вход по аккаунту


The Thermal Unfolding of Native Cytochrome c in the Transition from Solution to Gas Phase Probed by Native Electron Capture Dissociation.

код для вставкиСкачать
Mass Spectrometry of Proteins
DOI: 10.1002/anie.200500668
The Thermal Unfolding of Native Cytochrome c
in the Transition from Solution to Gas Phase
Probed by Native Electron Capture
Kathrin Breuker* and Fred W. McLafferty
The effect of hydration on protein structure, dynamics,
folding, and stability has been studied for decades and still
is a topic of active research.[1–4] Removal of solvent results in
the formation of stable gas-phase protein conformations that
can be dramatically different from the original native state,[5–9]
but so far nothing is known about the sequence of structural
changes of a native protein that is suddenly exposed to
vacuum. We address here the effect of hydration on structure
and stability, and report for the first time site-specific data on
the thermal unfolding of a native protein structure in the
transition from solution to gas phase. Our data support a
sequential unfolding mechanism dominated by the loss of
hydrophobic bonding.
We have shown recently that electrospray ionization
(ESI)[10] of aqueous solutions of ferric Cytochrome c, (FeIII)Cyt c, in concentrations favorable for the formation of
noncovalently bound homodimers ( 75 mm) produces unexpected backbone-cleavage products, a phenomenon we
termed “native electron capture dissociation” (NECD).[11]
Briefly, as an electrosprayed dimer ion is introduced into
the Fourier transform mass spectrometer (FTMS) and passes
the heated capillary for desolvation, one of its monomers
partially unfolds. This causes proton transfer from the
compact monomer II to the partially unfolded monomer I,
and induces a substantial charge asymmetry.[12] In turn, this
prompts intermolecular transfer of two electrons to the heme
of monomer I, one reducing the heme iron and the other
causing protein backbone cleavage (NECD) next to residues
in contact with the heme.[11, 13] Finally, the two cleavage
[*] Dr. K. Breuker
Institute of Organic Chemistry and
Center for Molecular Biosciences Innsbruck (CMBI)
Innsbruck University
Innrain 52a, 6020 Innsbruck (Austria)
Fax: (+ 43) 512-507-2892
Prof. Dr. F. W. McLafferty
Department of Chemistry and Chemical Biology
Baker Laboratory, Cornell University
Ithaca, NY 14853-1301 (USA)
[**] The authors thank the Austrian FWF and BMBWK (grants P15767,
T229 to KB) and the NIH (grant GM16609 to FWM) for generous
funding, and Huili Zhai, Cheng Lin, Xuemei Han, Robert Konrat,
Harry B. Gray, Harold Hwang, and Mi Jin for stimulating discussions.
Supporting information for this article is available on the WWW
under or from the author.
Angew. Chem. Int. Ed. 2005, 44, 4911 –4914
products and monomer II separate and are detected in the
trapped-ion cell of the FTMS.
Figure 1 shows the yield[14] of fragment ions from NECD
of two aqueous solutions of horse heart (FeIII)Cyt c (75 mm,
pH 5) versus backbone cleavage site; the capillary temperature was 31 8C.[15] Solutions A (for the experiment in
Figure 1 a) and B (for the experiment in Figure 1 b) were
prepared in the same way, except that A was stored overnight
and B was stored for three months. In agreement with the
proposed NECD mechanism,[11] protein backbone cleavage
occurs next to residues with noncovalent bonding to the heme
in the native structure. The NECD fragments from solution A
result from cleavage next to K13, Y48/T49, L68, K79/M80,
F82, and I85, while solution B gives additional products from
cleavage next to L35, T40, F46, N52, and W59.
Thus the dimers in solutions A and B appear to represent
different structures; that of solution B stabilizes more
protein–heme bonds in monomer I so that they are still
intact when NECD occurs.[16] The cleavage products from
solution A are consistent with monomer II binding near K79/
M80 of monomer I, and stabilization of its protein–heme
interactions through the hydrogen-bond network that
involves residues 49, 67, and 78–82 (Figure 2). The products
from solution B indicate binding near Y48/T49 of monomer I,
which in addition stabilizes its heme contacts of the residues
associated with the extensive hydrogen-bond network of 37,
38, 40, 49, 50, 52–55, 57, and 59. This binding close to Y48/T49
actually corresponds to that found in the asymmetric dimer
unit in crystals of (FeIII)Cyt c (horse heart) at low ionic
strength in which T47 and D50 of monomer I bind to V11,
Q12, and Q16 of monomer II.[17] Interestingly, the structure of
the crystal monomer corresponding to our partially unfolded
monomer I is somewhat more disordered than its counterpart.
