Universal florigenic signals triggered by FT homologues
regulate growth and flowering cycles in perennial
Eliezer Lifschitz1,* and Yuval Eshed2
1Department of Biology, Technion I.I.T. Haifa, 32000, Israel
2Department of Plant Sciences, Weizmann Institute of Science, Rehovot 76100, Israel
Received 23 March 2006; Accepted 5 July 2006
The transition from vegetative to floral meristems in
higher plants is programmed by the coincidence of
internal and environmental signals. Classic grafting
experiments have shown that leaves, in response to
changing photoperiods, emit systemic signals, dubbed
‘florigen’, which induce flowering at the shoot apex.
The florigen paradigm was conceived in photoperiod-
sensitive plants: nevertheless it implies that although
activated by different stimuli in different flowering
systems, the signal is common to all plants. Tomato
is a day-neutral, perennial plant, with sympodial and
modular organization of its shoots and thus with
reiterative regular vegetative/reproductive transitions.
SINGLE FLOWER TRUSS a regulator of flowering-time
and shoot architecture encodes the tomato orthologue
of FT, a major flowering integrator gene in Arabidopsis.
SFT generates graft-transmissible signals which com-
plement the morphogenetic defects in sft plants, sub-
stitute for light dose stimulus in tomato and for
contrasting day-length requirements in Arabidopsis
and MARYLAND MAMMOTH tobacco. It is discussed
how systemic signals initiated by SFT interact with
the SELF PRUNING gene to regulate vegetative to
reproductive (V/R) transitions in the context of two
flowering systems, one for primary apices and the
other for sympodial shoots.
Key words: Florigen, sympodial programme, tomato flowering,
Considerations of floral termination and
Growth habit in plants is described by two fundamental
models, monopodial and sympodial. Both growth habits are
found in families of the most primitive plants, such as
liverworts, mosses, and cycads (Bell, 1992). In annuals
with a simple monopodial shoot, a single vegetative phase
is replaced by a single reproductive phase, signalling the
completion of the life cycle. However, growth habits of
many other plants, particularly perennial trees, require the
cohabitation of vegetative and floral buds along their shoots
and thus the constant regulation of vegetative–reproductive
transitions at the whole-plant level.
Classical grafting experiments have shown that leaves,
in response to changing photoperiods, emit systemic sig-
nals, dubbed ‘florigen’, which induce flowering in vegeta-
tive shoot apical meristems (SAMs) (Chailakhyan, 1936;
Zeevaart, 1976). Genetic studies in Arabidopsis and rice
defined major pathways that transduce environmental sig-
nals to integrators, predominantly FLOWERING LOCUS T
(FT), which induce flowering in long-day and short-day
plants (reviewed in Mouradov et al., 2002; Boss et al.,
2004). However, the genetic components of ‘florigen’ and
their link to characterized flowering pathways remain
Tomato plants are photoperiod-insensitive perennials in
their native habitat, and exhibit perennial characteristics of
growth, even during one short seasonal cycle. The devel-
opmental versatility and architectural flexibility of tomato
are reflected in a plethora of gene mutations, affecting
single growth modules such as the primary shoot, or the
whole plant constitution. Several examples are shown
in Fig. 2 and Table 1. Advantage has been taken of the
* To whom correspondence should be addressed. E-mail: email@example.com
ª The Author . Published by Oxford University Press [on behalf of the Society for Experimental Biology]. All rights reserved.
For Permissions, please e-mail: firstname.lastname@example.org
Journal of Experimental Botany, Vol. 57, No. 13, pp. 3405–3414, 2006
Major Themes in Flowering Research Special Issue
doi:10.1093/jxb/erl106Advance Access publication 27 September, 2006
by guest on June 2, 2013
complex, but otherwise regular and predictable, develop-
mental pattern of the tomato shoot to investigate how the
meristems’ response to cycling vegetative and reproduc-
tive messages is regulated by specific flowering genes. A
broader perspective, however, is provided by the model
Arabidopsis in which extensive analyses of the shoot
constituents have been carried out in the context of an
annual monopodial shoot.
Growth and termination/flowering cycles in the
compound shoot of tomato
The primary shoot of wild-type tomato is terminated by the
first inflorescence (Fig. 1A, primary termination), after
8–12 leaves. Although the time to primary termination is
determined genetically, it is also responsive to environ-
mental conditions, particularly total daily light integrals
(Atherton and Harris, 1986). Subsequently, the apparent
main shoot consists of an upright array of reiterated lateral
branches called sympodial units (SUs), each with three
vegetative nodes and a terminal inflorescence. The first SU
(Fig. 1B, SU1) is subtended by the leaf just below the first
terminal inflorescence. It unites with the basal part of the
host leaf (HL), and due to vigorous growth extends the leaf
above the inflorescence and displaces the inflorescence
sideways. All subsequent SUs are also laterals, each arising
from the most proximal (third) axillary bud of the preced-
ing unit (Fig. 1B). The continuous aerial growth of the
wild-type tomato is therefore carried out by compound
shoots made up of SUs, whose inflorescences grow to
be positioned between the second and the third leaves of
the SUs which they terminate (Fig. 1C).
Further architectural elaboration is achieved by lateral
shoots released sequentially from the axils of the two more
distal mature leaves of each SU (Fig. 1C, LS). Such laterals
do not unite with the bases of their subtending leaves, and
thus appear as genuine axillary branches. Most signifi-
cant, unlike the proximal laterals which are fated to form
SUs and terminate regularly after three leaves, the distal
laterals, just like the primary shoots, are first terminated
after a variable number of leaves before converting to
the robust three-leaf sympodial pattern (Fig. 1C).
Floral meristems are always determinate, but in tomato,
in contrast to Arabidopsis, the sympodial habit dictates that
transition to flowering is synonymous with termination of
the vegetative apical meristem itself. Termination is used
here to describe a regulated developmental event in which
the vegetative growth of the SAM is replaced by a type of
growth committed to form a terminal differentiated organ
such as a flower, thorn or leaf. It is distinct, for example,
from a temporary arrest of a primordial axillary bud.
