XMAP215 polymerase activity is built by combining multiple tubulin-binding TOG domains and a basic lattice-binding region.

Abstract

XMAP215/Dis1 family proteins positively regulate microtubule growth. Repeats at their N termini, called TOG domains, are important for this function. While TOG domains directly bind tubulin dimers, it is unclear how this interaction translates to polymerase activity. Understanding the functional roles of TOG domains is further complicated by the fact that the number of these domains present in the proteins of different species varies. Here, we take advantage of a recent crystal structure of the third TOG domain from Caenorhabditis elegans, Zyg9, and mutate key residues in each TOG domain of XMAP215 that are predicted to be important for interaction with the tubulin heterodimer. We determined the contributions of the individual TOG domains to microtubule growth. We show that the TOG domains are absolutely required to bind free tubulin and that the domains differentially contribute to XMAP215's overall affinity for free tubulin. The mutants' overall affinity for free tubulin correlates well with polymerase activity. Furthermore, we demonstrate that an additional basic region is important for targeting to the microtubule lattice and is critical for XMAP215 to function at physiological concentrations. Using this information, we have engineered a "bonsai" protein, with two TOG domains and a basic region, that has almost full polymerase activity.

Full text

Page 1

XMAP215 polymerase activity is built by combining
multiple tubulin-binding TOG domains and
a basic lattice-binding region
Per O. Widlunda, Jeffrey H. Stearb, Andrei Pozniakovskya, Marija Zanica, Simone Rebera, Gary J. Brouhardc,
Anthony A. Hymana,1, and Jonathon Howarda,1
aMax Planck Institute of Molecular Cell Biology and Genetics, Pfotenhauerstrasse 108, 01307 Dresden, Germany;
Universität zu Berlin, Chausseestrasse 117, 10115 Berlin, Germany; and
Montréal, QC, Canada H3A 1B1
bInstitut für Biologie, Humboldt
cDepartment of Biology, McGill University, 1205 avenue Docteur Penfield,
Edited by J. Richard McIntosh, University of Colorado, Boulder, CO, and approved December 30, 2010 (received for review November 3, 2010)
XMAP215/Dis1 family proteins positively regulate microtubule
growth. Repeats at their N termini, called TOG domains, are impor-
tant for this function. While TOG domains directly bind tubulin
dimers, it is unclear how this interaction translates to polymerase
activity. Understanding the functional roles of TOG domains is
further complicated by the fact that the number of these domains
present in the proteins of different species varies. Here, we take
advantage of a recent crystal structure of the third TOG domain
from Caenorhabditis elegans, Zyg9, and mutate key residues in
each TOG domain of XMAP215 that are predicted to be important
for interaction with the tubulin heterodimer. We determined the
contributions of the individual TOG domains to microtubule
growth. We show that the TOG domains are absolutely required
to bind free tubulin and that the domains differentially contribute
to XMAP215’s overall affinity for free tubulin. The mutants’ overall
affinity for free tubulin correlates well with polymerase activity.
Furthermore, we demonstrate that an additional basic region is
important for targeting to the microtubule lattice and is critical
for XMAP215 to function at physiological concentrations. Using
this information, we have engineered a “bonsai” protein, with two
TOG domains and a basic region, that has almost full polymerase
activity.
C
cell shape, directing cellular movement, and mediating chromo-
some segregation and cell division. Although these polymeric fi-
laments have different structures and display different dynamics,
the cell regulates their assembly and disassembly in related ways.
Polymer growth is polar in both cases and occurs at the plus ends
of microtubules and the barbed ends of actin filaments. Both
polymers have specific nucleating proteins, assemble with the
help of polymerases, and disassemble with the aid of severing
proteins and depolymerases (1–4). How these various activities
coordinate to create the cytoskeleton is a central question in cell
