Review

The suspensor: not just suspending

the embryo
Tomokazu Kawashima and Robert B. Goldberg
Department of Molecular, Cell, and Developmental Biology, University of California, Los Angeles, CA 90095, USA

The suspensor is a terminally differentiated embryonic not understood. Because the suspensor is a terminally
region that connects the embryo to surrounding tissues differentiated and highly-specialized part of the embryo
during early seed development. Most seed-bearing plant that has no role in subsequent plant development, it has
embryos contain suspensor regions, which occur in a been able to evolve independently of the embryo proper
wide variety of sizes and shapes, and suspensor-like with respect to both form and function throughout the
structures are present in the embryos of some lower plant kingdom.
land plants. Recent technological advances, including With few exceptions (e.g., swamp wattle, Figure 1b), the
novel genomics approaches, have provided insights into embryos of most seed-bearing plants have suspensors,
the function of the suspensor and the DNA sequences suggesting that the suspensor is evolutionally conserved
that control suspensor-specific gene expression. The and that it plays an important role in plant embryogenesis.
molecular mechanisms controlling embryo basal-cell It is remarkable that well-preserved suspensors are
lineage specification and suspensor differentiation observed in gymnosperm seed fossils uncovered in Permian
events are also beginning to be illuminated. Here, we rocks found in Antarctica, indicating that highly differen-
summarize the role of the suspensor in plant embryo- tiated suspensors were a prominent part of higher plant
genesis and discuss future directions of suspensor embryos over 300 million years ago (mya) [15,16]. Some
biology, including the dissection of suspensor gene lower plant embryos, including those of mosses and other
regulatory networks. bryophytes, also possess a suspensor, or an embryonic
region that might perform similar functions (e.g., foot)
(Figure 1c) [5]. It remains unclear whether the suspensor
Suspensors are ancient and morphologically diverse
and foot evolved from a common ancestral cell type. Never-
embryo structures
theless, the suspensor is an ancient embryonic structure,
The suspensor (Figure 1a) was first discovered almost 170
and the processes by which suspensor cell fate is determined
years ago, but was originally thought to be a part of the
in higher plants may have already existed in lower plants,
pollen tube [1,2]. Further investigation of embryo devel-
which diverged from seed-bearing plants 450 mya [17].
opment in several plant species showed that the suspensor
is a unique embryonic region, connects the embryo to the
The suspensor plays an important role in embryo
seed coat, and is morphologically diverse throughout the
development
plant kingdom, in contrast with the embryo proper
The position and structure of the suspensor within the
(Figure 1b; Box 1) [3–8]. For example, the scarlet runner
embryo provide clues for its possible role(s) during embry-
bean (Phaseolus coccineus) suspensor, which has been a
ogenesis. For example, the suspensor pushes the embryo
classic model system for investigating suspensor function
proper into the endosperm cavity and connects the embryo
[9–12], has at least 200 cells and is massive compared with
proper to surrounding maternal and endosperm tissues -
the much smaller Arabidopsis (Arabidopsis thaliana) sus-
serving as a conduit for nutrients and growth regulators
pensor that is comprised of a simple file of seven cells
required for embryonic development (Figure 1a). Cellular
(Figure 1). Both, however, contain a uniform set of cells
structures that enhance the ability to transfer molecules
with a large basal cell at their micropylar ends (Figure 1).
from cell to cell are present in many suspensors, including
On the other hand, the orchid suspensor consists of only
(i) cell-wall ingrowths, (ii) haustorial outgrowths, and (iii)
one, or a few, large, highly vacuolated cells (Figure 1b).
numerous plasmodesmata [6,8,18]. In addition, the fact
Regardless of shape or size, the suspensor is derived
that the suspensor does not have a waxy cuticle layer
from the basal cell of the two-cell proembryo and is fully
facilitates direct communication with adjacent seed tissues
differentiated prior to the beginning of embryo proper
(e.g., seed coat, endosperm), in contrast with the embryo
maturation and the onset of dormancy (e.g., scarlet runner
proper that is surrounded by a cuticle that covers the
bean suspensor development, Figure 1a) [13,14]. During
protoderm surface [19,20].
seed maturation, the suspensor undergoes programmed
cell death and does not contribute to the development of
Transfer of nutrients and growth factors to the
next plant generation following seed germination (Box 1)
embryo proper
[6]. The mechanisms responsible for shifting the suspensor
What direct experimental evidence is there for the
from a differentiation program to a cell death program are
participation of the suspensor in transferring nutrients
Corresponding author: Goldberg, R.B. ([email protected]). and growth factors to the embryo proper? Experiments
1360-1385/$ – see front matter ß 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.tplants.2009.11.002 Available online 4 December 2009 23
 Review Trends in Plant Science Vol.15 No.1

