tassel-less1 Encodes a Boron Channel Protein Required for
Inflorescence Development in Maize
Regular Paper
April Leonard1, Beth Holloway1,3, Mei Guo2, Mary Rupe2, GongXin Yu2, Mary Beatty2,
Gina Zastrow-Hayes2, Robert Meeley2, Victor Llaca1, Karlene Butler1, Tony Stefani2,
Jennifer Jaqueth1 and Bailin Li1,*
1
DuPont Pioneer, 200 Powder Mill Road, Wilmington, DE 19880, USA
DuPont Pioneer, 7300 NW 62nd Ave, Johnston, IA 50131, USA
3
Present address: USDA, ARS Honeybee Breeding, Genetics and Physiology Laboratory, Baton Rouge, LA 70820, USA.
*Corresponding author: E-mail, [email protected]; Fax, +302-695-2728.
(Received October 11, 2013; Accepted February 14, 2014)
2
tassel-less1 (tls1) is a classical maize (Zea mays) inflorescence
mutant. Homozygous mutant plants have no tassels or very
small tassels, and ear development is also impaired. Using a
positional cloning approach, ZmNIP3;1 (a NOD26-like intrinsic protein) was identified as the candidate gene for tls1. The
ZmNIP3;1 gene is completely deleted in the tls1 mutant
genome. Two Mutator-insertional TUSC alleles of ZmNIP3;1
exhibited tls1-like phenotypes, and allelism tests confirmed
that the tls1 gene encodes ZmNIP3;1. Transgenic plants with
an RNA interference (RNAi) construct to down-regulate
ZmNIP3;1 also showed tls1-like phenotypes, further demonstrating that TLS1 is ZmNIP3;1. Sequence analysis suggests
that ZmNIP3;1 is a boron channel protein. Foliar application
of boron could rescue the tls1 phenotypes and restore the
normal tassel and ear development. Gene expression analysis
indicated that in comparison with that of the wild type or tls1
plants treated with boron, the transition from the vegetative
to reproductive phase or the development of the floral meristem is impaired in the shoot apical meristem of the tls1
mutant plants. It is concluded that the tls1 mutant phenotypes are caused by impaired boron transport, and boron is
essential for inflorescence development in maize.
Keywords: Boron Boron channel protein Tassel Tasselless1 Zea mays ZmNIP3;1.
Abbreviations: BAC. bacterial artificial chromosome; bd1,
branched silkless1; bif2, barren inflorescence2; HMW, high molecular weight; MPSS, massively parallel signature sequencing;
NIP, NOD26-like intrinsic protein; RFP, red fluorescent protein;
RGII, rhamnogalacturonan II; RNAi, RNA interference; SAM,
shoot apical meristem; SNP, single nucleotide polymorphism;
tls1, tassel-less1; TUSC, Trait Utility System for Corn.
Introduction
It has long been known that the micronutrient boron is essential
for vascular plant growth and development, and increasing
evidence has shown that it is also essential, or at least beneficial,
for a variety of bacterial and animal species (Gauch and Dugger
1954, Dell and Huang 1997, Blevins and Lukaszewski 1998,
Goldbach and Wimmer 2007, Tanaka and Fujiwara 2008). This
is especially significant in agriculture as boron deficiencies have
been reported in at least 80 countries in >130 different crop
species (Shorrocks 1997). Symptoms of boron deficiency in
plants are far ranging and can be exhibited as a lack of apical
dominance, abnormal flower development and fruit/seed set,
brittle leaves and stunted plant growth (Blevins and
Lukaszewski 1998, Tanaka and Fujiwara 2008, Lordkaew et al.
2011). Boron toxicity is also an issue in some regions of the
world including South Australia, the Middle East, the southern
coast of Peru, etc. where marine evaporates and marine argillaceous sediment are prevalent (Nable et al. 1997).
The roles boron plays in plant development appear to be
varied and complex, as evidenced by the array of phenotypes
seen in deficient plants. One of the best understood roles of
boron is its involvement in the cross-linking of two rhamnogalacturonan II (RGII) monomers (O’Neill et al. 2004). RGII is a component of pectin, which is important in cell wall structure (O’Neill
et al. 2001, O’Neill et al. 2004). This role may therefore explain why
boron-deficient plants can have brittle leaves (Blevins and
Lukaszewski 1998). Boron has also been shown to be important
in plasma membranes. The occurrence of boron deficiency has
been associated with aberrant ion fluxes and altered membrane
potential resulting from decreased ATPase activity (Cakmak and
Römheld 1997). Although the molecular mechanism underlying
these observations is not well understood, it is possible that the
ability of boron to bind membrane compounds containing cis-diol
groups, such as glycoproteins and glycolipids, is involved in membrane structure and, therefore, its function (Cakmak and Römheld
1997). Boron has also been implicated in a number of other plant
processes including photosynthesis, metabolism, sugar translocation, nitrogen fixation and others (Gauch and Dugger 1953, Gauch
and Dugger 1954, Cakmak and Römheld 1997, Dell and Huang
1997, Blevins and Lukaszewski 1998).
Plant Cell Physiol. 55(6): 1044–1054 (2014) doi:10.1093/pcp/pcu036, available online at www.pcp.oxfordjournals.org
! The Author 2014. Published by Oxford University Press on behalf of Japanese Society of Plant Physiologists.
All rights reserved. For permissions, please email: [email protected]
1044
Plant Cell Physiol. 55(6): 1044–1054 (2014) doi:10.1093/pcp/pcu036 ! The Author 2014.
TLS1 encodes a boron channel protein
Perhaps the most important process in which boron plays a
role is plant reproduction. Reproductive development can be affected at several different stages as a result of boron deficiency,
including inflorescence formation, pollen development and germination, pollen tube growth, and seed and fruit set (Dell and
Huang 1997). In addition, the requirement of a plant for boron is
generally much higher at this stage than during vegetative growth
(Gauch and Dugger 1954). Consistent with this notion, a recent
study in maize (Zea mays) showed that plants grown in borondeficient conditions have depressed grain yield while there is no
significant difference in the dry weight of vegetative tissues
(Lordkaew et al. 2011). Interestingly, it has been demonstrated
that the application of boron increases yield in a number of different species including almond (Nyomora et al. 1997), apple
(Wójcik et al. 1999), alfalfa (Dordas 2006), sour cherry (Hanson
1991), olive (Perica et al. 2001) and soybean (Schon and Blevins
1990). Despite these important implications, very little is known
about the exact role boron plays in reproductive development.