biology (5). This work focuses on assembly.
The main promoters of polymer growth are the XMAP215/Dis
family for microtubules and the formins for actin (4, 6–9). The
function of formins in actin polymerization is well characterized.
Formins have two key domains that are important for their activ-
ity, FH1 and FH2 (8, 10). While the FH2 domain is necessary
for binding to the barbed end of actin, repeats of polyproline
in the FH1 domain are required to interact with actin/profilin
complexes and recruit them to the barbed end (4, 11, 12).
Much less is known about how the regions of XMAP215
coordinate in promoting microtubule growth (13). Recent work
has shown that XMAP215 acts as a classic catalyst (14). At physi-
ological tubulin concentrations, XMAP215 is a tubulin poly-
merase that promotes incorporation of tubulin into the growing
plus end. However, in the absence of free tubulin, XMAP215
accelerates depolymerization of GMPCPP-stabilized microtu-
bules. Therefore, XMAP215 can act both as a polymerase and
ells assemble and disassemble actin filaments and microtu-
bules to carry out a vast array of functions, such as defining
a depolymerase, and its activity depends on the concentration
of its substrate, tubulin. However, how the various domains of the
XMAP215 protein contribute to these activities is not known.
Members of the XMAP215/Dis1 family are characterized by
a varying number of TOG domains at their N termini (Fig. 1).
Based on mutants in TOG domains that interfere with tubulin
binding (15) and protein activity (16–20), it has been proposed
that TOG binding to tubulin is required for its catalytic activity
(21); however, there is no proof for this idea. It is also not known
how the various properties of XMAP215—association with the
tubulin dimer, binding to the microtubule lattice and plus end,
diffusion along the lattice—depend on the TOG domains. We
have therefore sought to determine how the TOG domains, and
possibly other domains, contribute to microtubule polymerase
activity.
Results
TOG Domains are Required to Bind Free Tubulin and for Microtubule
Polymerization. Two key loops in TOG3 of Zyg9, the XMAP215/
Dis1 homolog in Caenorhabditis elegans, were previously identi-
fied as being important for interaction with free tubulin (15).
We mutated two conserved residues in the corresponding con-
served loops of all five TOG domains of XMAP215 to determine
their contribution to microtubule growth promotion (Fig. 1).
This TOG1-5AA mutant did not promote growth under any con-
ditions (Fig. 2A). We assayed for growth above 5 μM tubulin
because no growth is seen at or below this concentration in the
absenceofanygrowth-promotingfactorduetothehighfrequency
of catastrophes at low tubulin concentrations (14). We then as-
sayed this mutant for the other properties characteristic of the
wild-type protein: free tubulin binding, microtubule binding,
microtubule lattice diffusion, and tip tracking. The mutant’s
affinity for free tubulin was severely reduced, as shown by size
exclusion chromatography (Fig. 2B and C).
TOGs 1 and 2 are Critical for Polymerase Activity. These data demon-
strate that the TOG domains play an essential function in med-
iating XMAP215’s ability to bind free tubulin and in promoting
microtubule growth; however, they do not allow us to assess the
contribution of individual TOG domains. To address this ques-
tion, we generated a series of constructs in which TOG domains
Author contributions: P.O.W., J.H.S., A.P., G.J.B., A.A.H., and J.H. designed research;
P.O.W., J.H.S., A.P., and S.R. performed research; P.O.W., J.H.S., M.Z., G.J.B., A.A.H., and
J.H. analyzed data; and P.O.W., J.H.S., M.Z., A.A.H., and J.H. wrote the paper.
The authors declare no conflict of interest.
This article is a PNAS Direct Submission.
Freely available online through the PNAS open access option.
1To whom correspondence may be addressed. E-mail: hyman@mpi-cbg.de or howard@
mpi-cbg.de.
This article contains supporting information online at www.pnas.org/lookup/suppl/
doi:10.1073/pnas.1016498108/-/DCSupplemental.
www.pnas.org/cgi/doi/10.1073/pnas.1016498108 PNAS ∣ February 15, 2011 ∣ vol. 108 ∣ no. 7 ∣ 2741–2746
BIOCHEMISTRY

