Exploration of the functional conservation across major
phylogenic boundaries to define the ancestry of genetic
functions will also be important. Within this framework,
the study of biological innovations, such as the origin of
new genes or their neo- or sub-functionalization, will be of
particular relevance to legume biologists, considering the
restricted occurrence of nitrogen-fixing symbiosis and its
proven functional link to the much older (approximately
400 million years ago) and certainly more ubiquitous
association of plants with mycorrhizal fungi [19,20].
The availability of legume genomes opens an unusually
valuable treasure chest, where efficient sequestration of
nitrogen, phosphate and carbon has been hiding within the
dynamic structure of gene networks. Unwinding this biological
wealth might provide an important insight on how
to preserve nonrenewable resources and the health of the
environment at the same time as fulfilling the increasing
need for sustainable biomass production.
Acknowledgements
We thank Alex Molnar for preparing Figure 1. K.S. is supported by grants
from Agriculture and Agri-Food Canada and the National Science and
Engineering Research Council of Canada (NSERC, grant R3277A01) and
J.S. is supported by the Danish National Research Foundation.
References
1 Mendel, G. (1866) Versuche u¨ber Pflanzen-hybriden. Verh. Natforsch.
Ver. Bru¨nn 4, 3–47. English translation from web site http://
www.mendelweb.org/Mendel.html
2 Darwin, C.R. (1880) The Power of Movement in Plants, John Murray
3 Sato, S. et al. (2008) Genome structure of the legume, Lotus japonicus.
DNA Res. DOI: 10.1093/dnares/dsn008
4 Dolye, J.J. and Luckow, M.A. (2003) The rest of the iceberg. Legume
diversity and evolution in a phylogenetic context. Plant Physiol. 131,
900–910
5 Hellriegel, H, and Wilfarth, H. (1888) Untersuchungen u¨ber die
Stickstoffnahrung der Gramineen und Leguminosen. Beilageheft zu
der Zeitschrift des Vereins fu¨r Ru¨benzucker-Industrie des Deutschen
Reichs, 234 (see ‘1888’ at http://www.asm.org/membership/
index.asp?bid=16899 for more information)
6 Handberg, K. and Stougaard, J. (1992) Lotus japonicus, an
autogamous, diploid legume species for classical and molecular
genetics. Plant J. 2, 487–496
7 Barker, D.G. et al. (1990) Medicago truncatula, a model plant for
studying the molecular genetics of the Rhizobium-legume symbiosis.
Plant Mol. Biol. Rep. 8, 40–49
8 Kaneko, T. et al. (2000) Complete genome structure of the nitrogenfixing
symbiotic bacterium Mesorhizobium loti. DNA Res. 7, 331–338
9 Kato, T. et al. (2000) Complete structure of the chloroplast genome of a
legume, Lotus japonicus. DNA Res. 7, 323–330
10 Masayoshi, K. (2000) Lotus japonicus ‘Miyakojima’ MG-20: an earlyflowering
accession suitable for indoor handling. J. Plant Res. 113, 507–
509
11 Szczyglowski, K. et al. (1998) Nodule organogenesis and symbiotic
mutants of the model legume Lotus japonicus. Mol. Plant Microbe
Interact. 11, 684–697
12 Schauser, L. et al. (1998) Symbiotic mutants deficient in nodule
establishment identified after T-DNA transformation of Lotus
japonicus. Mol. Gen. Genet. 259, 414–423
13 Perry, J.A. et al. (2003) A TILLING reverse genetics tool and a webaccessible
collection of mutants of the legume Lotus japonicus. Plant
Physiol. 131, 866–871
14 Colebatch, G. et al. (2004) Global changes in transcription orchestrate
metabolic differentiation during symbiotic nitrogen fixation in Lotus
japonicus. Plant J. 39, 487–512
15 Nishimura, R. et al. (2002) HAR1 mediates systemic regulation of
symbiotic organ development. Nature 420, 426–429
16 Radutoiu, S. et al. (2003) Plant recognition of symbiotic bacteria
requires two LysM receptor-like kinases. Nature 2003, 585–592
17 Murray, J.D. et al. (2007) A cytokinin perception mutant colonized
by Rhizobium in the absence of nodule organogenesis. Science 315,
101–104
18 Choi, H-K. et al. (2004) Estimating genome conservation between crop
and model legume species. Proc. Natl. Acad. Sci. U. S. A. 101, 15289–
15294
19 Duc, G. et al. (1989) First report of non-mycorrhizal plant mutants
(MycÀ
) obtained in pea (Pisum sativum L.) and fababean (Vicia faba
L.). Plant Sci. 60, 215–222
20 Kistner, C. et al. (2005) Seven Lotus japonicus genes required for
transcriptional reprogramming of the root during fungal and
bacterial symbiosis. Plant Cell 17, 2217–2229
1360-1385/$ – see front matter. Crown Copyright ß 2008 Published by Elsevier Ltd.
