The Role of the COP/DET/FUS Genes in Light Control of
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
Plant Physiol. (1996) 112: 871-878
The Role of the COP/DET/FUS Genes in Light Control of
Arabidopsis Seedling Development’
Ning Wei and Xing-Wang Deng*
Department of Biology, Yale University, New Haven, Connecticut 06520-81 04
Light is vital to plant life, not only as an energy source In Arabidopsis, genes for five phytochromes, phytochrome
for photosynthesis but also as an important environmental A, B, C, D, and E, and a blue-light receptor, CRYl (or HY4),
signal regulating development and growth. Light affects have been isolated. Recent reviews of the photobiology and
almost every stage of plant development (Kendrick and molecular biology of photoreceptor action have been pub-
Kronenberg, 1994), including seedling development, which lished (Furuya 1993; Vierstra, 1993; Quail, 1994; McNellis
represents one of the most dramatic and best characterized and Deng, 1995; Quail et al., 1995; Chamovitz and Deng,
processes (von Arnim and Deng, 1996).In Arabidopsis tkali- 1996). In this Update we will emphasize the role of a group
ana, for example, the morphology of the embryo in the of negative regulators genetically identified as anstitutive
imbibing seed (d 1), as well as the emerging seedling from photomorphogenic (COP) or de-eiolated (DET) loci.
the seed coat (d Z), are minimally affected by light condi-
tions (Wei et al., 199417). Soon after, however, seedling
morphogenesis differs drastically, depending on the light A TENTATIVE CLASSlFlCATlON O F THE
environment (Fig. 1).Light-grown seedlings exhibit short COP/DET MUTA NTS
hypocotyls and open and expanded cotyledons. Cell-type Genetic screens have been fruitful in identifying regula-
differentiation and chloroplast development are soon es- tory components involved in light-controlled seedling de-
tablished, and photosynthetically related genes are highly velopment. Based on contrasting light- and dark-grown
expressed. The shoot apical meristem is activated to seedling morphology, two types of loss-of-function mu-
produce true leaves and the plants proceed with further tants were recovered. Mutations in positive regulators of
vegetative and reproductive growth soon thereafter. This photomorphogenesis result in a ky (kpocotyl elongated)
development pattern in light is known as photomorpho- phenotype, a partia1 etiolated morphology under defined
genesis. In contrast, when seedlings are grown in complete light conditions, whereas mutations in the negative regu-
darkness, they undergo a developmental program known lators result in a cop or det phenotype when grown in the
as skotomorphogenesis or etiolation, in which the cotyle- dark. The ky mutants include mutants of photoreceptors as
dons remain folded and undeveloped, while the hypocot- well as downstream signaling components such as hy5
yls rapidly elongate. The apical hook serves to protect (Koornneef et al., 1980), fhyl, and fhy3 (Whitelam et al.,
cotyledons and the quiescent shoot meristems as the seed- 1993).
ling elongates rapidly to reach for the light. Instead of The copldet-type mutants can be further classified into
developing chloroplasts, the cotyledon cells form etioplasts three general groups (Table I). Mutants in the first group
that can readily convert into chloroplasts when exposed to have so far been identified in severa1 Arabidopsis loci
light. This process is known as greening or de-etiolation. In (copl, detl, and cop8-15) (Chory et al., 1989; Deng et al.,
addition, etiolated seedlings display a very different gene 1991; Wei and Deng, 1992; Wei et al., 1994b; Miséra et al.,
expression pattern from that determined by light. After the 1994; Kwok et al., 1996). These mutants exhibit a light-
initial elongating growth, the seedlings come to a develop- independent pleiotropic phenotype: dark-grown seedlings
mental arrest in the continuous absence of light. resemble their light-grown siblings in overall morphology,
In higher plants, light-controlled physiological and de- cell and plastid differentiation, and expression pattern of
velopmental responses are mediated through at least three light-regulated genes. This group of mutants will be the
families of photoreceptors: phytochromes, cryptochromes, focus of this review.
or blue-light receptors, and UV-B receptors, depending on The second group of mutants have opened cotyledons
the wavelengths of light to which they are most sensitive. without apical hooks but display normal etioplast devel-
opment and elongated hypocotyls when grown in the dark
Our research is supported by grants from the U.S. Department
(cop2lamp1, cop3lkls1, and cop4) (Chaudhury et al., 1993;
of Agriculture (N.W.) and the National Science Foundation and Hou et al., 1993; Lehman et al., 1996). Whereas cop2 and
National Institutes of Health (X.-W.D.). X.-W.D. is a National cop3 do not significantly affect light-regulated gene expres-
Science Foundation Presidential Faculty Fellow. sion, cop4 shows moderate de-repression of nuclear-
* Corresponding author; e-mail xingwang.deng@yale.edu; fax encoded light-induced gene expression (CAB1, chlorophyll
1-203-432-3854. alb-binding protein) in the dark (Hou et al., 1993).
