Interior Surface Modification of Bacteriophage MS2

Article abstract of DOI:10.1021/ja031790q

An efficient strategy for the interior surface functionalization of MS2 viral capsids is reported, featuring a new hetero-Diels?Alder bioconjugation reaction. After virus isolation, the RNA genome was removed from the spherical particles by exposure to pH 11.8 conditions for a period of 4 h. Following this, 180 tyrosine residues on the interior surface of each empty capsid shell were modified by using a site-selective diazonium-coupling reaction. To attach additional functionality, the azo conjugate was reduced with Na₂S₂O₄ to afford an ortho-amino tyrosine derivative. Oxidation of this moiety with NaIO4 produced an o-iminoquinone on the protein surface, which was found to undergo an efficient hetero-Diels?Alder reaction with N-(4-aminophenyl)acrylamide. This four-step procedure can be carried out in under 4 h, reaches high levels of conversion, and yields the desired conjugates in >60% overall yield. Copyright

Full text of DOI:10.1021/ja031790q

Published on Web 03/09/2004
Interior Surface Modification of Bacteriophage MS2
Jacob M. Hooker, Ernest W. Kovacs, and Matthew B. Francis*
Department of Chemistry, UniVersity of California, Berkeley, California 94720-1460, and Material Science
DiVision, Lawrence Berkeley National Labs, Berkeley, California 94720-1460
Received December 17, 2003; E-mail: francis@cchem.berkeley.edu
There has been growing interest in the use of protein assemblies
as scaffolds for nanoscale materials.1 Viral capsids have received
particular attention in this regard, as they offer robust and
monodisperse structures in a range of sizes and shapes. For
inorganic materials, the inside surface of these structures has been
used to template the growth of metal oxide particles,2 and phage
libraries have been used to identify capsids that can recognize and
order nanocrystals.3 In terms of covalent modification, cysteine
residues introduced using site-directed mutagenesis have proven
useful for the positioning of dye molecules,4 gold nanoparticles,5
and carbohydrates6 on capsid exteriors. Engineered cysteines have
also been used to attach viruses to inorganic substrates.7 These
studies highlight the versatility of capsids as building blocks for
material construction.
We are particularly interested in the covalent modification of
the interior surface of viral capsids8 for the attachment of drug cargo,
site-isolated catalysts, and nucleation sites for crystal growth. To
achieve this, we have developed an efficient four-step strategy for
the interior functionalization of hollow capsid shells. This method
features a new hetero-Diels-Alder bioconjugation reaction for the
attachment of olefin substrates to selectively modified tyrosine
residues.
Bacteriophage MS2 was chosen as the target for these studies
because of its facile propagation and isolation from Hfr+ Escheri-
chia coli cultures. By following established protocols,9 30 mg of
highly pure virus can be obtained from each liter of infected broth.
The capsid shell comprises 180 sequence-identical protein mono-
mers assembled into an icosahedral arrangement that is 27 nm in
diameter. Of particular interest for internal modification is the
presence of 32 pores that are 1.8 nm in diameter, providing access
to the interior volume of the capsid, Figure 1a.10
Figure 1. Bacteriophage MS2. (a) Access to the capsid interior is provided
by 32 pores that are 1.8 nm in diameter. The positions of two modifiable
cys residues are indicated in red and yellow. (b) A cut-away view of the
capsid reveals tyr 85 (green) as an accessible residue on the interior surface
of each capsid monomer. (c) The location of tyr 85 is indicated in a “side-
on view” of two coat protein monomers. The position of these monomers
is indicated by the white outline in a and b.
Before interior functionalization, the RNA genome was removed
by exposing native MS2 to pH 11.8 conditions for 4 h. It is assumed
that the alkaline conditions degrade the RNA through phosphate
hydrolysis and reduce its affinity for the capsid by deprotonating
the associated lysine side chains, although the detailed process by
which the RNA fragments escape the capsid shell is currently under
investigation. Following alkaline hydrolysis, the empty capsids were
isolated via precipitation with over 90% protein recovery. The RNA
core found in native MS2 normally prevents the TEM staining agent
(UO2(OAc)2) from entering the capsid interior; however, after
removal of the genome, the center of the particles appears dark
due to penetration by the stain, Figure 2a,b. The absence of RNA
absorption at 260 nm confirms genome removal, Figure 2c. Stability
studies have indicated that the genome-free capsids do not disas-
semble in the pH range of 3-10 over a 12 h period, Figure 2d.
As the exterior surface already possesses cysteine residues that
have been modified in our lab,11 we were reluctant to use site-
directed mutagenesis to introduce new reactive sites. Instead, a
native residue, tyrosine 85, was targeted to provide 180 modification
sites on the interior surface, Figure 1b,c. Analysis of the crystal
Figure 2. Preparation of genome-free capsid shells. (a) TEM image of
RNA-containing MS2 capsids stained with UO2(OAc)2. (b) After exposure
to pH 11.8 for 4 h the RNA genome is removed, allowing the stain to enter
the capsid (for a and b, scale bar ) 50 nm). (c) The loss of RNA is
confirmed by the lack of absorption at 260 nm in samples isolated using
SEC. (d) The “empty” capsids are stable for 12 h over a wide pH range.
structure10 suggested that the three other tyrosines in each monomer
were significantly less surface accessible and therefore were not
expected to participate in bioconjugation reactions.
Using diazonium-coupling reactions, tyrosine 85 was modified
with high efficiency and selectivity. In particular, virtually complete
9
3718
J. AM. CHEM. SOC. 2004, 126, 3718-3719
10.1021/ja031790q CCC: $27.50 © 2004 American Chemical Society

