28 FEBRUARY 2014 VOL 343 SCIENCE www.sciencemag.org 982
These factors imply that the lateral and verti-
cal surfaces in the microtubules change shape
to form the ITT structure.
The inversion assembly involves spermine, which is a highly charged +4 cation and
is strongly attracted to negatively charged
tubulin. Spermine is a well-known “
condenser” of double-stranded DNA (dsDNA)
into toroids (7). In general, a single dsDNA
molecule will generically form toroids in the
presence of counterions of charge at least +3;
the effect is electrostatic in nature and does
not require a specific molecular binding (8, 9).
The high charge is required to make the electrostatic interactions dominate over entropy.
Initially, spermine causes the bundle formation of the microtubules. For bundles of stiff
biopolymers (e.g., dsDNA, actin, or microtubules), only divalent ions are needed (10, 11).
In all these cases, because the counterions
are localized near the charged polymers, they
lose a substantial amount of entropy, which
is compensated enthalpically by the strong
electrostatic interactions of the multivalent
ions. In the bundle geometry, the counter-
ions have more accessible volume and thus
lose less entropy than in the smaller-volume
toroidal case. Thus, a lower charge and elec-
trostatic interaction strength are needed for
bundles. That the ring formation leading to
the ITT structure must be induced by highly
charged spermine suggests that spermine is
being localized in a small volume, possibly
between the lateral contacts in the microtu-
bule, or even undergoes specific binding.
The work of Ojeda-Lopez et al. brings
together many aspects of hierarchical assem-
bly, including strong electrostatics, conforma-
tional switches, and nanoparticle-nanoparticle
interactions. How spermine produces the ITT
structure is still an open and challenging ques-
tion that involves interactions over multiple
length scales. The initial interaction between
spermine and tubulin can be studied with
atomistic simulations, but any shape change
in tubulin is a major challenge to simulate.
Understanding how spermine destabilizes
taxol-bound microtubules is likely to require
atomistic detail while also treating dynamics
on the scale of whole tubulin proteins, which
is not presently feasible. The path forward for
simulations is to attack separately at atom-
istic and coarse-grained scales and to push
them toward each other. More methodological
development is needed on the coarse-grained
side, and there are many opportunities here
and in other nanoparticle systems. For exam-
ple, new developments in synthesis are creat-
ing the patchy nanoparticles whose assembly
may be easier to understand (12). Theoretical
work on patchy particles is providing a funda-
mental base to study the complex behaviors
that occur in these systems (13). While under-
standing nanoparticle systems is complex and
very challenging, a foothold for learning the
right questions is developing.
References and Notes
1. M. A. Ojeda-Lopez et al., Nat. Mater. 13, 195 (2014).
2. H. Lodish et al., Molecular Cell Biology (Freeman, New
3. H. Hess, Annu. Rev. Biomed. Eng. 13, 429 (2011).
4. V. VanBuren, L. Cassimeris, D. J. Odde, Biophys. J. 89,
5. V. Hunyadi, D. Chrétien, H. Flyvbjerg, I. M. Jánosi, Biol.
Cell 99, 117 (2007).
6. S. Cheng, M. J. Stevens, Soft Matter 10, 510 (2013).
7. V. A. Bloomfield, Biopolymers 44, 269 (1997).
8. J. Widom, R. L. Baldwin, J. Mol. Biol. 144, 431 (1980).
9. M. J. Stevens, Biophys. J. 80, 130 (2001).
10. J. X. Tang, T. Ito, T. Tao, P. Traub, P. A. Janmey,
Biochemistry 36, 12600 (1997).
11. M. J. Stevens, Phys. Rev. Lett. 82, 101 (1999).
12. A. H. Gröschel et al., Nature 503, 247 (2013).
13. L. Rovigatti et al., Phys. Rev. Lett. 111, 168302 (2013).
Acknowledgments: Supported by the U. S. Department of
Energy, Office of Basic Energy Sciences, Division of Materials
Sciences and Engineering under Award KC0203010. Sandia
National Laboratories is a multiprogram laboratory managed
and operated by Sandia Corporation, a wholly owned subsidiary of Lockheed Martin Corporation, for the U.S. Department
of Energy’s National Nuclear Security Administration under
Mast cells initiate allergic reactions. These cells express a receptor (FcεRI) that binds to immunoglob-
ulin E (IgE). When antigen binds to recep-
tor-bound IgE, a wide range of responses
is triggered that together cause the symp-
toms of allergy. These include the release of
enzymes and a variety of bioactive chemi-
cals from granules, the generation of lipid-
derived inflammatory molecules, and the
secretion of multiple cytokines and chemo-
kines. However, not all responses are equally
induced. On page 1021 of this issue, Suzuki
et al. (1) unravel how a difference in the affin-
ity of antigen favors one response or another
by switching between intracellular signaling
pathways. Quantitative differences in antigen
affinity thus determine the quality of mast
FcεRI consists of a subunit (FcRα)
that binds to the Fc portion of IgE and two
Immunoreceptor tyrosine-based activation
motif (ITAM)–containing subunits (FcRβ
and FcRγ) that generate activation signals
when phosphorylated by Src family tyrosine
kinases (see the figure). Phosphorylation
occurs when plurivalent allergens (
allergy-evoking antigens) bind to receptor-bound
IgE antibodies and aggregate FcεRI.
Antigens bind to IgE antibodies with
Signaling Shifts in Allergy
variable affinities. Differences in binding
affinity translate into differences in bind-
ing kinetics. Antigens that bind with a low
affinity indeed dissociate more rapidly from
receptor-bound IgE than do antigens that
bind with a high affinity. This is a basis for
the “kinetic proofreading” concept, accord-
ing to which the strength of signals depends
on the duration of receptor engagement (2).
Consequently, low-affinity antigens induce
mast cell responses of a lower magnitude
than high-affinity antigens. All responses,
however, are not similarly affected. Unlike
cytokine secretion, chemokine secretion
was found to be of a similar or greater mag-
nitude when induced by low-affinity than
by high-affinity antigens (3). Suzuki et al.
explain this behavior, which the kinetic
proofreading concept does not account
for, by showing that the affinity of antigen
selects both the Src kinase and the trans-
membrane adapter protein that initiates and
organizes proximal signals in FcεRI sig-
nalosomes, respectively (signalosomes are
protein assemblies in which activation sig-
nals are generated). Consequently, different
A mechanism that senses antigen affinity
shifts cell signaling and leads to qualitatively
different inflammatory responses.
Institut Pasteur, 75015 Paris, France, and Centre
d’Immunologie de Marseille-Luminy, 13009 Marseille,
France. E-mail: firstname.lastname@example.org