871872 Wei and Deng Plant Physiol. Vol. 11 2, 1996
Table 1. Summary of negative regulatory loci of photomorphogenesis in Arabidopsis
Dark-Grown
COP/DET Gene Other Names Mutant Seedling Lethality Reis
Null Mutant
Morphology
COPl FUS1, EM8168 Pleiotropic Yes Deng et al., 1991
COP8 FUS8, EMB134 Pleiotropic Y es Wei et a\., 1994
COP9 FUS7, EMB143 Pleiotropic Yes Wei and Deng, 1992
COPlO FUS9, EM8 144 Pleiotropic Yes Wei et al., 1994
COPl I FUS6, EM873 Pleiotropic Yes Wei et al., 1994; Castle and Meinke, 1994
COP12 FUS12 Pleiotropic Y es Miséra et al., 1994; Kwok et al., 1996
COP13 FUSll Pleiotropic Y es Miséra et al., 1994; Kwok et al., 1996
COP14 FUS4 Pleiotropic Yes Miséra et al., 1994; Kwok et al., 1996
COP15 FUS5 Pleiotropic Yes , Miséra et al., 1994; Kwok et al., 7 996
DETl FUS2 Pleiotropic Yes Chory et al., 1989
COPZ AMPl, PTl Open cotyledons No Hou et al., 1993; Lehman et al., 1996; Chaudhury et al., 1993
COP3 HLSl Open cotyledons No Hou et al., 1993; Lehman et al., 1996
COP4 Open cotyledons Unknown Hou et al., 1993; Lehman et al., 1996
DET2 COP7 Short hypocotyl No Chory et al., 1991
DET3 Short hypocotyl Unknown Cabrera y Poch et al., 1993
PRCl Short hypocotyl Unknown Desnos et al., 1996
DIM Short hypocotyl Unknown Takahashi et al., 1995
CPD Short hvDocotvl No Szekeres et al., 1996
The primary seedling phenotype of the third group is a THE PLEIOTROPIC COP/DET LOCl ARE ALSO DEFINED
short hypocotyl when grown in darkness for less than 5 d BY THE FUSCA MUTATIONS
(det2, det3, dim,prcl, cpd) (Chory et al., 1991; Cabrera y Poch
et al., 1993; Takahashi et al., 1995; Desnos et al., 1996; A phenotype common to a11 severe alleles of the first
Szekeres et al., 1996). Significant cotyledon expansion and group of copldet mutants is the accumulation of purple
opening also occurred after extended dark growth (more pigment (anthocyanin) in the mature seed and young seed-
than 1 week) in most of those mutants (except p y c l ) , al- lings, a feature that was used for screening fusca mutants
though no sign of chloroplast development was observed. (Miiller, 1963; Castle and Meinke, 1994; Miséra et al., 1994).
In addition, true leaves can initiate and develop after ex- Detailed characterization of 12 available fusca loci sug-
tended growth in the dark. The short-hypocotyl phenotype gested that 10 of them show pleiotropic photomorphogenic
for the p r c l mutant is specifically associated with the dark- development in the dark (Miséra et al., 1994; Kwok et al.,
grown seedling, since the hypocotyl growth of the light- 1996), among which 6 are allelic to the pleiotropic COPI
grown mutant is still under normal light control. This DET loci (COPZ, DETZ, COP8-12) mentioned above and 4
indicates that the mechanism underlying hypocotyl elon- define nove1 loci (Kwok et al., 1996). Thus, a total of 10
gation in the dark is different from that in the light (Desnos pleiotropic COPIDETIFUS loci have been identified.
et al., 1996). When grown in darkness, the copldetlfis mutants exhibit
Unlike the first group of mutants, the second and third almost a11 aspects of the phenotype normally observed in
groups of mutants exhibit partia1 light-dependent seedling light-grown seedlings (Fig. 1). The dark-grown mutants
development in darkness. Since many other factors also affect exhibit short hypocotyls and open cotyledons without api-
different aspects of seedling morphogenesis, such as hor- cal hooks. Cotyledon cell enlargement and differentiation
mones, nutrients, and metabolites, it is likely that some loci in resemble those of wild-type, light-grown plants. Likewise,
these two groups may define regulatory components directly instead of etioplasts, the plastids develop into an interme-
related to those signaling pathways. For example, cop3 has diate form that looks like it is on its way to becoming a
turned out to be allelic to hooklessl ( k l s l ) (Lehman et al., 1996), chloroplast. Tissue-specific plastid differentiation is also
which was identified based on ethylene responsiveness, and affected, since greening and chloroplast development have
ampl (allelic to cop2) was isolated based on altered cytokinin also been observed in mutant roots under light conditions.
responsiveness (Chaudhury et al., 1993). cop4 lacks a normal This is also accompanied by ectopic expression of photo-
gravitropic response (Hou et al., 1993); therefore, COP4 may synthesis-related genes in roots and the chalcone synthase
also be involved in gravity response. More recently, it has gene in mesophyll cells of leaves (Chory and Peto, 1990).