The inflorescence shoots and leaves of the tomato
plant are also assembled from modular, determinate,
morphogenetic units. The tomato inflorescence can be
looked upon as a condensed compound shoot, consisting
of one-nodal sympodial units (see also Cronquist, 1988;
Table 1. Tomato mutants impaired in general or specific components of the compound shoot
Mutant or variant
No. of extra before
terminating flower (tmf) None First with single
flower, later, normal
Normal but abnormal
3 UnknownHareven et al., 1994
leafless (lfs) –10No UnknownMenda et al., 2004
self pruning (sp)None Normal, but epistatic to
from 3 to 0
CETS Pnueli et al., 1998
1–25–15 Unknown Present report
2–4Branched with bracts
2 Present report
compound inflorescence (s)
Quinet et al., 2006
Verbalov et al., 2002
macrocalyx (mc) 0–1 Large sepals, vegetative
Primary, sympodial and inflorescence shoots
1–23 UnknownPnueli et al., 1998
1–5 Leafy, excessively
3–4 LEAFY Molinero-Rosales
et al., 2004
Dielen et al., 2004
Lifschitz et al., 2006
Schmitz et al., 2002
single flower truss (sft)
aNumber of leaves relative to wild-type siblings or common tomato lines.
bIn these cases, sympodial buds are inhibited, and sympodial shoots might acquire a basal lateral shoot programme.
3406 Lifschitz and Eshed
by guest on June 2, 2013
Szymkowiak and Irish, 2005). Each SU is comprised of
a single modified leaf (bract) and is terminated by a single
flower. Subsequently, a new lateral of a single flower shoot
arises at the axil of the subtending bract. While bracts are
not visible in the cultivated tomato, in the sibling species
Solanum pennellii they are clearly associated with each
floral SU of the inflorescence (Fig. 1D). The potential of the
inflorescence to form sympodial shoots in which single
flowers are separated, or replaced by leaves, is manifested
in several mutants such as falsiflora, jointless, blind, single
flower truss, uniflora, or macrocalyx (Table 1; Szymko-
wiak and Irish, 2005). Thus, these mutants can be exploited
to study the consequences of altered vegetative/reproduc-
tive (V/R) balance in the inflorescence shoot as well,
providing an additional developmental context for the
evaluation of ‘floral’ genes. Since termination is synony-
mous with flowering, and both the compound shoots and
the inflorescence shoots are sympodial, gene mutations that
affect flowering time or sympodial patterns also determine
shoot and inflorescence architecture. Likewise, the com-
pound tomato leaf is composed of a terminal leaflet and
3–5 pairs of primary lateral leaflets which develop in a
basipetal order along the rachis. Leaflets are capable of
forming second-and third-order duplications of the basic
pattern late into maturation, and secondary intercalary late-
rals (foliols) may develop in the fully expanded leaf (Fig.
1E; Hareven et al., 1996). The compound leaf, like the
compound shoot, is therefore a chimera of mature and
developing leafy organs.
The most important decisions with respect to the
evolving architecture of the compound tomato shoots relate
to the interplay between three apical elements: the third leaf
and the terminal inflorescence of a given SU, and the
auxillary bud destined to form the next sympodial unit.
These are referred to, collectively, as the sympodial fork.
(Fig. 1A). Modifications in the developmental balance
among the three elements will disrupt the regularity of the
sympodial pattern and will change the shoot architecture.
The most prominent known genes regulating the tomato
architecture by their primary effect on the elements of the
sympodial fork are SINGLE FLOWER TRUSS (SFT, Kerr,
1982; Molinero-Rosales et al., 2004) and the SELF
PRUNING (SP, Yeager 1927; Pnueli et al., 1998). The
SP gene, a homologue of TFL1 and CEN (Pnueli et al.,
1998), promotes the indeterminate state of the apical
Fig. 1. The compound tomato shoot. (A–C) The gradual development of
the tomato compound shoot. (A) Primary termination of the shoot apical
meristem by an inflorescence (inf) is accompanied by immediate release
of apical dominance over the axillary bud of the uppermost host leaf
(HL). This axillary will give rise to the sympodial bud (SB). These three
elements, inf, HL, and SB, comprise the sympodial fork. (B, C) The rapid
development of the sympodial unit (SU1, circled) positions the host leaf
above the inflorescence which is pushed to the side. Basal axillaries later
develop into ‘juvenile’ lateral shoots (LS) that do not displace their host
leaf and terminate after a variable number of leaves before resuming
a regular sympodial habit. (D) The compound inflorescence shoot of
tomato is comprised of condensed sympodial units of a single leaf (bract)
and a terminal flower. The single leaf is usually absent in wild-type
tomato (left) but present in the sibling species, S. pennellii. Modifications
of the basic pattern are evident by the abrupt termination of the
inflorescence shoot in FIL?FT or elaborate branching in compound
inflorescence (s) backgrounds. (E) The compound tomato leaf (left) can
be much simpler, as in FIL?FT, or complex, as in 35S:KN1, depending
on various growth and termination signals.
Florigenic signals shape the compound tomato shoot 3407
by guest on June 2, 2013
meristems, whereas SFT, a homologue of FT, promotes
termination/flowering by triggering a signalling pathway
that abides by all the tenets of the florigen paradigm
(Lifschitz et al., 2006).
The tomato FT gene, SINGLE FLOWER TRUSS,
induces early primary flowering in day-neutral
tomato and tobacco
Orthologues of the FT gene accelerate flowering in the
long-day Arabidopsis and the short-day rice. In both cases
their effect on flowering time is mediated by CONSTANS
orthologues (Hayama and Coupland, 2004). It has been
shown that constitutive expression of the AtCO gene, or
of two tomato CO homologues, fail to enhance flowering-
time in day-neutral tomato and tobacco (Ben-Naim et al.,
2006). Overexpression of AtCO in potato caused a graft-
transmissible reduction in the response of tuberization to
photoperiod, but had no effect on flowering time (Martinez-
Garcia et al., 2002). Yet, overexpression of SP conditions
late-flowering in tomato, tobacco, and Arabidopsis (Pnueli
et al., 1998, 2001). The Arabidopsis FT gene was ex-
pressed in tomato and tobacco and it was found that it
induces extremely early flowering in both day-neutral
species. A putative orthologue of FT, formerly identified
as SP3D (Carmel-Goren et al., 2003), also induced extreme
precocious flowering in day-neutral tomato and tobacco
(Fig. 2A). Sequence analysis of four mutant alleles, and
complementation of sft by the constitutive expression of
SP3D, identified the tomato FT orthologue as represented
by the SFT gene, a late-flowering morphogenetic gene
(Lifschitz et al., 2006).