All rights reserved.
doi:10.1016/j.tplants.2008.08.001 Available online 30 August 2008
Techniques & Applications
Boosting tandem affinity purification of plant protein
complexes
Jelle Van Leene1
, Erwin Witters2,3
, Dirk Inze´1
and Geert De Jaeger1
1
Department of Plant Systems Biology, Flanders Institute for Biotechnology and Department of Molecular Genetics, Ghent
University, Technologiepark 927, 9052 Gent, Belgium
2
Center for Proteome Analysis and Mass Spectrometry, University of Antwerp, Groenenborgerlaan 171, 2020 Antwerp, Belgium
3
Flemish Institute for Technological Research, VITO, Boeretang 200, 2400 Mol, Belgium
Protein-interaction mapping based on the tandem affinity
purification (TAP) approach has been successfully
established for several systems, such as yeast and mammalian
cells. However, relatively few protein complex
purifications have been reported for plants. Here, we
highlight solutions for the pitfalls and propose a major
breakthrough in the quest for a better TAP tag in plants.
The rise of TAP
Over the past 20 years, a wide variety of methods have
been developed to explore protein interactions. Co-immunoprecipitation
or yeast two-hybrid were often the method
of choice, but the emergence of powerful, ultrasensitive
high-throughput mass spectrometry (MS), together with
the availability of comprehensive protein sequence repertoires,
has favored the development of methods relying
on in situ affinity purification of protein complexes.Corresponding author: De Jaeger, G. (geert.dejaeger@psb.ugent.be).
Update Trends in Plant Science Vol.13 No.10
517
Especially, the tandem affinity purification (TAP)
approach, based on the expression of a bait protein fused
to a double affinity tag (the TAP tag), has proven to be of
great value. The classical TAP tag consists of two immunoglobulin
G (IgG)-binding domains of protein A from
Staphylococcus aureus, a specific protease cleavage site
for elution by addition of the tobacco (Nicotiana tabacum)
etch virus (TEV) protease and a calmodulin-binding peptide
(CBP). Purification steps were optimized for highest
recovery while maintaining protein complex integrity.
TAP of protein complexes was first demonstrated in
yeast (Saccharomyces cerevisiae) [1] and was soon applied
in a wide variety of organisms, giving rise to high-quality
and comprehensive protein-interaction networks [2,3].
Nowadays, databases are filled with protein-interaction
data from TAP experiments, but in the plant research
field, the TAP approach considerably lags behind.
Here, we review and discuss the use of TAP in plants
and provide solutions to problems associated with the
technology.
TAP in plants: a brief overview
Until now, only a limited number of purifications from
plant material through TAP have been reported, and most
have been performed with the traditional yeast tag or with
a plant-adapted version, called the improved TAP tag
(TAPi) [4]. The latter contains the same modules as the
traditional yeast TAP tag but with an optimized codon
sequence for plants and an intron for higher gene expression
and without cryptic nuclear localization signals. Both
the traditional and the TAPi tags were used with success
for complex purifications from Arabidopsis thaliana [4–10]
and rice (Oryza sativa) [11,12]. A protein complex from
Arabidopsis was also purified with an alternative TAP tag
(TAPa), in which the CBP domain was replaced by a 9xMyc
and 6xHis sequence, preventing the non-specific purification
of endogenous calmodulin-binding proteins and
allowing purification of cation-dependent enzyme complexes
because no EGTA-containing buffers are required
[13]. Furthermore, the TEV cleavage site was replaced by
the more-specific and low-temperature-active human rhinovirus
3C protease cleavage site. Although this tag has
often been used as an epitope tag for protein gel blotting
[14] and in co-immunoprecipitation [14,15] or chromatin
immunoprecipitation experiments [15], only a single
protein complex has been characterized with the TAPa
tag [13]. In conclusion, in plants, the traditional yeast TAP
and the TAPi tags perform best so far.
