Characterization and RNA-seq analysis

of underperformer, an activation-tagged
potato mutant

Sukhwinder S. Aulakh, Richard

E. Veilleux, Allan W. Dickerman,
Guozhu Tang & Barry S. Flinn

Plant Molecular Biology

An International Journal on Molecular
Biology, Molecular Genetics and
Biochemistry

ISSN 0167-4412
Volume 84
Number 6

Plant Mol Biol (2014) 84:635-658

DOI 10.1007/s11103-013-0159-4

1 23
Your article is protected by copyright and all
rights are held exclusively by Springer Science
+Business Media Dordrecht. This e-offprint
is for personal use only and shall not be self-
archived in electronic repositories. If you wish
to self-archive your article, please use the
accepted manuscript version for posting on
your own website. You may further deposit
the accepted manuscript version in any
repository, provided it is only made publicly
available 12 months after official publication
or later and provided acknowledgement is
given to the original source of publication
and a link is inserted to the published article
on Springer's website. The link must be
accompanied by the following text: "The final
publication is available at link.springer.com”.

1 23
 Author's personal copy
Plant Mol Biol (2014) 84:635–658
DOI 10.1007/s11103-013-0159-4
Characterization and RNA-seq analysis of underperformer,
an activation-tagged potato mutant
Sukhwinder S. Aulakh · Richard E. Veilleux ·
Allan W. Dickerman · Guozhu Tang · Barry S. Flinn 
Received: 16 September 2013 / Accepted: 21 November 2013 / Published online: 4 December 2013
© Springer Science+Business Media Dordrecht 2013
Abstract  The potato cv. Bintje and a Bintje activation-
tagged mutant, underperformer (up) were compared.
Mutant up plants grown in vitro were dwarf, with abun-
dant axillary shoot growth, greater tuber yield, altered tuber
traits and early senescence compared to wild type. Under
in vivo conditions, the dwarf and early senescence pheno-
types of the mutant remained, but the up plants exhibited
a lower tuber yield and fewer axillary shoots compared to
wild type. Southern blot analyses indicated a single T-DNA
insertion in the mutant, located on chromosome 10. Initial
PCR-based gene expression studies indicated transcrip-
tional activation/repression of several genes in the mutant
flanking the insertion. The gene immediately flanking the
right border of the T-DNA insertion, which encoded an
uncharacterized Broad complex, Tramtrac, Bric-a-brac;
also known as Pox virus and Zinc finger (BTB/POZ)
domain-containing protein (StBTB/POZ1) containing an
Armadillo repeat region, was up-regulated in the mutant.
Global gene expression comparisons between Bintje and up
Electronic supplementary material  The online version of this
article (doi:10.1007/s11103-013-0159-4) contains supplementary
material, which is available to authorized users.
S. S. Aulakh · R. E. Veilleux · B. S. Flinn 
Department of Horticulture, Virginia Tech, Blacksburg,
VA 24061, USA
S. S. Aulakh · G. Tang · B. S. Flinn (*) 
The Institute for Sustainable and Renewable Resources (ISRR),
The Institute for Advanced Learning and Research (IALR),
Danville, VA 24540, USA
e-mail: barry.flinn@ialr.org
A. W. Dickerman 
Virginia Bioinformatics Institute (VBI), Virginia Tech,
Blacksburg, VA 24061, USA
using RNA-seq on leaves from 60 day-old plants revealed a
dataset of over 1,600 differentially expressed genes. Gene
expression analyses suggested a variety of biological pro-
cesses and pathways were modified in the mutant, includ-
ing carbohydrate and lipid metabolism, cell division and
cell cycle activity, biotic and abiotic stress responses, and
proteolysis.
Keywords  Solanum tuberosum · Potato · BTB/POZ
domain · Pleiotropic · Transcriptome · RNA-seq
Introduction
Advances in genomic technology have allowed the
sequencing and annotation of numerous plant genomes.
Potato (Solanum tuberosum) is the 30th plant species and
first asterid for which a genome sequence has been pro-
duced, providing significant new information on potato
gene models. The genome size of the doubled monoploid
potato (S. tuberosum Group Phureja DM 1-3 566 R44)
selected for sequencing is 844 megabases (Mb), and the
assembly and annotation of 86 % of this 844 Mb have pre-
dicted 39,031 protein coding genes and a 62.2 % repeti-
tive DNA content (The Potato Genome Sequencing Con-
sortium 2011). Similarly, in the closely related tomato
genome, 900 Mb of sequence were generated from inbred
cultivar Heinz 1706; the assembly and annotation of
760 Mb (84 %) of sequence predicted 34,727 protein cod-
ing genes (The Tomato Genome Consortium, 2012). Sev-
eral other valuable genetic resources are also available for
potato, including the transcriptome of the reference potato
genome (Massa et al. 2011), potato expressed sequence
tags (ESTs) from NCBI (249,761—dbEST release
130101) and microarray expression data (Kondrák et al.
13
 Author's personal copy
636
2011). The potato genome reveals 2,642 genes specific to
this large angiosperm clade and provides a platform for
genetic improvement of potato. Many traits of interest in
potato are quantitative in nature and access to the genome
sequence will simplify their characterization and deploy-
ment in cultivars (The Potato Genome Sequencing Consor-
tium 2011). However, as for all plant genomes, scientists
are faced with the enormous task of functional annota-
tion of the various genes. Mutant populations are a valu-
able resource for gene discovery, but few such resources
are available for potato (Fischer et al. 2008) slowing the
process of gene discovery and annotation. Therefore, the
challenge for the post-sequencing era is to create valuable
mutant resources in potato and to identify the biological
functions of the predicted genes. Various approaches can
be used to discover gene function, and the most direct is
to disrupt the gene(s) and analyze the consequences. How-
ever, gene disruption approaches also have limitations. In a
tetraploid crop such as potato, disruption of a single allele
may not result in phenotypic changes, as the three unal-
tered alleles may compensate.
Activation tagging is used for generating dominant or
gain-of-function mutations, complementing conventional
insertional mutagenesis and offering many advantages. In
this method, T-DNA or a transposable element contain-
ing multimerized Cauliflower Mosaic Virus (CaMV) 35S
enhancers (Hayashi et al. 1992; Suzuki et al. 2001) is
inserted into the genome. These enhancers can function in
either orientation and at a considerable distance from cod-
ing regions, and may cause transcriptional activation of
genes, resulting in dominant gain-of-function mutations.
Such gene activation may produce novel phenotypes, and
can aid in determining gene functions, including those of
redundant members of a gene family or genes essential for
development and survival of an organism. However, the
activation tag may also integrate into a gene, disrupting
gene function. Many genes are present in multiple copies
in the genome with partially redundant functions (Hiratsu
et al. 2003), and this aspect is even more pronounced in
polyploid species like potato, making functional gene anno-
tation difficult. However, the activation tagging approach
could prove useful, given that it activates the transcription
of flanking endogenous gene(s), and produces an obvious
phenotype. This tagging system has been used to identify
and characterize a variety of genes in Arabidopsis (Huang
et al. 2001; Kakimoto 1996; Kardailsky et al. 1999; Weigel
et al. 2000), petunia (Zubko et al. 2002), tomato (Mathews
et al. 2003), barley (Ayliffe et al. 2007) and poplar (Busov
et al. 2003; Harrison et al. 2007). The only prerequisite for
using activation tagging in any species is the need for at
least one genotype to be efficiently transformed. Hence,
activation tagging has been growing in popularity in diverse
non-model plant species (Busov et al. 2011) but its utility
13
Plant Mol Biol (2014) 84:635–658
has yet to be demonstrated in potato, a potential model crop
for autopolyploid species.
Potato is the world’s third most important crop after
wheat and rice, and the most widely grown vegetable crop,
with more than half of its global harvest produced in devel-
oping countries (http://faostat.fao.org/). The cultivated
potato (Solanum tuberosum L.) belongs to a large genus
that includes 160 tuber-bearing species, eight of which
are cultivated. Most commercial potato cultivars are auto-
tetraploid (2n = 4x = 48), having originated from diploid
ancestors in the Andean Mountains of South America. The
development of resources and methods to elucidate the
function of potato genes remains an important goal for the
realization of many of the potential benefits of the potato
genome sequence (The Potato Genome Sequencing Con-
sortium 2011) and generation and characterization of acti-
vation tagged potato lines could provide one such resource.
Another useful resource is knowledge of the transcrip-
tome. The main goal of whole transcriptome analysis is to
identify, characterize and catalogue all of the transcripts
expressed within a specific cell/tissue/organ, at a particu-
lar stage, with the potential to determine the correct splic-
ing patterns and structure of genes, and to quantify the dif-
ferential expression of transcripts under physiological and
pathological conditions (Costa et al. 2010). The arrival of
Next Generation Sequencing (NGS) platforms [Genome
Sequence FLX (Roche), Genome Analyzer IIx and HiSeq
2000 (Illumina), SOLiD 4 and 5500 Series Genetic Ana-
lyzer (Applied Biosystems)] over the last few years has
allowed sequencing capabilities and associated gene
expression profiling capabilities to achieve a new level.
In this paper, we report on the characterization of the
potato activation tagged mutant AT615, which we termed
underperformer (up), due to its general poor performance,
reduced organ size and early senescence compared to wild
type Bintje. We also report the results of RNA-seq analysis
to characterize gene expression changes on a global level
in the leaves of the mutant compared to wild type Bintje.
This analysis resulted in a unique dataset of 1,632 genes
that were differentially expressed between up and Bintje.
We attempted to present the broader picture of gene expres-
sion changes, derive the biological meaning from this data
and relate this to the role of activation tagging in the mutant
phenotype.
Methods
Generation and screening of potato activation tagged
mutants
The wild type Bintje and activation tagged AT 615 (up)
mutant plantlets were received from BioAtlantech, having
 Author's personal copy
Plant Mol Biol (2014) 84:635–658 637
been generated (Supplementary Materials and Methods) by
the Canadian Potato Genome Project (Regan et al. 2006).
Determination of T‑DNA copy number by Southern blot
analysis
Genomic DNA of Bintje and up was digested with PciI
and PsiI restriction enzymes and separated on a 0.8 %
agarose gel. Digested DNA was transferred to positive
charged Zeta probe nylon membrane (Bio-Rad; Hercules,
CA, USA) using downward capillary transfer and NaOH
buffer (Sambrook and Russell 2001). DNA was transferred
to the membrane and rinsed in 2X SSC, then placed on a
sheet of Whatman 3 MM filter paper and allowed to air dry
for 30 min. The probe (640 bp) used for hybridization was
amplified from the 4X35S enhancer region of the pSKI074
plasmid by PCR using primers (Supplementary Table S1)
and labeled using Amersham AlkPhos Direct Labeling
Reagents (GE Healthcare Life Sciences; Buckinghamshire,
UK Catalog No. RPN3682) following the manufacturer’s
protocol. Overnight hybridization with the probe was done
in Alkphos Direct hybridization buffer (Amersham GE) at
55 °C. After hybridization, the blot was washed to remove
excess probe and signal detected with CDP-Star (GE
Healthcare Life Sciences; Buckinghamshire, UK) chemilu-
minescent detection reagent. For final signal generation, the
blot was exposed to autoradiography film and the film was
developed with Kodak GBX developer and fixer solution.
In vitro plant growth and tuberization
For in vitro plant growth assays, Bintje and up plants were
grown from single node explants taken from 4 to 5 week
old plantlets. Explants were placed on MS basal medium
with vitamins (Phytotech, Lenexa, KS, USA) supplemented
with 3 % sucrose, pH 5.8 and solidified with 0.7 % agar.
Plants were grown under 16 h photoperiod, 24 ± 1 °C and
70–100  μE light intensity in a plant growth chamber and
data were recorded after 4 weeks.
For in vitro tuberization assays, stem sections, 3–5 cm
long, with 4–5 nodes per section were prepared from 4 to
5 week old in vitro Bintje and up plantlets by removing
all leaves. These stem sections were grown in 40 ml liq-
uid propagation medium (MS basal medium + 148 mg l−1
NaH2PO4  + 0.4 mg l−1 thiamine HCl + 100 mg l−1 ino-
sitol  + 3 % sucrose, pH 5.8) (Radouani 2003) at 16 h
photoperiod, 24 ± 1 °C and 70–100 μE for 3–4 weeks in
250 ml conical flasks capped with ventilated plugs. After
3–4 weeks, propagation medium was replaced with micro-
tuberization medium (MS basal + 8 % sucrose, 2 × the
amounts of NH4NO3, KH2PO4 and KNO3 and without
NaH2PO4, pH 5.8) (Radouani 2003). Culture flasks were
kept at 18–20 °C in the dark to induce tubers and data were
recorded after 30 days. All in vitro experiments were done
in randomized complete block design with three replicates