Figure 1. Development, diversity, and evolution of the suspensor. (a) Scarlet runner bean embryo development. Schematic images (lower row) were drawn from the bright-
field photographs (upper row) that were taken from Weterings et al. [14]. (b) Schematic images of embryos and suspensors from different flowering plant species. Green
and yellow colors indicate the embryo proper and suspensor, respectively. A.r. does not have an obvious suspensor structure. C.b. and T.m. suspensors possess haustorial
structures. L.a. and L.l. suspensor cells are multinucleated. Images were taken and adapted from Guignard in [8] for A.r.; Guignard in [5] for C.a., C.l., L.a., and L.l.;
Chamberlin et al. [71] for G.m.; Raddolph in [5] for Z.m.; Swamy in [8] for C.b. and S.p.; Nagl and Kuhner in [8] for T.m. (c) Suspensor and suspensor-like structures in the
plant kingdom. Phylogenetic tree was adapted from Bowe et al. [72] and Langdale [17]. Green, yellow, and blue colors indicate the embryo proper, suspensor, and foot,
respectively. Images were taken and adapted from Goldberg et al. [13] for Arabidopsis, and Raven et al. [73] for Lycopodium, Marchantia, Polypodium, and Pinus. a, apical
region of proembryo; ax, axis; b, basal region of proembryo; c, cotyledon; cc, central cell; ec, egg cell; ep, embryo proper; s, suspensor; sy, synergid. Bars = 100 mm.

with labeled metabolites, such as sucrose and polyamine, PIN7-mediated auxin transport is essential for embryo
in developing seeds demonstrated the direct movement of development [25]. Taken together, these data show that
molecules from the suspensor to the embryo proper in an the suspensor transports molecules involved in both nutri-
energy dependent manner [21,22]. More recent studies tion and growth regulation to the embryo proper to support
using transgenic plants have shown that the Arabidopsis embryo development. The cellular processes by which the
globular-stage suspensor is connected symplastically to suspensor transfers substances to the embryo proper
the embryo proper by visualizing the movement of green remain to be determined.
fluorescent protein (GFP) from the suspensor to the
embryo proper [23]. In addition, genes involved in trans- Biosynthesis of plant hormones
ferring molecules, such as the Arabidopsis AtSUC3 sucrose The suspensor has an abundance of organelles such as
transporter and the loblolly pine (Pinus taeda) PtNIP1;1 mitochondria, endoplasmic reticulum, and specialized
aquaglyceroporin, are up-regulated in the suspensor plastids [8,18]. In addition, the suspensor cells of many
[23,24]. Finally, auxin is transported from the suspensor plant species have undergone extensive endopolyploidiza-
to the embryo proper in Arabidopsis embryos by the sus- tion, possess polytene chromosomes - most notably in their
pensor-specific PIN7 auxin efflux carrier protein, and basal cells (e.g., scarlet runner bean; Figure 1a), and are