Here, we report the characterization of the tassel-less1 (tls1)
mutant. tls1 is a classical maize inflorescence mutant that was
identified as early as 1926 (Woodworth 1926) and, more recently, characterized and mapped (Albertsen et al. 1993).
Through a map-based cloning approach, it was determined
that TLS1 encodes a NOD26-like intrinsic protein (NIP),
ZmNIP3;1. This protein most closely matches boron channel
proteins from Arabidopsis and rice. Treating tls1 plants with a
foliar boron spray was able to rescue the phenotype, confirming
that TLS1 is related to boron transport and demonstrating that
boron is essential for inflorescence development.
Results
tls1 plants are defective in reproductive and
vegetative growth
The tls1 mutant has been described previously (Albertsen et al.
1993). In contrast to wild-type plants which display full, welldeveloped tassels (Fig. 1A), tls1 mutants exhibit a variety of
phenotypes ranging from small tassels (Fig. 1B) to plants that
completely lack a tassel (Fig. 1C). Ear development is also impaired in tls1 mutants. Severe mutants can exhibit multiple husks
(Fig. 1D) containing very small ears or no ears at all (Fig. 1E). Less
severe tls1 mutants can develop more normal looking ears, albeit
smaller than those of the wild type (Fig. 1E). Although visually
indistinguishable at early stages of development, mature plant
leaves are also affected. The leaves of mutant plants generally
appear narrower, but are also thicker, stiffer and more brittle in
texture (Fig. 1F). Severe mutants are also normally shorter than
less severe mutants and wild-type plants (not shown).
Candidate gene identification and tls1 reference
allele characterization
A map-based cloning approach was used to identify the gene
responsible for the tls1 phenotype. Seventy-five F2 individuals
from tls1 Mo17 were phenotyped and genotyped with 15
single nucleotide polymorphism (SNP) markers across the
long arm of chromosome 1, which were chosen based on previous mapping results (Albertsen et al. 1993). The identified
flanking markers, PHM5484-22 (222,430,393 bp, B73Ref_v2)
and PHM10765-46 (225,871,091 bp, B73Ref_v2) were subsequently used to genotype approximately 3,000 F2 individuals
to identify recombinants (Fig. 2). The recombinant plants were
scored for the tls1 phenotype, self-pollinated, and approximately 175 ears were harvested. The phenotype of each recombinant was determined or confirmed by progeny test. With the
use of additional markers and the recombinant lines, the tls1
interval was delimited to a five overlapping bacterial artificial
chromosome (BAC) interval between markers c0375b06_10
and c0260e13_35 (Fig. 2). This region has a very low level of
polymorphism between tls1 and Mo17, making it difficult to
narrow down the interval any further. All 10 annotated genes
within this region were sequenced in tls1 and Mo17, and, one of
them, annotated as NOD26-like integral membrane protein/
aquaporin/ZmNIP3;1 (hereafter referred to as ZmNIP3;1), was
unamplifiable in mutant individuals. As no major variations
were found in the other genes, ZmNIP3;1 became the apparent
candidate gene for tls1.
Since ZmNIP3;1 was unable to be PCR amplified in mutant
individuals, this gene is either completely missing in tls1 or has
major variations between the tls1 and Mo17 alleles. In order to
understand the nature of the mutation, a BAC library was
generated from homozygous tls1 plants and BAC clones spanning the tls1 interval were sequenced. It was determined that
although much of the tls1 BAC sequence aligned to the B73
reference genome sequences and the flanking markers,
c0375b06_10 and c0260e13_35, were present in the tls1 BAC
sequence, there was a large syntenic break between the B73
reference and the tls1 mutant. This break included highly repetitive sequence that appeared to be unique to tls1 in comparison with the B73 reference, as well as reference sequence
that was absent in the tls1 mutant. Although the exact sizes of
the insertions and deletions could not be determined accurately due to the overall highly repetitive nature of the region, it
was clear that the missing reference sequence included the
ZmNIP3;1 candidate gene. Therefore, the ZmNIP3;1 gene is
completely deleted in tls1, which may cause the tls1 mutant
phenotype.
Validation of the tls1 candidate gene
To validate ZmNIP3;1 as the candidate gene for tls1, two Trait
Utility System for Corn (TUSC) lines (Meeley and Briggs 1995),
P177F10, containing a Mutator (Mu) insertion in the promoter
region of ZmNIP3;1, and P30D5, containing an insertion in exon
2, were identified (Fig. 3A). Individual plants which were homozygous for the Mu insertion exhibited the tls1 phenotype, with
no or small tassels. To confirm the candidate gene further, an
allelism test was conducted by crossing plants heterozygous for
the Mu insertion to heterozygous tls1 plants. The resulting
Plant Cell Physiol. 55(6): 1044–1054 (2014) doi:10.1093/pcp/pcu036 ! The Author 2014.
1045
A. Leonard et al.
Fig. 1 tls1 mutant characterization. (A) A wild type plant with a normal-size tassel. (B) A tls1 plant with a mild mutant phenotype exhibiting a
small tassel. (C) A tls1 plant with a severe tassel-less mutant phenotype. (D) Multiple empty husks (arrows) of a tls1 mutant plant with a severe
no-tassel (NT) phenotype. (E) Comparison of ears representative of a wild-type (WT) plant, a tls1 plant with a mild small tassel (ST) phenotype
and the range seen in tls1 plants with a severe NT phenotype. (F) A leaf from WT, ST and NT plants harvested from mature flowering plants.