and each experiment was repeated at least three times.
Data were analyzed in JMP, version 9.0.0 (SAS Institute
Inc., Cary, NC, USA) and tables and graphs were drawn in
Microsoft Excel (2010).
Phenotyping under growth chamber and greenhouse
conditions
Phenotyping was done on plants grown in walk-in growth
chambers in the Department of Forest Resources and Envi-
ronmental Conservation, Virginia Tech. Ten equal size (by
weight) tubers of Bintje (average tuber weight 24 g) and up
(average tuber weight 22 g) were planted in 30 cm diam-
eter pots filled with commercial Miracle-Gro potting mix
(Scotts Miracle-Gro, Marysville, OH, USA). Standard
plant growth practices were followed. Plants were grown
under 16 h photoperiod; 22 °C day/16 °C night and 350 μE
light intensity and data were recorded 75 days after tuber
planting (45 days after plant emergence). Growing plants
were supported by a plastic net and manually watered twice
per week or as needed. Ratings for potato tuber character-
istics were based on the standardized rating codes detailed
in Appendix 2 of the NC Potato Variety Trial Report (http://
potatoes.ncsu.edu/pdf/NCPOTRPT05.pdf). The up mutant
was also compared to Bintje under greenhouse conditions.
Bintje and up plants were grown in the greenhouse of the
Institute for Sustainable and Renewable Resources (ISRR)
at the Institute for Advanced Learning and Research
(IALR), Danville, VA, USA. Twenty tubers each of wild
type and mutant (after equalizing the weight) were planted
directly into 30 cm diameter pots filled with commercial
Miracle-Gro potting mix. Standard plant growth practices
were followed. The greenhouse was set to 24–30 °C. Data
were recorded 75 days after planting (45 days after emer-
gence) and analyzed in JMP.
Identification of potato genomic regions flanking
the T‑DNA
We used the GenomeWalker universal kit (Clontech,
Mountain View; CA, USA) to identify and amplify the
genomic region flanking the T-DNA insertion. As the
pSKI074 vector has few unique sequences between the
T-DNA right border and the tetramerized enhancer region,
we selected the left border region of pSKI074 for genome
walking. Genomic DNA was extracted using Plant DNAzol
Reagent (Life Technologies; Grand Island, NY, USA) fol-
lowing manufacturer’s instructions. Regions flanking the
T-DNA were amplified via GenomeWalker following the
manufacturer’s instructions. Genomic DNA was digested
with single restriction enzymes that leave blunt ends
13
 Author's personal copy
638
(MscI, NaeI, ScaI and SspI) in separate tubes, followed
by ligation with GenomeWalker adapters to form Genom-
eWalker “libraries.” Gene specific primers and nested gene
specific primers were designed against the mannopine syn-
thase gene sequences and used with arbitrary primers 1
(AP1) and 2 (AP2) to amplify the potato genomic region
flanking the left border of the T-DNA (Supplementary
Table S1). The primary PCR amplification was done with
outer adapter primer (AP1), gene specific primer (GSP1)
and GenomeWalker library DNA as templates which
were followed by secondary PCR amplification using
nested adapter primer (AP2), nested gene specific primer
(GSP2) (Supplementary Table S1) and diluted primary
PCR product as template. Secondary PCR amplification
products were separated by electrophoresis and purified
using a gel extraction kit (Qiagen; Valencia, CA, USA);
cloned and sequenced using a CEQ 8800 genetic analysis
system (Beckman Coulter, Brea, CA, USA). Recovered
genomic sequences flanking the left border of the T-DNA
insertion sites were identified and used to perform Basic
Local Alignment Search Tool (BLAST) searches in the
GenBank database (accessed on 4-5-2010) of the National
Center for Biotechnology Information (NCBI), and in the
potato genome version 4.0.3 (http://potatogenomics.plan
tbiology.msu.edu/blast.html) with default search param-
eters (expected threshold 10, maximum number of align-
ments 100, max number of descriptions 100, word length
11, no filter and both strands). This allowed the T-DNA
insertion site to be located on Chromosome 10 (genomic
coordinates; chr10:674572) of the potato genome.
Approximately 400 kb of potato sequence upstream
and downstream of the insertion site were analyzed for
the presence of annotated gene models using the potato
genome browser (http://potato.plantbiology.msu.edu/cgi-
bin/gbrowse/potato). Sequence homology searches
and analyses were performed using the NCBI BLAST
server. Sequence alignments were carried out by the
CLUSTALW2 (Larkin et al. 2007) method and using
the EMBL server (http://www.ebi.ac.uk/Tools/msa/
clustalw2/). Identification of conserved domains within the
predicted protein sequence of the identified tagged gene
(StBTB/POZ domain-containing protein 1) was carried out
using InterProScan version 4.8 at the EMBL-EBI website
(http://www.ebi.ac.uk/Tools/pfa/iprscan/).
Preparation of RNA
Tuber samples from wild type and mutant plants were
used for RNA extraction and subsequent RT-PCR, while
foliar samples were used for RT-PCR and RNA-seq anal-
yses. Ten lateral leaves, 5–6 cm long and 2–3 cm broad,
which were at the same developmental stage (not fully
mature) harvested from ten plants, 60 days after planting
13
Plant Mol Biol (2014) 84:635–658
the tubers (about 30 days after emergence) were pooled to
form a composite sample for each genotype. Three repli-
cates of each composite sample were harvested, frozen in
liquid nitrogen and stored at −80 °C until needed. RNA
was extracted for RT-PCR analysis using the Concert Plant
RNA Reagent (Invitrogen, Carlsbad, CA) following manu-
facturer’s instructions. RNA was extracted for RNA-seq
analysis using a combination of Trizol Reagent (Invitrogen,
Carlsbad, CA, USA) extraction and the RNA mini kit (Qia-
gen, Valencia, CA, USA) purification.
Reverse transcriptase PCR analysis
DNase treatment of RNA was done with the DNA-free
kit (Ambion; Foster City, CA, USA) following manu-
facturer’s instructions. First strand cDNA was synthe-
sized using SuperScript III (Invitrogen; Carlsbad, CA,
USA) using 1 μg of total RNA. After cDNA synthesis
the final volume was adjusted to 100 μl of diluted cDNA
for each cDNA synthesis reaction. Reverse transcrip-
tion polymerase chain reaction (RT-PCR) was performed
using gene-specific primers (Supplementary Table S2).
Equal amounts of cDNA were used in each reaction and
Solanum tuberosum Elongation Factor 1-α (StEF1-α)
gene expression was used as control. Gel images were
acquired using an Alpha Innotech gel doc system (San
Leandro, CA, USA).
RNA‑sequencing
RNA for RNA-seq was extracted from leaves of plants that
had been growing in the chambers for 60 days using a com-
bination of Trizol Reagent extraction and RNA mini kit
purification methods as described earlier. For DNase treat-
ment we used Turbo DNA free protocol following manu-
facturer’s instruction with some modifications. Samples
with an RNA Integrity Number (RIN) value >8 were sub-
jected to cDNA synthesis. The Illumina TruSeq RNA sam-
ple preparation kit (low-throughput protocol) was used for
poly-A based mRNA enrichment and thermal mRNA frag-
mentation, followed by cDNA synthesis using 4 μg of total
RNA sample as starting material for each enrichment reac-
tion. The fragmented mRNA samples were used for cDNA
synthesis using reverse transcriptase (Super-Script II) and
random primers. The cDNA was further converted into
double stranded DNA using the reagent supplied in the kit
and resulting dsDNA used to prepare a library of template
molecules suitable for high-throughput DNA sequencing
using the TruSeq Illumina kit.
Sequencing (RNA-seq) was performed at the Research
Technology Support Facility (RTSF) of Michigan State
University using the Illumina Genome Analyzer (GAII)
and HiSeq 2000 system, which uses a massively parallel
 Author's personal copy
Plant Mol Biol (2014) 84:635–658 639
sequencing-by-synthesis four-dye approach to generate
billions of bases of high-quality DNA sequence per run.
A total of two biological replicates each consisting of two
technical replicates was sequenced. For one set of biologi-
cal replicates we used the GAII platform generating sin-
gle end 36 bp reads and for the second biological set we
used the HiSeq 2000 machine generating single end 50 bp
reads.
Bioinformatics and statistical analysis
Sequence reads were mapped to S. tuberosum Group
Phureja DM1-3 516-R44 (DM) superscaffolds using
Tophat v2.0.4 (Trapnell et al. 2009), which made use of
Bowtie. Reads were counted in regions of genes based on
the PGSC_DM_v3_2.1.10_pseudomolecule annotation.
Transcript abundance was expressed as raw read counts and
also in a normalized unit, reads per kilobase per million
reads (RPKM) as implemented in Cufflinks v2.0.2 (Trap-
nell et al. 2010), which allow comparison both within and
among samples. Transcripts with RPKM value >0.001 and
95 % confidence interval were considered for further analy-
sis. Differentially expressed genes were flagged as signifi-
cant based on multiple comparisons values, q. Only genes
with q values <0.05 were considered significant. Individual
datasets (GAII or HiSeq) as well as the combined dataset
(GA and HiSeq) were used for Cuffdiff analysis and the
resulting output files were scanned for significant differ-
entially expressed genes. A stricter statistical analysis was
performed using a differential expression analysis tool
based on negative binomial distribution (Anders and Huber
2010) and genes which had significantly different expres-
sion level at α < 0.05 level were considered. The hierarchi-
cal clustering of genes filtered using Cuffdiff analysis was
done using MapMan v3.5.1 software (Thimm et al. 2004).
Protein domain architecture analysis was done with Sim-
ple Modular Architecture Research Tool (SMART). To
carry out phylogenetic analysis of the BTB/POZ domain
containing protein family members from various plant spe-
cies, we used the StBTB/POZ1 protein sequence (Genbank
No. AGT37251) to extract BTB/POZ1 orthologs from
GenBank and Phytozome v9.1 (www.phytozome.net).
The deduced protein sequences from different plant spe-
cies were aligned using the Clustal W method in MegAlign
(DNAStar Lasergene 11.0) and a rooted phylogenetic tree
was constructed. The length of each pair of branch repre-
sents the distance between each sequence pair. The units at
the bottom of the tree indicate the number of substitution
events. Bootstrapping was used to test for consistency and
robustness of the tree created using ClustalW. The branch
patterns with greater values are statistically more likely to
occur and provide a measure of confidence in the optimum
tree structure.
Visualization of gene expression data
We used MapMan (version 3.5.1), which displays large data
sets onto diagrams (Maps) that symbolically depict areas
of biological function. Maps are diagrams of pathways or
processes in which experimental data are displayed. Map-
Man organizes data into functional categories (BINs)
which are themselves split into subcategories (subBINs).
Each Bin was given a corresponding numerical code (e.g.,
‘photosynthesis’  = 1) which can be extended in a hierar-
chical manner (e.g., the subBINs ‘light reactions’ = 1.1;
‘photorespiration’  = 1.2) and so on (Thimm et al. 2004).
The change of expression is displayed via false color code:
blue for increased and red for decreased expression. A set-
ting was selected that saturates at a value of 2 (22 = a four-
fold change). Potato mapping file Stub_PGSC_DM_v3.4
was downloaded from the MapMan website (http://mapm
an.gabipd.org/web/guest/mapmanstore). The mapping file
stores the association of genes, metabolites or proteins to
MapMan BINs to link experimental data with MapMan
ontology. An experimental data file with potato peptide
IDs (PGSC0003DMP4000xxxxx) and relative expression
levels (log2 fold values) was created and uploaded in the
MapMan dataset folder. The experimental data file con-
tained the measurements from the RNA-seq experiment in
the log2 fold changes between a treatment (mutant up) and
control (Bintje). MapMan not only displays a large data-
set but also incorporates statistical analysis. For example
deviation of response of items in a particular BIN from
the response of all other items is calculated by Wilcoxon
test, which is similar to Student’s t test. MapMan also has
a built-in multiple comparison test (Benjamini Hochberg)
which calculates corrected P values (Usadel et al. 2009).
Results
Generation and screening of potato activation tagged
mutants
The activation tagging approach was used by the Cana-
dian Potato Genome Project (Regan et al. 2006) to create
over 8,000 mutants of cv. Bintje using the activation tag-
ging vector pSKI074 (Weigel et al. 2000). The mutant line
AT615, which we designated underperformer (up), was
selected during an in vitro screen. After 1 month of in vitro
growth from single node explants, the up mutant was much
shorter than wild type Bintje and several new axillary stems
were formed, leading to branching, whereas wild type
Bintje exhibited a single main stem (Supplementary Fig.
S1). Transgenic plantlets were initially screened by poly-
merase chain reaction (PCR) to confirm the presence of the
T-DNA plasmid sequences in the plant (data not shown).
13
 Author's personal copy

640 Plant Mol Biol (2014) 84:635–658

Determination of T‑DNA copy number by Southern blot (Fig.  3). The number of stolons also increased for Bintje
analysis but not up.