The primary shoots of sft plants produce an inflorescence
after 15–20 leaves, as compared to the 8–12 leaves in their
wild-type siblings. The first organ in the terminal in-
florescence is usually a flower with an enlarged adaxial
sepal but the inflorescence is indeterminate, bearing leaves
instead of flowers (Molinero-Rosales et al., 2004). Unlike
in the wild-type tomato, this vegetative, terminal inflores-
cence (VI) of sft plants exerts partial apical dominance over
the presumptive sympodial bud, thus maintaining a pole
position (Fig. 2D, E). Moreover, due to its vigorous growth,
and the release of internal axillary shoots, the VI itself
becomes the main shoot. Occasionally, the VI shoot is
again terminated by a new VI, bearing one or two flowers.
Significantly, in a large-scale mutant hunt (Menda et al.,
2004), the three most extreme late-flowering mutant lines
were sft alleles. The sft mutation therefore represents a shift
in the vegetative/reproductive equilibrium within the
sympodial fork in favour of the vegetative state.
In independent lines of wild-type tomato plants express-
ing the 35S:SFT transgene, the first inflorescence arose
after three to five leaves, compared with 10–12 leaves of
their siblings (Lifschitz et al., 2006). Premature flowering
induced by SFT was associated with modifications in the
sympodial pattern. In lines where termination occurred
after three leaves, the prospective sympodial bud was
temporarily arrested and an inflorescence, with a sig-
nificantly reduced number of flowers, maintained a pole
position (Fig. 2A, B). When the proximal axillary
eventually emerged, it did not unite with the host leaf
petiole. However, subsequent SUs maintained the regular
three-nodal sympodial size.
Overexpression of SFT in the background of several
mutants which differed in the nature of their terminal
inflorescence,showed that allflowered asearly as wild-type
transgenic plants, but unlike sft plants, the developmental
fate of the terminating organs was not affected (illustrated
Fig. 2. Tomato shoots with altered SP-to-SFT ratios. (A) Extreme early flowering after three leaves (numbered), short internodes, small and less
complex leaves in 35S:SFT plants. In homozygous lines (inset), primary termination is accompanied by transient arrest of the sympodial bud release, as
illustrated in (B). (C) Early flowering stimulated by SFT is independent of FALSIFLORA (LEAFY orthologue) activity or floral identity. (D, E) Late
flowering, mixed floral and vegetative indeterminate inflorescence (VI) and delayed release of the prospective sympodial unit (PSU) in sft mutants. (F, G)
Gradual reduction in length of sympodial units results in the formation of consecutive inflorescences (inf) and arrest of vegetative growth conditioned
by a mutation in the SP gene.
3408 Lifschitz and Eshed
by guest on June 2, 2013
for falsiflora in Fig. 2C). Thus, the role of SFT is to confer
termination of growth and initiation of a terminal inflo-
rescence, but not to determine the identity of the ensuing
In Arabidopsis and tobacco, transitions to flowering are
associated with elongation of the inflorescence shoots, i.e.
bolting. Acceleration of flowering, by SFT or other genes,
like sp, in tomato is associated with shortening of the in-
ternodes. In additionto shorter internodes, there was overall
growth attenuation: smaller leaves, sometimes with a re-
duced number of lateral leaflets, much thinner stems, and
a faster growth rate. The seasonal life-cycle of 35S:SFT
transgenic tomato plants, may be completed within 9–10
Floral-promoting SFT signals are graft-
transmissible and complement all
developmental defects of sft mutant plants
Several indications have implied that the Arabidopsis FT
gene provides a genetic link between the systemic and the
ling CETS factor, it is not expressed in the SAM proper but
can be detected, upon induction, in shoot apices (SAP)
containing young leaves. Flowering is delayed in ft mutant
plants and when FT is over-expressed, flowering occurs
earlier with a determinate inflorescence (reviewed by Jack,
2004). FT is regulated by CONSTANS in both long and
short-day plants and grafting experiments in Arabidopsis
have shown that systemic induction of flowering by
CONSTANS is most likely mediated by FT (An et al.,
2004). Huang et al. (2005) recently showed that heat-
induction of FT in a single leaf is sufficient to promote
flowering, and that a fraction of the heat-induced FT
RNA is found in SAPs, suggesting that the FT mRNA
itself may be a florigenic agent.
The universality of the florigen paradigm has been de-
monstrated by interspecies grafting experiments (Zeevaart,
1976). Grafting results are independent of the validity of
promoters, the resolution of in situ hybridization, infer-
ences derived from the activation of upstream genes or
interpretations of clonal analysis. Due to the ease of
grafting, the photoperiod-independent flowering, the com-
pound shoot, and the perennial habit, the premise has been
tested that orthologues of the FT gene trigger a universal
florigenic signal in tomato.
In all reciprocal grafts between sft receptor and 35S:SFT
donor plants, sft receptor shoots produced normal flowers,
normal inflorescences, and normal sympodial architec-
ture, 3–5 weeks after grafting (Lifschitz et al., 2006).
Thus, flowering signals initiated by the SFT gene rescue
flowering-time and morphogenetic defects in sft mutant
plants by both endogenous expression and graft trans-
mission. The rescue of receptor sft in grafts required the
persistent emission of systemic SFT signals: formation
of normal SUs, inflorescences and flowers in receptor sft
shoots continued only as long as the 35S:SFT donor was
present. Complementation of sft by graft, using 35S:SFT
donor shoots, suggests that transcriptional auto-regulation
is not an obligatory component of the systemic regulation
of flowering by SFT (Lifschitz et al., 2006).