Fig. 2 Map-based cloning of tls1. The tls1 BAC interval and the ZmNIP3;1 candidate gene are indicated.
progeny were phenotyped and analyzed by PCR for the presence of a Mu insertion in ZmNIP3;1 and for the presence of at
least one copy of the wild-type ZmNIP3;1 allele (Fig. 3B). It was
found that the vast majority of plants which contained one
copy of the TUSC allele and one copy of the tls1 reference
allele exhibited the tls1 phenotype, while the other plants
were wild type (Table 1; Fig. 3C), indicating that ZmNIP3;1
and tls1 are allelic. The small number of discrepancies was
attributed to misclassification as a result of the incomplete
1046
penetrance of the phenotype and the fact that the gene-specific
primers used to amplify the wild type ZmNIP3;1 allele do not
amplify efficiently in the TUSC line, P177F10. Overall though,
the results of the allelism test validate that ZmNIP3;1 is the
causative gene for the tls1 mutant phenotype.
As an additional validation of the ZmNIP3;1 candidate gene,
transgenic RNA interference (RNAi) plants were produced in
order to test whether knocking-down ZmNIP3;1 expression can
result in the tls1 phenotype. Since the most obvious
Plant Cell Physiol. 55(6): 1044–1054 (2014) doi:10.1093/pcp/pcu036 ! The Author 2014.
TLS1 encodes a boron channel protein
Fig. 3 Allelism test for candidate gene validation. (A) Locations of Mutator insertion sites in the two TUSC alleles. (B) Diagram of allelism test
strategy and expected results. (C) Phenotypes of a wild-type plant and a tls1 plant from the allelism test.
Table 1 Genotypes and phenotypes of progeny from crosses between the tls1 reference mutant allele and two TUSC alleles
P30D5
tls1
genotypea
P177F10
Wild-type
genotypeb
tls1
genotypea
Wild-type
genotypeb
tls1 phenotype
7
1
11
0
Wild-type
phenotype
0
24
5
21
a
Individuals containing one copy of the Mu-insertional TUSC allele and one
copy of tls1 reference allele of ZmNIP3;1 (no amplification of ZmNIP3;1).
b
Individuals containing at least one intact copy of ZmNIP3;1.
phenotypes in tls1 are in the tassel and ear, a constitutive promoter (maize ubiquitin gene promoter) driving transgene expression was expected to target these two tissues. Twenty
individual T0 transgenic plants were generated, which contained single or multiple copies of the transgene. All of these
transgenic events were evaluated for their tassel phenotypes.
Among these transgenic events, a spectrum of tassel sizes, ranging from tassel-less, to very small or nearly normal tassel sizes,
was observed (Fig. 4; Supplementary Table S2). Six transgenic
plants that were tassel-less or had very small tassels, had mostly
3–4 copies of the transgene, and tended to have a relatively
higher level of transgene expression. Transgenic plants that had
larger or near normal tassel growth mostly contained a single
copy of the transgene and had no or a very low level of transgene expression (Supplementary Table S2). The level of native
ZmNIP3;1 expression in the leaves was too low to determine
the efficacy of RNAi silencing of the endogenous gene (data not
shown). However, it is likely that higher transgene expression is
more effective in silencing endogenous gene expression, resulting in the tassel-less phenotype.
Fig. 4 Tassel-less phenotype expression of transgenic corn plants
down-regulating ZmNIP3;1 expression. Each plant is an independent
transgenic event at T0. Shown are examples of transgenic plants with
varying degrees of tassel size reduction, from tassel-less to near normal
size tassel.
Besides the tassel phenotype, ear growth of the transgenic
plants was also affected. The effect on ear growth was largely
consistent with the effects on tassel growth. That is, those
plants that had a normal tassel also produced a normal ear,
whereas those that were tassel-less or had severe tassel
Plant Cell Physiol. 55(6): 1044–1054 (2014) doi:10.1093/pcp/pcu036 ! The Author 2014.
1047
A. Leonard et al.
reduction also had no ear or impaired ear growth such as no silk
(Supplementary Table S2).
ZmNIP3;1 is involved in boron transport
Sequence analysis revealed that the ZmNIP3;1 protein in maize
is highly similar to AtNIP5;1 from Arabidopsis and OsNIP3;1
from rice (Fig. 5). All of these proteins are members of the
class II NOD26-like intrinsic protein (NIP) family (Liu et al.
2009). NIP II proteins can be permeable to formamide, boric
acid, glycerol, urea, arsenite and other large solutes (Wallace
and Roberts 2005, Bhattacharjee et al. 2008, Bienert et al. 2008,
Liu et al. 2009). Of particular interest, the closely related proteins, AtNIP5;1 and OsNIP3;1, have been implicated in the
transport of boron (Takano et al. 2006, Hanaoka and Fujiwara
2007). The root, leaf and flower development of the Arabidopsis
nip5;1 mutant is affected by boron limitation (Takano et al.
2006). It is possible that ZmNIP3;1 in maize may also be
involved in boron transport. Consistent with this notion, it
has been shown that plants grown in boron-deficient conditions exhibit a phenotype that is very reminiscent of tls1 mutants (Lordkaew et al. 2011).
To test the hypothesis that ZmNIP3;1 is involved in boron
transport and the tls1 mutant phenotype is due to impaired
transport of boron, wild-type and tls1 plants were treated with
a foliar boron spray from V2 to V6 and the resulting phenotypes were observed. Wild-type plants treated with the boron
spray exhibited no significant differences in comparison with
the untreated wild-type plants (data not shown). However, the
difference between the treated and untreated tls1 plants was
striking. Visually, untreated mutant plants had very small tassels
with short branches, or completely lacked a tassel (Fig. 6B),
while the mutant plants that were treated with boron had
full tassels with many long branches (Fig. 6C) and appeared
very similar to the wild type (Fig. 6A). In addition, the ears of
mutant plants that were treated with boron appeared similar in
size to wild-type ears (Fig. 6D). Quantitatively, although the
number of tassel branches was significantly different between
wild-type and the treated mutant plants, the mean number of
tassel branches in the treated mutant was much closer to the
wild-type level than that of the untreated mutant (Table 2;
Fig. 6E). The length of the tassel branches was actually significantly greater in the treated mutant plants than in the wild
type (Table 2; Fig. 6F), and there was no significant difference
in ear length between the sprayed mutant and the wild-type
plants (Table 2; Fig. 6G). These results show that boron application to tls1 plants can rescue the mutant phenotype, confirming that ZmNIP3;1 is involved in boron transport and the
tls1 phenotype is due to impaired transport of boron. Directly
or indirectly, boron plays a crucial role for inflorescence development in maize. Indeed, the soil boron level affects the expression of the tls1 mutant phenotypes, with low boron levels
associated with more severe tls1 phenotypes (A. Durbak and
P. McSteen, personal communication). Nevertheless, differences in soil boron levels cannot fully explain the variations
in tls1 phenotypes, as such variations in expression could be
observed even under uniform soil conditions in the greenhouse.