To determine the T-DNA copy number in the up mutant,
we performed Southern blot analysis with genomic DNA,
using two restriction enzymes (PciI and PsiI) that cut
within the T-DNA, in the region just left of the tetramer-
ized enhancer sequences. Using a probe that corresponded
primarily to the enhancer region of pSKI074, we observed
a strong, single hybridizing band in the mutant for both
restriction enzymes, indicating the presence of a single
T-DNA insertion (Supplementary Fig. S2).
In vitro plant growth and tuberization
Phenotyping under growth chamber and greenhouse
conditions
Additional phenotype information was obtained for up
plants when grown ex vitro in soil, in addition to the traits
observed under in vitro conditions. Mutant plants were
modified in various above-ground and tuber characteristics
(Table 1; Figs. 4, 5, 6), had smaller sized leaves (Figs. 4a,
b, 5), produced long and fingerlike tubers (Figs. 4c, d, 5),
had small floral buds which did not open and dropped pre-
maturely (Fig. 4e–g), and exhibited early senescence of the

We multiplied Bintje and up plants from single node

explants and recorded data after 4 weeks of in vitro growth.
Mutant up plants exhibited a smaller mean stem length and
a larger mean number of axillary branches, which were
significantly different from the Bintje control. No signifi-
cant differences were found in number of roots per plant
(Fig. 1). In vitro tuberization experiments were conducted
to assess tuberization ability of the up mutant. All recorded
parameters (number of tubers/explant, number of stolons/
explant and tuber weight/explant) were significantly greater
for up (Fig. 2a) compared to wild type Bintje. We also
observed differences in tuber shape and size between Bintje
and up plants. Bintje tubers were round and more uniform
compared to the longer tubers of up (Fig. 2b). As cytokinin
is known to impact tuberization, kinetin effects on tuberiza-
tion were assessed with both Bintje and up, to see if any
variation in responsiveness existed in the mutant. In vitro
tuberization medium supplemented with 10 μM kinetin
improved the number and weight of tubers for both Bintje
and up, but the increase was somewhat greater for Bintje

Fig. 2  In vitro tuberization response of wild type Bintje and up

Fig. 1  In vitro plant growth of wild type Bintje and underper-
former (up) mutant plants. Single node explants from 4 to 5 week-
old potato plantlets of Bintje and the up mutant were placed on MS
medium with vitamins, and data recorded after 30 days. The values
are mean ± SE. Asterisks statistical significance of means in the
same parameter estimated using Tukey’s HSD test (*P < 0.05–0.005;
**P < 0.005–0.0005, ***P < 0.0005)
mutant plants. a Stem explants 3–4 cm long with 4–5 nodes were
cultured in liquid propagation medium for 3–4 weeks under 16 h
photoperiod and then transferred to high sucrose liquid tuberization
medium and transferred to the dark, and data recorded after an addi-
tional 30 days. The values are mean ± SE. Asterisks statistical signifi-
cance of means in the same parameter estimated using Tukey’s HSD
test (*P < 0.05–0.005; **P < 0.005–0.0005; ***P < 0.0005). b Pho-
tographic representation of the response of the Bintje and up mutant
30 days after plantlets were placed on liquid tuberization medium
(upper panel), and the harvested tubers of both (lower panel), show-
ing the different shape and size of the mutant tubers. Bar 2 cm

13
 Author's personal copy
Plant Mol Biol (2014) 84:635–658 641

misshapen, cracked and rotted tubers) and a lower specific

gravity (Table 1).

Identification of potato genomic regions flanking the

T‑DNA

Using GenomeWalker technology (Clontech), we posi-

tioned the T-DNA insertion on chromosome 10:674572
(Supplementary Fig. S3a). Further PCR and sequence
analysis of the left border T-DNA/plant DNA junction
site revealed that a vector backbone of 250 bp (7,802 to
8,053 sequence of pSKI074) outside of the T-DNA was
also found in the up mutant (data not shown). In a simi-
Fig. 3  The effect of kinetin on in vitro tuberization response of wild lar analysis of the right border T-DNA/plant DNA junc-
type Bintje and up mutant plants. Stem explants 3–4 cm long with tion site in the up mutant, we found that 22 bp of the
4–5 nodes were cultured in liquid propagation medium for 3–4 weeks T-DNA right border (4,273 to 4,294 sequence of pSKI074)
under 16 h photoperiod and then transferred to high sucrose liquid
tuberization medium without (control) or with 10 μm kinetin, trans-
and the same number of base pairs of the potato genome
ferred to the dark, and data recorded after an additional 30 days. The (chr10:674550..674571) were lost during T-DNA integra-
values are mean ± SE. Asterisks statistical significance of means in tion (data not shown). Using the potato genome browser
the same parameter estimated using Tukey’s HSD test (*P < 0.05– (http://potato.plantbiology.msu.edu/cgi-bin/gbrowse/
0.005; **P < 0.005–0.0005; ***P < 0.0005)
potato), we identified several gene model predictions on
either side of the T-DNA insertion. The first predicted
Table 1  Various phenotypic measurements for ex vitro-grown Bintje potato gene open reading frame flanking both right and left
and up plants borders of the T-DNA occurred approx. 5.2 and 3 kb from
Description Mean values + SE P value the T-DNA, respectively (Supplementary Fig. S3a).
Bintje up
Transcript analysis of potato genes flanking the T‑DNA
Plant height (cm) 72.9 ± 1.3 41.8 ± 1.4 <0.0001*
Chlorophyll content index 46.1 ± 1.7 40.8 ± 1.5 0.0229* We used reverse transcriptase PCR for transcript analysis
Tuber shape 4.3 ± 0.4 8.6 ± 0.2 <0.0001* of potato genes flanking the T-DNA with primers designed
Tuber color 6.6 ± 0.2 6.8 ± 0.2 0.6186 against the reading frame of each gene model prediction
No. of eyes/tuber 13.8 ± 0.9 35 ± 2.2 <0.0001* (Supplementary Table S2), and with cDNA made from
Tuber eye depth 8 ± 0.2 5.1 ± 0.2 <0.0001* leaf RNA. We analyzed approximately 400 kb regions
Tuber skin texture 7 ± 0.3 7.3 ± 0.3 0.5057 both upstream and downstream of the insertion site, and
Specific gravity 1.11 ± 0.006 1.08 ± 0.004 0.0205* observed up- and down-regulation of various potato genes
flanking both borders (Fig. 7). The most significant change
in gene expression in the mutant was a strong up-regula-
leaves (Fig. 4h, i). When freshly harvested tubers of Bintje tion of the first two predicted genes (conserved genes of
were replanted in soil without any treatment to break tuber unknown function), which started 5.2 and 11.5 kb from the
dormancy, they exhibited normal dormancy, with no spout- right border (Fig. 7a). Another gene whose expression was
ing observed within 40 days (Fig. 4j). In contrast, over the up-regulated in the mutant was a predicted SAM-dependent
same time period, 60 % of up tubers sprouted (Fig. 4j), methyltransferase, located 41 kb downstream of the right
indicating reduced tuber dormancy in the mutant. The border. In addition several genes within the 400 kb region
plant and tuber data (Table 1; Figs. 4, 5), indicated that of the right border were down-regulated (F-box family pro-
the up mutant exhibited significantly reduced plant size, tein, GA2-oxidase 1, P21-rho-binding domain-containing
smaller leaves and tubers and fewer tubers. Biomass and protein, adenosine deaminase, Dicer homolog, DNA-3
yield traits (aboveground fresh weight, total fresh weight, methyladenine glycosylase) in the mutant. In the region
total dry biomass weight, total tuber weight per plant) analyzed, one gene off of the left border of the T-DNA, a
were also significantly lower in the mutant compared to conserved gene of unknown function, was observed to be
wild type Bintje (Fig. 6). Other tuber qualitative traits highly up-regulated in the up mutant, whereas a Protein
(tuber shape, tuber color, tuber eye depth) also differed phosphatase 2C and another conserved gene of unknown
between the two genotypes (Table 1). The mutant had function were down-regulated in the mutant (Fig. 7b). This
more tubers that exhibited external defects (such as green, transcript analysis of the flanking genes on both sides of the

13
 Author's personal copy

642 Plant Mol Biol (2014) 84:635–658

Fig. 4  Phenotyping of Bintje and up mutant plants under growth
chamber and greenhouse conditions. a Bintje and b up plants growing
in a walk-in-chamber, 80 days after planting. c Bintje and d up tubers
after harvest. e Bintje and f up inflorescences. g A close up of repre-
sentative flower buds from Bintje (left) and up (right) plants. h Lack
of senescence observed 35 days after planting with Bintje plants,
while up plants (i) exhibited distinct senescence. j Tuber dormancy
phenotypes for Bintje (left) and up (right) 30 days after tubers were
planted without cold storage. Note the reduced tuber dormancy and
sprouting of the mutant tubers

Fig. 5  Growth characteris-
tics of Bintje and up plants
grown in a walk-in-growth
chamber. Plants were grown
from tubers directly planted
in soil after breaking the tuber
dormancy (cold storage for
45 days). Data were recorded
75 days after planting (DAP).
Tuber data were recorded at
final harvest (130 DAP). The
values are mean ± SE. The
asterisks indicate statistical
significance of means in the
same parameter estimated using
Tukey’s HSD test (*P < 0.05–
0.005; **P < 0.005–0.0005;
***P < 0.0005). L length, B
breadth

13
 Author's personal copy
Plant Mol Biol (2014) 84:635–658 643
Fig. 6  Yield traits of Bintje and up plants grown in walk-in-growth
chambers. Plants were grown from tubers directly planted in soil after
breaking the tuber dormancy (cold storage for 45 days). Biomass
data were recorded, 75 days after planting (DAP). Tuber data were
recorded at final harvesting (130 DAP). The values are mean ± SE.
The asterisks indicate statistical significance of means in the same
parameter estimated using Tukey’s HSD test (*P < 0.05–0.005;
**P < 0.005–0.0005; ***P < 0.0005)
T-DNA revealed that insertion of the T-DNA had resulted
in modification of the expression of many genes in the up
mutant, not only the first (or few) genes immediately adja-
cent to the right border and associated tetramerized enhanc-
ers. In order to more fully characterize gene expression
changes, we used RNA sequencing (RNA-seq) of leaf tran-
scripts to obtain a more comprehensive global gene expres-
sion profile of the mutant.
RNA‑sequencing
We generated >207 million RNA-seq reads in total, of which
183 million (88 %) were mapped to the potato genome
assembly PGSC_DM_v3_2.1.10_pseudomolecules.zip
using TopHat v2.0.4. Our combined dataset flagged 1,632
genes as showing significantly different expression lev-
els between the two genotypes. Of these differentially
expressed genes, 661 (40.5 %) were up-regulated and
971 genes (59.5 %) were down-regulated in the mutant.
Although the T-DNA insertion in the mutant was previ-
ously identified as occurring on chromosome 10, the dis-
tribution of up- and down-regulated genes followed a ran-
dom pattern across all 12 potato chromosomes (Fig. 8).
The greatest numbers of up- (n = 110) and down-regulated
(n  = 133) genes were found on chromosomes 9 and 2,
respectively.
Visualization of gene expression data
We used MapMan software to localize the 1,632 differen-
tially-expressed genes to various categories and subcatego-
ries based on GO annotation, and the genes were shown
as different elements for major plant pathways/processes
(Fig.  9). MapMan displays large datasets of metabolic
pathways or other processes as pictorial diagrams that sym-
bolically depict areas of biological function. The classifica-
tion was based on hierarchical functional categories (Bins,
subBins, individual enzymes). The major MapMan path-
ways were classified to represent metabolism, cell function,
regulation, cellular response, stress secondary metabolism,
large enzyme families and RNA protein synthesis. Table 2
summarizes each of these pathways, their various numbers
of up- and down-regulated genes, and the highest up-regu-
lated and down-regulated gene in each pathway classifica-
tion for the up mutant relative to wild type Bintje.
In the following sections, we will describe the major
biological pathways/processes, those most relevant to
our RNA-seq dataset and into which our differentially
expressed genes were clustered. The detailed Bin ids, Bin
description, peptide ids, gene annotation and log2 fold
change values are given in Supplementary Table S3. The
information linking gene ids with peptide ids is given in
Supplementary Table S4. We use the terms up- and down-
regulation in gene expression levels of mutant genes rela-
tive to wild type Bintje gene expression. The majority of
the genes identified were down-regulated in the up mutant.
Metabolism overview
Many genes were identified which were modified in their
expression patterns in the mutant and were associated with
metabolism (Fig. 9a, Supplementary Table S3). These
genes were primarily associated with photosynthesis [Bin:
1], major carbohydrate metabolism [Bin: 2], minor carbo-
hydrate metabolism [Bin: 3], cell wall synthesis, composi-
tion, modification and degradation [Bin: 10], lipid metabo-
lism [Bin: 11], amino acid metabolism [Bin: 13], secondary
metabolism [Bin: 16] and nucleotide metabolism [Bin: 23].
The majority of the identified photosynthesis genes related
to photosystem subunit proteins, ATP synthase activity,
electron flow (e.g., similar to PIF1 and NDF4) and the Cal-
vin cycle (e.g., similar to glyceraldehyde 3-phosphate dehy-
drogenase subunits and fructose-1,6,-bisphosphatase) were
up-regulated in the mutant. In addition, most of the identi-
fied major carbohydrate metabolism genes (e.g., similar to
SUS4, SPF1F and AMY1) and minor carbohydrate metab-
olism genes (e.g., similar to DIN10, GOLS1 and TPS11)
were also up-regulated in the mutant. Regarding cell wall
metabolism, a majority of the identified genes associated
with cell wall degradation (e.g., similar to pectate lyases,
BURP domain-containing proteins) and cell wall modi-
fication (e.g., similar to XTH27, EXP8 and EXPL1) were
also up-regulated, while a mix of genes encoding pectin
esterase family proteins were up- and down-regulated. In
contrast, a majority of the identified genes associated with
lipid metabolism and relating to fatty acid synthesis and
13
 Author's personal copy