been shown that DET2 and CPD (CYP90)encode enzymes in Moreover, the pattern of gene expression is similar to that
the biosynthetic pathway of brassinosteroids (Li et al., 1996; of light-grown siblings. Both nuclear- and chloroplast-
Szekeres et al., 1996); thus, the light regulatory network may encoded photosynthesis-related genes are de-repressed in
be interlinked with other signaling pathways and light might the dark and in dark-adapted plants. This pleiotropic phe-
exert its effect on an aspect of seedling morphology through notype implies that the light switch controlling photomor-
the action of hormones or other factors. Therefore, some of phogenic and skotomorphogenic programs is no longer
these less pleiotropic loci may represent components of the functional in these mutants, resulting in a constitutive de-
"cross-talk" between the light regulatory network and other fault photomorphogenic developmental pathway (McNel-
signaling pathways. lis and Deng, 1995).COP/DET/FUS Genes in Light Regulation 873
Within the context of this genetic model, it is not sur-
prising that all severe mutations also lead to a fusca phe-
notype, since anthocyanin accumulation is one of the light-
inducible traits that increases quantitatively with higher
light intensity. Therefore, a complete loss of function of
these repressive components would mimic the action of the
highest light intensity extreme, under which condition in-
creased anthocyanin accumulation would be expected. In
addition, some other characteristics of the photomorpho-
genic fusca, such as very short hypocotyl and severe growth
retardation, resemble those of highly photostressed seed-
lings. The lethality offuscn mutant alleles after the seedling
stage may result from the accumulated stress and/or may
indicate that, in addition to their function as repressers of
photomorphogenesis in darkness, the COP/DET/FUS loci
are essential for adult plant development in the light.
Whereas the pleiotropic COP/DET/FUS loci may be
involved also in other regulatory processes, several lines
of evidence support the conclusion that they are specif-
ically involved in the light control of development and
that they represent integral parts of the light regulatory
network. First, weak mutations in copl and detl loci
produce rather specific and striking defects in light re-
sponses while allowing for the completion of the life
cycle. Second, the only detectable phenotype of seedlings
overexpressing COP1 is a partial etiolated response un-
der specific light conditions (McNellis et al., 1994b). The
third and most recent research shows that overexpres-
sion of a COP1 N-terminal 282-amino acid fragment
(N282) results in a dominant negative effect specifically
Figure 1. Comparison of the phenotypic characters of a representative
associated with the photomorphogenic development of
pleiotropic cop mutant with wild-type seedlings. The light-grown wild-
type seedling (left), dark-grown wild-type seedling (middle), and a typ-
the seedlings (McNellis et al., 1996). This includes chlo-
ical dark-grown pleiotropic cop mutant seedling (right) are shown at the roplast-like plastid development and activation of nor-
top. The panels below illustrate the characteristics of the cotyledon mally light-induced genes in the dark, whereas the effect
surface cell morphology, plastid morphology, and expression patterns of on the expression of stress- and pathogen-inducible
two selected nuclear-encoded (RBCS) and plastid-encoded (PSBA) genes is minimal.
genes of the corresponding seedlings on the top.
FUNCTIONAL RELATIONSHIP OF THE PLEIOTROPIC
THE COP/DET/FUS GENES ARE ESSENTIAL FOR COP/DET/FUS GENES
REPRESSION OF PHOTOMORPHOGENIC
DEVELOPMENT IN DARKNESS
Two general hypotheses have been proposed to explain
the nearly identical phenotypes in all 10 pleiotropic cop/det/
The recessive nature of these pleiotropic cop/det/fus mu- fus mutants (Castle and Meinke, 1994; Misera et al., 1994;
tations is consistent with their wild-type gene products McNellis and Deng, 1995). One hypothesis is that each of
acting to repress photomorphogenesis in the dark and light the genes acts in parallel pathways and is independently
signals releasing such repressive activity. Although this required to repress photomorphogenic development, and
genetic model is only one way of interpreting the mutant the other hypothesis is that these genes work in proximity
phenotype, it has gained support from additional molecu- to each other in the same regulatory pathway. Recently,
lar and genetic studies. Transgenic Arabidopsis seedlings several lines of evidence supporting the latter hypothesis
overaccumulating COP1 show a partially etiolated pheno- have been reported, as described below.
type: longer hypocotyl under dark/light-cycled white light Genetically, synthetic lethality and phenotype enhance-
as well as continuous far-red or blue-light conditions, dem- ment between two weak and viable alleles of copl and detl
onstrating the suppressor activity of COP1 on photomor- loci have been observed (Ang and Deng, 1994), implying
phogenic development (McNellis et al., 1994b). Genetic that the two loci act in the same pathway. Biochemical
analysis of double mutants between the pleiotropic cop/det/ characterization of COP9 indicated that it exists in a com-
fus mutations and the photoreceptor mutations suggested plexed form (Wei et al., 1994b). In addition, the COP9
that these gene products act downstream of photoreceptors complex does not accumulate in cop8 and copll mutants,
(Fig. 2; Chory et al, 1989; Deng et al., 1991; Wei and Deng, which suggests that both COPS and COP11 are required for
1992; Misera et al., 1994; Wei et al., 1994b; Kwok et al., the formation or the stability of the complex. This obser-
1996). vation demonstrates that at least the three genes, COPS,874 Wei and Deng Plant Physiol. Vol. 1 1 2, 1996
Phytochromes + FHY1, FHY3 + ?i COPl, DET1 ?
