephate (DHAP), and the aldose glyceralde- stabilized by hydrogen bonding to the ly- sign—by optimizing the residue layers surhyde-3-phosphate (GAP). The secret of sine residue. The intermediate then col- rounding the immediate binding surfaces
catalysis by TIM lies in the precise orienta- lapses to give GAP and regenerates the en- (7). The stability of the resulting RBP varition of three critical amino acid residues zyme (4). The abstraction of the C1 proton ant, NovoTim1.2, approached that of na(glutamate, histidine, and lysine) in its ac- is energetically highly unfavorable, and the tive RBP. To further improve the catalytic
tive site and in controlled movements of enediol intermediate easily loses its phos- activity, the authors turned to the comthe protein chain during catalysis. The glu- phate by β-elimination. To avoid this side pletely different approach of directed evotamate residue abstracts a proton from C1 reaction, TIM uses a mobile loop as a lid to lution, which combines random mutageneof the substrate DHAP. The histidine then close the active site after the substrate is sis with in vivo selection. The evolved subbound. Thus, the generation of TIM activi- variants of NovoTim1.2, which contained
R. Sterner is at the Universität Regensburg, Institut
ty on a different protein scaffold is a de- additional amino acid substitutions at the
für Biophysik und Physikalische Biochemie, D-93040
protein surface remote from the active site,
manding task.
Regensburg, Germany. E-mail: reinhard.sterner@
Hellinga and his co-workers are pio- catalyzed the TIM reaction at a rate 105- to
biologie.uni-regensburg.de F. X. Schmid is at the
neers in computational protein design. 106-fold above background but with catUniversität Bayreuth, Laboratorium für Biochemie, DRecently, they succeeded in transforming alytic efficiency parameters kcat/KM about
95440 Bayreuth, Germany.
De Novo Design of an Enzyme
1916
25 JUNE 2004
VOL 304
SCIENCE
www.sciencemag.org
CREDIT: IRIS LAMBECK/UNIVERSITÄT REGENSBURG
E
PERSPECTIVES
three orders of magnitude below that for
wild-type TIM. This reduced efficiency
does not diminish the achievement, considering that native TIM is a kinetically perfect enzyme with a turnover that is limited
only by the diffusion-determined rate at
which substrate and enzyme encounter
each other (4).
Attempts to achieve the de novo design
of enzymes are not new (8). Antibodies
with catalytic activities have been obtained by using analogs of transition states
as antigens (9). Recently, Bolon and Mayo
used computational design to transform
the catalytically inert protein thioredoxin
into an esterase (10). However, both the
catalytic antibodies and the designed
thioredoxin are much less active than the
RBP variants transformed into TIM.
The new work by the Hellinga lab (2)
exemplifies the enormous power of computational biology, and illustrates how this
approach can be combined with directed
evolution. The latter is well suited to identify beneficial mutations far from the active site. Such mutations are difficult to
find by computation but important for the
fine-tuning of catalysis (11).
Recently, in equally exciting work, Baker
and co-workers designed from scratch a
small protein with a new folding topology
and verified its structure by x-ray crystallography (12). With the important contributions of these two studies as a foundation,
the design of tailored catalytic activities on
artificial proteins seems now to be within
reach. Such an achievement would be a
milestone on the path to synthetic biology,
with enormous potential for applications in
medicine and biotechnology.
References
1. C. Walsh, Nature 409, 226 (2001).
2. M. A. Dwyer, L. L. Looger, H. W. Hellinga, Science 304,
1967 (2004).
3. J. A. Gerlt, F. M. Raushel, Curr. Opin. Chem. Biol. 7, 252
(2003).
4. J. R. Knowles, Nature 350, 121 (1991).
5. L. L. Looger, M. A. Dwyer, J. J. Smith, H. W. Hellinga,
Nature 423, 185 (2003).
6. H. W. Hellinga, F. M. Richards, J. Mol. Biol. 222, 763
(1991).
7. M. A. Dwyer, L. L. Looger, H. W. Hellinga, Proc. Natl.
Acad. Sci. U.S.A. 100, 11255 (2003).
8. D. N. Bolon, C. A. Voigt, S. L. Mayo, Curr. Opin. Chem.
Biol. 6, 125 (2002).