In contrast to 35S:SFT, wild-type donors failed to
complement the sft phenotype. This failure was attributed
to the endogenous consumption of scarce SFT signals,
coupled with a low efficiency of transmission. Floral sup-
pressors, autonomous or systemic, which like SP (see
later) balance flowering promoting signals, and regulators
of transmission per se, may also play a role. The rescue of
flowering, under short days, by wild-type donors may be
difficult in Arabidopsis as well, but in species where
efficient auto-regulation or relay-amplification mechanisms
evolved, uninduced wild-type donors may work.
SFT generates universal florigenic signals
Variations in responses to florigenic stimulus, presumably
due to different components of a mechanism shared by all
plants, or to independent ‘florigens’ operating differentially
in different plants (Bernier, 1988), have been recorded. FT
orthologues induce, via endogenous expression, flowering
in mono and dicotyledonous plants, in long-day, short-day,
and day-neutral plants and the SFT gene triggers systemic
signals that rescue its own mutant phenotype in day-length
neutral tomato. Therefore, it was asked whether the sys-
temic rescue of sft represents a variable component of the
florigen mechanism, or one of its core conserved elements.
To answer this, the potential of a systemic, not merely an
endogenous, SFT signal, to substitute for light require-
light intensity-sensitive tomato background, was explored.
High irradiance requirements in tomato
In a given genetic constitution, the number of leaves to the
first inflorescence depends primarily on the total daily light
integrals during early growth, less on light-intensities
and practically not at all on day-length (Kinet, 1977, for
a comprehensive analysis, and reviews by Wittwer and
Aung, 1969; Picken et al., 1985). In cultivars such as
Money Maker, M82, or VFNT-cherry used in our ex-
periments, the number of leaves to primary termination,
under constant low or high daily light integrals, may vary
between 6 and 16 and the number of leaves per SU between
three and six. An extreme case of light-dose-dependent
flowering is conditioned by a recessive mutation in the
UNIFLORA (UF)gene. No othertomato gene displayssuch
a differential sensitivity to the light doses (Dielen et al.,
2004). Under given conditions (70–150 lmol?2s?1at
18–24 8C) one-third of the uf plants never flower, while the
rest formed the first, sometimes only flower, after 30–50
Florigenic signals shape the compound tomato shoot3409
by guest on June 2, 2013
leaves. Thus, uf plants allow the study of ‘flowering time’
in a virtually all-or-none situation. The single uf flower is
borne by a vegetative inflorescence shoot (termed pseudo-
shoot, PS, in Lifschitz et al., 2006) which is positioned
like a regular inflorescence, but in which subsequent
flowers are replaced by leaves. Strikingly, during the Israeli
summer, uf plants growing in an open field are barely dis-
tinguishable from their wild-type progenitors.
Under light conditions set to suppress flowering, uf
35S:SFT plants produced flowering PSs after 3–5 leaves,
and many more flowers replaced leaves along the PSs.
Likewise, uf receptors grown in non-permissive light were
stimulated to flower by 35S:SFT donors, demonstrating that
graft-transmissible SFT-borne signals substituted for the
high light-dose requirements, and converted leaf-primordia
of the uf PSs to flowers (Lifschitz et al., 2006).
Graft-transmissible signals generated in tomato
substitute for short-day stimulus in Maryland
In a seminal discovery, Garner and Allard (1920) identified
a recessive mutation which confers a short-day response
over a day-neutral background in the Maryland Mammoth
(MM) tobacco plants. Under long days, cv. Samsun plants
flower after 24–25 leaves while MM plants do not flower
at all (Fig. 3A). The tomato 35S:SFT transgene induced
similar early flowering effects in both tobacco strains under
long and short days. To see if the SFT systemic pathway
itself, not just SFT function, is mediated by the same
mechanism in tomato and the short-day tobacco, 35S:SFT
tomato donor shoots were grafted onto leaf petioles of long-
day-grown MM plants. The recipient MM plants flowered
3–4 weeks later, indicating that 35S:SFT signals, generated
in tomato and transmitted via leaf petioles, induced flower-
ing in MM apices under conditions in which they otherwise
never flower (Lifschitz et al., 2006; Fig. 3B). The long-
day character of MM is conditioned by a single recessive
allele. It would be of interest to see whether the require-
ment for a long-day stimulus by the diploid tobacco species
Nicotiana sylvestris can also be substituted by systemic
SFT, as well as whether alleles of the same gene are in-
volved in these opposite day-length responses in tobacco.
Systemic SFT signals substitute for the long-day
stimulus in Arabidopsis
Plants of independent Arabidopsis lines expressing SFT or
FT driven by the leaf-specific promoter BLS flowered under
short days after only five to seven leaves, as compared with
15–19 leaves of wild-type plants. Other leaf promoters
used to express FT in Arabidopsis had similar effects (Abe
et al., 2005). The BLS:SFT transgene induced early-
flowering responses and complemented the sft mutant; it
also induced early flowering in the short-day MM tobacco
lines (Lifschitz et al., 2006).
Complementation of sft by continuous graft-transmissible
signals from 35S:SFT donor shoots, or by endogenous
expression of BLS:SFT in leaves, indicates that systemic
SFT signals played their primary flowering role at the apical
meristems of the tomato. The dynamic temporal and spatial
expression of FT or SFT in Arabidopsis or tomato leaves
and the observation that elimination of young tomato leaves
enhances flowering (Leopold and Lam, 1960) indicate that
the source, mobility, distribution, and targets of SFT signals
in tomato are tightly regulated. Evidently, mis-expression
of SFT is insufficient to promote systemic flowering as
illustrated by the failure of leafless donors to enhance
flowering (our work) and by the failure of root-specific
expression of FT to promote flowering in Arabidopsis
(Abe et al., 2005).