--+ PRC1
Plastid development
Hypocotyl growth
COP8, COP9
LIGHT I)Cryptochromes COP10, COP11
(blue-light receptors) FUS12, FUS13 Activation of
FUS13, FUS14 -.+7 Phytohormone+ shoot meristem
UV-B receptors
-+ HY5
Other
Signals
2hways CotyledonILeave
Development
Figure 2. Diagram of hypothetical light-signaling cascade during seedling development in Arabidopsis. Light is perceived
by three families of photoreceptors: phytochromes, cryptochromes, and UV-B receptors. The signals are transmitted through
receptor-specific early-signaling components such as FHYI and FHY3 (phytochrome A-specific) and eventually merge to the
central switch as represented by 10 COP/DET/FUS products and HY5. The central processor then brings about the light
responses through interaction with putative response-specific effectors such as PRC1 and A T H I . It may also cross-talk with
phytohormone pathways to convey its effect in some responses. It should be emphasized that phytohormones are able to
influence overlapping biological processes during seedling development independently of light signals. The relative position
of signaling molecules including Ca2+,calmodulin, G proteins, and cGMP are undetermined in this model. The arrows
denote the flow of information and do not indicate positive or negative interactions in genetic terms. Question marks and
dashed arrows indicate speculative paths with no experimental evidence either positive or negative.
COP9, and COP11, act in the same pathway. Last, the COP9 they correspond to any pleiotropic COPIDETIFUS genes.
complex has been shown to play a role in nuclear import or Although immunoblot analysis with COPl antibodies in-
nuclear retention of GUSCOPl (Chamovitz et al., 1996), dicates that COPl is not part of the purified COP9 complex,
connecting COPl with the COP9 complex in the same it remains to be addressed whether COPl and others could
pathway . be part of the larger, light-labile version of the "dark"
COP9 complex, which was observed in the dark-grown
SOME OF THE PLEIOTROPIC COP/DET/FUS CENES Arabidopsis seedlings (Wei et al., 199410).
ARE DEFINED AS A MULTISUBUNIT Both immunostaining with the COP9 antibodies and his-
NUCLEAR COMPLEX tochemical staining of GUS-COP9 fusion protein suggest
that COP9 is nuclear localized in both Arabidopsis and
Two approaches have independently provided definitive cauliflower (Chamovitz et al., 1996; Staub et al., 1996).Since
evidence that multiple pleiotropic COPIDETIFUS genes COP9 exists only in the complexed form (Wei et al., 1994a),
encode subunits of the COP9 complex. Using antibodies the COP9 complex must be nuclear. It is interesting to note
raised against COPll (also called FUS6, Castle and Meinke, that, despite its lack of recognizable nuclear localization
1994), Staub et al. (1996) found that COP9 and COPll signals, GUS-COP9 can be correctly localized into nuclei of
co-fractionated in the same large size fractions (560 kD) root cells of transgenic plants. Therefore, it is possible that
and co-immunoprecipitated each other from total plant partia1 or full assembly of the COP9 complex, which would
extract. Furthermore, biochemical purification of the COP9 recruit other subunit(s) containing nuclear localization sig-
complex from cauliflower, a close relative of Arabidopsis nals, may be prerequired for its nuclear translocation. Con-
(Chamovitz et al., 1996),revealed that the complex is spher- sistent with this possibility, it has been observed that,
ical, about 12 nm in diameter, and contains 12 subunits whereas GUS-COP9 displays a nuclear staining in roots of
with equal molar amounts. Direct amino acid sequencing copl, d e t l , cop9, and copl0 mutants, it is cytoplasmic in cop8
of the individual proteins in the complex not only con- and copll (Chamovitz et al., 1996), both of which probably
firmed the identity of the expected COP9 subunit but also represent mutations in genes encoding subunits of the
revealed that the COPll is also a subunit in the purified COP9 complex.
complex. Together with the above biochemical observation
in Arabidopsis, co-purification of COP11 with COP9 from HOW DOES LICHT RECULATE
cauliflower provides the best proof that multiple pleiotro- COP/DET/FUS ACT I V ITY 1
pic COPIDETIFUS genes encode subunits of the same com-
plex. It will be interesting to determine the identities of The genetic model predicts that when light signals are
other subunits of the COP9 complex and to see whether perceived by photoreceptors they are transduced to abro-COf/DET/FUS Genes in Light Regulation 875
gate the suppressive activities of the COP/DET/ FUS the nuclear GUS-COP1 level appears to correlate quantita-
products over photomorphogenic development. Recent tively with the extent of suppression of photomorphogenic
cloning of four of these genes, COPl (Deng et al., 1992), development in hypocotyl cells. In root cells, however,
FUS6/ COPll (Castle and Meinke, 1994), DETl (Pepper GUS-COP1 is constitutively nuclear, which is consistent
et al., 1994), and COP9 (Wei et al., 1994a), provide tools with the established role of COPl in suppressing root
for molecular analysis of this light-inactivation mecha- chloroplast development in both light and darkness. There-
nism. Analysis of the expression levels of those gene fore, it is hypothesized that COPl acts inside the nucleus to
products showed that protein accumulation of COP1, repress photomorphogenesis and that light modulation of
COP9, and COPll and the mRNA level of DETl are not COPl activity involves a tissue-type-specific nucleocyto-
subject to light regulation and are ubiquitously present plasmic repartitioning (Fig. 3). Since the transgene encod-
in most, if not all, tissue types (McNellis et al., 1994a; ing the GUS-COP1 fusion protein can completely rescue a
Pepper et al., 1994; Wei et al., 1994a; Staub et al., 1996). nu11 mutant of copl and thus is functional (A.G. von Arnim
Therefore, light likely modulates their activities post- and X.-W. Deng, unpublished data), the localization pat-
translationally through protein-protein interaction, pro- tern of GUS-COP1 is likely to be biologically relevant.