TAP in plants: pitfalls and solutions
Approximately ten years after the proof of concept in yeast,
the few TAP data from plants demonstrate that problems
are associated with the method. Indeed, unlike in yeast,
efficient homologous recombination is not feasible in
higher plants. So, the endogenous protein and the tagged
counterpart will compete for complex assembly. To overcome
this pitfall, different strategies can be followed. The
TAP-tagged protein can be introduced into a mutant background,
where the endogenous protein is suppressed by
RNA interference [16] or is eliminated by transferred
DNA (T-DNA) insertion [13]. These complementation
approaches determine the functionality of the tagged
protein and increase the success rate of the purification
because more interactors are available for complex assembly
with the tagged protein. A more generic approach for
increasing competition is overexpression of the tagged bait,
a strategy used in all successful TAP reports in plants so
far. Another problem is false negative interactors, especially
when low-abundant complexes are studied. Because
proteins are present in a high dynamic range, varying from
only 10–100 copies to more than 107
copies per cell, and
because they cannot be amplified like polynucleotides by
PCR, the success rate of TAP depends on the amount of
protein complexes purified and the MS sensitivity. One
possibility for circumventing the problem of false negatives
is the combination of multiple TAP eluates from
parallel purifications [13]. Alternatively, the amount of
protein extract can be increased before purification. When
studying basic cell biological processes, plant cell suspension
cultures have a major advantage compared to whole
plants because they are fast growing and provide an
unlimited supply of synchronizable biological material.
Moreover, the PSB-D culture used previously [5] has a
ploidy level of 8C, meaning that more proteins are available
for complex assembly. The suspension culture is ideal
for investigating the cell cycle [5], but it can also be used to
isolate complexes involved in other fundamental processes,
such as primary metabolism, gene expression or
cell-wall synthesis.
The quest for a better TAP tag
Despite the valuable strategies described above, it was
clear that a major leap forward would only be possible
through further optimization of complex purification.
Therefore, we evaluated different TAP tags for plant cells
(Figure 1a). In line with the TAPa tag [13], we replaced the
CBP part in the traditional TAP tag with linear peptide
epitopes to reduce background. Although the final TAP
eluates obtained with this SFZZ tag (Figure 1a) looked
often much ‘cleaner’ on gel, the purified amount of complexes
was systematically low and only a few interacting
proteins could be sequenced (see supplementary Table S1).
Probably the TAPa tag also has to deal with this low
complex yield because six different TAP eluates had to
be pooled to identify the COP9 signalosome complex [13].
Despite a layer of background proteins sticking to the
calmodulin resin, in our laboratory, the best results with
respect to complex yield were, until recently, always
obtained with the traditional or the TAPi tags [4]. Background
proteins sticking to the resins and other false
positives from non-specific binding to complexes after
protein extract preparation can be determined by mock
and exogenous protein purifications using, for example,
green fluorescent protein (GFP) or b-glucuronidase. This
list of proteins is then systematically subtracted from the
original prey list. To get rid of bait-specific false positives
absent in the control TAP list, it is also valuable to repeat
purifications and to give more confidence to interactions
that were confirmed in multiple experiments [5] or that
achieved the best protein identification scores. Assigning
confidence scores to interactions by integrating interaction
data with other data sources is also rewarding, a method
Update Trends in Plant Science Vol.13 No.10
518
often applied in prediction of protein–protein interactions
[17,18].