C O M M U N I C A T I O N S
Scheme 1. Functionalization of Tyrosine 85a
analysis of the modified fragment after trypsin digestion indicates
that the nitrogen atom remains in the product and that a subsequent
oxidation15 occurs to afford benzoxazine 8 as the final reaction
product. Although BnONH2 reacts with 5 to form oximes, no such
adducts are formed for the Diels-Alder product, further suggesting
that 9 is not observed. Assignment of the regiochemistry of the
reaction product is underway. As indicated by SEC and TEM
analysis, no morphological changes were observed for the capsids
after this modification procedure.
Starting with “empty” MS2 capsids, this efficient four-step
procedure can be carried out in less than 4 h and offers an
orthogonal modification strategy to more commonly used lysine-
and cysteine-targeted reactions. Under mild reaction conditions, this
reaction can reach high levels of conversion with exceptional protein
recovery. A more thorough exploration of this new hetero-Diels-
Alder bioconjugation reaction is in progress, as are efforts to build
targeted drug delivery systems and other core/shell materials from
dual-surface modified MS2 capsids.
a (a) 2 (5 equiv), pH 9, 4 °C, 15 min; (b) Na2S2O4 (100 mM), pH 7.2,
rt, 2 h, 80-85% protein recovery, 2 steps; (c) NaIO4 (100 µM), pH 6.5,
followed by (d) 6 (10 mM), 2 h, 75% protein recovery.
Acknowledgment. We gratefully acknowledge the Lawrence
Berkeley National Laboratory, Materials Science Division (U.S.
Department of Energy Contract No. DE-AC03-76SF00098), and
the Center for New Directions in Organic Synthesis for financial
support. CNDOS is supported by Bristol-Myers Squibb as a
Sponsoring Member and Novartis Pharma as a Supporting Member.
J.M.H. was supported by a predoctoral fellowship from the NSF.
Supporting Information Available: Experimental procedures and
characterization data for all intermediates. This material is available
free of charge via the Internet at http://pubs.acs.org.
Figure 3. Analyses of protein modification reactions. (a) Intact capsid
samples isolated using SEC were disassembled and characterized by
MALDI-TOF MS. All m/z ratios are within 0.04% (5 amu) of the expected
values. (b) After subjecting azo-adduct 3 to trypsin, the digest fragments
were identified by MALDI-TOF MS. The expected mass for the modified
fragment was observed, confirming the site selectivity for Y85. No other
modified fragments could be identified. (c) High-resolution MALDI-TOF
MS analysis of the analogous trypsin digest fragment for 8.
References
(1) For examples, see: Niemeyer, C. M. Angew. Chem., Int. Ed. 2001, 40,
4128.
(2) (a) Douglas, T.; Young, M. Nature 1998, 393, 152. (b) Shenton, W.;
Douglas, T.; Young, M.; Stubbs, G.; Mann, S. AdV. Mater. 1999, 11,
253. (c) Douglas, T.; Young, M. AdV. Mater. 1999, 11, 679. (d) Douglas,
T.; Strable, E.; Willits, D.; Aitouchen, A.; Libera, M.; Young, M. AdV.
Mater. 2002, 14, 415. (e) Dujardin, E.; Peet, C.; Stubbs, G.; Culver, J.
N.; Mann, S. Nano Lett. 2003, 3, 413. (f) Knez, M.; Bittner, A. M.; Boes,
F.; Wege, C.; Jeske, H.; Maiss, E.; Kern, K. Nano Lett. 2003, 3, 1079.
(3) (a) Whaley, S. R.; English, D. S.; Hu, E. L.; Barbara, P. F.; Belcher, A.
M. Nature 2000, 405, 665. (b) Lee, S.-W.; Mao, C.; Flynn, C. E.; Belcher,
A. M. Science 2002, 296, 892. (c) Mao, C. B.; Flynn, C. E.; Hayhurst,
A.; Sweeney, R.; Qi, J. F.; Georgiou, G.; Iverson, B.; Belcher, A. M. P.
Natl. Acad. Sci. U.S.A. 2003, 100, 6946.
(4) (a) Wang, Q.; Lin, T.; Tang, L.; Johnson, J. E.; Finn, M. G. Angew. Chem.,
Int. Ed. 2002, 41, 459. (b) Wang, Q.; Kaltgrad, E.; Lin, T.; Johnson, J.