24
Review Trends in Plant Science Vol.15 No.1

Box 1. Unique suspensor features proper development during the early stages of embryogen-
esis and (ii) the presence of the suspensor can be substi-
Embryonic origin
The suspensor is a part of the embryo and is derived from the basal tuted by the application of exogenous GAs. These results
cell of the two-cell proembryo (Figure 1a) [13]. The suspensor is confirm the importance of GA production in the suspensor
present in most higher plant species, and suspensor-like structures for embryo development, at least for the scarlet runner
(i.e., foot) are found in lower plants, including mosses (Figure 1c). bean [32–35].
Morphological diversity
Do the suspensors of all plant embryos synthesize GAs?
Suspensor structure and size vary widely in flowering plants. Several studies have identified the presence of bioactive
Suspensors range from a single cell in orchids to a massive GAs in suspensors of several different species in addition to
collection of cells in the scarlet runner bean (Figure 1). Suspensors the scarlet runner bean [36,37]. What about the small
can also contain tiers of multi-nucleated cells that form a syncytium Arabidopsis suspensor (Figure 1c)? Recent experiments
(Figure 1b). Many suspensors have giant basal cells at their
micropylar end, that is the site of maximum metabolic activity [62]. have shown that GA3-ox4, a gene encoding the last enzyme
required for GA biosynthesis, GA3-oxidase, is active in the
Endoreduplication and polytene chromosomes endosperm, but not in the suspensor during early
Suspensor cells can be polyploid and/or undergo extensive Arabidopsis seed development [38]. Related GA3 genes
endoreduplication. A novel feature of some suspensors, particularly
(GA3-ox1, GA3-ox2, and GA3-ox3) are also inactive in
in legumes, is the presence of giant polytene chromosomes [8,12].
The DNA content per nucleus in the scarlet runner bean suspensor, the suspensor [38]. This suggests that GA biosynthesis
for example, increases 8,000-fold from the embryo proper is not a common feature of all suspensors, and that there
boundary towards the large basal cell (Figure 1a) and is contained are differences in suspensor function among plant species.
in the polytene chromosomes [63]. Suspensor polytene chromo- What functions are shared by the suspensors of all plant
somes have giant puffs analogous to those observed in Drosophila
embryos, and what species-specific suspensor activities are
reflecting foci of high transcriptional activity [12].
coupled with specialized suspensor morphologies
Programmed cell death (Figure 1b) remain to be investigated.
The suspensor undergoes programmed cell death when the embryo
proper enters the maturation phase and does not contribute to the Using genomics to uncover suspensor function
next plant generation [64]. Suspensors have been used as a model
to investigate programmed cell death in higher plants, and markers
One way to answer the question of what are common and
of eukaryotic programmed cell death, such as DNA fragmentation, specialized suspensor functions among different species is
nuclear degradation, and caspase-like activities, have been to isolate suspensor mRNAs from diverse plants and com-
observed [65–70]. pare their transcriptional profiles. This, of course, is chal-
lenging because of the size, accessibility, and limited
temporal presence of the suspensor during the plant life
often multinucleated forming a syncytium (e.g., Lathyrus cycle. Newly emerging genomics and laser-capture micro-
angustifolia; Figure 1b) (Box 1) [8,12,18]. These features dissection (LCM) technologies, however, have changed this
generally reflect high levels of cellular biosynthetic and opened up a new era in understanding suspensor
activity, and are markers for terminally differentiated biology [4].
regions, such as the suspensor, that produce specialized Recently, our laboratory sequenced 300,000 ESTs
products over a short period of time. In fact, classical obtained from hand-dissected globular-stage scarlet run-
experiments carried out with the giant scarlet runner bean ner bean giant suspensors (Figure 2) (GenBank Accession
suspensor (Figure 1a) over a generation ago, showed that it Numbers GD372310-GD660862 and CA899870-
has higher levels of RNA and protein synthesis in com- CA916678) [4]. In collaboration with John Harada’s labora-
parison with the embryo proper during the early stages of tory at UC Davis, we also used LCM and Affymetrix
development [9–11]. GeneChip technologies to profile (i) soybean (Glycine
What is known about specific molecules that are syn- max (L.) Merr.) suspensor mRNAs at three developmental
thesized in the suspensor as a consequence of its elevated stages [(globular -GEO Series GSM147936-GSM14938);
metabolic activity? Several plant hormones, including gib- (heart - GEO Series GSM182032-GSM182034); and (coty-
berellic acid (GA), auxin, cytokinin, and abscisic acid, are ledon - GEO Series GSM191077-GSM191078)] [4], and (ii)
found in the suspensor of several species [6]. For example, Arabidopsis suspensor mRNA at the globular stage (GEO
the giant scarlet runner bean suspensor (Figure 1a) has Series GSM284386-GSM284387) (Figure 2). These are the
been used as a system to investigate suspensor hormone first comprehensive sets of genomics data to show (i) the
activity because it can be hand-dissected from embryos at number of mRNAs required for suspensor development, (ii)
the globular stage and separated from the embryo proper the diverse functions carried out by the suspensors of
[4,9,26–28]. The scarlet runner bean suspensor has the different plant species, and (iii) the types of transcription
highest level of free indole-3-acidic acid (IAA) in the early factors represented in the suspensor. Soybean and Arabi-
embryo, suggesting that it is the site of auxin synthesis dopsis suspensors contain at least 10,000 and 15,000
[29]. In addition, scarlet runner bean suspensor extracts diverse mRNAs during early embryo development,
are able to synthesize GAs [30], and mRNAs encoding all respectively (Figure 2a) - a significant proportion (20%)
enzymes in the GA biosynthetic pathway are highly preva- of which are in the ‘‘unclassified’’ category (Figure 2b). This
lent in the scarlet runner bean suspensor compared with indicates that a large number of genes are required for
the embryo proper [4,31]. Experimental manipulation of suspensor development and that many novel, yet undis-
scarlet runner bean embryonic regions in culture demon- covered, biological processes take place in the suspensor.
strated that (i) the suspensor is required for the embryo Taken together, comparative genomics from these, and

25
Review Trends in Plant Science Vol.15 No.1

Figure 2. Soybean, scarlet runner bean, and Arabidopsis suspensor mRNA profiles. (a) Numbers of mRNAs and ESTs detected by GeneChip and 454 high-throughput DNA
sequencing studies, respectively. (b) Functional classes of proteins encoded by mRNAs present in globular-stage Arabidopsis, soybean, and scarlet runner bean
suspensors. TF, transcription factor.