Fig. 5 Protein sequence alignment of ZmNIP3;1 with its most closely related proteins from rice (OsNIP3;1) and Arabidopsis (AtNIP5;1).
ZmNIP3;1 and OsNIP3;1, 88.8% similar and 84.3% identical. ZmNIP3;1 and AtNIP5;1, 76.6% similar and 67.3% identical.
1048
Plant Cell Physiol. 55(6): 1044–1054 (2014) doi:10.1093/pcp/pcu036 ! The Author 2014.
TLS1 encodes a boron channel protein
Fig. 6 Effects of boron application on tls1 mutant phenotypes. (A) Wild-type tassel. (B) A tassel of a tls1 plant not treated with boron. (C) A
tassel of a tls1 plant treated with boron. (D) Comparison of immature ears from a wild-type plant (WT) and a tls1 plant that was treated with
boron (tls1 +B). (E–G) Histograms with the distribution of tassel branch count (E), tassel branch length (F) and ear length (G) of wild-type (WT,
10 individuals), tls1 mutants (29 individuals) and tls1 mutants treated with boron (26 individuals).
Plant Cell Physiol. 55(6): 1044–1054 (2014) doi:10.1093/pcp/pcu036 ! The Author 2014.
1049
A. Leonard et al.
Table 2 Effect of foliar boron treatment of tls1 mutants on tassel branch number, tassel branch length and ear length
Wild type
tls1 + B
tls1
Branch number
Mean ± SE
tls1 + B (P-value)a
tls1 (P-value)a
13.20 ± 0.79
0.026
<0.001
10.62 ± 0.62
–
<0.001
2.38 ± 0.56
–
–
Branch length
Mean ± SE
tls1 + B (P-value)a
tls1 (P-value)a
12.73 ± 0.49
<0.001
<0.001
14.84 ± 0.32
–
<0.001
5.84 ± 0.34
–
–
Ear length
Mean ± SE
tls1 + B (P-value)a
tls1 (P-value)a
13.70 ± 0.98
0.708
<0.001
14.23 ± 0.78
–
<0.001
3.48 ± 0.45
–
–
a
P-value calculated using t-test function in Excel.
Therefore, tls1 represents an example of incomplete
penetrance.
To determine if the boron levels in the seeds from the
boron-treated tls1 plants could have a maternal effect on the
expression of the tls1 phenotype in later generations, the recovered tls1 mutant plants were self-pollinated and progeny
were grown out to observe their phenotypes. Half of the
plants were untreated and the other half were treated with
boron spray as described above. The untreated progeny exhibited the tls1 mutant phenotype, while the tassel phenotype
of the boron-treated progeny was similar to that of wild-type
plants (data not shown). This indicates that the recovery of the
phenotype is not heritable and boron application is necessary
at each generation to rescue the phenotype.
Expression analysis
To examine the temporal and spatial expression pattern of the
tls1 gene, ZmNIP3;1 was queried in a massively parallel signature sequence (MPSS) database consisting of libraries of 17 bp
sequence tags from cDNAs isolated from >200 diverse maize
tissues and developmental stages (Muszynski et al. 2006). While
ZmNIP3;1 is expressed in floral tissues, most notably in the silks,
it is also expressed in other tissue types throughout the plant,
albeit at a low level in most tissues (Fig. 7). This may indicate
that ZmNIP3;1 functions in multiple stages of plant development in addition to inflorescence development. This notion is
consistent with the observation that tls1 plants exhibit aberrations in both vegetative and reproductive growth (Fig. 1)
(Albertsen et al. 1993).
An RNA-seq experiment was conducted to identify genes, as
well as developmental and biochemical pathways affected by
the tls1 mutation. Leaf and the shoot apical meristem (SAM)
tissues were taken from V5 tls1 plants, wild-type siblings and
tls1 plants that had been treated with the foliar boron spray as
described previously. Analysis of the results from the leaf tissue
indicated that gene expression was only altered as a result of
boron spray treatment, i.e. no notable differences in expression
patterns between tls1 mutant plants and the wild-type sibling.
However, two interesting classes of genes were observed when
examining the results of gene expression in the SAM: genes that
1050
Fig. 7 Expression pattern of ZmNIP3;1. Relative ZmNIP3;1 expression
abundance in parts per 10 million in different maize tissues determined by the MPSS technology.
were highly expressed in the wild-type and the sprayed mutant
plants but expressed at a low level in the untreated tls1 plants
(Supplementary Fig. S1A) and genes that were strongly expressed in the tls1 mutant but expressed at a low level in wildtype and treated mutant plants (Supplementary Fig. S1B). A
total of 368 genes were expressed at higher levels in the SAM of
wild-type and boron-treated tls1 plants than in that of the
untreated tls1 mutant plants, with a maximal obtainable fold
change between any of the three genotypes (effect size) of
at least three (Supplementary Fig. S1A; Supplementary
Table S3). Interestingly, five of the genes within the top 10
effect sizes were found to be MADS-box genes involved in inflorescence development (Ambrose et al. 2000, Thompson et al.
2009, Ciaffi et al. 2011, Zhang et al. 2012). It was also found
that genes involved in auxin pathways were enriched in this
gene set, including barren inflorescence2 (bif2), which regulates
axillary meristem development in maize inflorescence
Plant Cell Physiol. 55(6): 1044–1054 (2014) doi:10.1093/pcp/pcu036 ! The Author 2014.