9. P. G. Schultz, J. Yin, R. A. Lerner, Agnew. Chem. Int. Ed.
41, 4427 (2002)
10. D. N. Bolon, S. L. Mayo, Proc. Natl. Acad. Sci. U.S.A. 98,
14274 (2001).
11. M. Bocola et al., Chembiochem 5, 214 (2004).
12. B. Kuhlman et al., Science 302, 1364 (2003).
M AT E R I A L S C I E N C E
Spinning Continuous
Fibers for Nanotechnology
Yuris Dzenis
anotubes of carbon and other materials are arguably the most fascinating materials playing an important
role in nanotechnology today. Their unique
mechanical, electronic, and other properties are expected to result in revolutionary
new materials and devices. However, these
nanomaterials, produced mostly by synthetic bottom-up methods, are discontinuous objects, and this leads to difficulties
with their alignment, assembly, and processing into applications. Partly because of
this, and despite considerable effort, a viable carbon nanotube–reinforced supernanocomposite is yet to be demonstrated.
Advanced continuous fibers produced a
revolution in the field of structural materials and composites in the last few decades
as a result of their high strength, stiffness,
and continuity, which, in turn, meant processing and alignment that were economically feasible. Fiber mechanical properties
are known to substantially improve with a
decrease in the fiber diameter. Hence, there
is a considerable interest in the development of advanced continuous fibers with
nanoscale diameters. However, conventional mechanical fiber spinning techniques cannot produce fibers with diameters smaller than about 2 µm robustly. Most
commercial fibers are several times that di-
N
The author is in the Department of Engineering
Mechanics, University of Nebraska–Lincoln, Lincoln,
NE 68588–0526, USA. E-mail: [email protected]
ameter, owing to the trade-offs between the
technological and economic factors.
Electrospinning technology enables
production of continuous polymer nanofibers from polymer solutions or melts in
high electric fields. When the electric force
on induced charges on the polymer liquid
overcomes surface tension, a thin polymer
jet is ejected. The charged jet is elongated
and accelerated by the electric field, undergoes a variety of instabilities, dries, and is
deposited on a substrate as a random
nanofiber mat. The first patent on the
process was awarded in 1934; however,
outside of the filter industry, there was little interest in the electrospinning or electrospun nanofibers, until the mid-1990s
(1). Since that time, the process attracted
rapidly growing interest triggered by potential applications of nanofibers in the
nanotechnology. The publication rate has
nearly doubled annually, reaching about
200 articles in 2003. Over a hundred synthetic and natural polymers were electrospun into fibers with diameters ranging
from a few nanometers to micrometers (see
the figure, panel A).
The main advantage of this top-down
nanomanufacturing process is its relatively
low cost compared to that of most bottomup methods. The resulting nanofiber samples are often uniform and do not require
expensive purification (panels B and C).
Unlike submicrometer-diameter whiskers,
inorganic nanorods, carbon nanotubes, and
www.sciencemag.org
SCIENCE
VOL 304
nanowires, the electrospun nanofibers are
continuous. As a result, this process has
unique potential for cost-effective electromechanical control of fiber placement and
integrated manufacturing of two- and
three-dimensional nanofiber assemblies.
In addition, the nanofiber continuity may
alleviate, at least in part, concerns about
the properties of small particles (2).
Nanofibers are expected to possess high
axial strength combined with extreme flexibility. The nanofiber assemblies may feature very high open porosity coupled with
remarkable specific surface area. Yet, these
assemblies would possess excellent structural mechanical properties. Uses of
nanofibers in composites, protective clothing, catalysis, electronics, biomedicine (including tissue engineering, implants, membranes, and drug delivery), filtration, agriculture, and other areas are presently being
developed. Clearly, there is a growing interest in the process, but the results reported to date are centered mostly on the empirical production and the proposed uses of
polymer nanofibers. At the same time,
thorough understanding of the mechanisms
of jet formation and motion is needed for
the development of robust methods of
process control. Analysis of the electrospinning process is complicated by electromechanical coupling, nonlinear rheology,
and unusual jet instabilities. Some progress
was recently made on modeling of jet initiation (3, 4). Steady-state spinning was
modeled in the nonlinear rheologic regime
important for polymer jets (5, 6).
Experimental observations and modeling
of bending (or whipping) instability (7, 8)
produced a major breakthrough in process
analysis. It substantially improved our understanding of the jet motion and removed
an early controversy in the electrospinning
studies over the interpretation of long–ex-
25 JUNE 2004
1917