other, plant species (Figure 1) should provide novel Asymmetrical distribution of transcripts from the zygote
insights into previously unknown roles of the suspensor to the basal cell
in embryo development and enable suspensor functions to Are there mRNAs that are distributed asymmetrically
be investigated across the plant kingdom (Figure 1c). within the egg and/or zygote? Experiments with apical
and basal cells isolated from maize eggs fertilized in vitro
Differentiation of the suspensor from the basal cell demonstrated the presence of basal-cell specific mRNAs.
lineage One class of basal cell-specific mRNAs is present in the egg
Suspensor development precedes that of the embryo proper cell, whereas another class accumulates in the zygote
(Figure 1a), and the suspensors of many plants become fully following fertilization [44]. These results suggest that
differentiated a few divisions after basal cell formation (e.g., mRNAs present in the egg and zygote can asymmetrically
tobacco and Arabidopsis) [39,40]. Furthermore, suspensor accumulate within the basal cell following zygotic division.
cells are direct clonal descendants of the basal cell. Thus, it is Whether this is a result of (i) an asymmetric distribution of
possible that the molecular mechanisms that control sus- pre-existing transcripts to the basal cell, (ii) de novo syn-
pensor-specific gene activity are directly linked to the pro- thesis of basal-cell-specific mRNAs, and/or (iii) selective
cesses that establish basal cell fate. turnover of basal-cell-specific mRNAs in the apical cell is
How does the basal-cell lineage become specified? Both not known [44].
autonomous and position-dependent signaling hypotheses The Arabidopsis WUSCHEL-RELATED HOMOEOBOX
have been proposed for how apical and basal cells of the two- TRANSCRIPTION FACTOR8 (WOX8)/STIMPY-LIKE
cell plant embryo might become specified to follow different (STIPL) mRNA is also distributed asymmetrically to the
developmental pathways [14,41,42]. One hypothesis pro- basal cell from the zygote and is highly prevalent in the
poses that there is an asymmetrical distribution of morpho- suspensor at the globular stage [41,45]. WOX9/STIP mRNA,
genetic factors (e.g. transcription factors) within the egg cell a close WOX8/STIPL mRNA relative, is also up-regulated in
and/or the zygote, and after zygotic division the apical and the basal cell, although, in comparison with WOX8/STIPL
basal cells obtain different sets of factors that are able to mRNA, it is less prevalent in the suspensor [41,45]. A wox8/
program distinct developmental fates [14,41]. By contrast 9 (stip/stipl) double mutant results in embryo lethality,
with this cell-autonomous process, cell–cell signaling events including the failure to establish an auxin gradient from
have been proposed in which signals derived from adjacent the basal cell lineage to the embryo proper (see online
seed tissues (e.g., seed coat and/or endosperm) interact with Supplementary Material, Table S1 for a list of Arabidopsis
the basal region of the zygote and/or the basal cell, and these genes that affect suspensor development when mutated)
signals trigger a cascade of events leading to the differen- [41,45,46]. Genes related to WOX8 and WOX9 in maize (e.g.,
tiation of the basal cell into the suspensor [42]. Although it is ZmWOX9A, ZmWOX9B, ZmWOX9C) [47] and the scarlet
still not clear how the basal cell is specified, recent findings runner bean (e.g., PcWOX9-like) [4] have suspensor-specific
suggest that both cell-autonomous and position-dependent transcriptional patterns, and WOX-related genes are found
processes may be essential for the establishment of the basal in the moss, Physcomitrella patens, that has a suspensor-
cell lineage and suspensor development [41–43]. like foot (Figure 1c) [22,48], suggesting a conserved role for

26
Review Trends in Plant Science Vol.15 No.1

WOX8 and WOX9 genes in suspensor development across the suspensor (Figure 3c) [51]. In addition, mutagenesis
the plant kingdom. Taken together, these results show that experiments showed that this regulatory module contains
the basal cell can accumulate specific mRNAs, including at least three short sequence motifs that are required for
those encoding transcription factors, and that WOX8 and suspensor transcription during early embryo development
WOX9 genes play an important role in basal cell specifica- (Figure 3a and 3c) [51]. These motifs consist of two similar
tion and subsequent suspensor differentiation. 10-bp flanking sequences, designated as 10-bp (50 -
GAAAAGCGAA-30 ) and 10-bp-like (50 -GAAAAACGAA-30 ),
Cell-cell signaling plays a role in suspensor and a central 6-bp sequence referred to as Region 2 (50 -
differentiation TTGGT-30 ) (Figure 3c) [51]. Knocking out each of these
Evidence for the importance of cell–cell signaling events in motifs either eliminated or reduced significantly transcrip-
basal cell specification and suspensor differentiation has tion in the suspensor [51]. What transcription factors
been obtained from studies with the Arabidopsis YODA interact with these suspensor cis-regulatory elements
(YDA) and SHORT SUSPENSOR (SSP) genes that encode and the connection, if any, between the G564 suspensor
mitogen-activated protein kinase kinase (MPKK) kinase transcriptional machinery and the SSP-YDA signaling
[42] and interleukin-1 receptor-associated kinase/Pelle- cascade remain to be determined (Figure 3c).
like kinase [43] proteins, respectively. Mutant yda and There are many thousands of genes that are active in
ssp zygotes do not elongate and produce a small basal cell the suspensor (Figure 2a), only a small fraction of which
following zygotic division that fail to develop into a sus- encode suspensor-specific mRNAs [4]. Whether G564 sus-
pensor [42,43] (see online Supplementary Material, Table pensor cis-regulatory elements are used by other suspen-
S1). In both cases, suspensor-specific marker genes are not sor-specific genes is not yet known. For example, it is
expressed [42,43]. possible that there are several suspensor gene regulatory
SSP acts upstream of YDA and appears to activate the networks that operate in parallel within the basal cell
YDA MPKK kinase pathway in the zygote and basal cell lineage, and only a subset employ G564 regulatory motifs.
through a unique paternal parent-of-origin effect [43]. Sequences related to the 10-bp motif that are present in
Surprisingly, SSP mRNA is transferred from the sperm other suspensor-specific upstream regions are able to sub-
cell to the egg, and then translated into SSP proteins in the stitute for the 10-bp motif to program transcription in the
zygote following fertilization [43]. Together, these results suspensor [51]. Whether these 10-bp-like motifs operate in
indicate that the SSP-YDA signaling pathway plays a the context of their own promoters remains to be deter-
major role in suspensor differentiation events, and that mined.
basal cell specification is, in part, paternally regulated. Recently, we observed that the suspensor-specific pine
Mitogen-activated protein kinases, MPK3 and MPK6, that PtNIP1;1 aquaglyceroporin gene upstream region [52] can
are downstream in the YDA pathway operating in stomatal activate transcription of the GUS reporter gene in tobacco
cells have been shown to phosphorylate the SPEECHLESS globular-stage embryos (Figure 3b) (T. Kawashima, PhD
transcription factor in vitro, and mpk3/mpk6 double thesis, University of California, Los Angeles, 2009). This
mutants fail to develop suspensor (see online Supple- observation suggests that suspensor gene regulatory net-
mentary Material, Table S1) [49,50]. How the SSP-YDA works are conserved in both angiosperms and gymnos-
pathway intersects with transcription factors that are perms, which diverged from a common ancestor over 300
important for suspensor differentiation and the identity mya (Figure 3a and 3b). The precise nature of suspensor
of signals within the seed that trigger this pathway during gene regulatory networks, how they are activated in the
early embryo development remain to be determined. basal cell lineage after zygotic division, and how they
originated during plant evolution remain major unan-
cis-regulatory sequences that activate suspensor- swered questions.
specific transcription
How are specific sets of genes activated in the suspensor Suspensor identity is controlled in part by the
once basal cell fate becomes established? We utilized the embryo proper
G564 gene of the scarlet runner bean to identify suspensor Is communication between the suspensor and embryo
cis-regulatory sequences [14,51]. G564 mRNA encodes a proper unidirectional, or does the embryo proper transfer
protein with unknown function, accumulates asymmetri- signals to the suspensor that are important for its de-
cally within the basal cells of a three-cell proembryo, and is velopment and/or function? Classical embryological stu-
highly prevalent in the suspensor at the globular stage dies in several species showed that destruction and/or
(Figure 1a) [14]. The G564 upstream region is able to alteration of the embryo proper results in abnormal sus-
activate suspensor-specific transcription of the E. coli b- pensor enlargement, and, in the case of Eranthis (winter
glucuronidase (GUS) reporter gene in transgenic tobacco aconite), a new embryo proper regenerates from the sus-
and Arabidopsis embryos shortly after fertilization, indi- pensor [6,53,54]. Similarly, mutations in genes that cause
cating that both the temporal and spatial transcriptional defects in Arabidopsis embryo proper development can
machinery responsible for G564 transcription in the sus- result in the suspensor acquiring an embryo proper-like
pensor is conserved in flowering plants (Figure 3a) [14,51]. state (e.g., raspberry1, sus1) and, in the extreme, generate
Functional dissection of the G564 upstream region in a new embryo proper (e.g., twn1) (see online Supple-
transgenic tobacco plants uncovered a 54-bp positive regu- mentary Material, Table S2) [7,55].
latory module located within a 150-bp tandem repeat that These observations indicate that the suspensor has the
is both necessary and sufficient to activate transcription in potential to follow an embryo proper-like pathway, and