644 Plant Mol Biol (2014) 84:635–658

Fig. 7  Reverse transcriptase PCR analysis of various potato genes of the arrow indicates the orientation of the predicted open reading
flanking the right border (a) and left border (b) of the activation tag frame, and the distance of each predicted gene from the T-DNA inser-
T-DNA insertion of the up mutant. RNA was extracted from leaves of tion is indicated in kb. Information on primers used for RT-PCR is
60 day old plants grown in a walk-in-growth chamber. The direction given in Supplementary Table S2

13
 Author's personal copy
Plant Mol Biol (2014) 84:635–658 645
Fig. 8  Chromosomal distribution of genes showing modified leaf
expression patterns in the up mutant compared to wild type Bintje.
The 1,632 significantly differentially expressed genes identified using
RNA-seq were grouped according to their location on the potato
chromosomes. Leaf RNA from 60 day old plants was used for RNA-
seq. Significance cutoff was based on q < 0.05 values
elongation (e.g., similar to KAS1, KCS5 and CER2) and
lipid degradation (e.g., similar to triacylglyerol lipases)
were down-regulated in the mutant, although some up-
regulation of lysophospholipase and fatty acid desaturease
(FAD2) gene expression was noted in the mutant.
Several genes associated with amino acid metabolism
were identified, and of those associated with amino acid
synthesis, many were down-regulated in the mutant (e.g.,
similar to tryptophan synthase, tryptophan-associated C2
domain-containing protein, DHS1 and ASN1). Similarly,
several identified genes associated with amino acid deg-
radation (e.g., similar to fumarylacetoacetase and proline
oxidase) were also down-regulated. In the area of second-
ary metabolism, the majority of the identified genes in the
mutant associated with terpenes, flavonoids, phenylpro-
panoids and phenolics were down-regulated (e.g., simi-
lar to APR3, MVA1, HMG1, TPS1, 4CL, CAD8, CAD9,
TT4). Finally, a variety of genes associated with nucleo-
tide metabolism was also modified in their expression pro-
files in the mutant, with almost all (e.g., similar to CTP
synthase, thymidine kinase, deoxyuridine 5′-triphosphate
nucleotidohydrolase and the ribonucleotide reductase
RNR1) being down-regulated.
Cell function overview
In addition to metabolism, differentially expressed genes
and their affiliation with cell function processes were
identified (Fig. 9b, Supplementary Table S3), with vari-
ous processes (e.g., cell division and cell cycle, DNA
synthesis, regulation of transcription) specified. Approxi-
mately one-third of the genes could not be assigned to
any functional category, and so were grouped together
[Bin: 35.2]. Amongst the up-regulated genes, the high-
est-fold increases in expression levels (7.3 and 6.8
log2 fold) were observed for a gene involved in regu-
lation of transcription [PGSC0003DMG400031871;
PGSC0003DMP400055149] and for a homeobox tran-
scription factor family gene [PGSC0003DMG400014020]
encoding peptide PGSC0003DMP400024694 [Bins: 27.3.9
and 27.3.22, respectively]. Amongst the down-regulated
genes in the mutant, the gene which showed the great-
est level of down-regulation (−5.3 log2 fold) was found
to be a glutaredoxin gene [PGSC0003DMG400002360;
PGSC0003DMP400004213] [Bin: 21.4].
Genes assigned to cell organization, cell division and cell
cycle, DNA/RNA synthesis/processing/repair
A closer look at these various functional classifications
(Fig.  9b, Supplementary Table S3) revealed the majority
of identified genes associated with cell organization [Bin:
31.1], cell division [Bin: 31.2] and cell cycle [Bin: 31.3]
were down-regulated in their expression (e.g., similar
to various kinesins, the mitotic spindle checkpoint gene
MAD2 and the regulator of chromatin condensation gene
RCC1). Similarly, numerous identified genes associated
with DNA synthesis [Bin: 28.1], and various genes asso-
ciated with RNA processing [Bin: 27.1], RNA synthesis
[Bin: 27.2], and DNA repair [Bin: 28.2] were down-regu-
lated, including those similar to various histone proteins,
DNA-directed DNA polymerases and transducer family
proteins.
Genes assigned to stress
Several genes associated with biotic and abiotic stress
(Bin: 20) were modified in their expression patterns in the
mutant, with very distinct patterns associated with biotic
or abiotic stress-associated genes (Fig. 9b, Supplementary
Table S3). The majority of the identified genes associated
with biotic stress (e.g., similar to various pathogenesis-
related PR proteins, thaumatin and osmotin-like proteins,
HIN1 and JAZ1) was down-regulated in the mutant, while
the majority of those genes associated with abiotic stress
was up-regulated (e.g., similar to those coding for germin-
like proteins, allergen family proteins, heat shock proteins,
SAG20 and COR413-PM2).
Genes assigned to regulation of transcription
One of the largest groups of genes identified represented
those associated with the regulation of transcription [Bin:
27.3]. A variety of the identified transcription factor classes
was primarily down-regulated in gene expression in the
mutant (Fig. 9b, Supplementary Table S3), with several
13
 Author's personal copy

646 Plant Mol Biol (2014) 84:635–658

13
 Author's personal copy
Plant Mol Biol (2014) 84:635–658 647
◂Fig. 9  MapMan overview display of differentially-expressed genes
assigned to a metabolism and b cell function. Changes in leaf tran-
script levels at 60 DAP were measured from leaves harvested from
activation tagged up mutant and wild type Bintje plants. Transcript
abundance was expressed in reads per kb of exon model per million
mapped reads (RPKM) and converted to relative (up/Bintje) log2
fold values. Relative log2 fold values were visualized using MapMan
software. The change in expression levels were displayed via false
color code: blue for increased and red for decreased expression in
the mutant. Each square represents one gene (peptide). A color set-
ting was selected in which it saturates at a value of 2 (22  = a four-
fold change). Genes with significantly (q < 0.05) different expres-
sion levels were displayed. Number near each cluster corresponds to
MapMan BIN ids. Grey dots were genes not present in our dataset.
Detailed information for each data point including BIN ids, BIN
description, peptide ids, gene annotation and log2 fold values are
given in Supplementary Table S3
AP2/EREBP members (e.g., genes similar to cytokinin
response factor CRF4, AINTEGUMENTA and AINTEG-
UMENTA-LIKE), GATA transcription factor family mem-
bers, YABBY family members (e.g., genes similar to YAB1
and YAB3), GRAS family members (e.g., genes similar to
SCR), C2H2 zinc finger family members, Squamosa pro-
moter binding protein family members (e.g., genes similar
to SPL8 and SPL9), WRKY family members (e.g., genes
similar to WRKY21 and WRKY70), TCP family members
(e.g., genes similar to TCP4 and TCP12), general tran-
scription factor family members (e.g., genes similar to
growth regulating factors GRF1, GRF2 and GRF3), chro-
matin remodeling factors (e.g., gene similar to DDM1),
DNA methyltransferases (e.g., genes similar to CMT3
and MET1) and SET-domain family members (e.g., genes
similar to SUVH5 and SUVH6) showing this pattern. Other
identified transcription factor classes displayed a combina-
tion of up- or down-regulation in the mutant, such as bHLH
family members (e.g., genes similar to UNE10 and SPCH,
respectively), DOF zinc finger family members (e.g., genes
similar to CDF3 and OBP1, respectively), Homeobox fam-
ily members (e.g., genes similar to HDG8 and BLH11,
respectively), MYB family members (e.g., genes similar to
MYB111 and MYB106, respectively), bZIP family members
(e.g., genes similar to TGA1 and bZIP44, respectively) and
MADS family members (e.g., genes similar to AGL8 and
SVP, respectively). The only transcription regulation fam-
ily member class where the modified genes were all up-reg-
ulated were those of the identified AUX/IAA family (e.g.,
genes similar to IAA29 and SHY2).
Genes assigned to development
In the development function category (Bin: 33), many of
the identified genes in the mutant were down-regulated
(Fig. 9b, Supplementary Table S3), including various stor-
age protein genes (e.g., similar to patatin-like proteins
PLP1, PLP6, PLP9), genes for squamosa promoter-binding
proteins (e.g., similar to SPL9, SPL8, SPL14) and numer-
ous other genes (e.g., similar to a gene encoding a tropomy-
osin-related protein, the gene for protodermal factor PDF1,
the gene OFP8 for an ovate-family protein, and a gene for
a dem-related protein). A few identified genes were up-reg-
ulated in the mutant, such as the flowering locus gene FT,
those encoding various nodulin family proteins (e.g., genes
similar to MtN21 and MtN3), and those encoding tetraspa-
nins (e.g., similar to the genes TET7 and TET8).
Genes assigned to hormones
Genes assigned to various hormone-associated functions
(Bin: 17) were modified in their expression profiles, with
various up- and down-regulation patterns in the mutant
(Fig.  9b, Supplementary Table S3). Many auxin-induced/
regulated/response/activated and transduction genes were
primarily up-regulated (e.g., similar to auxin response
family genes, PIN5, DFL1 and SAUR12) in the mutant,
although a small proportion was down-regulated (e.g., sim-
ilar to the indole 3-acetic acid-amido synthetase GH3.1). In
addition, a few cytokinin synthesis/degradation-associated
genes (e.g., similar to the cytokinin oxidase/dehydrogenase
gene CKX1 and the isopentenyl transferase 3 gene IPT3)
were also up-regulated in the mutant. Genes associated
with ethylene synthesis/degradation/signal transduction
were both up-regulated (e.g., similar to 2OG-Fe(II) oxy-
genase, 2-oxoglutarate-dependent dioxygenase genes) and
down-regulated (e.g., similar to senescence-related SRG1
and ethylene response factor ERF 1 and ERF15 genes), as
were gibberellin synthesis/degradation/induced/regulated
genes (e.g., similar to the ENT-Kaurenoic acid hydroxylase
gene KAO2, a GA-responsive protein gene, a GASA1 gene,
and GA3- and GA20-oxidases). Some jasmonate metabo-
lism and signal transduction-associated genes were modi-
fied in the mutant (e.g., up-regulation of an allene oxide
synthase gene, and down-regulation of JAZ1), as were
some genes associated with salicylic acid synthesis/degra-
dation (e.g., up-regulation of genes similar to an SAM car-
boxymethyltransferase gene, and down-regulation of IAA
carboxymethyltransferase gene IAMT1).
Genes assigned to regulation
When considering genes assigned to a regulation func-
tion (Bin: 30), a majority of genes involved in signal-
ing was down-regulated in the mutant (Fig. 9b, Sup-
plementary Table S3), including those for various
leucine-rich repeat receptor kinases (e.g., genes similar
to FLS2 and IMK2), DUF26 receptor kinases (e.g., genes
similar to WAK2, LYM1 and ARK3), calcium signaling
molecules (e.g., genes similar to those for various calm-
odulin-binding proteins, like IQD6, IQD18 and NPGR1)
13

648

13
Table 2  Summary of RNA-seq gene expression changes in foliar tissues of the up mutant relative to the Bintje control as identified using MapMan
Major MapMan No. of major No. of differentially No. of genes No. of genes Highest up-regulated gene Highest down-regulated gene
pathways MapMan BINs expressed genes up-regulated down-regulated [PGSC0003DMP4000..; log2- [PGSC0003DMP4000..; log2-
in mutant in mutant fold change] fold change]