Neither of the NECD spectra shows fragments from
cleavage in the terminal regions, that is, next to F10, L94, and
L98, or in the 18–34 segment. This suggests that the initial
unfolding causing charge asymmetry in either dimer structure
involves separation of the terminal helices, rupture of the
H18 Fe coordinative bond,[18] and unfolding of the associated
W-loop. Proton transfer to the newly exposed residues causes
charge asymmetry, the extent of which can be calculated from
the sum of the average charge values for the quasi-complementary fragment ions c48 and y56, + 8.5 (Figure 3 a), and the
average monomer ion charge of + 7.2 at 50 % yield (representing all monomer II, as all monomer I ions are fragmented). Taking into account the two electrons for heme iron
reduction and backbone cleavage, these + 8.5 and + 7.2
values correspond to + 10.5 and + 5.2 charges, respectively,
prior to electron transfer. Thus a 2.0:1 charge asymmetry
drives NECD for both of these solutions at pH 5, independent
of temperature (Figure 3 a), whereas the total NECD yield
goes up from 4 % at 28.5 8C to 23 % at 42 8C and then down to
15 % at 45.5 8C for solution A, and from 21 % at 29 8C to 54 %
at 40 8C to 30 % at 48 8C for solution B (Figure 3 b).
A 75 mm solution at pH 4 gave cleavage products from the
same sites as in Figure 1 a, except none next to K13. The sums
of the fragment-ion charges were + 12.5 at 27.5 and 29.5 8C,
indicating the presence of more highly charged dimeric
NECD precursors at pH 4. Although heating lowered the
2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Figure 1. NECD fragment-ion yields (Y) versus backbone cleavage site from ESI of aqueous solutions of horse heart (FeIII)Cyt c (75 mm, pH 5). The
samples were prepared by dissolving the lyophilized protein in pure water and storing a) overnight at 4 8C, b) at 4 8C for three months; capillary
temperature 31 8C. Ellipses: residues with native heme contacts (filled black where NECD was observed). Upper left: Structure of native (FeIII)Cyt c in solution;[23] heme: dark gray; contacting residues giving rise to NECD: same colors as in (a) and (b). Gray areas: unfolding free energies
of (FeIII) Cyt c in solution relative to the native state.[21, 22]
Figure 3. NECD of (FeIII)Cyt c solutions versus capillary temperature:
a) summed average charge states (n) of c48 and y56, b) total yields (Y).
Solution A, pH 4 (*), pH 5 (&); solution B, pH 5 (~).
Figure 2. Schematic representation of hydrophobic (thin solid lines),
hydrogen (thick dashed lines), and coordinative (thick solid lines)
bonds (< 3.5 H) of heme-contacting residues in native horse heart
(FeIII)Cyt c.[23] The T40–heme interaction (thin dashed line) is unspecific
but within 2.9 H. N52 and W59 are connected by hydrogen bonds to
the same heme propionate 7 (p7) oxygen atom, and T49 and T78 are
connected by hydrogen bonds to the same heme propionate 6 (p6)
oxygen atom.
fragment-ion charge of + 12.5 to the + 8.5 value of the
solutions at pH 5 (Figure 3 a), the charge asymmetry before
electron transfer changed little with temperature; the average
value of + 8.9 for monomers at 29.5 8C indicates a charge
asymmetry of 14.5:6.9 (2.1:1) before electron transfer.
Increasing temperature effects a dramatic yield increase from
2 % at 27.5 8C to 32 % at 29.5 8C, falling to 4 % at 45 8C
(Figure 3 b).
This strong positive and negative effect of heating on
NECD yields for both dimer solutions, one at both pH 4 and
5, without a change in the required charge asymmetry,
indicates several competitive dissociations of the solution
dimer structure. The lowest energy pathway must be the
direct dissociation of the intermonomer bond, as no dimer
ions were detected in the NECD spectra measured under any
conditions. The increasing yield of NECD cleavage products
2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2005, 44, 4911 –4914
with heating is then consistent with an increasingly competitive unfolding of monomer I and more dimer ions reaching
the critical charge asymmetry for NECD before the intermonomer bond dissociates; NECD of solution B reaches a 50 %
fragment-ion yield at 40 8C, corresponding to complete
dissociation of all monomers I. Higher temperatures open
up yet another pathway, the competitive unfolding of monomer II, which increases the proportion of dimer ions whose
charge asymmetry stays below the critical level. Note that the
less favored unfolding of monomer II is consistent with the
asymmetric dimer structure of (FeIII)Cyt c crystals,[17] in which
unfolding of the terminal helices of monomer II would be
restricted by the V11, Q12, and Q16 binding in the dimer
interface (Figure 1). However, an asymmetric dimer structure
is not necessarily a requirement for asymmetric unfolding;
many gas-phase dimer ions exhibit this phenomenon,
although none as yet has shown NECD.[12, 19, 20]
The branching ratio of products from different cleavage
sites can also be strongly affected by temperature, as
illustrated for solution A in Figure 4. For example, the yield
Figure 4. Branching ratio of NECD fragments from cleavage at sites i
versus j, (Yi/Yj), for solution A, pH 5: a) Y47/Y48 (^), Y49/Y48 (~), Y50/Y48
(~), Y51/Y48 (*), (Y47 + Y48 + Y49 + Y50 + Y51)/Y79 (&); b) Y82/Y47 (&),
Y83/Y68 (~), Y84/Y83 (~), Y12/Y84 (^), Y68/Y82 (*).