27
 Review Trends in Plant Science Vol.15 No.1

Figure 3. Conservation of suspensor gene regulatory networks among seed-bearing plants. (a) Accumulation of G564 mRNA in scarlet runner bean embryos, and
suspensor GUS activity in Arabidopsis and tobacco embryos carrying a G564/GUS transgene. Phylogenetic tree was adapted from Paterson et al. [74] and Bowe et al. [72].
Scarlet runner bean G564 mRNA in situ images and G564/GUS tobacco and Arabidopsis embryo photos were taken from Weterings et al. [14] and Kawashima et al. [51],
respectively. (b) GUS activity in transgenic tobacco embryos carrying a PtNIP1;1/GUS gene. Loblolly Pine image was taken from http://bioimages.vanderbilt.edu/,
photographed by Steve Baskauf. (c) A schematic model of the scarlet runner bean G564 upstream region taken from Kawashima et al. [51]. Suspensor cis-regulatory
sequences were identified by functional G564 promoter/GUS gene fusion and mutagenesis experiments in transgenic tobacco plants [51]. Unknown transcription factors X
and Y bind the 10-bp/10-bp-like motifs and 5’-TTGGT-3’ Region 2 motif, respectively. Transcription factors X and Y are hypothesized to form a complex (X-Y-X) that recruits
the basal transcriptional machinery to activate G564 transcription in the suspensor. The identity of putative X and Y transcription factors is not yet known. It is also possible
that there is an additional short sequence required for suspensor-specific transcription within the 54-bp suspensor regulatory module. a, apical region of proembryo; b,
basal region of proembryo; ep, embryo proper; s, suspensor. Bars = 50 mm.

suggest that the embryo proper sends an ‘inhibitory’ signal play important roles in establishing suspensor form and
to the suspensor that is required to maintain its differ- function. The development of in vitro culture systems that
entiated state [6,7,56]. Alternatively, it is also possible that generate somatic embryos with suspensors in both angios-
there is a critical balance of growth regulators in the entire perms (e.g., Brassica napus) [58] and gymnosperms (e.g.,
embryo that maintains the developmental state of both, Japanese larch - Larix leptolepis) [59–61] should permit the
the suspensor and the embryo proper [57]. Disruption of dissection of suspensor molecular processes, including
these signals by defects in the embryo proper might cause regulatory networks, in a high-throughput manner. There
the suspensor to take on an embryo proper-like fate similar are many questions yet to be answered in suspensor
to the induction of somatic embryos from differentiated biology. For example, what are the precise molecular
cells in vitro [55]. The precise mechanism and the nature of events that enable the basal cell to differentiate into a
putative signals by which the embryo proper influences suspensor and what are the gene regulatory networks that
suspensor cell fate remain a mystery. control this process? We are now in a new era of possibility
to answer these questions and explore this unique embryo
Future Perspectives region. One thing is clear, however: the suspensor does a
Rapidly evolving genomic technologies have opened up lot more than simply ‘suspend’ the embryo.
new opportunities to investigate the role and evolution
of the suspensor in the plant kingdom. It is now possible to Acknowledgements
investigate gene activity in the suspensor of almost any We express our gratitude to all the current and former members of our
plant species and uncover novel genes and pathways that laboratory that have contributed significantly to our thinking about