Page 2

were individually or pair-wise mutated. We tested the ability of
these combinations of TOG mutants to promote microtubule
growth using two different assays. First, we determined the activ-
ities of the various mutants at a fixed tubulin concentration of
5 μM. No growth was seen from microtubule seeds with 5 μM
tubulin alone (as noted above). However, when full length
XMAP215 was added, we saw a dose dependent increase in
growth rate that reached a maximum at approximately 200 nM
protein. All functional point mutants reached maximum growth
at this concentration but with significantly different maximum
growth rates (Vmax) (Fig. 3A). And second, we measured the
microtubule growth rate over a range of tubulin concentrations
at a fixed XMAP215 concentration of 200 nM. The growth rates
increased linearly with tubulin concentration for all mutants;
however, the mutants displayed a strikingly different contribution
to polymerization activity (Fig. 3B). The two assays gave consis-
tent results: The mutation of TOGs 3∕4 and TOG 5 had marginal
effects on activity. TOG 1 and TOG 2 contributed strongly to the
activity; and the double mutation of TOG 1 and 2 resulted in a
protein with minimal polymerase activity.
Polymerase Activity Correlates with Binding to Free Tubulin Dimers.
Considering that mutation of all TOG domains in the full length
protein prevented interaction with free tubulin dimers, we
wanted to see how competent our array of mutants were to bind
free tubulin. As was done for the wild-type GFP-tagged protein,
we determined this by size exclusion chromatography. As a mea-
sure of tubulin binding, we computed amount of tubulin bound
per XMAP215 (Figs. S1 and S2). In order to compare tubulin
binding to growth, we determined the Vmaxfor all mutants by
repeating the growth experiments at saturating XMAP215 con-
centrations (200 and 400 nM). Their abilities to bind tubulin
fell in a range between that of the wild-type protein and the
TOG1-5AA mutant (Fig. S1). In fact, the amount of tubulin
bound per XMAP215 was proportional to the polymerase activity
(Fig. 3C). Therefore, all TOG domains contribute to both the
affinity of XMAP215 for the tubulin dimer and polymerization
activity, suggesting that the affinity for tubulin plays an important
role in the polymerization mechanism (see Discussion).
Efficiency of Polymerase Activity Increases with Increasing Lattice
Affinity. Since TOGs 1 and 2 showed the most significant contri-
bution to activity, we asked whether they were sufficient to pro-
mote microtubule growth. We expressed an XMAP215 fragment
containing just TOGs 1 and 2 in Escherichia coli. This fragment
had little polymerase activity at 200 nM protein, where we see
maximal growth with wild-type protein (Fig. 3A). We therefore
attempted to determine what features displayed by the wild-type
protein were absent with the TOG12 fragment. The TOG12 frag-
ment was still able to bind tubulin dimers (Fig. 4A) and microt-
ubule ends (Fig. 4D); however, it had a severely reduced affinity
for the microtubule lattice (Fig. 4B). We therefore tested whether
addition of a microtubule-binding domain would enhance the
XMAP215 (Xl), ch-TOG (Hs), Msps (Dm), MOR1 (At)
Stu2p (Sc), Dis1p (Sp)
TOG1
TOG2
XMAP215/Dis1/TOG family
W21A
K102A
W292A
K373A
W610A
K691A
W870A
K950A
F1250A
K1335A
N
C
N
C
N
C
TOG1
N
C
Zyg-9 (Ce)
XMAP215TOG1-5AA
TOG2
TOG3TOG4TOG5
TOG1 TOG2TOG3
wild-type TOG domain
inactive TOG domain
lattice binding domain
2065
1415
888
Fig. 1.
proteins. The human, frog, fly, plant, worm, and yeast homologues are
shown. (B) Positions of the point mutations in XMAP215TOG1-5AA.
XMAP215 and its homologues. (A) The three major classes of TOG
Absorbance (mAu)
Volume (mL)
XMAP1-5AA + Tubulin
XMAP1-5AA
Tubulin
Growth rate (dimers/s)
Tubulin (µM)
6.5 7.0 7.5 8.0 8.59.0 9.510.0
0
10
20
30
40
50
XMAP215 + Tubulin
XMAP215
Tubulin
Absorbance (mAu)
Volume (mL)
Growth rate (µm/min)
A
B
0246810 12
0
25
50
75
100
125
150
0.0
0.9
1.8
2.8
3.7
4.6
5.5