tein modification, subcellular localization, or a combina- The light control over GUS-COP1 nucleocytoplasmic
tion of these processes. partitioning has the following implications: First, COPl
Aside from the COP9 complex (both COP9 and COPll), exclusion from the nucleus under light may represent a
examination of the subcellular localization of fusion pro- mechanism for disabling COPl's suppressive function in
teins between GUS reporter and COPl or DETl suggested photomorphogenesis. In the dark COPl acts as a nuclear
that they are also nuclear regulators (Pepper et al., 1994; regulator, which likely turns off the photomorphogenic
von Arnim and Deng, 1994). Whereas the nuclear location program by interacting with some yet unknown targets. By
of the COP9 complex and DETl seems to be independent of reducing its abundance in the nucleus, light makes COPl
light conditions, the nucleocytoplasmic partitioning of less effective in interacting with its intended targets; thus,
COPl is regulated by light in a cell-type-dependent man- it becomes less able to suppress photomorphogenic devel-
ner (von Arnim and Deng, 1994; Fig. 3). In hypocotyl cells opment. By regulating the relative nuclear abundance of
of stably transformed Arabidopsis seedlings and epidermal COP1, this mechanism could easily achieve a quantitative
cells of onion bulb, the GUS-COP1 protein is enriched in suppression of photomorphogenic development under dif-
the nucleus in darkness and is excluded from the nucleus ferent intensities of light. Furthermore, this model also
under constant white light. When a dark-grown seedling is provides a framework for examining the functional role of
exposed to light, the protein slowly repartitions from a other pleiotropic COPIDETIFUS genes (Fig. 2). For exam-
nuclear to a cytoplasmic location and vice versa for a ple, GUS-COP1 was unable to localize to nuclei of dark-
light-grown seedling transferred into darkness. Moreover, grown hypocotyl cells in those pleiotropic cop mutants
defective in the COP9 complex (Chamovitz et al., 1996).
This observation suggests that the COP9 complex is in-
volved in COPl nuclear translocation or nuclear retention
in darkness. This role of the COP9 complex would be
consistent with the phenotype of their mutations.
The slow kinetics of the GUS-COP1 nuclear localization
during a light-dark transition (von Arnim and Deng, 1994)
would imply that its changed nuclear abundance is un-
likely to be the primary cause of the initial development
switch, but more likely it is a way of maintaining COPl in
an inactive state. It is possible that when a dark-grown
seedling is exposed to light COPl may be initially tran-
siently inactivated by a still unknown mechanism. Subse-
quent exclusion of COPl from the nucleus would serve as
an efficient way to prevent its suppressive action on pho-
tomorphogenic development under constant light. Mean-
while, the COP1-interactive protein CIPl (Matsui et al.,
1995), which displays a cytoskeleton-type cellular distribu-
tion pattern, has the potential to be involved as a cytoplas-
Figure 3. Hypothetical model illustrating the relationship of the mic anchor protein in regulating access of COPl to the
pleiotropic COP/DET/FUS gene products. All of the gene products
nucleus through protein-protein interactions.
are proposed to act within the nucleus and together lead to repres-
sion of photomorphogenic seedling development. Light signals dis- IMPLICATIONS FOR MULTIPLE FUNCTIONAL
rupt those nuclear interactions, their ability to repress photomorpho-
MODULES IN COPl
genesis, and a concomitant repartitioning of COPl between the
nucleus and cytosol. Note that the dashed arrows between the COPl The central role of COPl as a light-inactivatible repressor
in nuclear and cytosolic compartments indicate the direction of of photomorphogenesis warrants some discussion about
concentration change and not necessarily the direction of movement the structural features of its encoded protein. Whereas
of the C O P l itself. COP9, DET1, and FUSGICOPI 1 encode nove1 proteins that876 Wei and Deng Plant Physiol. Vol. 112, 1996
have no obviously identifiable homology to other known gene expression is independent of chloroplast develop-
proteins or motifs, COPl has a nove1 combination of the ment and the light induction is transient. It is interesting
following recognizable motifs: a ring-finger zinc-binding that mutations in copZ and detl disrupt its light-dependent
domain at the amino-terminal region, followed by a coiled- expression, resulting in the accumulation of ATHZ mRNA
coil helix structure, and several WD-40 repeats at the in the dark. Many proteins with homeodomains are known
carboxyl-terminal (Deng et al., 1992; McNellis et al., 1994a). as transcriptional regulators, capable of controlling devel-
The coiled-coil and WD-40 repeats are likely surfaces for opmental patterns through modulating meristematic cell
mediating protein-protein interactions, whereas the zinc division activities at the site of primordia formation (Hake
finger domain has the potential to interact with either or et al., 1995). Accordingly, the ATHZ gene contains a Pro-
both proteins and nucleic acids. The multiple interactive rich region and two Ser / Thr-rich regions that could func-
motifs of COPl provide a physical basis for its being a tion as transcriptional activation domains (Quaedvlieg et
central player in interacting with multiple upstream and al., 1995). Therefore, it is possible that ATHZ could be one
downstream components in the light regulatory cascades. of the downstream targets of COP/DET/FUS proteins,
Examination of the available ethyl methanesulfonate- which, when induced in the light, activates a chloroplast-
induced recessive copl alleles (McNellis et al., 1994a) re- independent morphogenic program such as cell division in
vealed that subtle changes in the WD-40 repeat regions of shoot meristems. Similarly, the expression of Arabidopsis
COPl protein lead to a severe phenotype similar to com- ATHB-2 and ATHB-4, both of which encode proteins con-
plete loss-of-function mutations. However, the copl-4 mu- taining homeodomains and Leu zipper motifs, are strongly
tation, which results in a COPl protein with only the 282 induced by far-red-rich light treatment (Carabelli et al.,
N-terminal amino acids (N282) at about 10% of wild-type 1993, 1996). However, it has not been reported whether
level and a complete remova1 of the WD-40 domain, pro- copldetljius mutations affect the gene expression of ATHB-2
duces one of the weakest phenotypic defects. This indicates and ATHB-4.