Metabolism 58 196 94 102 Tetrapyrolle synthesis gene Secondary

glutamyl-tRNA reductase metabolism.phenylpropanoids
[55149; 3.9] [47121; −4.8]
Cell function 22 1,441 560 881 Regulation of transcription Not assigned.unknown
[55149; 7.3] [28930; −6.5]
Regulation 22 452 178 274 RNA.regulation of transcription. Redox.glutaredoxins
unclassified [04,213; −5.3]
[55149; 7.3]
Cellular response 14 163 66 97 Development.unspecified Redox.glutaredoxins
[40404; 5.2] [04213; −5.3]
Stress 24 322 145 177 Stress.abiotic.unspecified Redox.glutaredoxins
[2994; 5.1] [04213; −5.3]
Secondary metabolism 11 44 10 34 Secondary metabolism.simple Secondary
phenols metabolism.phenylpropanoids
[35339; 3.7] [47121; −4.8]
Large enzyme families 12 110 38 72 Misc.GDSL-motif lipase Misc.nitrilases, *nitrile lyases,
Author's personal copy

[38664; 2.7] berberine bridge enzymes,

reticuline oxidases, troponine
reductases
[44963; −5.3]
RNA protein synthesis 8 141 67 74 Protein.degradation.ubiquitin. Protein.degradation.ubiquitin.
E3.SCF.FBOX E3.SCF.FBOX
[35351; 3] [47928; −4.6]
Plant Mol Biol (2014) 84:635–658
 Author's personal copy
Plant Mol Biol (2014) 84:635–658 649
and G-protein-associated signaling (e.g., genes similar to
ROPGEF1, ROPGEF2, ROPGEF5 and RABA1D). How-
ever, some signaling-associated genes were up-regulated
(e.g., genes similar to the lectin protein kinase family, the
histidine phosphotransfer kinase AHP3 and calcium-bind-
ing protein PBP1).
Genes assigned to protein synthesis, amino acid activation,
protein modification and protein degradation
Many of the genes associated with protein synthesis (Bin:
29.2) and modification (Bin: 29.4) that were modified in
expression in the mutant were down-regulated (Fig. 9b,
Supplementary Table S3), including those for various ribo-
somal structural proteins (e.g., genes encoding 40S ribo-
somal protein S15A and the 60S ribosomal proteins L26
and L38), post-translational protein modifiers (e.g., similar
to the histidine kinase gene AUR1, the cyclin-dependent
kinase genes CDBK1 and CDBK2, genes for numerous
protein kinase family proteins). However, some genes were
up-regulated in the mutant (e.g., similar to genes encod-
ing protein phosphatase 2C, genes encoding octicosapepti
de/Phox/Bem1p (PB1) domain-containing proteins, and the
protein kinase gene WAG1).
In contrast to protein synthesis and modification, many
genes associated with protein degradation (Bin: 29.5) were
up-regulated in the mutant (Fig. 9b, Supplementary Table
S3), including genes encoding numerous subtilases and
other serine proteases, cysteine proteases, aspartate pro-
teases and a metalloprotease (e.g., genes similar to XSP1,
SCPL48 and XCP1). In addition, various ubiquitination
pathway proteins were altered in their expression profiles
(up- or down-regulated) in the mutant, including vari-
ous ubiquitin conjugating enzymes (e.g., genes similar to
UBC9, UBC19, UBC36 and UBC37). Various members
of the E3 ubiquitin ligase classes were also modified, with
approximately two-thirds of the SCF E3 ligases modi-
fied in their expression being down-regulated (e.g., simi-
lar to genes ASK4, VB1, SLY2), one-half of the RING E3
ligases down-regulated (e.g., similar to genes VIM1, PUB8,
BRCA1), and our tagged gene, which classified as an appar-
ent member of the Cullin BTB/POZ E3 ligases, being
up-regulated.
Genes assigned to redox and metal handling
The majority of the genes altered in their expression pro-
files associated with redox (Bin: 21) was up-regulated in
the mutant (Fig. 9b, Supplementary Table S3), and included
thioredoxin members (e.g., similar to genes ACHT4 and
ATHM4), genes for various glutaredoxin family protein
members, and genes encoding a superoxide dismutase
(e.g., similar to gene FSD2), and a SOUL heme-binding
family protein member. A few genes were down-regulated
in the mutant, and included additional glutaredoxin family
protein members, a dehydroascorbate reductase gene (e.g.,
similar to DHAR1), a protein disulfide isomerase gene
(e.g., similar to PDI6) and a thioredoxin-disulfide reductase
gene (e.g., similar to NTRC).
A few genes associated with metal handling/binding,
chelation and storage (Bin: 15) were modified in their
expression profiles in the mutant (Fig. 9b, Supplementary
Table S3), with a nicotianamine synthase gene (e.g., simi-
lar to NAS3) strongly up-regulated, and genes encoding
a metal ion binding protein and a heavy metal-associated
domain containing protein down-regulated.
Genes assigned to transport
The majority of transport-associated genes (Bin: 34) altered
in their expression profiles was up-regulated in the mutant
(Fig.  9b, Supplementary Table S3). Many of these genes
encoded transporters for a variety of compounds, includ-
ing sugars, amino acids, ammonium, calcium, sulphate,
phosphate, potassium, nucleotides, peptides/oligopeptides,
unspecified cations and anions (e.g., genes similar to car-
bohydrate transporter MSS1, amino acid transporter AAP3,
ammonium transporter AMT1, sulfate transporter AST91,
phosphate transporter ATPT2, purine permease PUP1, cop-
per transporter COPT1, peptide transporter PEPT3, potas-
sium transporter KT2, calcium exchanger CAX7). However,
some transporter classes exhibited a greater proportion of
down-regulation of gene expression in the mutant, such
as genes for members of the ABC transporters/multidrug
resistance systems, members of the MATE efflux protein
family and members of the SEC14 cytosolic factor protein
family.
Genes assigned to large enzyme families
Many of the genes encoding large enzyme families (Bin:
26) altered in their expression profiles were down-regulated
in the mutant (Fig. 9b, Supplementary Table S3). These
genes represented enzyme families such as UDP gluco-
syl and glucoronyl transferases (e.g., similar to GT72B1),
beta 1,3 glucan hydrolases (e.g., similar to BGL2),
nitrilases/nitrile lyases/berberine bridge enzymes/reticuline
oxidases/troponine reductases (e.g., similar to MES10),
cytochrome P450 s (e.g., similar to CYP71B34), peroxi-
dases (e.g., similar to PER12), dynamins (e.g., similar to
ADL4) and GDSL-motif lipases. In contrast to the above,
relatively few large enzyme families displayed an up-
regulation of their identified genes, such as invertase/pec-
tin methylesterase inhibitor family proteins (e.g., similar
to C/VIF1) and glutathione S transferases (e.g., similar to
ATGSTU25).
13
 Author's personal copy
650
Tuber gene expression
While the primary focus of our gene expression analyses
was the above-ground tissues, our observed tuber pheno-
types of elongated tuber shape and early sprouting were
similar to those reported by Morris et al. (2006), who
modified tuber isoprenoid biosynthesis via the plastid MEP
(2-C-methyl-d-erythritol) pathway. As Morris et al. (2006)
characterized DXS (1-deoxy-d-xylulose 5-phosphate syn-
thase), PSY (phytoene synthase) and PDS (phytoene desat-
urase) transcript levels in their study, we carried out a simi-
lar assessment of these transcripts for both wild type Bintje
and up tubers (Fig. 10). Using RT-PCR, we noted increased
levels of DXS and PSY transcripts in tubers of the mutant,
which was similar to that reported by Morris et al. (2006)
in their transgenics exhibiting elongated and early sprout-
ing tuber phenotypes.
A BTB/POZ domain‑containing protein
Our data showed that the first two potato genes flank-
ing the right border were significantly up-regulated in
the up mutant. In the potato genome browser, these two
genes (PGSC0003DMG401011239 and PGSC0003DMG
402011239) are annotated as two separate conserved genes
of unknown function (Supplementary Fig. S3a, b). RT-
PCR results indicated that both genes were up-regulated
(Fig.  7a), and RNA-seq reads corresponding to this gene
region (Supplementary Fig. S3c) covered two annotated
potato gene models PGSC0003DMG402011239 and
PGSC0003DMG401011239 (PGSC0003DMP400019889).
The raw reads from the Illumina GAII platform indicated
similar, greater read counts for these two predicted genes
for the mutant underperformer compared to Bintje (Sup-
plementary Fig. S3c), illustrating the up-regulation of tran-
script level in the mutant, and the low level of expression
in Bintje. The corresponding gene expression value for this
Fig. 10  Reverse transciptase PCR analysis of various potato isopre-
noid biosynthesis pathway genes in tubers of wild type Bintje and the
up mutant. RNA was extracted from freshly harvested mature potato
tubers. Primer information used in RT-PCR is given in Supplemen-
tary Table S2. EF-1α potato elongation factor 1-α (control), DXS
1-deoxy-d-xylulose 5-phosphate synthase, PSY phytoene synthase,
PDS phytoene desaturase
13
Plant Mol Biol (2014) 84:635–658
Fig. 11  Analysis of the identified StBTB/POZ domain-containing ▸
protein 1 (StBTB/POZ1) sequence. a Nucleotide sequence encoding
the predicted open reading frame (ORF). b Amino acid translation of
the predicted ORF, showing the location of the C-terminal BTB/POZ
domain (red highlight) and the N-terminal ARM repeat domain
(black highlight). c InterProScan results showing the presence of the
BTB/POZ and ARM repeat domains
region was found to be 2.3-fold (log2 value) greater in the
mutant, providing secondary verification of our PCR results.
The potato gene models in the PGSC database showed
little physical separation between the two genes mod-
els (Supplementary Fig. S3a, b), approximately 100 bp.
Based on these observations, and the similar level of
expression noted for these two genes in the up mutant, we
hypothesized that the PGSC potato gene model was incor-
rect. Recent models generated by the Tomato Genome
Annotation Group (ITAG) also suggest that the potato
(Sotub10g006050.1.1), tomato (Solyc10g005600.2.1) and
tobacco (NbS00001319g0217.1) gene model predictions
comprise a single gene containing these exons (Supplemen-
tary Fig. S3b). Interestingly, our RNA-seq transcript reads
were similar to the predicted exon reads, except for the
reads which aligned with the second intron of potato gene
(PGSC0003DMG401011239; chr10:665860..666280).
The RNA-seq tracks in the PGSC database showed simi-
lar reads in the corresponding region (Supplementary Fig.
S3b). A further database search using the 421 bp long
sequence covering these reads (chr10:665860..666280) as
query revealed that this sequence has a secondary, perfect
hit (3.9e−88; 100 % identity) in a region of chromosome 12
that is not annotated for PGSC loci but houses ITAG gene
models for tomato and potato (ITAG-Sotub12g020170.1.1;
chr12:35542420..35542540). One interpretation is that
these RNA-seq reads are misaligned with our conserved
gene of unknown function on chromosome 10 and actu-
ally represent expression of the ITAG gene model on chro-
mosome 12. The PGSC RNA-seq data also indicated that
the gene expression pattern is similar across many differ-
ent potato tissues (Supplementary Fig. S3b). Furthermore,
while our RNA-seq reads across the predicted exons are
more abundant in the up mutant when compared with
wild type Bintje, our RNA-seq reads for the second intron
region show little difference in abundance between Bintje
and up (Supplementary Fig. S3b), providing further sup-
port for the idea that these reads are misaligned with this
gene model.
To confirm that the two predicted PGSC gene mod-
els actually did represent a single gene, we used prim-
ers designed against the start and stop codons based on
the ITAG gene model Sotub10g006050.1.1, and carried
out RT-PCR using leaf cDNA from the mutant. If the
ITAG gene models were correct, an amplification prod-
uct of approximately 3 kb would result. Our PCR results
 Author's personal copy
Plant Mol Biol (2014) 84:635–658 651