TLS1 encodes a boron channel protein
(McSteen and Hake 2001). In addition to other floral development genes, enriched biological processes as determined by the
Gene Ontology Fisher’s exact test include anatomical structure
morphogenesis, cell communication, gibberellin regulation and
biosynthesis, and polar auxin transport (data not shown). In
addition, 955 genes were expressed at lower levels in the SAM of
wild-type and boron-treated tls1 plants than in that of the
untreated tls1 mutant plants (Supplementary Fig. S1B;
Supplementary Table S4). Among these are genes involved
in photosynthesis, light responses, and carbon fixation and metabolism. This observation suggests that the transition from the
vegetative to reproductive phase or the development of the
floral meristem is impaired in the SAM of the tls1 mutant
plants.
Discussion
In the current study, we report the cloning of the maize mutant
tls1, and determined that the deletion of ZmNIP3;1, a putative
boron channel protein, in the tls1 genome is responsible for the
mutant phenotypes. Boron application could rescue the tls1
mutant phenotypes, demonstrating that boron is essential for
inflorescence development in maize. It has long been known
that boron is an essential micronutrient for plants (Warington
1923). In boron-limiting conditions, plants can exhibit a variety
of symptoms including stunted plant growth, brittle leaves and
shoots, impaired inflorescence development, and poor fruit and
seed set. Because of the number of different phenotypes displayed by boron-deprived plants, the exact role that boron
plays has been difficult to elucidate. Despite the uncertainties,
it is known for many diverse plant species that the boron requirement is much higher in reproductive stages (Gauch and
Dugger 1954, Shorrocks 1997, Blevins and Lukaszewski 1998,
Lordkaew et al. 2011). These notions were supported by the
current study. It is interesting to note that the expression analysis indicated that ZmNIP3;1 is expressed most strongly in the
silks (Fig. 7), which may be indicative of an exceptionally strong
requirement for boron in this tissue type. If, in other species,
analogous tissue types express a ZmNIP3;1 homolog most
strongly as well, this may explain why treating plants with
boron increases seed and fruit yield and quality (Schon and
Blevins 1990, Hanson 1991, Nyomora et al. 1997, Wójcik et al.
1999, Perica et al. 2001, Dordas 2006).
In Arabidopsis, two NIP II proteins have been identified
which have been implicated in boron transport. AtNIP5;1 is
important for boron uptake during boron-limiting conditions.
It could also facilitate the transport of water (Takano et al.
2006). AtNIP6;1 is also involved in boron transport but is impermeable to water (Tanaka et al. 2008). Interestingly, these
two boron transporters were found to be expressed in different
tissue types; AtNIP5;1 was expressed in roots while AtNIP6;1
was expressed predominantly in nodal regions of shoots, especially the phloem region of vascular tissues (Takano et al. 2006,
Tanaka et al. 2008). In the present study, it was found that
ZmNIP3;1 is involved in boron transport (Fig. 6) and is expressed in multiple tissue types (Fig. 7). These analyses, together with the lack of a close maize homolog of AtNIP6;1,
may suggest that ZmNIP3;1 is functionally equivalent to
AtNIP5;1 and AtNIP6;1 from Arabidopsis. Indeed, a phylogenetic analysis of NIPs in plants indicated that both AtNIP5;1 and
AtNIP6;1 are closely related to ZmNIP3;1 (Liu et al. 2009).
In this study, an RNA-seq experiment was conducted to identify genes and biological processes affected by the tls1 mutation
to help determine the role boron plays in inflorescence development. Rather than identifying one main pathway, many genes
involved in inflorescence development were induced in the SAM
of boron-treated tls1 plants as compared with the untreated
mutant (Supplementary Table S3). In addition to several
MADS-box genes involved in inflorescence development
(Ambrose et al. 2000, Thompson et al. 2009, Ciaffi et al. 2011,
Zhang et al. 2012), other important inflorescence genes were also
induced including: bif2, involved in auxin transport (Wu and
McSteen 2007); the photoperiod proteins, CONSTANS and
FLOWERING LOCUS T (Turck et al. 2008); and branched silkless1,
a transcription factor involved in spikelet meristem identity
(Chuck et al. 2002), among others. On the other hand, genes
involved in vegetative processes, such as photosynthesis, are expressed at higher levels in the SAM of untreated tls1 plants as
compared with those of boron-treated tls1 and wild-type plants
(Supplementary Table S4). It seems that the transition from
vegetative to reproductive phase or the development of the
floral meristem is impaired in the SAM of the tls1 mutant plants.
Although the exact nature of the involvement of boron in
inflorescence development cannot yet be determined, the current study may have important agricultural implications. In the
present study, it was shown that ZmNIP3;1 can be downregulated transgenically and that the level of RNAi transgene
expression may correlate with the severity of the tls1 phenotype
(Fig. 4; Supplementary Table S2). Down-regulation of
ZmNIP3;1 in a tissue-specific manner may result in plants that
lack tassels but have fewer pleiotropic effects associated with tls1
plants, specifically plants with normal ears. This tassel suppression would probably result in the reallocation to the ear of resources that would normally be needed for tassel development
(Duncan et al. 1967, Hunter et al. 1969). These plants would be
expected to perform better under normal and stress conditions,
such as nitrogen or water stress. They may also benefit from
reduced shading from tassels (i.e. increased rate of photosynthesis) normally experienced by plants (Duncan et al. 1967, Hunter
et al. 1969). In addition, since foliar boron applications could
rescue the tls1 mutant phenotypes, this scheme could be employed to maintain tls1 lines in a seed production setting.
Materials and Methods
Plant material and map-based cloning of tls1
The tls1 mutant was obtained from Marc Albertsen (Albertsen
et al. 1993). An F2 mapping population was generated by
Plant Cell Physiol. 55(6): 1044–1054 (2014) doi:10.1093/pcp/pcu036 ! The Author 2014.