that the N282 protein itself retains significant activity, Since photomorphogenesis is a complex process involv-
whereas imperfections in the WD-40 repeat regions are ing many cell-type-specific and organ-specific morpho-
detrimental to this function. Consistent with this idea, genic responses, the pleiotropic COPIDETIFUS products
when the N282 portion of COPl is expressed in transgenic probably have multiple downstream effectors. For exam-
seedlings, it produced a dominant negative phenotype that ple, COP /DET/ FUS may transduce light signals and influ-
is partially de-etiolated in the dark, with hypocotyl elon- ente an aspect of morphogenic development by interacting
gation that is hypersensitive to light inhibition (McNellis et with phytohormone pathways and / or other regulatory
al., 1996). The ability of the N282 protein to mask endoge- pathways. In fact, some of the negative regulators listed in
nous COPl function could be interpreted to mean that the second and third groups in Table I may represent those
N282 itself may contain the modules responsible for inter- downstream effectors controlling specific aspects of the
acting with downstream targets of COPl and that the morphogenic responses. The recent findings by Szekeres et
phenotype results from depletion of COPl downstream al. (1996) and Li et al. (1996) that DET2 and CPD are
targets due to the presence of high levels of N282. involved in the brassinosteroid .biosynthetic pathway and
that application of brassinolide can partially compensate
CANDIDATES FOR DOWNSTREAM EFFECTORS OF the hypocotyl phenotype of most copldetlfus mutants in an
COP/DET/FUS PROTEINS allele-specific manner would be consistent with this spec-
ulation. However, it is also possible that light and brassi-
As nuclear regulators, any COP/DET/FUS protein has nosteroids independently affect hypocotyl growth during
the potential to directly affect the expression pattern of seedling development. To distinguish between these hy-
genes responsible for photomorphogenesis or skotomor- potheses, it is necessary to address whether light affects the
phogenesis. This can be achieved in a variety of ways. For cellular level (production/ degradation), perception, and
example, a given COP/DET/FUS protein or complex can signal transduction of brassinosteroids, or whether photo-
modulate the expression pattern of light-regulated genes morphogenic copldetlfisca mutants have a lower than nor-
by directly interacting with the light-responsive promoters, mal level of these molecules.
or specific transcription factors, which bind to light-
responsive promoters, or by regulating the expression of
COP/DET/FUS MAY REPRESENT A CROUP
genes encoding the transcription factors. These possibilities
OF EVOLUTIONARILY CONSERVED
can be investigated by identifying either direct protein-
DEVELOPMENTAL REGULATORS
protein interactive partners (Matsui et al., 1995) or their
potential DNA-binding sites. Although no conclusive evi- Although the COPIDETIFUS genes were identified by
dente for either of the possibilities has been provided yet, their role in light-regulated development in Arabidopsis,
several transcription factors whose expression is regulated homologous genes have been found in other organisms,
by light could represent potential candidates. including human, mouse, and Caenorhabditis elegans,
One appealing candidate is ATH1, which encodes a ho- mostly through genome sequencing (Chamovitz and Deng,
meodomain protein whose expression is transiently in- 1995). Although these findings have not necessarily pro-
duced by light in Arabidopsis seedlings (Quaedvlieg et al., vided additional clues for the specific functions of these
1995; Fig. 2). However, unlike many photosynthesis-related genes, it suggests that they may play roles in the develop-
genes such as C A B (chlorophyll alb-binding protein), ATHZ ment of other organisms. It is particularly worth notingCOP/DET/FUS Genes in Light Regulation 877
that Arabidopsis COP9 and COP11, t h e only t w o known Castle L, Meinke D (1994) A F U S C A gene of Arabidopsis encodes a
subunits of the COP9 complex, h a v e a 67 a n d 65% similar- novel protein essential for plant development. Plant Cell 6 25-41
ity t o t h e h u m a n counterparts, respectively, across t h e Chamovitz DA, Deng X-W (1995) The novel components of the
Arabidopsis light signaling pathway may define a group of
entire length of the proteins. This degree of conservation general developmental regulators shared by both animal and