1-5AA
XMAP215
Tubulin
60 s
2µm
2µm
2µm
6.57.07.5 8.08.59.0 9.510.0
0
10
20
30
40
50
Tubulin
1-5AA
XMAP215
C
60 s
60 s
Fig. 2.
and does not bind tubulin. (A) XMAP215-GFP shows a fivefold increase in
tubulin growth at all tubulin concentrations examined while XMAP215-
TOG1-5AA-GFP does not. Error bars represent SEM (N ≥ 15 for each point).
Kymographs show a representative single microtubule from each experi-
ment. Trend lines project back to the critical concentration in the absence
of catastrophes. Rhodamine-labeled GMPCPP seeds are in red. Alexa488
labeled tubulin is in green. (B) XMAP215-GFP binds tubulin. Chromatography
experiments showing A280 over elution volume. Three traces are shown in
each graph: XMAP215-GFP alone in green, tubulin alone in blue, XMAP215-
GFP in combination with tubulin in red. (C) XMAP215TOG1-5AA-GFP does
not bind tubulin. Chromatography experiments showing A280 over elution
volume. Three traces are shown in each graph: XMAP215TOG1-5AA-GFP
alone in green, tubulin alone in blue, XMAP215-GFP in combination with
tubulin in red.
XMAP215TOG1-5AA-GFP does not promote microtubule growth
2742
∣
www.pnas.org/cgi/doi/10.1073/pnas.1016498108 Widlund et al.

Page 3

activity of the TOG12 fragment at lower concentrations. We
decided to use the K-loop of the kinesin KIF1A to target the
TOG12 fragment to the microtubule lattice. We chose this loop
because it is a simple basic region that has been reported to
effectively target KIF1A and another kinesin, MCAK, to the
microtubule lattice (22, 23), and we wanted to exclude other
0
0.9
1.8
2.8
3.7
4.6
Growth rate (dimers/s)
Protein concentration (nM)
02468 1012
0
25
50
75
100
125
150
0.0
0.9
1.8
2.8
3.7
4.6
5.5
Growth rate (mm/min)
Growth rate (dimers/s)
Tubulin concentration (mM)
XMAP215
TOG34AA
TOG5AA
TOG1AA
TOG2AA
N
C
N
C
N
C
N
C
N
C
N
C
Growth rate (mm/min)
A
B
0100200300 400
0
25
50
75
100
125
TOG12AA
WT
12AA
34AA
5AA
1AA
2AA
1-5AA
Tubulin Bound per XMAP215
C
Growth rate (mm/min)
0.0 0.1 0.2 0.3 0.40.50.6
0
1
2
3
4
WT
12AA
34AA
5AA
1AA
2AA
1-5AA
Fig. 3.
(A) Microtubule growth rate by XMAP215-GFP and all point mutants with
increasing XMAP215 concentration at 5 μM tubulin. Error bars represent
SEM (N ≥ 10 for each point). (B) Growth promotion by XMAP215-GFP and
all point mutants with increasing tubulin concentration. Error bars represent
SEM (N ≥ 10 for each point). (C) Maximum average growth rate of individual
XMAP215 mutants plotted against their ability to bind tubulin as determined
from chromatograms (Figs. S1 and S2). The maximum average growth rate
was determined in at least four separate experiments at either 200 or
400 nM protein. Because no significant difference was seen between 200
and 400 nM protein for each construct, the rates were averaged (N ≥ 10
for each experiment). Error bars represent SEM. Tubulin binding was mea-
sured in duplicate.
Characterization of all XMAP215 TOG domain point mutants.
TOG12
TOG12+
TOG12+++
10nM 50nM100nM200nM 400nM 1000nM
Protein concentration (nM)
Growth rate (dimers/s)
B
C
D
Time
30 s
2µm
6µm
7.58.08.5
Volume (mL)
9.0 9.5 10.0
0
10
20
30
40
50
Absorbance (mAu)
TOG12 + Tubulin
TOG12
Tubulin
A
4µm
E
TOG12-GFP TOG12+++-GFP
TOG12+++
TOG12+
TOG12
XMAP215
0500 1000 15002000
0
20
40
60
80
100
120
0.0
0.7
1.5
2.2
3.0
3.7
4.4
Growth rate (µm/min)
Fig. 4.
Chromatography experiments showing A280 over elution volume. Three
traces are shown in each graph: TOG12 alone in green, tubulin alone in blue,
TOG12 in combination with tubulin in red. (B) The Kif2A lattice-binding
domain increases affinity of TOG12 for the microtubule lattice. Fragments
were incubated with rhodamine-labeled GMPCPP-stabilized microtubules
with increasing protein concentration as indicated. The merged image shows
microtubules in red and GFP in green. (C) Microtubule growth rate with
increasing TOG12, TOG12+, and TOG12+++ concentration at 5 μM tubulin.
Error bars represent SD (N ≥ 10 for each point). (D) TOG12 binds the plus end
of stabilized GMPCPP seeds. Rhodamine-labeled GMPCPP seeds are in red.
TOG12-GFP is in green. (E) TOG12+++ GFP tracks growing microtubule plus
ends. Rhodamine-labeled GMPCPP seeds are in red. TOG12+++GFP is in
green.
Construction of a minimal polymerase. (A) TOG12 binds tubulin.
Widlund et al.PNAS
∣
February 15, 2011
∣
vol. 108
∣
no. 7
∣
2743
BIOCHEMISTRY