1051
A. Leonard et al.
crossing homozygous tassel-less (tls1) plants to the maize
inbred line Mo17. Based on previous mapping results
(Albertsen et al. 1993), 75 F2 individuals were genotyped with
15 SNP markers distributed across the long arm of chromosome
1 to identify markers flanking tls1. The two flanking markers,
PHM5484-22 and PHM10765-46, were then used to screen approximately 3,000 F2 individuals to identify recombinants.
Recombinant individuals were self-pollinated whenever possible and approximately 175 F3 ears were harvested. The F3
families were grown and phenotyped to confirm the phenotype
of each F2 line. Tissues from single F2 plants or pools of eight F3
individuals from each family were used for DNA isolation and
genotyping. Additional markers were developed from the tls1
interval and fine mapping was carried out as described (Jiang
et al. 2012). All marker and primer information is listed in
Supplementary Table S1.
Allelism test for candidate gene validation
To validate the tls1 candidate gene, two lines with a Mu insertion in ZmNIP3;1 were identified from the TUSC population
(Meeley and Briggs 1995). Heterozygous TUSC individuals were
crossed to heterozygous tls1 individuals for an allelism test
(Fig. 3B). The progeny were classified as being either positive
or negative for a Mu insertion as well as being with or without
the wild-type ZmNIP3;1 allele. The presence of a Mu insertion
was determined by PCR using a Mu-specific primer in combination with a ZmNIP3;1-specific primer, and the presence of the
wild-type ZmNIP3;1 allele was determined by PCR using genespecific primers (Fig. 3B). Gene-specific primers to test for the
presence of the wild-type ZmNIP3;1 allele were designed such
that they did not amplify ZmNIP3;1 efficiently in the TUSC
background or a Mu insertion would prevent amplification.
All primers are listed in Supplementary Table S1.
BAC library construction and screening
Leaves from approximately 200 etiolated tls1 seedlings were harvested and fast-frozen. High molecular weight (HMW) DNA was
isolated as described (Luo and Wing 2003), with some modifications. The resulting agarose-embedded HMW DNA was partially
digested with HindIII and a BAC library was constructed on the
pCC1BAC vector, following the manufacturer’s recommendations (Epicentre). A total of 73,728 recombinant colonies were
picked in 384-well plates and gridded at high density in a
Hybond-N+ charged nylon membrane. Clones were screened
and identified by hybridization using immunological detection
of PCR digoxigenin (DIG)-labeled probes (Roche) according to
the manufacturer’s protocol. Positive BACs were retrieved and
confirmed by direct plus/minus colony PCR. Probes specific for
the tls1 region were labeled from PCR fragments amplified using
the primers listed in Supplementary Table S1.
BAC de novo sequencing
BAC DNA from positive clones was purified using the Qiagen
Large Construct system. Sequencing random shear sublibraries
1052
were constructed on pBluescript SK+ following standard procedures (Song et al. 2001). Plasmids were amplified directly
from 70 ml arrayed single-colony cultures using Templiphi (GE
Healthcare Life Sciences) and sequenced in an ABI 3730XL DNA
sequencer using ABI PRISM BigDye (Applied BioSystems). Base
calling, quality assessment, assembly and validation were performed using the phred, phrap and exgap software (Ewing and
Green 1998).
Phenotype rescue with boron spray
Wild-type and homozygous tls1 mutant individuals from the F2
mapping population were selfed to increase the seed, and the
resulting F3 progeny were used for this analysis. Half of the
individuals from each genotype were left untreated, while the
other half were treated once a week from V2 to V6 with a foliar
boron spray containing 1.2 g l 1 of a soluble boron powder
consisting of 62% B2O2 and 20.5% elemental boron (www.
jrjohnson.com, SKU: 19-0031J). The boron solution was applied
to the plants via a spray bottle until liquid was present on the
upper surface of all the leaves. Recovered mutant plants were
self-pollinated and seed was collected for a progeny test in
which half the plants were untreated and the other half were
treated with boron spray in the manner described above.
RNA-seq experiment and data analysis
Plants from the F2 mapping population of tls1 crossed to Mo17
were used for gene expression analysis. Twelve plants each of
the tls1 mutant, the wild type and the tls1 mutant treated with
boron spray were grown to the V5 stage. The treated tls1
mutant plants were sprayed from V2 until the time of sampling
as described above. Three tissue types were sampled: the
youngest fully expanded leaf, immature leaves (not fully expanded) and the SAM. At sampling time, for both leaf tissue
types, three individuals from each genotype/treatment type
were pooled together, resulting in four biological replicates.
For the SAM, all 12 individuals for each genotype/treatment
were sampled together in order to obtain enough tissue for
analysis. All tissue was sampled and immediately frozen on
dry ice. The tissue was then used for RNA-seq and further
analysis.
Total RNAs were isolated from frozen maize (Z. mays) leaf
and floral tissues by use of the Qiagen RNeasy kit for total RNA
isolation (Qiagen). Sequencing libraries from the resulting total
RNAs were prepared using the TruSeq mRNA-Seq kit and
protocol from Illumina, Inc. and sequenced on the Illumina
HiSeq 2000 system with Illumina TruSeq SBS v3 reagents. On
average, 10 million 50 bp sequences were generated for each
sample. The resulting sequences were trimmed based on quality scores and bowtie aligned (Langmead et al. 2009) to a
DuPont Pioneer proprietary maize gene set and normalized
to RPKtM (Mortazavi et al. 2008).
The generated RPKtM data matrix was visualized and analyzed in Genedata Analyst software. Specifically, two algorithms
in the Analyst were applied: effect size to identify differentially
Plant Cell Physiol. 55(6): 1044–1054 (2014) doi:10.1093/pcp/pcu036 ! The Author 2014.
TLS1 encodes a boron channel protein
expressed genes because of lacks of replicates, and K-mean
clustering to identify genes with specific expression patterns.
Function enrichment analysis was performed with Gene
Ontology Fisher’s exact test in the Analyst and Pathway
Studio (Ariadne Genomics, Inc.) to identify significantly overrepresented gene groups (biological processes or metabolic/
signal transduction pathways).