may imply a homologous cellular function. It seems possi- plant kingdoms. Cell 82: 353-354
ble that t h e pleiotropic COPIDETIFUS genes represent a n Chamovitz DA, Deng X-W (1996) Illuminating signals. Curr Opin
evolutionarily conserved g r o u p of developmental regula- Plant Biol (in press)
tors, which i n higher plants w e r e recruited t o regulate Chamovitz DA, Wei N, Osterlund MT, von Arnim A, Staub JM,
Matsui M, Deng X-W (1996) The Arabidopsis COP9 complex, a
photomorphogenic development. Perhaps t h e knowledge novel multisubunit nuclear regulator involved in light control of
gained f r o m t h e studies of these pleiotropic COPIDETIFUS a developmental switch. Cell 8 6 115-121
genes will provide guidance for understanding t h e func- Chaudhury AM, Letham S, Craig S, Dennis ES (1993) ampl: a
tion of their h u m a n counterparts. mutant with high cytokinin levels and altered embryonic pat-
tern, faster vegetative growth, constitutive photomorphogen-
esis, and precocious flowering. Plant J 4: 907-916
CONCLUDING REMARKS Chory J, Nagpal P, Peto CA (1991) Phenotypic and genetic anal-
ysis of det2, a new mutant that affects light-regulated seedling
Although t h e initial genetic identification of t h e pleiotro- development in Arabidopsis. Plant Cell 3: 445-459
pic COPIDETIFUS genes w a s reported m o r e t h a n 30 years Chory J, Peto CA (1990) Mutations in the DETl gene affect cell-
ago (Miiller, 1963), the recent rediscovery of this class of type-specific expression of light-regulated genes and chloroplast
genes i n photomorphogenic screens resulted i n significant development in Arabidopsis. Proc Natl Acad Sci USA 8 7 8776-
advances i n o u r understanding of their physiological a n d 8780
Chory J, Peto C, Feinbaum R, Pratt L, Ausubel F (1989)Avabidopsis
cellular functions. Genetic, molecular, a n d cell biological fhaliana mutant that develops as a light-grown plant in the
studies h a v e not only revealed the molecular identity of absence of light. Cell 58: 991-999
four of these genes b u t also led t o m a n y insights concern- Deng XW (1994) Fresh view of light signal transduction in plants.
ing their roles i n light-regulated development, including Cell 76: 423-426
cellular localization of action, possible light regulation, and Deng XW, Caspar T, Quail PH (1991) copl: a regulatory locus
involved in light-controlled development and gene expression
functional interactions. These advances should provide fer- in Arabidopsis. Genes Dev 5: 1172-1182
tile g r o u n d for future studies designed t o elucidate the Deng XW, Matsui M, Wei N, Wagner D, Chu AM, Feldmann KA,
overall regulatory network and t o understand the molecu- Quail PH (1992)COP1, an Arabidopsis regulatory gene, encodes
lar interactions involved. Many obvious questions need t o a protein with both a zinc-binding motif and a G beta homolo-
be answered: W h a t a r e t h e interaction partners of the gene gous domain. Cell 71: 791-801
Desnos T, Orbovic V, Bellini C, Kronenberger J, Caboche M,
products already identified? W h a t a r e their biochemical Traas J, Hofte H (1996) Procusfel mutants identify two distinct
interactions a n d activities? How do they regulate their genetic pathways controlling hypocotyl cell elongation, respec-
d o w n s t r e a m targets? H o w a r e light signals from multiple tively, in dark- and light-grown Arabidopsis seedlings. Develop-
receptors integrated t o modulate their activities? How do ment 122: 683-693
Furuya M (1993) Phytochromes: their molecular species, gene
they relate to those light-signaling components s u c h a s
families, and functions. Annu Rev Plant Physiol Plant Mo1 Biol
trimeric G proteins, Caz+ / calmodulins, a n d cGMP (Bowler 4 4 617-646
a n d Chua, 1994), which a r e defined biochemically. Indeed, Hake S, Char BR, Chuck G, Foster T, Long J, Jackson D (1995)
t h e most exciting time i s y e t t o come. , Homeobox genes in the functioning of plant meristems. Philos
Trans R SOCLond-Biol Sci 350 45-51
Hou Y, von Arnim AG, Deng X-W (1993) A new class of Arabi-
ACKNOWLEDCMENTS dopsis constitutive photomorphogenic genes involved in regu-
We would like to thank Professor Arthur Galston and Dr. Daniel lating cotyledon development. Plant Cell 5: 329-339
Chamovitz for reading and commenting on this manuscript. Kendrick RE, Kronenberg GHM (1994) l'hotomorphogenesis in
Plants. Kluwer, Dordrecht, The Netherlands
Koornneef M, Rolff E, Pruitt CJP (1980) Genetic control of light-
Received May 22, 1996; accepted August 2, 1996. inhibited hypocotyl elongation in Avabidopsis thaliana (L.)