13
 Author's personal copy
652
generated a 3 kb fragment (Supplementary Fig. S4), indi-
cating that the two predicted genes of the PGSC potato
model actually comprised a single gene. To obtain the
complete transcript sequence, we used 5′ and 3′ RACE
to identify a full-length transcript, representing an open
reading frame of 3,051 bp (Fig. 11a; Genbank Acces-
sion No. KC979135), encoding a predicted protein of
1,017 amino acids (Fig. 11b). A BLAST analysis of the
predicted protein yielded hits to several hypothetical/
predicted plant orthologs (castor bean, soybean, cucum-
ber, grape, tomato), representing a BTB/POZ domain-
containing protein (Supplementary Fig. S5). As such, our
potato gene (StBTB/POZ domain-containing protein 1)
yielded a predicted translation product with a BTB/POZ
(Broad complex, Tramtrac, Bric-a-brac; also known as
Pox virus and Zinc finger) conserved domain near the
C-terminus of the predicted protein, as well as an Arma-
dillo (ARM) repeat domain towards the N-terminus
(Fig. 11b, c).
During this study, our focus on endogenous expres-
sion of StBTB/POZ1 was on foliar tissues, and our results
indicated a relatively low level of endogenous expression
in wild type potato, compared to the up mutant. Given the
novelty of this gene, to further assess normal expression of
StBTB/POZ1 in potato, an analysis of PGSC RNA-seq data
was carried out, and indicated that this gene is expressed
in many tissues/organs under various conditions (Supple-
mentary Fig. S3b; data not shown). Using leaf expression
as the baseline, low levels of StBTB/POZ1 expression was
also noted for stem, flowers and young/mature tubers. In
contrast, greater expression was noted for roots, stolons,
sprouting tubers and fruit. RNA-seq data also indicated that
StBTB/POZ1 expression was induced under certain stress
conditions. While wounding, heat shock and mannitol treat-
ments were non-inductive, salt stress induced StBTB/POZ1
expression, and Phytophthora infestans treatment sig-
nificantly enhanced abundance (data not shown). These
RNA-seq profiles can be accessed through the Spud DB
Genome Browser (http://potato.plantbiology.msu.edu/cgi-
bin/gbrowse/potato/), viewing Landmark or Region
chr10:659000..670000.
Discussion
The activation tagged mutant underperformer (up) was
initially identified by its dwarf and axillary branching
phenotype under in vitro growth conditions. Further in
vitro and in vivo studies revealed additional mutant phe-
notypes, including loss of flowering (characterized by
small floral buds and premature floral bud abscission),
early senescence, reduced tuber dormancy and changes in
tuber size, tuber shape, tuber number and number of tuber
13
Plant Mol Biol (2014) 84:635–658
eyes. These suggested that the single insertion resulted
in numerous pleiotropic effects, causing a range of phe-
notypes. Our in vivo and in vitro growth data supported
each other, with the exception of the number and weight
of tubers and number of axillary branches. In vitro plants
of the up mutant produced more axillary branches; how-
ever we observed fewer axillary branches in the mutant up
plants during in vivo plant growth compared to wild type
Bintje This trait may explain, to some extent, the greater
number of tubers in up, and the subsequently greater tuber
weight per mutant under in vitro conditions. Under green-
house and growth chamber experiments, up mutant plants
exhibited earlier senescence, resulting in less time for
tuber formation and bulking, possibly causing lower tuber
number and overall reduction in yield. Observed differ-
ences in mutant phenotypes between in vitro and ex vitro
growth conditions could be due to environmental factors
(e.g., gaseous environment, in vitro sucrose supply, tem-
perature and light) to which the mutant may be sensitive,
which are less controlled in soil grown plants compared to
in vitro plants.
In this study, reverse transcriptase PCR results showed
both up- and down-regulation of genes on both sides of the
T-DNA insertion. In almost all cases of activation tagging
using the CaMV 35S enhancer, the gene immediately adja-
cent to the right border has been up-regulated (Weigel et al.
2000; Ahn et al. 2007; Wan et al. 2009), leading eventually
to the observed phenotype(s). In our study, the first two pre-
dicted genes adjacent to the right border were strongly up-
regulated in the up mutant. On closer analysis, it became
apparent that the potato gene model was incorrect, and that
these two predicted open reading frames actually repre-
sented a single open reading frame. This was much more
in line with the gene model predictions generated by ITAG.
Hence, the up-regulated gene adjacent to the right border
identified here, located on chromosome 10 and predicted
to be a BTB/POZ domain-containing protein, is believed to
be the actual tagged gene ultimately leading to the down-
stream phenotypes.
In addition to the above gene, the PCR-based gene
expression analysis of the predicted genes extending up
to 400 kb from the left and right border of the insertion
revealed that several other genes were modified in their
expression profiles, not only the first gene adjacent to the
right border/enhancers. Generally, most activation tag
studies have assessed gene expression in close proximity
to the insertion (Weigel et al. 2000; Imaizumi et al. 2005;
Wan et al. 2009) and have not considered chromosomal
locations further upstream or downstream of the inser-
tion. However, gene expression profiling of the Arabidop-
sis activation tagged mutant K3571 revealed 1,273 genes
with greater than twofold RNA abundance compared to
control. Among 244 genes in a 984 kb region off of the
 Author's personal copy
Plant Mol Biol (2014) 84:635–658 653
right border, 201 genes were up-regulated and six were
down-regulated (Ahn et al. 2007). Several explanations for
the activation of multiple genes were suggested, such as:
(1) the scanning action of an activator bound to the 35S
enhancer, which scans the genome and activates the genes
in its path until it meets with an insulator (Bell and Felsen-
feld 1999), (2) chromatin changes and re-arrangements
may occur due to the T-DNA insertion, reducing gene
repression, and (3) the T-DNA insertion serves to differ-
entially express a nearby gene, such as a transcription fac-
tor, which has a subsequent impact on downstream genes.
At present, it is unknown how many of these explanations
may have contributed to the observed up gene expression
changes.
Visualization and assessment of gene expression data
A more comprehensive comparison of global gene expres-
sion between the Bintje control and the up mutant using
RNA-seq revealed that more than 1,600 genes, represent-
ing loci across all 12 chromosomes, were differentially
expressed in foliar tissues of the mutant. Using Map-
Man software (Thimm et al. 2004), we viewed the data
at different levels of resolution, from the global level, the
level of discrete processes, or down to the single gene
level. The overview of expression levels indicated that the
total number of genes down-regulated was greater than
the total number of genes up-regulated in the up mutant.
These differentially expressed genes sorted into numer-
ous categories, representing different metabolic and func-
tional pathways, and indicated that multiple perturbations
of normal plant gene expression and functional/metabolic
pathways were taking place within the mutant. This is
not surprising, and can be expected to vary depending on
where the tagged locus fits into various pathways within
the plant. For example, an activation-tagged Arabidop-
sis line overexpressing a MYB transcription factor gene
(PAP1) was found to over-express 38 genes (Tohge et al.
2005). In contrast, ectopic expression in an activation-
tagged Arabidopsis line of the ORE7/ESC gene, an AT-
hook DNA binding protein, resulted in 1,096 genes show-
ing at least a two-fold change in gene expression in the
mutant compared to wild type (Lim et al. 2007). Hence,
our large number of observed gene expression changes
can be viewed as reasonable. In addition, while many
genes were altered in their expression profiles in foliar
samples of our mutant, previous studies have indicated
tissue specific modification of gene expression in activa-
tion-tagged mutants (Weigel et al. 2000). Therefore, while
we were able to catalog the gene expression changes in
leaf tissues using RNA-seq, it is probable that a different
set of differentially-expressed genes may occur in other
tissues.
Gene expression, pathway modifications, and up mutant
characteristics
The mutant phenotypes noted here represented those occur-
ring above-ground (dwarf stature, slow growth, prema-
ture floral bud abscission, premature leaf senescence) and
below-ground (elongated tuber shape, reduced tuber dor-
mancy and early sprouting, other tuber quality traits). Our
foliar gene expression profiling results indicated numer-
ous gene expression perturbations for different pathways.
Various genes in the photosynthesis pathway, hormone
metabolism, abiotic and biotic stress pathways, starch and
sucrose metabolism, minor carbohydrate metabolism, cell
wall degradation proteins, proteolysis pathway, and redox
reaction pathway and transport were up-regulated in the
mutant. Furthermore, there appeared to be a shift in car-
bon metabolism, with increased gene expression towards
carbohydrate metabolism, and reduced gene expression
towards lipid metabolism in the mutant. In contrast, numer-
ous genes involved in DNA synthesis, cell division and cell
cycle, cell organization, development, transcription factors,
signaling, protein modification and secondary metabolism
were down-regulated in the mutant. At this stage, it is dif-
ficult to assess the contribution of each modified gene to
the observed phenotypes of the mutant, although there are
many changes in the mutant as a result of our activation tag
insertion. As cell division and cell expansion are key con-
tributors to overall plant size and form, the strong down-
regulation observed for DNA synthesis, cell cycle and cell
division genes in the up mutant are expected to be key
contributors to the phenotype. We can also identify vari-
ous genes modified in their expression, and shown in other
studies to be associated with similar phenotypes. For exam-
ple, several Aintegumenta/Aintegumenta-like genes were
down-regulated in the up mutant. Given the importance of
these genes in regulating cell division and associated flo-
ral organ and shoot development (Krizek 2009; Mizukami
and Fischer 2000), their reduced expression in the mutant
could contribute to the smaller stature and reduced floral
bud development observed. In addition, our results indi-
cated the up-regulation of DFL1 in the up mutant, whose
over-expression has previously been described as causing
inhibition of shoot elongation, and subsequent dwarfness
in Arabidopsis (Nakazawa et al. 2001). We also noted that
several members of the Growth Regulating Factor (GRF)
gene family were down-regulated in the mutant, and these
transcription factors play a role in the regulation of GA-
induced internode/stem elongation (van der Knaap et al.
2000). Similarly, plant β-1,3-glucanases, involved in path-
ogen defense as well as a wide range of normal develop-
mental processes like cell division and cell wall remodeling
(Doxey et al. 2007) were also down-regulated in the up
mutant.
13
 Author's personal copy
654
Given the significant role of hormones in controlling
plant development, it was not surprising that various hor-
mone-associated pathways appeared modified in the up
mutant. For example, auxins are closely connected with
numerous plant growth and development processes (e.g.,
lateral root differentiation and apical dominance), and they
are also known to induce the expression of several genes
like DFL1 and SAURs (Nakazawa et al. 2001). In addition,
the overexpression of the IAMT1 gene, which encodes an
indole-3-acetic acid (IAA) carboxyl methyltransferase that
converts IAA to methyl-IAA ester (MeIAA), affects auxin-
regulated processes (Qin et al. 2005). In the present study,
DFL1 (discussed above) and SAUR were up-regulated,
whereas IAMT1 was down-regulated in the up mutant, sug-
gesting modifications in auxin response potential.
The phytohormones auxin and ethylene, and gradients
associated with these also play a role in abscission and
senescence (Basu et al. 2013; Taylor and Whitelaw 2001),
traits which were noted to occur prematurely in the up
mutant. Auxin also inhibits transcription of various genes
associated with both processes (Noh and Amasino 1999;
Tucker et al. 2002) and the auxin response factors, ARF1
and ARF2, have been shown to regulate senescence and