of fragments from cleavage at site 12, Y12, versus that from
cleavage at site 84, Y84, decreases with increasing temperature, revealing that fewer dimers had retained the K13–heme
interaction than the I85–heme interaction at higher temperatures. Y68/Y82 does not change with temperature, suggesting a
similar stability for the L68–heme and F82–heme interactions.
The entire data set gives the order of stability, and the reverse
order of unfolding: K79/M80 > Y48/T49 > F82 = L68 > I85 >
K13; that for solution B is Y48/T49 = W59 = K79/M80 >
F46 = N52 = F82 > T40 > L68 = I85 > L35 > K13 (see the
Supporting Information).
From this order of unfolding, the most stable heme
binding sites of monomer I for both the solution A and B
dimers are those postulated to be close to the dimer binding
interface (Figure 2). However, both dimers have initially
unfolded in the same manner as indicated by the lack of
fragments in the terminal and the 18–34 regions, and by the
same charge asymmetry for both dimers. As shown in Figure 2
Angew. Chem. Int. Ed. 2005, 44, 4911 –4914
(solid lines), these initially unfolded regions are generally
stabilized by hydrophobic bonding, in sharp contrast to the
hydrogen-bond stabilization (dashed lines, Figure 2) of the
least unfolded protein–heme bonding of monomer I. As
previously postulated,[5–9] removal of water should weaken
hydrophobic bonding (no solvent to avoid) but should also
strengthen hydrogen bonding (no solvent to compete with).
Of further interest, this solution-to-gas phase stability
order is essentially the reverse of the unfolding in solution.
Englander and co-workers recently characterized five cooperative folding–unfolding units (“foldons”) in native Cyt c,
whose free energies of unfolding are shown in Figure 1
(K13 > L68 = L35 > W59 > K79/M80 = F82 = I85 > T40 =
F46 = Y48/T49 = N52).[21, 22] The most stable foldon in solution comprises the terminal helices that are held together by
hydrophobic bonding; NECD found no bonding to the heme
at F10, L94, and L98; bonding at K13 was found to be the next
least stable. In solution, extensive hydrophobic bonding from
the terminal helices (V20 with F10, Y97, L98, A101; L32 with
L98) also stabilizes the 18–34 W-loop, as part of the second
most stable foldon. In this region NECD finds L35 and L68 as
the next weakest heme binding partners, with six other
cleavage products missing. On the other hand, among the
most stable interactions in the gas phase is the Y48/T49–heme
contact, which is part of the least stable solution foldon.
Dehydration of the native protein structure, plus the stabilization by monomer II, almost inverts the stabilities of the
individual interactions in solution.
In conclusion, site ordering for the unfolding of a native
protein in the transition from solution to gas phase greatly
favors hydrophobic bonds, even those that are the last to
unfold in solution. Surrounding such a bond by vacuum
instead of water of course weakens it; intermediate effects of
this type have been observed with protein conformations in
nonpolar solutions.[24] The main requirement for NECD of the
Cyt c dimers appears to be fast initial unfolding of a peripheral
region, so that monomeric native proteins with peripheral
hydrophobic bonding in extended structures should also be
investigated for NECD. Although here NECD yields could be
seriously reduced by heme absence or competitive unfolding
of another part of the protein, competitive dissociation of
intermolecular noncovalent bonds (e.g. dimer interface)
would be no problem. The occurrence of NECD should also
be sought in the dissociations of gaseous ion dimer structures
that exhibit asymmetric charge partitioning.[12, 19, 20]
Experimental Section
This study utilized a 6-Tesla FTMS described previously.[25] Ions
formed by ESI entered the instrument through a heated capillary
where NECD occurs;[11] the products were transferred into the ion
cell (< 10 9 Torr) through quadrupole ion guides. ESI utilized homemade emitters with a tip inner diameter of 5 mm, flow rate of 200–
500 nL min 1, and spray potential of 1 kV. Horse heart Cyt c (Sigma,
USA) was dissolved in pure water, with pH adjustment by addition of
acetic acid.