Copyright Clearance Center: 0032-0889/ 96111210871/ 08. Heyhn. Z Pflanzenphysiol 100: 147-160
Kwok SF, Piekos B, Miséra S, Deng X-W (1996) A complement of
LITERATURE ClTED 10 essential and pleiotropic Arabidopsis genes are necessary for
suppression of photomorphogenesis in darkness. Plant Physiol
Ang LH, Deng XW (1994) Regulatory hierarchy of photomorpho- 110: 731-742
genic loci: allele-specific and light-dependent interaction be- Lehman A, Black R, Ecker JR (1996) HOOKLESSZ, an ethylene
tween the HY5 and COPl loci. Plant Cell 6 613-628 response gene, is required for differential cell elongation in the
Bowler C, Chua NH (1994) Emerging themes of plant signal Arabidopsis hypocotyl. Cell 85: 183-194
transduction. Plant Cell 6: 1529-1541 Li J, Nagpal P, Vitart V, McMorris TC, Chory J (1996) A role for
Cabrera y Poch HL, Peto CA, Chory J (1993) A mutation in the brassinosteroids in light-dependent development of Arabidop-
Arabidopsis DET3 gene uncouples photoregulated leaf develop- sis. Science 272: 398-401
ment from gene expression and chloroplast biogenesis. Plant J 4: Matsui M, Stoop CD, von Arnim AG, Wei N, Deng XW (1995)
671-682 Arabidopsis COPl protein specifically interacts in vitro with a
Carabelli M, Morelli G, Whitelam G, Ruberti I (1996) Twilight- cytoskeleton-associated protein, CIP1. Proc Natl Acad Sci USA
zone and canopy shade induction of Athb-2 homeobox gene in 92: 42394243
green plants. Proc Natl Acad Sci USA 93: 3530-3535 McNellis TW, Deng X-W (1995) Light control of seedling mor-
Carabelli M, Sessa G, Baima S, Morelli G, Ruberti I (1993) The phogenetic pattern. Plant Cell 7: 1749-1761
Arabidopsis Athb-2 and -4 genes are strongly induced by far-red McNellis TW, Torii KU, Deng X-W (1996) Expression of an N
rich light. Plant J 4: 469-479 terminal fragment of COPl confers dominant-negative suppres-878 Wei and Deng Plant Physiol. Vol. 112, 1996
sion of seedling photomorphogenic development in transgenic Szekeres M, Németh K, Koncz-Kálmán Z, Mathur J, Kaus-
Arabidopsis. Plant Cell 8: 1491-1503 chmann A, Altman T, Rédei GP, Nagy F, Schell J, Koncz C
McNellis TW, von Arnim AG, Araki T, Komeda Y, Misera S, (1996) Brassinosteroids rescue the deficiency of CYP90, a cyto-
Deng XW (1994a) Genetic and molecular analysis of an allelic chrome P450, controlling cell elongation and de-etiolation in
series of copl mutants suggests functional roles for the multiple Arabidopsis. Cell 85: 171-182
protein domains. Plant Cell 6: 487-500 Takahashi T, Gasch A, Nixhizawa N, Chua N-H (1995) The
McNellis TW, von Arnim AG, Deng XW (199413)Overexpression DIMINUTO gene of Arabidopsis is involved in regulating cell
of Arabidopsis COPl results in partia1 suppression of light- elongation. Genes Dev 9: 97-107
mediated development: evidence for a light-inactivable repres- Vierstra RD (1993) Illuminating phytochrome functions. Plant
sor of photomorphogenesis. Plant Cell 6 1391-1400 Physiol 103: 679-684
Miséra S, Miiller AJ, Weiland-Heidecker U, Jurgens G (1994) The von Arnim AG, Deng XW (1994) Light inactivation of Arabidopsis
FUSCA genes of Arabidopsis: negative regulators of light re- photomorphogenic repressor COPl involves a cell-specific reg-
sponses. Mo1 Gen Genet 244 242-252 ulation of its nucleocytoplasmic partitioning. Cell 79: 1035-1045
Miiller AJ (1963) Embryonentest zum nachweis rezessiver letal- von Arnim AG, Deng XW (1996) Light control of seedling devel-
faktoren bei Arabidopsis thaliana. Biol Zentralbl 82: 133-163 opment. Annu Rev Plant Physiol 47: 215-243
Pepper A, Delaney T, Washburn T, Poole D, Chory J (1994)DET1, Wei N, Chamovitz DA, Deng XW (1994a) Arabidopsis COP9 is a
a negative regulator of light-mediated development and gene component of a novel signaling complex mediating light control
expression in Arabidopsis, encodes a novel nuclear-localized
of development. Cell 78: 117-124
protein. Cell 78:109-116
Wei N, Deng X-W (1992) COP9: a new genetic locus involved in
Quaedvlieg N, Dockx J, Rook F, Weisbeek P, Smeekens S (1995)
light-regulated development and gene expression in Arabidop-
The homeobox gene ATHl of Arabidopsis is derepressed in pho-
sis. Plant Cell 4: 1507-1518
tomorphogenic mutants copl and detl, Plant Cell 7: 117-129
Quail PH (1994) Photosensory perception and signal transduction Wei N, Kwok SF, von Arnim AG, Lee A, McNellis TW, Piekos B,
in plants. Curr Opin Genet Dev 4: 652-661 Deng XW (199413) Arabidopsis COP8, COP10, and COP11 genes
Quail PH, Boylan MT, Parks BM, Short TW, Xu Y, Wagner D are involved in repression of photomorphogenic development in
(1995) Phytochromes: photosensory perception and signal trans- darkness. Plant Cell 6 629-643
duction. Science 268: 675-680 Whitelam GC, Johnson E, Peng J, Caro1 P, Anderson ML, Cowl
Staub JM, Wei N, Deng X-W (1996) Evidence for COP11 as a JS, Harberd NP (1993) Phytochrome A nu11 mutants of Arabi-
component of the nuclear localized COP9 complex in Arabidop- dopsis display a wild-type phenotype in white light. Plant Cell
sis. Plant Cell (in press) 5: 757-768You can also read