28
Review Trends in Plant Science Vol.15 No.1

suspensor biology. We especially thank Nester Apuya and Koen 24 Ciavatta, V.T. et al. (2001) An aquaglyceroporin is abundantly
Weterings who initiated our work on scarlet runner bean suspensor expressed early in the development of the suspensor and the
genes. Suspensor research in our laboratory has been funded by grants embryo proper of loblolly pine. Plant Physiol. 127, 1556–1567
from the National Science Foundation Plant Genome Program, the 25 Friml, J. et al. (2003) Efflux-dependent auxin gradients establish the
United States Department of Energy, the United States Department of apical-basal axis of Arabidopsis. Nature 426, 147–153
Agriculture, Ceres, Inc. and the Nakajima Foundation. 26 Alpi, A. et al. (1979) Identification of gibberellin A1 in the embryo
suspensor of Phaseolus coccineus. Planta 147, 225–228
Appendix A. Supplementary data 27 Alpi, A. et al. (1975) Growth-regulator levels in embryo and suspensor
of Phaseolus coccineus at 2 stages of development. Planta 127, 153–162
Supplementary data associated with this article can be
28 Piaggesi, A. et al. (1989) Gibberellins in embryo-suspensor of Phaseolus
found, in the online version, at doi:10.1016/j.tplants. coccineus seeds at the heart stage of embryo development. Plant
2009.11.002. Physiol. 91, 362–366
29 Picciarelli, P. et al. (2001) Endogenous auxins and embryogenesis in
References Phaseolus coccineus. Aust. J. Plant Physiol. 28, 73–78
1 Meyen, F.J.F. (1840) Noch einige Worte über den Befruchtungsakt und 30 Ceccarelli, N. et al. (1981) Gibberellin biosynthesis in Phaseolus
die Polyembryonie bei den höheren Pflanzen. In Haude und Spenersche coccineus suspensor. Z Pflanzenphysiol 102, 37–44
Buchhaus, Berlin 31 Solfanelli, C. et al. (2005) Expression of two genes encoding gibberellin
2 Nikiticheva, Z.I. (2006) Suspensor. In Embryology of Flowering Plants: 2- and 3-oxidases in developing seeds of Phaseolus coccineus. Plant Cell
Terminology and Concepts (Batygina, T.B., ed.), pp. 198–202, Science Physiol. 46, 1116–1124
Publishers 32 Cionini, P.G. et al. (1976) Suspensor, gibberellin and invitro
3 Guignard, L. (1882) Recherches d’embryogenie vegetale comparee.1st development of Phaseolus coccineus embryos. Planta 131, 115–117
Memoire: Legumineuses. Ann. Sci. Nat. Bot. serie VI 12, 5–166 33 Yeung, E.C. and Sussex, I.M. (1979) Embryogeny of Phaseolus
4 Le, B.H. et al. (2007) Using genomics to study legume seed coccineus - suspensor and the growth of the embryo-proper invitro.
development. Plant Physiol. 144, 562–574 Z Pflanzenphysiol 91, 423–433
5 Wardlaw, C.W. (1955) Embryogenesis in Plants, Methuen 34 Brady, T. and Walthall, E.D. (1985) The effect of the suspensor and
6 Yeung, E.C. and Meinke, D.W. (1993) Embryogenesis in angiosperms: gibberellic acid on Phaseolus vulgaris embryo protein content. Dev.
development of the suspensor. Plant Cell 5, 1371–1381 Biol. 107, 531–536
7 Schwartz, B.W. et al. (1997) Development of the suspensor: 35 Walthall, E.D. and Brady, T. (1986) The effect of the suspensor and
differentiation, communication, and programmed cell death during gibberellic acid on Phaseolus vulgaris embryo protein synthesis. Cell
plant embryogenesis. In Advances in Cellular and Molecular Biology Differ. 18, 37–44
of Plants, Volume 4, Cellular and Molecular Biology of Plant Seed 36 Picciarelli, P. and Alpi, A. (1987) Embryo-suspensor of Tropaeolum
Development (Larkins, B.A. and Vasil, I.K., eds), pp. 53–72, Kluwer majus - identification of gibberellin-A63. Phytochemistry 26, 329–330
Academic Publishers 37 Picciarelli, P. et al. (1991) Gibberellins in suspensor, embryo and
8 Raghavan, V. (2006) Life and times of the suspensor - cell signaling endosperm of developing seeds of Cytisus laburnum. Phytochemistry
between the embryo and suspensor. In Double Fertilization, pp. 81-100, 30, 1789–1792
Springer 38 Hu, J. et al. (2008) Potential sites of bioactive gibberellin production
9 Walbot, V. et al. (1972) Macromolecular synthesis during plant during reproductive growth in Arabidopsis. Plant Cell 20, 320–336