floral organ abscission in Arabidopsis (Ellis et al. 2005).
In addition, these traits are also induced by reduced avail-
ability of carbon assimilate (Marcelis et al. 2004). As men-
tioned above, RNA-seq data indicated that carbohydrate
metabolism and auxin response potential appeared modi-
fied in the mutant, and may contribute to the early floral
bud abscission and leaf senescence observed.
Various expression profiling studies have characterized
the genes associated with pedicel/floral organ abscission
and senescence (Breeze et al. 2011; Cai and Lashbrook
2008; Nakano et al. 2013). During these processes, expres-
sion of genes associated with different functional path-
ways are modified. These include genes associated with
transcriptional regulation (e.g., Myb transcription factors,
ERF ethylene response factors, ARF auxin response fac-
tors), receptor-like kinases (e.g., FLS2, wall associated
kinase), cell wall degrading, modifying and modeling pro-
teins (e.g., cellulose synthase, expansin, invertase/pectin
methylesterase inhibitor, pectate lyase, polygalacturonase,
XTH endotransglucosylase/hydrolase), hormone-associ-
ated regulators for biosynthesis, action and transport (e.g.,
allene oxide synthase, auxin efflux carrier, indole-3-acetic
acid-amido synthetase, gibberellin oxidase), lipid metabo-
lism (e.g., GDSL-motif lipase/hydrolase family), stress
response (e.g., osmotin, pathogenesis-related protein) and
proteolytic degradation (e.g., subtilase, ubiquitin conjugat-
ing enzyme). RNA-seq analysis of the up mutant in this
present study revealed that many of the same pathways
and gene types were modified in the mutant. As examples
of the types of genes listed above, foliar tissues of the up
13
Plant Mol Biol (2014) 84:635–658
mutant displayed up- and down-regulation of various Myb
transcription factors, up-regulation of numerous proteases
(serine, cysteine, aspartate, metallo), various ubiquitin-con-
jugating enzymes and E3 ligases, several pectate lyases and
polygalacturonases, allene oxide synthase, down-regulation
of several GDSL-motif lipases, osmotin-like proteins. The
modification of the same types of pathways and genes in
the up mutant would be expected to contribute to the earlier
abscission and senescence characteristics.
In consideration of the tuber phenotypes observed for
the up mutant, several of the pathways noted as being mod-
ified, based on RNA-seq data, could have direct implica-
tions for the observed phenotype. Carbohydrate metabo-
lism perturbations can have distinct effects on potato tubers
(Frommer and Sonnewald 1995; Morris et al. 2006). For
example, transgene-mediated reduction of plastid ATP/
ADP-transporter activity and consequent reduction in tuber
starch content results in an elongated tuber phenotype
(Tjaden et al. 1998). Our RNA-seq results indicated that
gene expression in the up mutant was modified for various
genes associated with major (starch, sucrose) and minor
carbohydrate metabolism, as well as photosynthesis, and
this may have contributed to the overall tuber phenotype.
Several phytohormones are known to contribute to the
control of tuber dormancy and sprouting (Suttle 2004).
ABA is needed for dormancy maintenance, and declines
during dormancy progression, while sensitivity to cytokinin
increases as dormancy progresses, and cytokinin increase
is associated with dormancy release and sprouting. In con-
trast, GA is a negative regulator of tuberization (Bachem
et al. 2001; Xu et al. 1998). The knockdown of a steroid
dehydrogenase transcript (GANGLY) in potato resulted in
a modification of active GA levels, and also elongated tuber
and sprouting phenotypes (Bachem et al. 2001), similar to
the observations noted for the up mutant. We noted that
expression of various GA-regulated, GA-responsive and
GA homeostasis genes were modified in the up mutant and
could impact overall tuber phenotype.
A study by Morris et al. (2006) modified tuber iso-
prenoid biosynthesis via the plastid MEP (2-C-methyl-
d-erythritol) pathway, which resulted in significantly elon-
gated tubers, as well as early sprouting. These phenotypes
closely resembled phenotypes observed with the up mutant.
In addition, when we assessed DXS (1-deoxy-d-xylulose
5-phosphate synthase), PSY (phytoene synthase) and PDS
(phytoene desaturase) transcript levels for both wild type
Bintje and up tubers, we noted increased levels of DXS and
PSY transcripts in the mutant, and no difference for PDS
transcript levels, results which paralleled those of Morris
et al. (2006), suggesting that similar mechanisms are asso-
ciated with the up tuber characteristics. The MEP pathway
provides intermediates for the phytohormones cytokinin,
gibberellic acid (GA) and abscisic acid (ABA), and Morris
 Author's personal copy
Plant Mol Biol (2014) 84:635–658 655
et al. (2006) noted enhanced levels of cytokinin, trans-zea-
tin riboside in tubers of their transgenic potatoes. While we
did not quantify cytokinin levels in up tubers, we did note
that cytokinin metabolism gene expression (cytokinin oxi-
dase/dehydrogenase, isopentenyl transferase 3) was modi-
fied in the mutant. All of the above results further suggest
that some modulation of plant phytohormone signaling
and response was associated with the various above- and
below-ground up mutant phenotypes.
We have described several potential genetic contribu-
tions to the up tuber phenotype above. van Eck et al. (1994)
described multiple alleles that control tube shape, with the
round (Ro) locus, the major QTL for shape, mapping to
chromosome 10, the same chromosome on which our activa-
tion tag is located. It remains to be determined if the tag has
directly impacted the Ro locus, or if the long tuber pheno-
type of up is due to pleiotropic effects of the activation tag.
The candidate tagged gene: a BTB/POZ domain‑ and ARM
repeat domain‑containing protein
Based on its location off of the right border of the T-DNA
and adjacent to the CaMV 35S enhancers, as well as the
clearly up-regulated gene expression in foliar tissues of the
up mutant, the candidate activation tagged gene is believed
to encode a predicted protein with a C-terminal BTB/POZ
domain, as well as an N-terminal Armadillo (ARM) repeat
domain, which we named StBTB/POZ domain contain-
ing protein 1 (StBTB/POZ1). BLAST analyses indicated
close similarity to several hypothetical/predicted proteins,
indicating that StBTB/POZ1 represents a novel, yet-to-be
characterized protein. The BTB/POZ domain-containing
proteins play an important role, possibly affecting global
repression of transcription factors (Hur et al. 2001; Liu
et al. 2011), plant defense (Boyle et al. 2009; Qu et al.
2010), protein degradation (Yuasa et al. 2008) and cell
death (Sadanandom et al. 2008). The BTB/POZ and ARM
repeat domains are both known as protein–protein inter-
action domains (Perez-Torrado et al. 2006; Tewari et al.
2010), suggesting that the protein product of StBTB/POZ1
may be modulating its effects through key protein–protein
associations. Two Arabidopsis proteins exhibiting a simi-
larly-organized N-terminal ARM repeat domain and C-ter-
minal BTB/POZ domain, although with little sequence
similarity (8e−05) to StBTB/POZ1, were previously iden-
tified as Class V BTB domain-containing proteins. These
were shown to interact via their BTB domain with Cullin
3A (Dieterle et al. 2005), thus flagging these proteins as
components of the multi-subunit CUL3-BTB E3 ubiquitin
ligases. These results suggest that StBTB/POZ1 is also a
member of the CUL3-BTB E3 ubiquitin ligases, an impli-
cation which is supported by our MapMan analyses that
placed StBTB/POZ1 within this pathway.
The process by which StBTB/POZ1 could lead to
the complex phenotype manifested by the up mutant is
unknown. However, in the CUL3-BTB multi-subunit E3
ligase, the BTB/POZ domain interacted with CUL3 (Diet-
erle et al. 2005), and the ARM repeat domain would be
expected to interact with other proteins, providing sub-
strate specificity for the target protein to be ubiquitinated.
The two previously mentioned Arabidopsis proteins
encompassing the ARM and BTB/POZ domains (Dieterle
et al. 2005) encoded ARIA (Kim et al. 2004) and ABAP1
(Masuda et al. 2008). ARIA represents a component of the
ABA signaling pathway, while ABAP1 is involved in DNA
replication and cell proliferation. In both cases, the ARM
repeats interacted with transcription factors; ARIA with the
ABA response element binding factor ABF2 (Kim et al.
2004), and ABAP1 with several members of the NAC, AP2
and TCP transcription factor families (Masuda et al. 2008).
Hence, StBTB/POZ1 action may involve the modulation of
various transcription factors, with the potential to alter the
expression of many genes and pathways in the up mutant.
Overexpression of ABAP1 limited DNA replication during
mitosis, and decreased cell proliferation, which could be
considered somewhat similar to the reduced levels of cell
cycle-associated and cell division-associated gene expres-
sion noted for the StBTB/POZ1-overexpressing up mutant.
Several BTB/POZ domain-containing proteins are interac-
tion partners with the Cullin component of the E3 ubiqui-
tin ligase complex (Thomann et al. 2005) and several plant
proteins, including those associated with plant hormone
signaling (Santner and Estelle 2010) are targeted for deg-
radation via the E3 ubiquitin ligase complex. In addition,
the strong up-regulation of various protease classes in the
up mutant suggest that proteolysis plays a significant role
in the mutant, observations which may help to explain
the molecular basis of the multiple phenotypic changes
observed, as well as the apparent broad gene expression
and metabolic pathway perturbations in the up mutant.
This paper has characterized an activation tagged potato
mutant with several interesting phenotypic traits, at the
developmental, growth form and global gene expression
levels. Based on all of our information, we have put forward
a candidate gene (StBTB/POZ1) believed to be responsible
for this. To confirm this hypothesis, transgenic experiments
taking two different approaches are in progress. In one
approach, the StBTB/POZ1 ORF has been cloned behind
a standard 35S CaMV promoter and introduced into wild
type Bintje, to see if over-expression of StBTB/POZ1 reca-
pitulates the up mutant phenotype. In a second approach,
an RNAi construct targeting StBTB/POZ1 will be intro-
duced into the up mutant, to see if this will cause a rever-
sion of the mutant back to a wild type form. The results of
these experiments will confirm if StBTB/POZ1 is truly the
tagged gene, responsible for the up mutant phenotype.
13
 Author's personal copy
656
Acknowledgments  This work represents a portion of S. Aul-
akh’s doctoral dissertation. We want to thank Jared Carter (Depart-
ment of Horticulture, Virginia Tech) for help with growth chamber
experiments, Robin Buell and Brieanne Vaillancourt, (Department of
Plant Biology, Michigan State University) for assistance with RNA-
sequencing, and Kavita Aulakh for help with image preparation. This
work was funded through Special Grants (2003-38891-02112, 2008-
38891-19353 and 2009-38891-20092) from the United States Depart-
ment of Agriculture, Hatch Project VA-135853, and operating funds
from the Commonwealth of Virginia to the Institute for Advanced
Learning and Research.
References
Ahn JH, Kim J, Yoo SJ, Yoo SY, Roh H, Choi JH, Choi MS, Chung
KS, Han EJ, Hong SM, Jung SH, Kang HJ, Kim BK, Kim MD,
Kim YK, Kim YH, Lee H, Park SH, Yang JH, Yang JW, Yoo
DH, Yoo SK, Lee JS (2007) Isolation of 151 mutants that have
developmental defects from T-DNA tagging. Plant Cell Physiol
48:169–178
Anders S, Huber W (2010) Differential expression analysis for
sequence count data. Genome Biol 11:R106
Ayliffe MA, Pallotta M, Langridge P, Pryor AJ (2007) A barley activa-
tion tagging system. Plant Mol Biol 64:329–347
Bachem CW, Horvath B, Trindade L, Claassens M, Davelaar E, Jordi
W, Visser RG (2001) A potato tuber-expressed mRNA with
homology to steroid dehydrogenases affects gibberellin levels
and plant development. Plant J 25:595–604
Basu MM, González-Carranza ZH, Azam-Ali S, Tang S, Shahid AA,
Roberts JA (2013) The manipulation of auxin in the abscission
zone cells of Arabidopsis flowers reveals that indoleacetic acid
signaling is a prerequisite for organ shedding. Plant Physiol
162:96–106
Bell AC, Felsenfeld G (1999) Stopped at the border: boundaries and
insulators. Curr Opin Genet Dev 9:191–198
Boyle P, Le Su E, Rochon A, Shearer HL, Murmu J, Chu JY, Fobert
PR, Després C (2009) The BTB/POZ domain of the Arabidop-
sis disease resistance protein NPR1 interacts with the repression
domain of TGA2 to negate its function. Plant Cell 21:3700–3713
Breeze E, Harrison E, McHattie S, Hughes L, Hickman R, Hill C,