Page 4

activities that were potentially present in the region surrounding
the native microtubule-lattice-binding domain. Indeed, addition
of one K-loop increased the affinity of the TOG12 fragment to
the microtubule lattice; addition of three repeats of this domain
further increased microtubule-binding activity (Fig. 4B), similar
to that of the wild-type protein (see below). We then assayed the
ability of these fusion proteins to promote microtubule growth.
While the TOG12 fragment alone showed activity at or above
400 nM protein, the TOG12+ and TOG12+++ were active at
much lower concentrations. Strikingly, all fragments appear to
have a maximum growth rate of approximately 3 μm∕min but
reach this maximum at very different protein concentrations
(Fig. 4C). Furthermore, their activities correspond very well with
the fragments’ affinities for the microtubule lattice. These experi-
ments define a minimal “bonsai” polymerase, namely a TOG12
fragment with a strong microtubule-binding domain, which be-
haves very similar to the wild-type protein. It binds tubulin and
themicrotubulelattice,promotesfastgrowthinthelownMrange,
and is able to track growing microtubule tips (Fig. 4E and
Movie S1).
The XMAP215 Microtubule-Lattice-Binding Domain Lies Between TOG4
and TOG5. Because the TOG12 fragment depended on a micro-
tubule-lattice-binding domain to function, we suspected that
the native XMAP215 has a microtubule-lattice-binding domain.
A region with high affinity (KD< 1 μM) for microtubules was
mapped to a region between residues 1150 and 1325 (13, 24). This
region includes part ofa region between TOG4 andTOG5as well
as part of TOG5 itself. We wanted to know if TOG5 or any of the
TOGs are involved in lattice binding. A fragment containing
TOG1-4 (residues 1–1081) bound poorly to the microtubule lat-
tice, consistent with published observations (Figure 5). We made
an additional fragment containing the region up to TOG5 (resi-
dues 1–1235); it bound microtubules similar to wild type (Fig. 5).
Taken together with the published analyses, our experiments
suggest that the microtubule-binding domain resides in the region
between residues 1150 and 1235, a region that excludes TOG5.
This region is basic, with a predicted pI of 9.8. This is consistent
with what is seen in XMAP215 homologues. The Saccharomyces
cerevisiae homologue of XMAP215, Stu2, has a basic linker after
the TOG domains, which has been shown to bind to the micro-
tubule lattice (17, 21). This basic region is also found in Schizo-
saccharomyces pombe Dis1 (25). This region combined with the
TOG domains cooperate to promote robust microtubule growth
at nanomolar concentrations of protein.
Discussion
We have shown that the ability of XMAP215 to efficiently cata-
lyze the incorporation of tubulin dimers into a microtubule is
dependent on tubulin binding and microtubule-lattice-binding.
We further demonstrate that these activities are mediated by
functionally distinct domains. Multiple TOG domains are neces-
sary to increase affinity for the tubulin dimer. Mutants in which
these interactions are disrupted are able to efficiently target the
microtubule lattice but are impaired in their capacity to promote
the incorporation of tubulin dimers into a growing microtubule
end. We have also identified a microtubule-lattice-binding do-
main on XMAP215, localized between TOGs 4 and 5. Deletion
of this basic region strongly inhibits the association of XMAP215
with the microtubule lattice. However, protein fragments lacking
this domain are still able to promote robust microtubule growth
when artificially targeted to the microtubule lattice.
The microtubule plus end can be thought of as an enzyme for
the incorporation of tubulin. This enzyme is inefficient, however,
because the growth rate in pure tubulin is well below the diffusion
limit: The association rate of GTP-tubulin for the individual pro-
tofilament plus ends is only 0.3 μM−1s−1, about 20 times smaller
than that of ATP-actin for individual protofilament barbed ends
in an actin filament (26). It is likely that only a fraction of the
tubulin dimers that collide with the plus end become stably incor-
porated into the microtubule lattice. XMAP215 can be thought of
as a nonessential activator of the microtubule end that increases
the fraction of tubulin dimers that successfully incorporate into
the microtubule lattice (14, 27). When sufficient XMAP215 is
added to saturate the plus end, the association rate is increased
fivefold to 1.5 μM−1s−1(per protofilament plus end).
In this work, we demonstrate that removal of functional TOG
domains affects the ability of XMAP215 to increase the associa-
tion rate for tubulin to the microtubule plus end. Accordingly,
all of the TOG domain point mutants we described showed a
lowered maximal growth rate as well as an association rate that
falls between 0.3 and 1.5 μM−1s−1per protofilament end. At a
fixed tubulin concentration, addition of increasing amounts of
XMAP215 resulted in a dose response that achieved maximal
growth rate at approximately 200 nM protein, suggesting that
the plus ends are saturated with XMAP215 at this concentration.
All of the point mutants reached their maximal growth rate at
approximately 200 nM protein, consistent withthe idea that these
proteins are able to bind to the microtubule lattice and target to
the plus end similar to the wild-type protein. We therefore con-
clude that the TOG domains and the tubulin affinity they confer
determinethemaximalgrowthrate(νmax)atanyfixedtubulincon-
centration and the decreased capacity for promoting microtubule
growth observed with these mutants can be attributed to the fact
that ends are saturated with a less effective polymerase (Fig. 6A).
XMAP215 requires a region between TOGs 4 and 5 to target
to the microtubule lattice. This positively charged linker may bind
to the E-hooks of microtubules, because it has been shown that
XMAP215 binds poorly to microtubules whose E-hooks have
been removed by subtilisin treatment (14). Removal of the micro-
tubule-binding domain affects the ability of the TOG domains of
XMAP215 to work at lower concentrations. The processivity of
XMAP215 at the microtubule plus end can be attributed to a
combination of tubulin binding and incorporation into the lattice
followed by lattice diffusion to the new end via the interaction
MTs
GFP Merge
TOG1-4
TOG1-4+
WT
TOG12
C
N
N
N
N
2µm
TOG1-5AA
C
N
1235
1081
560
Fig. 5.
TOG5. 200 nM XMAP215-GFP and various deletion mutants were incubated
with rhodamine-labeled GMPCPP-stabilized microtubules and imaged. The
left image shows the microtubules. The center image shows the GFP signal.
The right image shows the merged image with microtubules in red and GFP
in green.
The microtubule-lattice-binding domain lies between TOG4 and
2744
∣
www.pnas.org/cgi/doi/10.1073/pnas.1016498108Widlund et al.