embryogeny - rates of rna synthesis in Phaseolus coccineus embryos 39 Soueges, R. (1920) Embryogenie des Solanacees: Developpement de
and suspensors. Dev. Biol. 29, 104–111 l’embryon chez les Nicotiana. C. R. Acad. Sci. Paris 170, 1125–1127
10 Sussex, I. et al. (1973) Biosynthetic activity of suspensor of Phaseolus 40 Mansfield, S.G. and Briarty, L.G. (1991) Early embryogenesis in
coccineus. Caryologia 25, 261–272 Arabidopsis thaliana.2. the developing embryo. Can. J. Bot. 69,
11 Clutter, M. et al. (1974) Macromolecular-synthesis during plant 461–476
embryogeny - cellular rates of rna-synthesis in diploid and polytene 41 Haecker, A. et al. (2004) Expression dynamics of WOX genes mark cell
cells in bean embryos. J. Cell Biol. 63, 1097–1102 fate decisions during early embryonic patterning in Arabidopsis
12 Nagl, W. (1974) Phaseolus suspensor and its polytene chromosomes. Z thaliana. Development 131, 657–668
Pflanzenphysiol 73, 1–44 42 Lukowitz, W. et al. (2004) A MAPKK kinase gene regulates extra-
13 Goldberg, R.B. et al. (1994) Plant embryogenesis: zygote to seed. embryonic cell fate in Arabidopsis. Cell 116, 109–119
Science 266, 605–614 43 Bayer, M. et al. (2009) Paternal control of embryonic patterning in
14 Weterings, K. et al. (2001) Regional localization of suspensor mRNAs Arabidopsis thaliana. Science 323, 1485–1488
during early embryo development. Plant Cell 13, 2409–2425 44 Okamoto, T. et al. (2005) Identification of genes that are up- or down-
15 Smoot, E.L. and Taylor, T.N. (1986) Evidence of simple polyembryony regulated in the apical or basal cell of maize two-celled embryos and
in permian seeds from antarctica. Am. J. Bot. 73, 1079–1081 monitoring their expression during zygote development by a cell
16 Taylor, T.N. and Taylor, E.L., (1993) Origin and evolution of the seed manipulation- and PCR-based approach. Plant Cell Physiol. 46, 332–338
habit. In The Biology and Evolution of Fossil Plants, pp. 460–484, 45 Wu, X. et al. (2007) Combinations of WOX activities regulate tissue
Prentice Hall proliferation during Arabidopsis embryonic development. Dev. Biol.
17 Langdale, J.A. (2008) Evolution of developmental mechanisms in 309, 306–316
plants. Curr. Opin. Genet. Dev. 18, 368–373 46 Breuninger, H. et al. (2008) Differential expression of WOX genes
18 Zhukova, G.Y. (2006) Ultrastructural characteristics of suspensor. In mediates apical-basal axis formation in the Arabidopsis embryo.
Embryology of Flowering Plants: Terminology and Concepts. Dev. Cell 14, 867–876
(Batygina, T.B., ed.), pp. 202–208, Science Publishers 47 Nardmann, J. et al. (2007) WOX gene phylogeny in Poaceae: a
19 Rodkiewicz, B. et al. (1994) Chlorophyll and cutin in early comparative approach addressing leaf and embryo development.
embryogenesis in capsella, Arabidopsis, and stellaria investigated Mol. Biol. Evol. 24, 2474–2484
by fluorescence microscopy. Sex Plant Reprod. 7, 287–289 48 Richardt, S. et al. (2007) PlanTAPDB, a phylogeny-based resource of
20 Yeung, E.C. et al. (1996) Embryology of cymbidium sinense: embryo plant transcription-associated proteins. Plant Physiol. 143, 1452–1466
development. Ann. Bot. 78, 105–110 49 Wang, H.C. et al. (2007) Stomatal development and patterning are
21 Yeung, E.C. (1980) Embryogeny of phaseolus - role of the suspensor. Z regulated by environmentally responsive mitogen-activated protein
Pflanzenphysiol 96, 17–28 kinases in Arabidopsis. Plant Cell 19, 63–73
22 Nagl, W. (1990) Translocation of putrescine in the ovule, suspensor and 50 Lampard, G.R. et al. (2008) Arabidopsis stomatal initiation is
embryo of Phaseolus coccineus. J. Plant Physiol. 136, 587–591 controlled by MAPK-mediated regulation of the bHLH
23 Stadler, R. et al. (2005) Cell-to-cell movement of green fluorescent SPEECHLESS. Science 322, 1113–1116
protein reveals post-phloem transport in the outer integument and 51 Kawashima, T. et al. (2009) Identification of cis-regulatory sequences
identifies symplastic domains in Arabidopsis seeds and embryos. Plant that activate transcription in the suspensor of plant embryos. Proc.
Physiol. 139, 701–712 Natl. Acad. Sci. U. S. A. 106, 3627–3632