Kiddle S, Kim YS, Penfold CA, Jenkins D, Zhang C, Morris
K, Jenner C, Jackson S, Thomas B, Tabrett A, Legaie R, Moore
JD, Wild DL, Ott S, Rand D, Beynon J, Denby K, Mead A,
Buchanan-Wollaston V (2011) High-resolution temporal profil-
ing of transcripts during Arabidopsis leaf senescence reveals
a distinct chronology of processes and regulation. Plant Cell
23:873–894
Busov VB, Meilan R, Pearce DW, Ma C, Rood SB, Strauss SH (2003)
Activation tagging of a dominant gibberellin catabolism gene
(GA 2-oxidase) from poplar that regulates tree stature. Plant
Physiol 132:1283–1291
Busov V, Yordanov Y, Gou J, Meilan R, Ma C, Regan S, Strauss S
(2011) Activation tagging is an effective gene tagging system in
Populus. Tree Genet Genom 7:91–101
Cai S, Lashbrook CC (2008) Stamen abscission zone transcrip-
tome profiling reveals new candidates for abscission control:
enhanced retention of floral organs in transgenic plants overex-
pressing Arabidopsis ZINC FINGER PROTEIN2. Plant Physiol
146:1305–1321
Costa V, Angelini C, De Feis I, Ciccodicola A (2010) Uncovering the
complexity of transcriptomes with RNA-Seq. J Biomed Biotech-
nol 2010:853916
Dieterle M, Thomann A, Renou JP, Parmentier Y, Cognat V, Lem-
onnier G, Müller R, Shen WH, Kretsch T, Genschik P (2005)
13
Plant Mol Biol (2014) 84:635–658
Molecular and functional characterization of Arabidopsis Cullin
3A. Plant J 41:386–399
Doxey AC, Yaish MW, Moffatt BA, Griffith M, McConkey BJ (2007)
Functional divergence in the Arabidopsis beta-1,3-glucanase gene
family inferred by phylogenetic reconstruction of expression
states. Mol Biol Evol 24:1045–1055
Ellis CM, Nagpal P, Young JC, Hagen G, Guilfoyle TJ, Reed JW
(2005) AUXIN RESPONSE FACTOR1 and AUXIN RESPONSE
FACTOR2 regulate senescence and floral organ abscission in
Arabidopsis thaliana. Development 132:4563–4574
Fischer L, Lipavska H, Hausman JF, Opatrny Z (2008) Morphologi-
cal and molecular characterization of a spontaneously tuberizing
potato mutant: an insight into the regulatory mechanisms of tuber
induction. BMC Plant Biol 8:117
Frommer WB, Sonnewald U (1995) Molecular analysis of carbon par-
titioning in solanaceous species. J Exp Bot 46:587–607
Harrison EJ, Bush M, Plett JM, McPhee DP, Vitez R, O’Malley B,
Sharma V, Bosnich W, Séguin A, MacKay J, Regan S (2007)
Diverse developmental mutants revealed in an activation-tagged
population of poplar. Can J Bot 85:1071–1081
Hayashi H, Czaja I, Lubenow H, Schell J, Walden R (1992) Activation
of a plant gene by T-DNA tagging: auxin-independent growth in
vitro. Science 258:1350–1353
Hiratsu K, Matsui K, Koyama T, Ohme-Takagi M (2003) Dominant
repression of target genes by chimeric repressors that include
the EAR motif, a repression domain, in Arabidopsis. Plant J
34:733–739
Huang SS, Cerny RE, Bhat DS, Brown SM (2001) Cloning of an
Arabidopsis patatin-like gene, STURDY, by activation T-DNA
tagging. Plant Physiol 125:573–584
Hur MW, Lee DK, Kwon HS (2001) Molecular mechanism of tran-
scription repression by HDAC and BTB/POZ-domain transcrip-
tion factors. FASEB J 15:A517
Imaizumi R, Sato S, Kameya N, Nakamura I, Nakamura Y, Tabata S,
Ayabe S, Aoki T (2005) Activation tagging approach in a model
legume, Lotus japonicus. J Plant Res 118:391–399
Kakimoto T (1996) CKI1, a histidine kinase homolog implicated in
cytokinin signal transduction. Science 274:982–985
Kardailsky I, Shukla VK, Ahn JH, Dagenais N, Christensen SK,
Nguyen JT, Chory J, Harrison MJ, Weigel D (1999) Activation
tagging of the floral inducer FT. Science 286:1962–1965
Kim S, Choi HI, Ryu HJ, Park JH, Kim MD, Kim SY (2004) ARIA,
an Arabidopsis arm repeat protein interacting with a transcrip-
tional regulator of abscisic acid-responsive gene expression,
is a novel abscisic acid signaling component. Plant Physiol
136:3639–3648
Kondrák M, Marincs F, Kalapos B, Juhász Z, Bánfalvi Z (2011)
Transcriptome analysis of potato leaves expressing the tre-
halose-6-phosphate synthase 1 gene of yeast. PLoS ONE
6:e23466
Krizek B (2009) AINTEGUMENTA and AINTEGUMENTA-LIKE6
act redundantly to regulate Arabidopsis floral growth and pattern-
ing. Plant Physiol 150:1916–1929
Larkin MA, Blackshields G, Brown NP, Chenna R, McGettigan
PA, McWilliam H, Valentin F, Wallace IM, Wilm A, Lopez R,
Thompson JD, Gibson TJ, Higgins DG (2007) Clustal W and
Clustal X version 2.0. Bioinformatics 23:2947–2948
Lim PO, Kim Y, Breeze E, Koo JC, Woo HR, Ryu JS, Park DH, Bey-
non J, Tabrett A, Buchanan-Wollaston V, Nam HG (2007) Over-
expression of a chromatin architecture-controlling AT-hook pro-
tein extends leaf longevity and increases the post-harvest storage
life of plants. Plant J 52:1140–1153
Liu Q, Yao F, Wang M, Zhou B, Cheng H, Wang W, Jin L, Lin Q,
Wang JC (2011) Novel human BTB/POZ domain-containing zinc
finger protein ZBTB1 inhibits transcriptional activities of CRE.
Mol Cell Biochem 357:405–414
 Author's personal copy
Plant Mol Biol (2014) 84:635–658 657
Marcelis LF, Heuvelink E, Hofman-Eijer LR, Bakker JD, Xue LB
(2004) Flower and fruit abortion in sweet pepper in relation to
source and sink strength. J Exp Bot 55:2261–2268
Massa AN, Childs KL, Lin H, Bryan GJ, Giuliano G, Buell CR
(2011) The transcriptome of the reference potato genome Sola-
num tuberosum Group Phureja clone DM1-3 516R44. PLoS ONE
6:e26801
Masuda HP, Cabral LM, De Veylder L, Tanurdzic M, de Almeida
Engler J, Geelen D, Inzé D, Martienssen RA, Ferreira PC,
Hemerly AS (2008) ABAP1 is a novel plant Armadillo BTB
protein involved in DNA replication and transcription. EMBO J
27:2746–2756
Mathews H, Clendennen SK, Caldwell CG, Liu XL, Connors K,
Matheis N, Schuster DK, Menasco DJ, Wagoner W, Lightner J,
Wagner DR (2003) Activation tagging in tomato identifies a tran-
scriptional regulator of anthocyanin biosynthesis, modification,
and transport. Plant Cell 15:1689–1703
Mizukami Y, Fischer RL (2000) Plant organ size control: AINTEGU-
MENTA regulates growth and cell numbers during organogen-
esis. Proc Natl Acad Sci USA 97:942–947
Morris WL, Ducreux LJ, Hedden P, Millam S, Taylor MA (2006)
Overexpression of a bacterial 1-deoxy-D-xylulose 5-phosphate
synthase gene in potato tubers perturbs the isoprenoid metabolic
network: implications for the control of the tuber life cycle. J Exp
Bot 57:3007–3018
Murashige T, Skoog F (1962) A revised medium for rapid growth and
bio-assays with tobacco tissue cultures. Physiol Plant 15:473–497
Nakano T, Fujisawa M, Shima Y, Ito Y (2013) Expression profiling of
tomato pre-abscission pedicels provides insights into abscission
zone properties including competence to respond to abscission
signals. BMC Plant Biol 13:40
Nakazawa M, Yabe N, Ichikawa T, Yamamoto YY, Yoshizumi T,
Hasunuma K, Matsui M (2001) DFL1, an auxin-responsive GH3
gene homologue, negatively regulates shoot cell elongation and
lateral root formation, and positively regulates the light response
of hypocotyl length. Plant J 25:213–221
Noh YS, Amasino RM (1999) Identification of a promoter region
responsible for the senescence-specific expression of SAG12.
Plant Mol Biol 41:181–194
Perez-Torrado R, Yamada D, Defossez PA (2006) Born to bind:
the BTB protein–protein interaction domain. BioEssays
28:1194–1202
Qin G, Gu H, Zhao Y, Ma Z, Shi G, Yang Y, Pichersky E, Chen H,
Liu M, Chen Z, Qu LJ (2005) An indole-3-acetic acid carboxyl
methyltransferase regulates Arabidopsis leaf development. Plant
Cell 17:2693–2704
Qu N, Gan W, Bi D, Xia S, Li X, Zhang Y (2010) Two BTB proteins
function redundantly as negative regulators of defense against
pathogens in Arabidopsis. Botany 88:953–960
Radouani A (2003) Effect of plant growth regulators, plantlet attrib-
utes and nutrient medium on in vivo and in vitro potato tuberiza-
tion. In: Agronomiques Science (ed) Rabat. Institut Agronomique
& Vétérinaire Hassan II, Morocco
Regan S, Gustafson V, Mallubhotla S, Chakravarty B, Bagchi M,
Siahbazi M, Rothwell C, Sardana R, Goyer C, Audy P, Li X-Q,
Wang-Pruski G, De Koeyer D, Flinn B (2006) Finding the perfect
potato: using functional genomics to improve disease resistance
and tuber quality traits. Can J Plant Path 28:S247–S255
Sadanandom A, Mesmar J, Yang C, Ewan R, Carr C, O’ Donnell E
(2008) The cell death regulator AtPUB17 directly interacts with
the BTB/POZ domain transcriptional repressor, AtBTB1 to con-
trol disease resistance in plants. Comp Biochem Physiol A Mo
Integr Physiol 150:S178–S179
Sambrook J, Russell DW (2001) Molecular cloning: a laboratory
manual. Cold Spring Harbour Laboratory Press, Cold Spring
Harbour
Santner A, Estelle M (2010) The ubiquitin-proteasome system regu-
lates plant hormone signaling. Plant J 61:1029–1040
Suttle J (2004) Physiological regulation of potato tuber dormancy. Am
J Potato Res 81:253–262
Suzuki Y, Uemura S, Saito Y, Murofushi N, Schmitz G, Theres K,
Yamaguchi I (2001) A novel transposon tagging element for
obtaining gain-of-function mutants based on a self-stabilizing Ac
derivative. Plant Mol Biol 45:123–131
Taylor JE, Whitelaw CA (2001) Signals in abscission. New Phytol
151:323–340
Tewari R, Bailes E, Bunting KA, Coates JC (2010) Armadillo-repeat
protein functions: questions for little creatures. Trends Cell Biol
20:470–481
The Potato Genome Sequencing Consortium (2011) Genome
sequence and analysis of the tuber crop potato. Nature
475:189–194
The Tomato Genome Consortium (2012) The tomato genome
sequence provides insights into fleshy fruit evolution. Nature
485:635–641
Thimm O, Bläsing O, Gibon Y, Nagel A, Meyer S, Krüger P, Selbig
J, Müller LA, Rhee SY, Stitt M (2004) MAPMAN: a user-driven
tool to display genomics data sets onto diagrams of metabolic
pathways and other biological processes. Plant J 37:914–939
Thomann A, Dieterle M, Genschik P (2005) Plant CULLIN-based
E3s: phytohormones come first. FEBS Lett 579:3239–3245
Tjaden J, Möhlmann T, Kampfenkel K, Henrichs G, Neuhaus HE
(1998) Altered plastidic ATP/ADP-transporter activity influences
potato (Solanum tuberosum L.) tuber morphology, yield and com-
position of tuber starch. Plant J 16:531–540
Tohge T, Nishiyama Y, Hirai MY, Yano M, Nakajima J, Awazuhara
M, Inoue E, Takahashi H, Goodenowe DB, Kitayama M, Noji M,
Yamazaki M, Saito K (2005) Functional genomics by integrated
analysis of metabolome and transcriptome of Arabidopsis plants
over-expressing an MYB transcription factor. Plant J 42:218–235
Trapnell C, Pachter L, Salzberg SL (2009) TopHat: discovering splice
junctions with RNA-Seq. Bioinformatics 25:1105–1111
Trapnell C, Williams BA, Pertea G, Mortazavi A, Kwan G, van Baren
MJ, Salzberg SL, Wold BJ, Pachter L (2010) Transcript assembly
and quantification by RNA-Seq reveals unannotated transcripts
and isoform switching during cell differentiation. Nat Biotechnol
28:511–515
Tucker ML, Whitelaw CA, Lyssenko NN, Nath P (2002) Functional
analysis of regulatory elements in the gene promoter for an
abscission-specific cellulase from bean and isolation, expression,
and binding affinity of three TGA-type basic leucine zipper tran-
scription factors. Plant Physiol 130:1487–1496
Usadel B, Poree F, Nagel A, Lohse M, Czedik-Eysenberg A, Stitt
M (2009) A guide to using MapMan to visualize and compare
Omics data in plants: a case study in the crop species, maize.
Plant Cell Environ 32:1211–1229
van der Knaap E, Kim JH, Kende H (2000) A novel gibberellin-
induced gene from rice and its potential regulatory role in stem
growth. Plant Physiol 122:695–704
van Eck HJ, Jacobs JM, Stam P, Ton J, Stiekema WJ, Jacobsen E
(1994) Multiple alleles for tuber shape in diploid potato detected
by qualitative and quantitative genetic analysis using RFLPs.
Genetics 137:303–309
Wan S, Wu J, Zhang Z, Sun X, Lv Y, Gao C, Ning Y, Ma J, Guo Y,
Zhang Q, Zheng X, Zhang C, Ma Z, Lu T (2009) Activation tag-
ging, an efficient tool for functional analysis of the rice genome.
Plant Mol Biol 69:69–80
Weigel D, Ahn JH, Blázquez MA, Borevitz JO, Christensen SK,
Fankhauser C, Ferrándiz C, Kardailsky I, Malancharuvil EJ, Neff
MM, Nguyen JT, Sato S, Wang ZY, Xia Y, Dixon RA, Harrison
MJ, Lamb CJ, Yanofsky MF, Chory J (2000) Activation tagging
in Arabidopsis. Plant Physiol 122:1003–1013
13
 Author's personal copy

658 Plant Mol Biol (2014) 84:635–658

Xu X, van Lammeren AA, Vermeer E, Vreugdenhil D (1998) The Zubko E, Adams CJ, Machaekova I, Malbeck J, Scollan C, Meyer P
role of gibberellin, abscisic acid, and sucrose in the regulation of (2002) Activation tagging identifies a gene from Petunia hybrida
potato tuber formation in vitro. Plant Physiol 117:575–584 responsible for the production of active cytokinins in plants. Plant
Yuasa S, Onizuka T, Shimoji K, Ohno Y, Hara M, Ogawa S, Fukuda J 29:797–808
K (2008) The Novel BTB/POZ protein, Rhobtb3, regulates mito-
chondrial function via the ubiquitin proteasome system. Circul
Res 103:E53–E54

13