A new TAP tag for plants: the GS tag
In our continuous search for an ideal TAP tag for plants,
we recently evaluated the GS tag [19], which combines
two IgG-binding domains of protein G with a streptavidin-binding
peptide, separated by two TEV cleavage
sites. This tag, developed to study mammalian protein
complexes, has been reported to give a tenfold increase in
bait recovery compared to the traditional TAP tag. We
adapted the GS protocol for plant cells (see online supplementary
Box S1) and tested background levels by comparing
two mock and two GFP purifications with the
traditional TAP tag (Figure 1c). Background levels,
counted as the average number of proteins identified in
two experimental repeats, dropped from 62 to 8 and from
87 to 11 proteins for mock and GFP purifications, respectively,
making MS analysis much less labor intensive and
the identification of genuine protein interactions easier,
Figure 1. Evaluation of alternative TAP tags in Arabidopsis cell suspension culture. (a) Overview of tested TAP tags. Abbreviations: 3xFlag, three copies of Flag tag; ProtA,
immunoglobulin G (IgG)-binding domain of protein A; ProtG, IgG-binding domain of protein G; SBP, streptavidin-binding peptide; StrepII, StrepII tag. (b) Higher bait
expression with GS tag fusions: 50 mg, 10 mg and 2 mg of total protein extracts of cultures expressing CKS1-tag, CDKA;1-tag or GFP-tag were separated via SDS–PAGE. Both
TAP and GS tag fusions were analyzed via immunoblotting with antibodies against bait proteins. For CKS1 and CDKA;1, endogenous protein levels are also shown. (c)
Comparison of final eluates with the traditional TAP and GS tag protocols: final eluates were precipitated, separated on 4–12% NuPAGE gradient gels and visualized with
Coomassie G. Background levels were analyzed with mock and GFP tag purifications. Proof of concept was demonstrated for CKS1 (At2g27960) and CDKD;2 (At1g66750).
Preys identified via MALDI-TOFTOF and confirmed in multiple experiments are indicated (see supplementary Table S2). The number of purifications used to determine the
confirmed interactors is shown in parentheses. Asterisks indicate bait proteins. (d) Schematic representation of the cloning strategy for GS fusions to the C terminus or (e)
to the N terminus of the bait protein. Abbreviations: ccdB, toxic killer gene for negative selection; KmR, neomycin phosphotransferase II gene for selection of transformed
plant cells; LB, left border for T-DNA insertion; RB, right border for T-DNA insertion; Sp, streptomycin and spectinomycin resistance gene; TT, cauliflower mosaic virus 35S
transcription terminator.
Update Trends in Plant Science Vol.13 No.10
519
especially with low-abundant complexes. An additional
benefit of the GS tag is the higher cellular concentration
levels of the bait protein (Figure 1b) and the concomitantly
higher complex incorporation and yield, as shown
by the stronger A-type cyclin-dependent kinase (CDKA;1)
band in the CDK subunit 1 (CKS1) GS-tag purification
(Figure 1c). As a final proof of the GS tag superiority,
we present results obtained with two cell-cycle baits,
CKS1 and the D-type CDK-activating kinase CDKD;2
(Figure 1c). Only the experimentally confirmed interactors
are represented and compared with those obtained
with the traditional TAP tag [5] (see also online supplementary
Table S2). For CKS1, most of the interactors
confirmed previously in seven purifications with the
traditional TAP tag were found with the GS protocol with
only two purifications. In addition, some new interesting
interactions could be detected with the GS tag only. The
known partners of CDKD;2, the H-type cyclin (CycH;1)
and CDK-activating kinase assembly factor ‘me´nage a`
trois’ 1 (MAT1), previously discovered in four purifications
with the traditional TAP tag, were identified in
two purifications with the GS protocol. Moreover, using
the GS tag we demonstrated that, as in rice [11], the
Arabidopsis CDKD;2 is part of the transcription factor
IIH (TFIIH) complex, because ultraviolet hypersensitive
6 (UVH6) and a TFIIH-complex-related transcription
factor co-purified. Furthermore, CDKD;2 might link regulation
of cell division with nucleotide biosynthesis
because of co-purification of three phosphoribosyl diphosphate
synthetases (PRSs).
Conclusion and perspectives
We have shown that the GS tag outperforms the
traditional TAP tag in plant cells, both concerning specificity
and complex yield. Recently, we replaced the TEV
protease cleavage sites in the GS tag with the rhinovirus
3C cleavage site for improved protein complex stability
during purification. Combined with the latest and most
sensitive MS technology, this tag should bring protein
complex analysis in plants to its full bloom. Cloning with
these tags is compatible with the Gateway system [20]
(Figure 1d,e), and vectors for C- or N-terminal cloning are
available at http://www.psb.ugent.be/gateway/.