T-DNA constructs and maize transformation
T-DNA constructs and plant transformation Gateway technology (Invitrogen) were used for vector construction. An RNAi
vector was made to silence the ZmNIP3;1 expression. The vector
was designed such that a ZmNIP3;1/ADH1 intron/ZmNIP3;1
(reverse complementary) hairpin composed of 1,678 bp of the
ZmNIP3;1 transcript (the sequence is given in the
Supplementary data) was integrated between the maize ubiquitin promoter (ProZmUBI) and the Sorghum bicolor gamma
kafarin (GKAF) terminator and co-integrated with JT vectors
using Gateway technology as previously described (Unger et al.
2001, Cigan et al. 2005). The plasmid also contained a red
florescent protein (RFP) (Clonetech Laboratories), driven by
an aleuronic-specific promoter, barley (Hordeum vulgare) lipid
transfer protein, PromHvLTP2 (Opsahl-Sorteberg et al. 2004) to
serve as a marker. The construct ProZmUBI:ZmNIP3;1 RNAi
was introduced into Agrobacterium strain LBA4404 and used
to transform maize embryos of GS3xGaspe Flint from DuPont
Pioneer germplasm, a cross between the Hi-II and Gaspe Flint.
Independent transgenic events were generated, and the copy
number of transgene insertion was determined based upon the
DNA copy number of the RFP maker. For real-time PCR analysis
to measure transgene expression, a Taqman reverse transcription kit (Applied Biosystems) was used as described (Guo et al.
2010). The PCR primers used for transgene expression are listed
in Supplementary Table S1. Transgene expression levels of
ZmNIP3;1 were measured relative to the endogenous reference
eIF4g, a maize eukaryotic initiation factor gene (GenBank accession No. NP_568534) with the Ct method as described by
the manufacturer.
Supplementary data
Supplementary data are available at PCP online.
Funding
This work was
Biotechnology.
supported
by
DuPont
Agricultural
Acknowledgments
We thank Mike Muszynski, Simone Stahringer, Marianna Faller
and Nathan Uhlmann for providing technical support, and
Paula McSteen for helpful discussion.
Disclosures
The authors have no conflicts of interest to declare.
References
Albertsen, M., Trimnell, M. and Fox, T. (1993) Description and mapping of the tassel-less (TLS1) mutation. Maize Genet. Coop. Newslett
6: 51–52.
Ambrose, B.A., Lerner, D.R., Ciceri, P., Padilla, C.M., Yanofsky, M.F. and
Schmidt, R.J. (2000) Molecular and genetic analyses of the Silky1
gene reveal conservation in floral organ specification between eudicots and monocots. Mol. Cell 5: 569–579.
Bhattacharjee, H., Mukhopadhyay, R., Thiyagarajan, S. and Rosen, B.P.
(2008) Aquaglyceroporins: ancient channels for metalloids. J. Bio. 7: 33.
Bienert, G.P., Thorsen, M., Schussler, M.D., Nilsson, H.R., Wagner, A.,
Tamas, M.J. et al. (2008) A subgroup of plant aquaporins facilitate
the bi-directional diffusion of As(OH)3 and Sb(OH)3 across membranes. BMC Biol. 6: 26.
Blevins, D.G. and Lukaszewski, K.M. (1998) Boron in plant structure
and function. Annu. Rev. Plant Biol. 49: 481–500.
Cakmak, I. and Römheld, V. (1997) Boron deficiency-induced impairments of cellular functions in plants. Plant Soil 193: 71–83.
Chuck, G., Muszynski, M., Kellogg, E., Hake, S. and Schmidt, R.J. (2002)
The control of spikelet meristem identity by the branched silkless1
gene in maize. Science 298: 1238–1241.
Ciaffi, M., Paolacci, A.R., Tanzarella, O.A. and Porceddu, E. (2011)
Molecular aspects of flower development in grasses. Sex. Plant
Reprod. 24: 247–282.
Cigan, A.M., Unger-Wallace, E. and Haug-Collet, K. (2005)
Transcriptional gene silencing as a tool for uncovering gene function in maize. Plant J. 43: 929–940.
Dell, B. and Huang, L. (1997) Physiological response of plants to low
boron. Plant Soil 193: 103–120.
Dordas, C. (2006) Foliar boron application improves seed set, seed
yield, and seed quality of alfalfa. Agron. J. 98: 907.
Duncan, W.G., Williams, W.A. and Loomis, R.S. (1967) Tassels and the
productivity of maize. Crop Sci. 7: 37–39.
Ewing, B. and Green, P. (1998) Base-calling of automated sequencer
traces usingPhred. II. Error probabilities. Genome Res. 8: 186–194.
Gauch, H.G. and Dugger, W. Jr (1953) The role of boron in the translocation of sucrose. Plant Physiol. 28: 457.
Gauch, H.G. and Dugger, W.M. (1954) The Physiological Action of
Boron in Higher Plants: A Review and Interpretation. University
of Maryland, Agricultural Experiment Station.
Goldbach, H.E. and Wimmer, M.A. (2007) Boron in plants and animals:
is there a role beyond cell-wall structure?. J. Plant Nutr. Soil Sci. 170:
39–48.
Guo, M., Rupe, M.A., Dieter, J.A., Zou, J., Spielbauer, D., Duncan, K.E.
et al. (2010) Cell Number Regulator1 affects plant and organ size in
maize: implications for crop yield enhancement and heterosis. Plant
Cell 22: 1057–1073.
Hanaoka, H. and Fujiwara, T. (2007) Channel-mediated boron transport in rice. Plant Cell Physiol. 48: S227.
Hanson, E.J. (1991) Sour cherry trees respond to foliar boron applications. HortScience 26: 1142–1145.
Hunter, R., Daynard, T., Hume, D., Tanner, J., Curtis, J. and
Kannenberg, L. (1969) Effect of tassel removal on grain yield of
corn (Zea mays L.). Crop Sci. 9: 405–406.
Plant Cell Physiol. 55(6): 1044–1054 (2014) doi:10.1093/pcp/pcu036 ! The Author 2014.
1053
A. Leonard et al.
Jiang, F., Guo, M., Yang, F., Duncan, K., Jackson, D., Rafalski, A. et al.