E.; Finn, M. G. Chem. Biol. 2002, 9, 805. (c) Wang, Q.; Lin, T.; Johnson,
J. E.; Finn, M. G. Chem. Biol. 2002, 9, 813.
(5) (a) Raja, K. S.; Wang, Q.; Gonzalez, M. J.; Manchester, M.; Johnson, J.
E.; Finn, M. G. Biomacromolecules 2003, 4, 472. (b) Wang, Q.; Raja, K.
S.; Janda, K. D.; Lin, T. W.; Finn, M. G. Bioconjugate Chem 2003, 14,
38.
(6) Raja, K. S.; Wang, Q.; Finn, M. G. ChemBioChem 2003, 4, 1348.
(7) Smith, J. C.; Lee, K. B.; Wang, Q.; Finn, M. G.; Johnson, J. E.; Mrksich,
M.; Mirkin, C. A. Nano Lett. 2003, 3, 883.
(8) The interior functionalization of a heat shock protein cage has recently
been reported: Flenniken, M. L.; Willits, D. A.; Brumfield, S.; Young,
M. J.; Douglas, T. Nano Lett 2003, 3, 1573.
(9) (a) Davis, J. E.; Sinsheimer, R. L. J. Mol. Biol. 1963, 6, 203. (b) Cargile,
B. J.; McLuckey, S. A.; Stephenson, J. L., Jr. Anal. Chem. 2001, 73, 1277.
(10) Based on analysis of the crystal structure from Valegaard, K.; Liljas, L.;
Fridborg, K.; Unge, T. Nature 1990, 345, 36.
coupling was achieved in 15 min through exposure to 5 equiv of
diazonium salt 2, Scheme 1. Analysis of modified capsids isolated
using size exclusion chromatography indicated the presence of a
new absorption band at 355 nm, and MALDI-TOF MS analysis of
the protein monomers confirmed single modification, Figure 3a.
Both methods verify that >95% conversion was achieved with
excellent protein recovery.12 The selectivity of this reaction for
tyrosine 85 was confirmed through trypsin digest analysis, Figure
3b.
Unfortunately, diazonium salts lacking the nitro substituent
attained lower levels of conversion, thus limiting the range of
functionality that can be installed using this method. To overcome
this problem, the azo bond of 3 was reduced with sodium
dithionite13 to afford o-amino tyrosine derivative 4. This reaction
reaches full conversion in 2 h at room temperature and can be
followed by the absorbance loss at 355 nm. This functional group
can then be oxidized to o-imino-quinone 5 with NaIO4, affording
a highly reactive functional group for further elaboration.
The reactivity of this functional handle was tested in the context
of a Diels-Alder reaction. By screening a series of dienophiles
under a variety of reaction conditions, it was found that acrylamide
6 gave particularly efficient conjugation, with conversion levels
exceeding 90% in 2 h at room temperature, as determined by
MALDI-TOF MS analysis, Figure 3a. No other modification sites
were identified after trypsin digest analysis, and exposure of
“empty” capsid 1 to 6 and NaIO4 yielded no reaction. Although
two Diels-Alder products (7 and 9) can be expected, current data
are more consistent with initial adduct 7.14 High-resolution MS
(11) This has been accomplished in high yield, using disulfide exchange
reactions, and will be reported separately.
(12) All bioconjugation reactions experience protein losses due to denaturation,
precipitation, and adsorption to chromatography resins. These losses are
included in our reported product yields.
(13) Jaffe, C. L.; Lis, H.; Sharon, N. Biochemistry 1980, 19, 4423.
(14) For examples of hetero-Diels-Alder reactions with orthoquinones, see:
Nair, V.; Mathew, B. Tetrahedron 1999, 55, 11017 and references therein.
(15) Similar oxidations have been observed under a variety of conditions: (a)
Chacun-Lefevre, L.; Buon, C.; Bouyssou, P.; Coudert, G. Tetrahedron
Lett. 1998, 39, 5763. (b) Bourlot, A. S.; Guillaumet, G.; Merour, J. Y. J
Heterocyclic Chem 1996, 33, 191.
JA031790Q
9
J. AM. CHEM. SOC. VOL. 126, NO. 12, 2004 3719