Acknowledgements
We thank Giulio Superti-Furga and Tilmann Bu¨ rckstu¨ mmer for
providing the plasmid encoding the GS tag, all members of the
Functional Proteomics group at the Plant Systems Biology Department,
including Emilie De Witte, and all members of the Centre for Proteomics
and Mass Spectrometry at Antwerp University for technical assistance,
and Martine De Cock for help in preparing the manuscript. This work
was supported by grants from the Institute for the Promotion of
Innovation through Science and Technology in Flanders (Generisch
Basisonderzoek aan de Universiteiten [GBOU] grant no. 20193) and from
the European Commission in the 6th Framework Programme
(Agronomics; LSHG-CT-2006–037704).
Supplementary data
Supplementary data associated with this article can be
found, in the online version, at doi:10.1016/j.tplants.2008.
08.002.
References
1 Rigaut, G. et al. (1999) A generic protein purification method for protein
complex characterization and proteome exploration. Nat. Biotechnol.
17, 1030–1032
2 Gavin, A.C. et al. (2006) Proteome survey reveals modularity of the
yeast cell machinery. Nature 440, 631–636
3 Krogan, N.J. et al. (2006) Global landscape of protein complexes in the
yeast Saccharomyces cerevisiae. Nature 440, 637–643
4 Rohila, J.S. et al. (2004) Improved tandem affinity purification tag and
methods for isolation of protein heterocomplexes from plants. Plant J.
38, 172–181
5 Van Leene, J. et al. (2007) A tandem affinity purification-based
technology platform to study the cell cycle interactome in
Arabidopsis thaliana. Mol. Cell. Proteomics 6, 1226–1238
6 Van Aken, O. et al. (2007) Mitochondrial type-I prohibitins of
Arabidopsis thaliana are required for supporting proficient
meristem development. Plant J. 52, 850–864
7 Xing, D. and Chen, Z. (2006) Effects of mutations and constitutive
overexpression of EDS1 and PAD4 on plant resistance to different
types of microbial pathogens. Plant Sci. 171, 251–262
8 Zhao, Q. et al. (2008) Two distinct interacting classes of nuclear
envelope-associated coiled-coil proteins are required for the tissuespecific
nuclear envelope targeting of Arabidopsis RanGAP. Plant
Cell 20, 1639–1651
9 Takahashi, N. et al. (2008) The DNA replication checkpoint aids
survival of plants deficient in the novel replisome factor ETG1.
EMBO J. 27, 1840–1851
10 Batelli, G. et al. (2007) SOS2 promotes salt tolerance in part by
interacting with the vacuolar H+
-ATPase and upregulating its
transport activity. Mol. Cell. Biol. 27, 7781–7790
11 Rohila, J.S. et al. (2006) Protein–protein interactions of tandem affinity
purification-tagged protein kinases in rice. Plant J. 46, 1–13
12 Abe, M. et al. (2008) Identification of dynamin as an interactor of rice
GIGANTEA by tandem affinity purification (TAP). Plant Cell Physiol.
49, 420–432
13 Rubio, V. et al. (2005) An alternative tandem affinity purification
strategy applied to Arabidopsis protein complex isolation. Plant J.
41, 767–778
14 Kim, W.Y. et al. (2007) ZEITLUPE is a circadian photoreceptor
stabilized by GIGANTEA in blue light. Nature 449, 356–360
15 Zentella, R. et al. (2007) Global analysis of della direct targets in
early gibberellin signaling in Arabidopsis. Plant Cell 19, 3037–
3057
16 Forler, D. et al. (2003) An efficient protein complex purification method
for functional proteomics in higher eukaryotes. Nat. Biotechnol. 21, 89–
92
17 Geisler-Lee, J. et al. (2007) A predicted interactome for Arabidopsis.
Plant Physiol. 145, 317–329
18 Cui, J. et al. (2008) AtPID: Arabidopsis thaliana protein interactome
database – an integrative platform for plant systems biology. Nucleic
Acids Res. 36, D999–D1008
19 Burckstummer, T. et al. (2006) An efficient tandem affinity purification
procedure for interaction proteomics in mammalian cells. Nat. Methods
3, 1013–1019
20 Karimi, M. et al. (2007) Recombinational cloning with plant Gateway
vectors. Plant Physiol. 145, 1144–1154
1360-1385/$ – see front matter. ß 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tplants.2008.08.002 Available online 2 September 2008
Update Trends in Plant Science Vol.13 No.10
520