SFT is not alone: the SELF PRUNING inhibitory
function dominates the switch to termination
during the sympodial cycles
The pattern of termination in sympodial systems is species-
specific. Whereas in wild-type tomato there are three leaves
Fig. 3. Differential transmission of systemic signals. (A, B) Maryland Mammoth tobacco plants will never flower under long days, but when an
35S:SFT tomato shoot was grafted on its leaf petiole (arrow in B), flowering was evident three weeks later. (C) No morphological alterations are evident
in wild-type tomato receptor grafted on the TKN2-misexpressing dominant mutation Curl donor.
3410 Lifschitz and Eshed
by guest on June 2, 2013
per SU, there are only two in the sibling species Solanum
pennellii. In petunia, a close relative of tomato, SUs consist
of one leaf and one terminal flower.
A major modification of the robust sympodial regularity
is referred to as ‘determinate’ and ‘indeterminate’ and
describes the growth habits of mutant sp and wild-type SP
plants, respectively. The introduction of the ‘determinate’
(sp) character resulted in plants with a bushy constitution
and nearly homogeneous fruit setting which facilitated
mechanical harvesting and thus revolutionized the tomato
industry (Rick, 1978). In sp plants, the number of leaves per
SU in the main and lateral shoots decreases progressively
with age, from three to two to one until, eventually, the last
SU generates only an inflorescence and the shoot system
terminates with two consecutive inflorescences (Fig. 2F,
G). SP therefore maintains the three-nodal system by
inhibiting precocious termination of the sympodial apical
meristems. Constitutive expression of SP delays primary
termination, increases the number of leaves per SU and
induces indeterminate, partially
(Pnueli et al., 1998). Environmental conditions such as
low-light integrals may also increase the number of leaves
per SU, but significantly, no growth conditions in tomato
are known to reduce the number of leaves in SUs to
a regular one or two.
SP is a functional homologue of TFL1 and a member of
the CETS family of signalling/adaptor factors (Pnueli et al.,
1998). The range of CETS-interacting proteins and their
molecular identity suggest they function as adaptors in
signalling or transcription complexes (Yeung et al., 1999;
Pnueli et al., 2001). One class of CETS partners, 14-3-3
adaptors, participate in many, sometimes unrelated cell
processes, by interacting with different signalling mole-
cules or with the same factor in different cellular contexts.
Similarly, the nature and range of interacting proteins
suggest analogous functions for SP and SFT in different
signalling pathways. Such roles are reflected in the range of
pleiotropic effects which are manifested as changes in the
growth/termination balance, but not in organ identity. For
example: sp, like 35S:SFT, is epistatic to falsiflora (fa), the
tomato LEAFY gene, with respect to termination patterns,
while fa is epistatic with respect to floral identity. sp is also
a suppressor of indeterminacy and leaf formation in in-
florescences of several late-flowering mutants (E Lifschitz,
unpublished data; but see also Pnueli et al., 1998;
Szymkoviak and Irish, 2005). Moreover, in a context-
dependent manner, sp is an ‘enhancer’ of primary flowering
time as well. sft fa (Molinero-Rosales et al., 2004) and sft
uf (our observation) never flower, but formation of
inflorescences is partially rescued in triple mutant combi-
nations with sp (E Lifschitz, Y Eshed, unpublished data).
Extreme SFT doses, which induce primary termination
after three leaves, do not alter the regularity of the growth/
termination cycles in the compound shoot. Conversely, the
gradual but continuous reduction of leaf number per SU
in sp mutants is not associated with premature primary
termination, suggesting that SFT and SP are not simply
Termination and the SFT/SP ratio in the
All sft alleles produce a first flower in a VI and all mutant
alleles of FT eventually flower in Arabidopsis. The
termination of SUs in sp plants is gradual, rather than
abrupt, and SAM termination in tfl1 Arabidopsis mutants
is suppressed under short days. SFT/FT and SP/TFL1 are,
therefore, not required for determination of the vegetative
or reproductive states of the SAMs. How, therefore, do
systemic termination signals by SFT, and growth-
promoting/anti-termination functions of SP, regulate the
fate of the primary apex?
Flowering in tomato is induced very early and evocation
has been estimated to occur in seedlings with two to four
expanding leaves (Picken et al., 1985). One possibility is
that SFT is increasingly up-regulated in the primary
vegetative shoot relative to SP and that SP only becomes
expressed at high levels in the axillary and sympodial buds.
This would predict a result similar to that observed;
a primary SAM that is indifferent to sp but highly
sensitive to SFT and secondary shoots that are relatively
refractory to increased levels of SFT but sensitive to
a reduction in SP.
SFT to SP ratios regulate termination and apical
dominance in the sympodial fork
Termination of the SUs uniquely after three leaves in
co-ordination with the discriminatory release of the most
proximal bud from apical dominance, requires the renewed
balancing of contrasting growth and termination signals
by the SAMs in each nodal segment of the SU. The age-
dependent termination of SUs by spimplies that to maintain
regularity, the anti-termination function of SP, is constantly
upgraded at the whole-plant level. Thus, to ensure the
robustness of the sympodial sequence, primary termination
establishes a graded SFT to SP threshold ratio, in which the
inhibitory effects of SP become dominant. How the SP
function is annulled after three leaves remains unknown,
but since regularity, under extremely high SFT levels can
be changed from three to two, additional factors beside
SP, must be involved.
The morphogenetic alterations conditioned by sp and sft
are ultimately manifested as changes in local and global
apical dominance. In wild-type SUs, SP maintains the
indeterminate state of the SAM, which in turns suppresses
the release of the first two axillary shoots. When the
function of SP in the SAMis compromised, either locally or
by systemic SFT signals, the prospective sympodial bud is
released. In the new primordial SU, the SFT to SP ratio is
Florigenic signals shape the compound tomato shoot 3411
by guest on June 2, 2013
presumably lower again by an unknown mechanism, and
the cycle repeats. As long as SP is functional, regularity at
any threshold level is maintained. Auxin is the most likely
mediator of indeterminacy and apical dominance within
SUs by SP (Pnueli et al., 2001). SP and SFT belong to
the same protein family, they interact with the same classes
of factors, and only a few amino acids distinguish their
floral-promotion and floral-inhibition functions (Hanzawa
et al., 2005; Ahn et al., 2006). Thus, the floral inhibitory
function of SP is probably mediated by systemic inhibi-
tory roles too (Lang et al., 1977). However, since SP is
expressed in the youngest leaf primordia and in the SAMs
of sympodial units, its systemic effects may be short-range
and confined to a modified source–sink track within the
sympodial fork itself.