29
Review Trends in Plant Science Vol.15 No.1

52 Ciavatta, V.T. et al. (2002) A promoter from the loblolly pine PtNIP1;1 63 Brady, T. (1973) Feulgen cytophotometric determination of DNA
gene directs expression in an early-embryogenesis and suspensor- content of embryo proper and suspensor cells of Phaseolus coccineus.
specific fashion. Planta 215, 694–698 Cell Differ. 2, 65–75
53 Haccius, B. (1963) Restitution in acidity-damaged plant embryos - 64 Bozhkov, P.V. et al. (2005) Programmed cell death in plant
regeneration or regulation? Phytomorphology 107–115 embryogenesis. Curr. Top. Dev. Biol. 67, 135–179
54 Haccius, B. and Reichart, H. (1964) Restitutionserscheinungen an 65 Smertenko, A.P. et al. (2003) Re-organisation of the cytoskeleton
pflanzlichen Meristemen nach Röntgenbestrahlung. II. Adventiv- during developmental programmed cell death in Picea abies
embryonie nach Samenbestrahlung von Eranthis hiemalis. Planta embryos. Plant J. 33, 813–824
355–372 66 Filonova, L.H. et al. (2000) Two waves of programmed cell death occur
55 Yadegari, R. and Goldberg, R.B. (1997) Embryogenesis in during formation and development of somatic embryos in the
Dicotyledonous Plants. In Advances in Cellular and Molecular gymnosperm, Norway spruce. J. Cell Sci. 113, 4399–4411
Biology of Plants, Volume 4, Cellular and Molecular Biology of Plant 67 Bozhkov, P.V. et al. (2005) Cysteine protease mcII-Pa executes
Seed Development (Larkins, B.A. and Vasil, I.K., eds), pp. 3–52, programmed cell death during plant embryogenesis. Proc. Natl.
Kluwer Academic Publishers Acad. Sci. U. S. A. 102, 14463–14468
56 Marsden, M.P.F. and Meinke, D.W. (1985) Abnormal-development of 68 Bozhkov, P.V. et al. (2004) VEIDase is a principal caspase-like activity
the suspensor in an embryo-lethal mutant of Arabidopsis thaliana. Am. involved in plant programmed cell death and essential for embryonic
J. Bot. 72, 1801–1812 pattern formation. Cell Death Differ. 11, 175–182
57 Umehara, M. and Kamada, H. (2005) Development of the embryo 69 Wredle, U. et al. (2001) DNA fragmentation and nuclear degradation
proper and the suspensor during plant embryogenesis. Plant during programmed cell death in the suspensor and endosperm of Vicia
Biotechnol. 22, 253–260 faba. Int. J. Plant Sci. 162, 1053–1063
58 Supena, E.D. et al. (2008) Regeneration of zygotic-like microspore- 70 Lombardi, L. et al. (2007) DNA degradation during programmed cell
derived embryos suggests an important role for the suspensor in early death in Phaseolus coccineus suspensor. Plant Physiol. Biochem. 45,
embryo patterning. J. Exp. Bot. 59, 803–814 221–227
59 Umehara, M. et al. (2005) Two stimulatory effects of the peptidyl 71 Chamberlin, M.A. et al. (1994) Early endosperm, embryo, and ovule
growth factor phytosulfokine during somatic embryogenesis in development in Glycine-Max (L) merr. Int. J. Plant Sci. 155, 421–436
Japanese larch (Larix leptolepis Gordon). Plant Sci. 169, 901–907 72 Bowe, L.M. et al. (2000) Phylogeny of seed plants based on all three
60 Umehara, M. et al. (2005) Identification of a novel factor, vanillyl genomic compartments: Extant gymnosperms are monophyletic and
benzyl ether, which inhibits somatic embryogenesis of Japanese Gnetales’ closest relatives are conifers. Proc. Natl. Acad. Sci. U. S. A.
larch (Larix leptolepis Gordon). Plant Cell Physiol. 46, 445–453 97, 4092–4097
61 Umehara, M. et al. (2007) Identification of a factor that 73 Raven, P.H. et al. (2005) Bryophytes, seedless vascular plants, and
complementarily inhibits somatic embryogenesis with vanillyl gymnosperms. In Biology of Plants (Seventh Edition), pp. 345-433, W.
benzyl ether in Japanese larch. In Vitro Cell Dev.-Plant 43, 203–208 H. Freeman and Company
62 Yeung, E.C. and Clutter, M.E. (1979) Embryogeny of Phaseolus 74 Paterson, A.H. et al. (2004) Ancient polyploidization predating
coccineus - ultrastructure and development of the suspensor. Can. J. divergence of the cereals, and its consequences for comparative
Bot. 57, 120–136 genomics. Proc. Natl. Acad. Sci. U. S. A. 101, 9903–9908

Plant Science Conferences in 2010

The 5th EPSO Conference: Plants for Life
18–22 April, 2010
Olos Polar Center (Lapland), Finland
http://www.epsoweb.org/catalog/conf2010.htm

21st International Conference on Arabidopsis Research

6–10 June 2010
Yokohama, Japan
http://arabidopsis2010.psc.riken.jp/

BANFF Conference on Plant Metabolism

24–28 June, 2010
Banff, Canada
http://www.ucalgary.ca/plantmetabolism/

XVII Congress of the Federation of European Societies of Plant Biology (FESPB)

4–9 July, 2010
Valencia, Spain
http://www.geyseco.es/fespb/principal.php?seccion=welcome

30