Page 5

between the lattice-binding domain and the E-hooks. We showed
that a TOG12 construct became effective at lower concentrations
when the affinity for the microtubule lattice was increased by
the addition of a nonnative microtubule-binding domain. TOG12
is still able to bind and diffuse along the microtubule lattice with-
out this domain and can even bind the microtubule plus end at
low nanomolar concentrations but needs to be at a much higher
concentration in solution to saturate the growing plus end. As
microtubule-lattice-binding domains are added to this fragment,
its affinity for the microtubule is increased resulting in a higher
flux of XMAP215 to the growing plus end (Fig. 6B). A conse-
quence of increased flux is that plus ends become saturated at
lower concentrations of protein in solution. We propose that
XMAP215 and the TOG12+++ fragment target to and maintain
association with the growing plus end by a mechanism that is
similar to their association with the microtubule lattice, namely
through basic regions. Once targeted to the end, the tubulin
affinity determines the maximal growth rate (νmax) at any fixed
tubulin concentration (Fig. 6A).
The TOG12+++ bonsai protein displayed characteristics very
similar to wild-type XMAP215. We therefore define this as a
minimal polymerase. It binds and diffuses on the microtubule
lattice, tracks polymerizing and depolymerizing plus ends, binds
tubulin, and promotes growth exclusively at the plus end. The
maximal growth rate was 3 μm∕min at 5 μM tubulin as opposed
to 4 μm∕min for wild-type XMAP215. The absence of TOGs 3, 4,
and 5 can account for this difference as mutation of these TOGs
resulted in mutants that showed similar maximal growth rates.
The KDfor this construct was also slightly higher compared to
wild-type due to differences in affinities of the nonnative and
native microtubule-lattice-binding domains.
Because the nondimerized TOG12 is sufficient to promote
robust microtubule growth, we consider this further evidence
against growth promotion by addition of oligomers to the micro-
tubule plus end (19, 28, 29). TOG12 could bind at most an oli-
gomer of 2 dimers but not 5–7 dimers as required by a shuttle
model. Instead, we argue that the additional TOG domains
provide additional affinity for individual tubulin dimers. To be a
catalyst, XMAP215 must bind tubulin with a high affinity but
also be able to release once the tubulin has been incorporated.
Therefore, multiple binding sites within the same molecule that
have high off rates would be ideal. The combined avidity of the
multiple TOG domains results in strong overall binding, while the
high off rates of individual TOG domains could allow for quick
release.
As with the formins in actin polymerization, we have now
separated domains that are critical for XMAP215 function.
Formins have FH1 and FH2 domains, whereas XMAP215 has
TOG domains and a basic lattice-binding region. There are, how-
ever, some notable differences. While formins use polyproline
repeats in FH1 domains to recruit multiple G-actin monomers,
TOG domains are used to increase affinity for one tubulin dimer.
Furthermore, while FH2 domains link formins to the barbed end,
the basic lattice-binding domain does not bind the end specifi-
cally. Instead, TOG domains are critical for catalysis while the
basic lattice-binding domain is important for targeting to the
microtubule end. We propose that the processive polymerization
of XMAP215 is a combination of tubulin binding from solution
and incorporation into the lattice, followed by lattice diffusion
to the new end via the interaction between the lattice-binding
domain and the E-hooks (Fig. 6C). The TOG12 fragment with
low lattice-binding activity can still associate specifically with
the ends of GMPCPP-stabilized microtubules (Fig. 4D), suggest-
ing that the TOG domains recognize a surface that is not exposed
on the microtubule lattice. We found that none of our mutants
separated microtubule end binding and tip-tracking activity from
tubulin binding, suggesting that those contacts that are made to
free tubulin are the same as those made to tubulin that has newly
been incorporated into the microtubule lattice. Therefore, con-
sistent with the idea that XMAP215 is a catalyst (14), TOG
domains bind to free tubulin, incorporated tubulin, and the tran-
sition state between the two in a similar way. We propose that it is
the affinity for this transition state that determines the maximal
growth rate (νmax) at a fixed tubulin concentration. Furthermore,
our observation that the growth rate at any XMAP215 protein
concentration correlates with the affinity of that protein for the
microtubule lattice implies that XMAP215 associates with the
microtubule end in a similar way that it binds to the microtubule
lattice. Thus, the presumed electrostatic interaction with the
lattice also occurs at the ends. We have incorporated these data
into the previous kinetic model (see SI Text).
Materials and Methods
Plasmid Construction. A pFastBac construct with XMAP215-GFP was described
previously (14). A fragment containing all five TOGs with the following
mutations was synthesized: W21A, K102A, W292A, K373A, W610A, K691A,
W870A, K950A, F1250A, K1335A. This EcoRI, KasI fragment was cloned
into the XMAP215-GFP vector to give XMAP215-TOG1-5AA-GFP. Subsequent
combinations of TOG mutations were made by switching TOGs between
plasmids using unique restriction sites between them: StuI, NotI, AatII, AgeI.
TOG1-4GFP, TOG1-4+GFP, ΔTOG1-4GFP, and ΔTOG1-4+GFP were made by
amplifying a PCR fragment lacking the region to be deleted, followed
DpnI digestion and subsequent ligation of the blunt ended fragment (linear
plasmid) to circularize the plasmid. TOG12 is the same as fragment 1 de-
scribed in ref. 16. All remaining fragments were derived from this plasmid
by mutagenesis described above. The TOG12 “bonsai” construct (30) was
made by inserting the following sequence into a SalI site a the 3′ end:
tcgacaataagaacaaaaagaaaaagaaagtcgacaataagaacaaaaagaaaaagaaagtcgac.
XMAP215 concentrationXMAP215 concentration
Growth rate
A
B
Growth rate
C
1. Lattice diffusion to end2. Stabilization of collision complex
.
3. Isomerization4. Tubulin release
νmax
∆ KD
T + XT
k’ [T]
+1
-1
k’
T XTn
XTn + 1
k’+2
k’-2
n
k-
Tn + 1
k+[X]
00
00
∆
Fig. 6.
result in mutants that have lowered maximal growth rates (νmax) at any fixed
tubulin concentration. The graph shows theoretical dose response of a pro-
tein with increasing affinities for the tubulin dimer that lead to increasing
affinities for the transition state (compared to the microtubule end alone):
α ¼ 2 in red, α ¼ 5 in green, and α ¼ 10 in blue. β ¼ 1, ½T? ¼ 0.1K1(see SI Text).
(B) Mutation or removal of the microtubule-binding domain in XMAP215
results in constructs that have the same νmaxbut a higher KD. The graph
shows theoretical dose response of a protein with a constant νmaxand an
altered konfor microtubule-lattice binding: 4x reduced in red, 2x reduced in
green, not reduced in blue. β ¼ 1, ½T? ¼ 0.1K1(see SI Text). (C) Model of
TOG12+++ on the plus end of a microtubule. (1) Diffusion to the end via
the lattice-binding domain. (2) TOG12+++ stabilizes the incoming dimer.
(3) TOG12+++ remains bound to the incorporated dimer. (4) Release of the
dimer. The transitions between each state are described by the reaction
scheme in the SI Text.
XMAP215 as a catalyst. (A) Mutation or removal of TOG domains
Widlund et al.PNAS
∣
February 15, 2011
∣
vol. 108
∣
no. 7
∣
2745
BIOCHEMISTRY