The two-flowering systems model
Inactivation of SP results in the complete collapse of the
sympodial regularity but is inconsequential in the timing of
termination of primary apices (Fig. 2; Pnueli et al., 1998).
Although tomato strains differ with respect to their primary
termination, all maintain the subsequent robust three-leaf
pattern. Under conditions of high irradiance where the
number of leaves to the first inflorescence may be reduced
by 50%, the sympodial rhythm is still maintained. Genetic
factors also distinguish primary meristems. In terminating
flower (tmf) the primary shoot is terminated by a solitary
abnormal flower, but subsequent lateral shoots are perfectly
normal. Thus primary shoots of tomato are extremely
sensitive, while SUs are relatively refractory to changes
in genes activity with the exception of SP where the
situation is reversed (Fig. 2). The primary apex may be
preferentially sensitive to SFT because it is laid down in
embryogenesis prior to build-up of inhibiting factors
like SP. Interestingly, Furr et al. (1947), reported on a
gap of several years between first and subsequently
flowering in perennial trees. Indeed, the responses of
Arabidopsis plants to FT and TFL1 (Ratcliff et al., 1998)
and expression of SFT/FT under different promoters in
Arabidopsis clearly distinguishes between the response
of primary and lateral shoots (E Lifschitz, unpublished
data). Inevitably, flowering in both species and the func-
tion of SFT/SP (FT/TFL1) is regulated, and has to be con-
sidered in the framework of two systems, one for primary
shoots and the other for sympodial in tomato and laterals
The molecular components of the florigen
pathway in tomato and Arabidopsis
Yeast-two-hybrid screens uncovered four different SP
interacting proteins (SIPs): A NIMA-like protein kinase
(SPAK) involved in cell division, 14-3-3 adaptor proteins,
a bZIP G-box (SPGB) factor, and an SP-specific interactor,
SIP4. SPAK also interacts with the 14-3-3s which in turn
also interact with SPGB and SIP4. SP and 14-3-3 share
a SPAK-interacting site. With the exception of SIP4, other
SIPs interact also with TFL1, CEN, and FT (Pnueli et al.,
2001). Abe et al. (2005) and Wigge et al. (2005) showed
that one Arabidopsis homologue of SPGB is encoded by
the late-flowering gene FD, and that FD is partially
required for the proper function of FT. RNA expression
data have suggested that FT is expressed primarily in
mature leaves whereas FD is expressed predominantly in
the SAM. This separation of expression territories pro-
moted the hypothesis that to induce flowering, FT primary
products must travel from leaves to SAMs (Abe et al.,
2005). Concomitantly, Huang et al. (2005) developed an
ingenious experimental protocol to induce FT, driven by
a heat-shock promoter, in a single leaf of Arabidopsis.
Using this method, a leaf-induced FT RNA was detectable
in the SAM proper and several 1-mm leaf primordia,
several hours after local heat induction. FT RNA may thus
function as a florigenic signal. It becomes important,
therefore, to determine whether the moving RNA is
functional and being translated in the target SAM proper,
is distributed in the other parts of the plant as expected
of a florigenic substance, and whether the FT RNA remain-
ing in the induced leaves, is unable to induce flowering if
its movement is prevented.
In tomato, we failed to detect 35S:SFT- or FIL:FT-born
transcripts beyond graft unions in rescued sft shoots
(Lifschitz et al., 2006). Considerations of universality
and differences in the assay sensitivities notwithstanding,
other differences between tomato and Arabidopsis may
be important. Although the two are not extremely diver-
gent, one is perennial and sympodial the other annual
and monopodial. The inductive system of Arabidopsis
may be more permissive to systemic FT signals such that
a single burst is sufficient for the single switch. Lasting
systemic induction in the robust, day-neutral, and cycling
flowering system of tomato, requires persistent emission
of SFT-triggered signals. In the tomato experimental sys-
tem, the source and target were separated by graft unions,
by considerable distance and, significantly, by genotype.
Moreover, if movement of florigens abides by the sink–
source rules, the organization of the sympodial shoot in
quasi-autonomous units may constitute another important
At another level, the tomato genes encoding the SPGB
and 14-3-3 SP-interacting proteins are expressed in all
leaves and throughout development, potentially making
it unnecessary for SFT RNA to travel (Lifschitz et al.,
2006). Two FD-like genes have been identified in Arabi-
dopsis, but their differential affinity to FT and TFL1 is still
debated (Abe et al., 2005; Wigge et al., 2005). It is possible
therefore that our yeast two hybrid screen missed a
SPGB homologue, which is expressed solely in the SAM.
3412 Lifschitz and Eshed
by guest on June 2, 2013
Currently it is assumed that the termination versus growth
effects of SFT and SP in the vegetative meristems of the
leaves,stems, andSAMs, requireinteractions with the same
factors. And since meristematic activity in tomato leaves
continues well into maturation, it follows that the FD-like
function is required in leaves as well. In addition, FD
transcripts have been detected in expanding Arabidopsis
leaves, and in agreement with this, a curled leaf phenotype,
stimulated by 35S:FT or 35S:SFT is largely dependent on
FD activity; although this, it may be argued, is due to the
expression of FD in leaf primordia only (Teper-Bamnolker
and Samach, 2005).
There are additional uncertainties and inconsistencies
in our current understanding of the floral-integrator genes
and their targets. The partial suppression of FT-induced
flowering by fd mutants led to a model whereby FD and
FT complexes co-operately activate floral genes, such as
AP1 (Wigge et al., 2005). However, ap1 or leafy Arabi-
dopsis mutants are not late-flowering and fd plants
misexpressing FT flower at least as early as the wild-type,
albeit not as early as 35S:FT (Abe et al., 2005).