(2012) Mutations in an AP2 transcription factor-like gene
affect internode length and leaf shape in maize. PLoS One 7:
e37040.
Langmead, B., Trapnell, C., Pop, M. and Salzberg, S.L. (2009) Ultrafast
and memory-efficient alignment of short DNA sequences to the
human genome. Genome Biol. 10: R25.
Liu, Q., Wang, H., Zhang, Z., Wu, J., Feng, Y. and Zhu, Z. (2009)
Divergence in function and expression of the NOD26-like intrinsic
proteins in plants. BMC Genomics 10: 313.
Lordkaew, S., Dell, B., Jamjod, S. and Rerkasem, B. (2011) Boron deficiency in maize. Plant Soil 342: 207–220.
Luo, M. and Wing, R.A. (2003) An improved method for plant BAC
library construction. Methods Mol. Biol. 236: 3–20.
McSteen, P. and Hake, S. (2001) barren inflorescence2 regulates axillary
meristem development in the maize inflorescence. Development
128: 2881–2891.
Meeley, B. and Briggs, S. (1995) Reverse genetics for maize. Maize
Genet. Coop. Newslett. 69: 67–67.
Mortazavi, A., Williams, B.A., McCue, K., Schaeffer, L. and Wold, B.
(2008) Mapping and quantifying mammalian transcriptomes by
RNA-Seq. Nat. Methods 5: 621–628.
Muszynski, M.G., Dam, T., Li, B., Shirbroun, D.M., Hou, Z.,
Bruggemann, E. et al. (2006) delayed flowering1 encodes a basic
leucine zipper protein that mediates floral inductive signals at the
shoot apex in maize. Plant Physiol. 142: 1523–1536.
Nable, R., Bañuelos, G. and Paull, J. (1997) Boron toxicity. Plant Soil 193:
181–198.
Nyomora, A.M.S., Brown, P.H. and Freeman, M. (1997) Fall foliarapplied boron increases tissue boron concentration and nut set
of almond. J. Amer. Soc. Hort. Sci. 122: 405–410.
O’Neill, M.A., Eberhard, S., Albersheim, P. and Darvill, A.G. (2001)
Requirement of borate cross-linking of cell wall rhamnogalacturonan II for Arabidopsis growth. Science 294: 846–849.
O’Neill, M.A., Ishii, T., Albersheim, P. and Darvill, A.G. (2004)
Rhamnogalacturonan II: structure and function of a borate crosslinked cell wall pectic polysaccharide. Annu. Rev. Plant Biol. 55:
109–139.
Opsahl-Sorteberg, H.G., Divon, H.H., Nielsen, P.S., Kalla, R., HammondKosack, M., Shimamoto, K. et al. (2004) Identification of a 49-bp
fragment of the HvLTP2 promoter directing aleurone cell specific
expression. Gene 341: 49–58.
Perica, S., Brown, P.H., Connell, J.H., Nyomora, A.M.S., Dordas, C., Hu, H.
et al. (2001) Foliar boron application improves flower fertility and
fruit set of olive. HortScience 36: 714–716.
1054
Schon, M.K. and Blevins, D.G. (1990) Foliar boron applications increase
the final number of branches and pods on branches of field-grown
soybeans. Plant Physiol. 92: 602–607.
Shorrocks, V. (1997) The occurrence and correction of boron deficiency. Plant Soil 193: 121–148.
Song, R., Llaca, V., Linton, E. and Messing, J. (2001) Sequence, regulation, and evolution of the maize 22-kD alpha zein gene family.
Genome Res. 11: 1817–1825.
Takano, J., Wada, M., Ludewig, U., Schaaf, G., von Wiren, N. and
Fujiwara, T. (2006) The Arabidopsis major intrinsic protein NIP5;1
is essential for efficient boron uptake and plant development under
boron limitation. Plant Cell 18: 1498–1509.
Tanaka, M. and Fujiwara, T. (2008) Physiological roles and transport
mechanisms of boron: perspectives from plants. Pflugers Arch. 456:
671–677.
Tanaka, M., Wallace, I.S., Takano, J., Roberts, D.M. and Fujiwara, T.
(2008) NIP6;1 is a boric acid channel for preferential transport of
boron to growing shoot tissues in Arabidopsis. Plant Cell 20:
2860–2875.
Thompson, B.E., Bartling, L., Whipple, C., Hall, D.H., Sakai, H.,
Schmidt, R. et al. (2009) Bearded-ear encodes a MADS box transcription factor critical for maize floral development. Plant Cell 21:
2578–2590.
Turck, F., Fornara, F. and Coupland, G. (2008) Regulation and identity
of florigen: FLOWERING LOCUS T moves center stage. Annu. Rev.
Plant Biol. 59: 573–594.
Unger, E., Betz, S., Xu, R.-j. and Cigan, A.M. (2001) Selection and orientation of adjacent genes influences DAM-mediated male sterility in
transformed maize. Transgenic Res. 10: 409–422.
Wallace, I.S. and Roberts, D.M. (2005) Distinct transport selectivity
of two structural subclasses of the nodulin-like intrinsic protein
family of plant aquaglyceroporin channels. Biochemistry 44:
16826–16834.
Warington, K. (1923) The effect of boric acid and borax on the broad
bean and certain other plants. Ann. Bot. 37: 629–672.
Wójcik, P., Cieslinski, G. and Mika, A. (1999) Apple yield and fruit quality
as influenced by boron applications. J. Plant Nutr. 22: 1365–1377.
Woodworth, C. (1926) Heritable characters of maize. J. Hered. 17:
405–412.
Wu, X. and McSteen, P. (2007) The role of auxin transport during
inflorescence development in maize (Zea mays, Poaceae).
Am. J. Bot. 94: 1745–1755.
Zhang, Z., Li, H., Zhang, D., Liu, Y., Fu, J., Shi, Y. et al. (2012)
Characterization and expression analysis of six MADS-box genes
in maize (Zea mays L.). J. Plant Physiol. 169: 797–806.
Plant Cell Physiol. 55(6): 1044–1054 (2014) doi:10.1093/pcp/pcu036 ! The Author 2014.