Page 6

The GFP tag was introduced on a NotI fragment into the 3′ end after the
microtubule-binding domain.
Protein Expression and Purification. Full length XMAP215, XMAP215-GFP, all
full length point mutants, TOG1-4GFP, TOG1-4+GFP, ΔTOG1-4GFP, and
ΔTOG1-4+GFP were expressed in SF+ cells using the Bac-to-Bac system from
Invitrogen essentially as described previously (14) except baculovirus infected
insect cell (BIIC) stocks were used (31) (see SI Text). All remaining constructs
were expressed in E. coli BL21 with plasmid pRARE (see SI Text).
Tubulin and Microtubule Preparation. Porcine brain tubulin was purified as
described (32). Labeling of cycled tubulin with Alexa Fluor 488 or TAMRA
(Invitrogen) was performed as described (33). GMPCPP microtubules were
grown as described (34).
Imaging. The total-internal-reflection fluorescence imaging was performed
with a setup described previously (14, 34, 35). The setup incorporates an
Andor DV887 iXon camera on a Zeiss Axiovert 200 M microscope using a Zeiss
100X/1.45 a Plan-FLUAR objective. Standard filter sets were used to visualize
tetramethylrhodamine, Alexafluor 488, and GFP.
Assay Conditions. The preparation of silanized cover glasses and perfusion
chambers was previously described (14, 34, 35). Reaction channels were first
rinsed with BRB80: 80 mM PIPES at pH 6.9, 1 mM MgCl2, and 1 mM EGTA.
Reaction channels were incubated with either 1% antirhodamine antibody
(Invitrogen) or 50 μg∕mL neutravidin (Sigma) in BRB80 for 5 min, followed by
1% pluronic F127 (Sigma) in BRB80 for 5 min, and finally rhodamine-labeled
or rhodamine and biotinylated, GMPCPP-stabilized microtubule seeds for
15 min. Channels were washed once with BRB80 and once with imaging
buffer (IB): BRB80 supplemented with 75 mM KCl, 0.1 mg∕ml BSA, 1% β-mer-
captoethanol, 40 mM glucose, 40 mg∕ml glucose oxidase, and 16 mg∕ml
catalase. We used an objective heater (Zeiss) to warm the sample to 35°C.
Microtubule growth at a fixed tubulin concentration and increasing
XMAP215 concentration was done with 4.5 μM unlabeled tubulin, 0.5 μM
Alexa Fluor 488 tubulin, XMAP215-GFP (0–200 nM), and 1 mM GTP. Microtu-
bule growth with increasing tubulin concentration and fixed XMAP215 con-
centration was done with 6, 7, 8 or 7, 8, and 9 μM tubulin by combining
varying amounts of 6 μM tubulin (5.5 μM unlabeled tubulin, 0.5 μM Alexa
Fluor 488 tubulin) and 9 μM tubulin (8 μM unlabled, 1.0 μM Alexa Fluor
488 tubulin), 200 nM XMAP215, and 1 mM GTP.
Size Exclusion Chromatography. Size exclusion chromatography was carried
out similar to ref. 14. Briefly, a Tosoh TSKgelG5000PWXL column was equili-
brated in 25 mM TrisHCl pH 7.5, 75 mM NaCl, 1 mM MgCl2, 1 mM EGTA, 0.1%
Tween20, 1 mM DTT. XMAP215 (5.7 μM) and tubulin (14.7 μM) or the equiva-
lent buffer in case of single protein injection were mixed with 0.2 mM GTP in
50 μL total volume, incubated for 10 min on ice and then injected onto the
Tosoh TSKgelG5000PWXL size exclusion column. For the TOG12 binding
experiment, 15 μM TOG12 and15 μM tubulin were used in50 μLtotal volume.
Data Analysis. Microtubule growth measurements were performed in Meta-
morph (Universal Imaging). Images were processed using Metamorph and
Image J. Curve fitting was done in OriginPro (Origin Lab). Tubulin binding
was determined using the heights of the XMAP215, XMAP215:Tubulin and
Tubulin peaks (Fig. S2).
ACKNOWLEDGMENTS. We thank J. Al-Bassam and S. Harrison for helpful
discussions; D. Drechsel, B. Borgonovo, and R. Lemaitre for advice and
technical assistance; and C. Gell for help with microscopy. We thank members
of the Hyman and Howard laboratories for advice and discussions. P.O.W.
was supported by a European Molecular Biology Organization long-term
fellowship, J.H.S. was supported by the National Institutes of Health National
Research Service Award program and the Deutsche Forschungsgemeinschaft,
G.J.B. acknowledges support from the Natural Sciences and Engineering
Research Council of Canada (Grant 372593). M.Z. is supported by the Inter-
national Human Frontier Science Program Organization. This work was
funded by the Max Planck Society.
1. Howard J, Hyman AA (2007) Microtubule polymerases and depolymerases. Curr Opin
Cell Biol 19:31–35.
2. Pantaloni D, Le Clainche C, Carlier MF (2001) Mechanism of actin-based motility.
Science 292:1502–1506.
3. Desai A, Mitchison TJ (1997) Microtubule polymerization dynamics. Annu Rev Cell Dev
Biol 13:83–117.
4. Paul A, Pollard T (2009) Review of the mechanism of processive actin filament elonga-
tion by formins. Cell Motil Cytoskeleton 66:606–617.
5. Goode BL, Drubin DG, Barnes G (2000) Functional cooperation between the micro-