In tomato, Kim et al. (2001) reported that leaf morphology
defects by Mouse ears (Me) and Curl (Cu), both dominant
mutations in the TKn2 gene are graft transmissible and that
such a systemic effect is associated with Me transcripts
crossing the graft junctions. In more than 120 grafts in
tomato and tobacco, no support could be found for the
systemic transmission of the Me, Curl, or Kn1 morpholog-
encodes a fusion transcript with almost all of its 59 end
comprised of PFP, a gene for a protein involved in fructose
metabolism (Parnis et al., 1997). The relevance of the
translocated chimeric PFP-TKn2Kn1 RNA transcripts for
normal development is thus questionable. Similarly, no
support for long-distance transmission of microRNA-
mediated alterations could be provided by tomato or
tobacco grafting (Alvarez et al., 2006). Taken together, a
model is currently favoured in which a downstream sys-
temic pathway is initiated by cell-autonomous functions of
SFT RNA. From this perspective, the options of systemic
SFT polypeptides, intercellular signal transduction path-
way, or a moving secondary metabolite are equally
Systemic acceleration of flowering and
termination are pleiotropic functions of
TFT to SP (FT to TFL1) ratios
It has been shown that mis-expression of SFT attenuates
growth of intercalary peripheral and plate meristems and all
these features are greatly enhanced in an sp background.
Apical meristems were induced to form terminal inflor-
escences, or were temporarily or permanently arrested,
by high levels of systemic SFT signals, and these were
also enhanced in an sp background (Lifschitz E and Eshed
Y unpublished data). Thus, all responses stimulated by the
SFT to SP balance involve changes in growth, suggesting
that floral transition and growth attenuation, instead of
being the consequence of one another, are two facets of the
same cellular responses. Since all the facets of the SFT/SP
ratio are also graft-transmissible, it is speculated that
growth and termination are targets for florigen-compatible
signals and that boosting flowering is a pleiotropic effect of
FT orthologues (Lifschitz et al., 2006). If growth were the
primary target of SFT, the systemic signals might involve
conditioning of the apical meristem via a finely regulated
temporal change in cell-proliferation patterns, providing the
context/time required for the vegetative-to-reproductive
We thank Arnon Brand for drawings and John Paul Alvarez for
SEM, live images and disagreements. This work was funded by
HFSPO, ISF and EU (EL) and by DIP and ISF (YE).
Abe M, Kobayashi Y, Yamamoto S, Daimon Y, Yamaguchi A,
Ikeda Y, Ichinoki H, Notaguchi M, Goto K, Araki T. 2005.
FD, a bZIP protein mediating signals from the floral pathway
integrator FT at the shoot apex. Science 309, 1052–1056.
Ahn JA, Miller D, Winter VJ, Banfield MJ, Lee JH, Yoo SY,
Henz SR, Brady RL, Weigel D. 2006. A divergent external loop
confers antagonistic activity on floral regulators FT and TFL1.
EMBO Journal 25, 605–614.
Alvarez JP, Pekker I, Goldshmidt A, Blum E, Amsellem Z,
Eshed E. 2006. Endogenous and synthetic micro RNAs stimulate
simultaneous, efficient and localized regulation of multiple targets
in diverse species. The Plant Cell 18, 1134–1151.
An H, Roussot C, Suarez-Lopez P, et al. 2004. CONSTANS acts
in the phloem to regulate a systemic signal that induces photoperi-
odic flowering of Arabidopsis. Development 131, 3615–3626.
Atherton JG, Harris J. (eds) 1986. The tomato crop. London:
Chapman and Hall.
Bell PR. 1992. Green plants: their origin and diversity. Cambridge:
Cambridge University Press.
Ben-Naim O, Eshed R, Parnis A, Teper-Bamnolker P, Shalit A,
Coupland G, Samach A, Lifschitz E. 2006. The CCAAT binding
factor can mediate interactions between CONSTANS-like proteins
and DNA. The Plant Journal 46, 462–476.
Bernier G. 1988. The control of floral evocation and morphogenesis.
Annual Review of Plant Physiology and Plant Molecular Biology
Boss PK, Bastow RM, Mylne JS, Dean C. 2004. Multiple pathways
in the decision to flowering: enabling promoting and resetting.
The Plant Cell 16, S18–S31.
Carmel-Goren L, Liu TS, Lifschitz E, Zamir D. 2003. The SELF
PRUNING gene family in tomato. Plant Molecular Biology 52,
Florigenic signals shape the compound tomato shoot 3413
by guest on June 2, 2013
Chailakhyan MK. 1936. New facts in support of the hormonal
theory of plant development. Doklady Proceedings of the Acad-
emy of Sciences, USSR 13, 79–83.
Cronquist A. 1988. The evolution and classification of flowering
plants. Lawrence, KS: Allen Press.
Dielen V, Quinet M, Chao J, Batoko H, Havelange A, Kinet J-M.
2004. Uniflora, a pivotal gene that regulates floral transition and
meristem identity in tomato (Lycopersicon esculentum). The New
Phytologist 161, 393–400.
Furr JR, Cooper WC, Reece PC. 1947. An investigation of flower
formation in adult and juvenile citrus trees. American Journal of
Botany 34, 1–9.
Garner WW, Allard HA. 1920. Effect of the relative length of day
and night and other factors of the environment on growth and
reproduction in plants. Journal of Agricultural Research 18,
Hanzawa Y, Money T, Bradley D. 2005. A single amino acid
converts a repressor to an activator of flowering. Proceedings of
the National Academy of Sciences, USA 102, 7748–7753.
Hareven D, Gutfinger T, Pnueli L, Bauch L, Cohen O, Lifschitz
E. 1994. The floral system of tomato. Euphytica 79, 235–243.
Hareven D, Gutfinger T, Parnis A, Eshed Y, Lifschitz E. 1996.