Received: February 22, 2005
Published online: July 6, 2005
2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Keywords: hydration · mass spectrometry · native electron
capture dissociation · noncovalent interactions · protein
M. F. Jarrold, Annu. Rev. Phys. Chem. 2000, 51, 179 – 207.
J. Woenckhaus, Int. J. Mass Spectrom. 2002, 213, 9 – 24.
F. Franks, Biophys. Chem. 2002, 96, 117 – 127.
C. N. Pace, S. Trevino, E. Prabhakaran, J. M. Scholtz, Philos.
Trans. R. Soc. London Ser. B 2004, 359, 1225 – 1235.
F. W. McLafferty, Z. Q. Guan, U. Haupts, T. D. Wood, N. L.
Kelleher, J. Am. Chem. Soc. 1998, 120, 4732 – 4740.
C. S. Hoaglund-Hyzer, A. E. Counterman, D. E. Clemmer,
Chem. Rev. 1999, 99, 3037 – 3079.
D. M. Horn, K. Breuker, A. J. Frank, F. W. McLafferty, J. Am.
Chem. Soc. 2001, 123, 9792 – 9799.
K. Breuker, H.-B. Oh, D. M. Horn, B. A. Cerda, F. W. McLafferty, J. Am. Chem. Soc. 2002, 124, 6407 – 6420.
H.-B. Oh, K. Breuker, S. K. Sze, Y. Ge, B. K. Carpenter, F. W.
McLafferty, Proc. Natl. Acad. Sci. USA 2002, 99, 15 863 – 15 868.
J. B. Fenn, M. Mann, C. K. Meng, S. F. Wong, C. M. Whitehouse,
Science 1989, 246, 64 – 71.
K. Breuker, F. W. McLafferty, Angew. Chem. 2003, 115, 5048 –
5052; Angew. Chem. Int. Ed. 2003, 42, 4900 – 4904.
J. C. Jurchen, E. R. Williams, J. Am. Chem. Soc. 2003, 125, 2817 –
NECD of (FeII)Cyt c yields the same products but involves
transfer of only a single electron.
The yield was calculated as [fragments]/([fragments] + [molecular ions]); dimer ions were not detected in the mass spectra.
Measured at the capillary orifice, the temperature inside is
estimated to be two times this value.
Many of the cleavage sites found for solution B are far apart on
the monomer structure of Figure 1, so that it is only possible for a
fraction of these sites to participate in the intermolecular
bonding of a single dimer structure. Thus it is implausible that
a direct electron transfer from monomer II to the cleavage site
causes NECD. Multiple dimer structures are also less likely for
solution B than for solution A, as the B dimers should have far
more time to isomerize to a minimum-energy structure.
R. Sanishvili, K. W. Volz, E. M. Westbrook, E. Margoliash,
Structure 1995, 3, 707 – 716.
Losing this coordination site will increase the electron affinity of
the heme.
C. Versluis, A. v. d. Staaij, E. Stokvis, A. J. R. Heck, J. Am. Soc.
Mass Spectrom. 2001, 12, 329 – 336.
N. Felitsyn, E. N. Kitova, J. S. Klassen, J. Am. Soc. Mass
Spectrom. 2002, 13, 1432 – 1442.
M. M. G. Krishna, Y. Lin, J. N. Rumbley, S. W. Englander, J. Mol.
Biol. 2003, 331, 29 – 36.
H. Maity, M. Maity, S. W. Englander, J. Mol. Biol. 2004, 343,
223 – 233.
L. Banci, I. Bertini, H. B. Gray, C. Luchinat, T. Reddig, A.
Rosato, P. Turano, Biochemistry 1997, 36, 9867 – 9877.
T. Knubovets, J. J. Osterhout, A. M. Klibanov, Biotechnol.
Bioeng. 1999, 63, 242 – 248.
R. A. Zubarev, D. M. Horn, E. K. Fridriksson, N. L. Kelleher,
N. A. Kruger, M. A. Lewis, B. K. Carpenter, F. W. McLafferty,
Anal. Chem. 2000, 72, 563 – 573.
2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2005, 44, 4911 –4914
Без категории
Размер файла
186 Кб
solutions, thermal, cytochrome, unfolding, native, capture, electro, probes, gas, transitional, dissociation, phase
Пожаловаться на содержимое документа