tubule and actin cytoskeletons. Curr Opin Cell Biol 12:63–71.
6. Gard DL, Kirschner MW (1987) A microtubule-associated protein from Xenopus eggs
that specifically promotes assembly at the plus-end. J Cell Biol 105:2203–2215.
7. Gard DL, Kirschner MW (1987) Microtubule assembly in cytoplasmic extracts of
Xenopus oocytes and eggs. J Cell Biol 105:2191–2201.
8. Goode BL, Eck MJ (2007) Mechanism and function of formins in the control of actin
assembly. Annu Rev Biochem 76:593–627.
9. Kinoshita K, Habermann B, Hyman AA (2002) XMAP215: A key component of the
dynamic microtubule cytoskeleton. Trends Cell Biol 12:267–273.
10. Sagot I, Rodal AA, Moseley J, Goode BL, Pellman D (2002) An actin nucleation mechan-
ism mediated by Bni1 and profilin. Nat Cell Biol 4:626–631.
11. Romero S, et al. (2004) Formin is a processive motor that requires profilin to accelerate
actin assembly and associated ATP hydrolysis. Cell 119:419–429.
12. Xu Y, et al. (2004) Crystal structures of a Formin Homology-2 domain reveal a tethered
dimer architecture. Cell 116:711–723.
13. Gard DL, Becker BE, Josh Romney S (2004) MAPping the eukaryotic tree of life:
Structure, function, and evolution of the MAP215/Dis1 family of microtubule-
associated proteins. Int Rev Cytol 239:179–272.
14. Brouhard GJ, et al. (2008) XMAP215 is a processive microtubule polymerase. Cell
132:79–88.
15. Al-Bassam J, Larsen NA, Hyman AA, Harrison SC (2007) Crystal structure of a TOG
domain: Conserved features of XMAP215/Dis1-family TOG domains and implications
for tubulin binding. Structure 15:355–362.
16. PopovAV, et al. (2001) XMAP215 regulatesmicrotubule dynamics through twodistinct
domains. EMBO J 20:397–410.
17. Wang PJ, Huffaker TC (1997) Stu2p: A microtubule-binding protein that is an essential
component of the yeast spindle pole body. J Cell Biol 139:1271–1280.
18. van Breugel M, Drechsel D, Hyman A (2003) Stu2p, the budding yeast member of
the conserved Dis1/XMAP215 family of microtubule-associated proteins is a plus
end-binding microtubule destabilizer. J Cell Biol 161:359–369.
19. Slep K, Vale R (2007) Structural basis of microtubule plus end tracking by XMAP215,
CLIP-170, and EB1. Mol Cell 27:976–991.
20. Bellanger JM, Gonczy P (2003) TAC-1 and ZYG-9 form a complex that promotes micro-
tubule assembly in C. elegans embryos. Curr Biol 13:1488–1498.
21. Al-Bassam J, van Breugel M, Harrison SC, Hyman A (2006) Stu2p binds tubulin and
undergoes an open-to-closed conformational change. J Cell Biol 172:1009–1022.
22. Ovechkina Y, Wagenbach M, Wordeman L (2002) K-loop insertion restores microtu-
bule depolymerizing activity of a “neckless” MCAK mutant. J Cell Biol 557–562.
23. Okada Y, Hirokawa N (1999) A processive single-headed motor: Kinesin superfamily
protein KIF1A. Science 283:1152–1157.
24. Spittle C, Charrasse S, Larroque C, Cassimeris L (2000) The interaction of TOGp with
microtubules and tubulin. J Biol Chem 275:20748–20753.
25. Nakaseko Y, Nabeshima K, Kinoshita K, Yanagida M (1996) Dissection of fission yeast
microtubule associating protein p93Dis1: Regions implicated in regulated localization
and microtubule interaction. Genes Cells 1:633–644.
26. Pollard TD (1986) Rate constants for the reactions of ATP- and ADP-actin with the ends
of actin filaments. J Cell Biol 103:2747–2754.
27. Segel IH (1975) Enzyme Kinetics: Behavior and Analysis of Rapid Equilibrium and
Steady-State Enzyme Systems (Wiley, New York).
28. Kerssemakers JW, et al. (2006) Assembly dynamics of microtubules at molecular reso-
lution. Nature 442:709–712.
29. Cassimeris L, Gard D, Tran PT, Erickson HP (2001) XMAP215 is a long thin molecule that
does not increase microtubule stiffness. J Cell Sci 3025–3033.
30. Ciferri C, et al. (2008) Implications for kinetochore-microtubule attachment from the
structure of an engineered Ndc80 complex. Cell 133:427–439.
31. Wasilko DJ, et al. (2009) The titerless infected-cells preservation and scale-up (TIPS)
method for large-scale production of NO-sensitive human soluble guanylate cyclase
(sGC) from insect cells infected with recombinant baculovirus. Protein Expres Purif
65:122–132.
32. Ashford AJ, Anderson SSL, Hyman AA (1998) Preparation of Tubulin from Bovine Brain
(Academic, San Diego) p 8.
33. Hyman A, et al. (1991) Preparation of modified tubulins. Methods Enzymol
196:478–485.
34. Gell C, et al. (2010) Microtubule dynamics reconstituted in vitro and imaged by single-
molecule fluorescence microscopy. Methods Cell Biol 95:221–245.
35. Helenius J, Brouhard G, Kalaidzidis Y, Diez S, Howard J (2006) The depolymerizing
kinesin MCAK uses lattice diffusion to rapidly target microtubule ends. Nature
441:115–119.
2746
∣
www.pnas.org/cgi/doi/10.1073/pnas.1016498108Widlund et al.