The making of a compound leaf: genetic manipulation of leaf
architecture in tomato. Cell 84, 735–744.
Hayama R, Coupland G. 2004. The molecular basis of diversity
in the photoperiodic flowering responses of Arabidopsis and rice.
Plant Physiology 135, 677–684.
Huang T, Bohlenius H, Eriksson S, Parcy F, Nilsson O. 2005.
The mRNA of the Arabidopsis gene FT moves from leaf to shoot
apex and induces flowering. Science 309, 1694–1696.
Jack T. 2004. Molecular and genetic mechanisms of floral control.
The Plant Cell 16, S1–S17.
Kerr EA. 1982. Single flower truss ‘sft’ appears to be on
chromosome 3. Tomato Genetics Cooperative Reports 32, 31.
Kim M, Canio W, Kessler S, Sinha N. 2001. Developmental
changes due to long-distance movement of a homeobox fusion
transcript in tomato. Science 293, 287–289.
Kinet JM. 1977. Effect of light conditions on the development of the
inflorescence in tomato. Scientia Horticulturae 6, 15–26.
Lang A, Chilakhian MK, Frolova LA. 1977. Promotion and
inhibition of flower formation in a day-neutral plant in graft with
a short-day plant and a long-day plant. Proceedings of the National
Academy of Sciences, USA 74, 2412–2416.
Leopold AC, Lam SL. 1960. A leaf factor influencing tomato
flowering. Proceedings of the American Society of Horticultural
Sciences 76, 543–547.
Lifschitz E, Eviatar T, Rozman A, Shalit A, Goldshmidt A,
Amsellem Z, Alvarez JP, Eshed Y. 2006. The tomato FT
ortholog triggers systemic signals that regulate growth and
flowering and substitute for diverse environmental stimuli.
Proceedings of the National Academy of Sciences, USA 103,
Martinez-Garcia JF, Virgos-Soler A, Prat S. 2002. Control of
photoperiod-regulated tuberization in potato by the Arabidopsis
flowering-time gene CONSTANS. Proceedings of the National
Academy of Sciences, USA 99, 15211–15216.
Menda N, Semel Y, Peled D, Eshed Y, Zamir D. 2004. In silico
screening of a saturated mutation library of tomato. The Plant
Journal 38, 861–872.
Molinero-Rosales N, Latorre A, Jamilena M, Lozano R. 2004.
SINGLE FLOWER TRUSS regulates the transition and main-
tenance of flowering in tomato. Planta 218, 427–434.
Mouradov A, Cremer F, Coupland G. 2002. Control of flowering
time: interacting pathways as a basis for diversity. The Plant Cell
Parnis A, Cohen O, Gutfinger T, Hareven D, Zamir D,
Lifschitz E. 1997. The dominant developmental mutants of
tomato, Mouse-ear and Curl, are associated with distinct modes
of abnormal transcriptional regulation of a Knotted gene. The
Plant Cell 9, 2143–2158.
Picken AJF, Hurd RG, Vince-Prue D. 1985. Lycopersicon
esculentum. In: Halevy AH, ed. CRC handbook of flowering,
Vol. III. Boca Raton: CRC Press, 330–346.
Pnueli L, Carmel-Goren L, Hareven D, Gutfinger T, Alvarez J,
Ganal M, Zamir D, Lifschitz E. 1998. The SELF-PRUNING
gene of tomato regulates vegetative to reproductive switching of
sympodial meristems and is the ortholog of CEN and TFL1.
Development 125, 1979–1989.
Pnueli L, Gutfinger T, Hareven D, Ben-Naim O, Ron N, Adir N,
Lifschitz E. 2001. Tomato SP-interacting proteins define a con-
served signaling system that regulates shoot architecture and
flowering. The Plant Cell 13, 2687–2702.
Quinet M, Dubois C, Goffin M, Chao J, Dielen V, Batoko H,
Boutry M, Kinet JM. 2006. Characterization of tomato (Solanum
lycopersicum L.) mutants affected in their flowering time and in
the morphogenesis of their reproductive structure. Journal Exper-
imental Botany 57, 1381–1390.
Ratcliff O, Amaya I, Vincent C, Rothstein S, Carpenter E,
Coen E, Bradley D. 1998. A common mechanism controls the life
cycle and architecture of plants. Development 125, 1609–1615.
Rick CM. 1978. The tomato. Scientific American 239, 76–87.
Schmitz G, Tillmann E, Carriero F, Fiore C, Cellini F, Theres K.
2002. The tomato Blind gene encodes a MYB transcription factor
that controls the formation of lateral meristems. Proceedings of the
National Academy of Sciences, USA 99, 1064–1069.
Szymkowiak EJ, Irish EE. 2005. JOINTLESS suppresses sympo-
dial identity in inflorescence meristems of tomato. Planta 223,
Teper-Bamnolker P, Samach A. 2005. The flowering integrator
FT regulates SEPALLATA3 and FRUITFULL accumulation in
Arabidopsis leaves. The Plant Cell 17, 2661–2675.
Medrano D, Drake R, Schuch W, Giovannoni J. 2002. A
MADS-box gene necessary for fruit ripening at the tomato
ripening-inhibitor (rin) locus. Science 296, 343–346.
Wigge PA, Kim MC, Jaeger KE, Busch W, Schmid M,
Lohmann JU, Weigel D. 2005. Integration of spatial and temporal
information during floral induction in Arabidopsis. Science 309,
Wittwer SH, Aung LM. 1969. Lycopersicon esculentum Mill. In:
Evans LT, ed. The induction of flowering: some case histories.
Melbourne: Macmillan of Australia, Chapter 19.
Yeager AF. 1927. Determinate growth in the tomato. The Journal of
Heredity 18, 263–265.
Yeung K, Seitz T, Li S, et al. 1999. Suppression of Raf-1 ki-
nase activity and MAP kinase signaling by RKIP. Nature 401,
Zeevaart JAD. 1976. Physiology of flower formation. Annual
Review of Plant Physiology 27, 321–348.
3414 Lifschitz and Eshed
by guest on June 2, 2013