Sie sind auf Seite 1von 9

Plant Physiol.

(1 995) 107: 797-805

Phytochrome A Overexpression in Transgenic Tobacco'

Correlation of Dwarf Phenotype with High Concentrations of Phytochrome in Vascular


Tissue and Attenuated Gibberellin Levels

Emily T. Jordan, Peggy M. Hatfield, David Hondred2, Manuel Talon3, Jan A. D. Zeevaart, and
Richard D. Vierstra*
Department of Horticulture, University of Wisconsin, Madison, Wisconsin 53706 (E.T.J., P.M.H, D.H., R.D.V.);
and Michigan State University-Department of Energy Plant Research Laboratory, Michigan State University, East
Lansing, Michigan 48824 (M.T., J.A.D.Z.)

imately 120-kD polypeptide. The regulatory functions of


Phytochromes are a family of related chromoproteins that regu- phytochromes require that they switch between two pho-
late photomorphogenesis in plants. Ectopic overexpression of the tointerconvertible forms, a Pr form that primarily absorbs
phytochrome A in severa1 plant species has pleiotropic effects, red light and a Pfr form that primarily absorbs far-red light.
including substantial dwarfing, increased pigmentation, and de- They are initially synthesized as Pr, which in most cases is
layed leaf senescence. We show here that the dwarf response is biologically inactive (Liscum and Hangarter, 1993; Ken-
related to a reduction i n active gibberellins (CAs) in tobacco (Nico- drick and Kronenberg, 1994).Upon photoconversion to Pfr,
fiana fabacum) overexpressing oat phytochrome A under the con-
phytochromes become biologically active and govern a
trol of the cauliflower mosaic virus (CaMV) 35s promoter and can
diverse array of physiological and developmental re-
be suppressed by foliar applications of gibberellic acid. I n trans-
genic seedlings, high concentrations of oat phytochrome A were
sponses that span the life cycle of plants, including initia-
detected in stem and petiole vascular tissue (consistent with the tion of seed germination, chloroplast biogenesis, regulation
activity of the CaMV 35s promoter), implicating vascular tissue as of stem growth, floral induction, and timing of senescence.
a potential site of phytochrome A action. To examine the efficacy of In many cases, these responses can be cancelled when Pfr is
this cellular site, oat phytochrome A was also expressed using photoconverted back to Pr. As a result of this unique
Arabidopsis chlorophyll a/b-binding protein (CAB) and the Arabi- interconversion between inactive Pr and active Pfr, phyto-
dopsis ubiquitin ( U B Q 7 ) promoters. Neither promoter was as effec- chromes operate as light-regulated, reversible switches in
tive as CaMV 35s i n expressing phytochrome in vascular tissue or in photomorphogenesis. The mechanismb) by which phyto-
inducing the dwarf phenotype. Collectively, these data indicate that chromes control morphogenesis is unclear, but recent mod-
the spatial distribution of ectopic phytochrome i s important in els have implicated protein kinase signaling cascades, het-
eliciting the dwarf response and suggest that the phenotype is
erotrimeric G-protein activation, and calcium/calmodulin
invoked by elevated levels of the far-red-absorbing form of phyto-
(Vierstra, 1993; Bowler et al., 1994).
chrome within vascular tissue repressing CA biosynthesis.
Attempts to elucidate the molecular mechanism(s) of
phytochrome action have focused, in part, on a biochemical
characterization of the photoreceptors and on identifying
Phytochromes are a family of regulatory photoreceptors structural domains that distinguish Pfr from Pr. However,
in plants that measure both the quantity and quality of this approach has been hampered by the lack of a relevant
available light (for reviews, see Quail, 1991; Vierstra, 1993; in vitro assay of Pfr activity (Quail, 1991) and by the
Kendrick and Kronenberg, 1994). They are dimeric, cyto- complicated assembly of the chromoprotein that is difficult
plasmic proteins, with each subunit consisting of a linear to reproduce with high yields in vitro or in microorganisms
tetrapyrrole chromophore covalently linked to an approx- (Wahleithner et al., 1991; Edgerton et al., 1993; Furuya and
Song, 1994). To overcome these limitations, one strategy
Supported in part by grants from the U.S. Department of now used is to ectopically express phytochrome genes in
Energy (DE-FG02-88ER13968) and the Research Division of the transgenic plants (for review, see Cherry and Vierstra,
University of Wisconsin-College of Agriculture and Life Sciences 1994). In a variety of plant species, expression of unmodi-
(Hatch 2858) to R.D.V and a National Institute of Health Cellular fied PHY genes results in the accumulation of chromopro-
and Molecular Biology Fellowship to E.T.J. teins, which are photochemically indistinguishable from
Present address: ICI Seeds, 2369 330th Street, Box 500, Slater, the endogenous photoreceptors. When expressed to suffi-
IA 50244.
cient levels, the introduced phytochromes are also func-
Present address: Department of Citriculture, Instituto Valen-
ciano de Investigaciones Agrarias, E-46113 Moncada, Valencia, tional and impart a light-exaggerated phenotype, charac-
Spain.
* Corresponding author; e-mail vierstra@macc.wisc.edu; fax Abbreviations: CAL?, chlorophyll alb binding protein; CaMV,
1-608-262-4743. cauliflower mosaic virus.
797
798 Jordan et al. Plant Physiol. Vol. 107, 1995

terized by reduced hypocotyl/stem elongation, decreased Here we provide additional evidence supporting the
apical dominance, and enhanced pigmentation in plants regulation of GA biosynthesis by phytochrome. We show
grown under natural light conditions (Boylan and Quail, that tobacco (Nicotiana tabacum) plants, dwarfed by over-
1989, 1991; Kay et al., 1989; Keller et al., 1989; Cherry et al., expression of oat phytochrome A under the corkrol CaMV
1991; Wagner et al., 1991; Boylan et al., 1994). This pheno- 355 promoter (Keller et al., 1989; Cherry et al., 1991), have
type, when used as an in vivo assay of phytochrome action substantially lower levels of active GAs. Foliar application
coupled with site-directed mutagenesis, has been a power- of GA, was able to reverse the dwarf phenotype. In these
ful tool for mapping domains important for assembly and transgenic plants, high concentrations of oat phytochrome
action. For example, functional domains near the extreme A were found in vascular tissue, implicating vascular tis-
N and C termini whose deletions impair activity or atten- sue as a potential cellular site for altering GA levels. Com-
uate the activity of the endogenous photoreceptors (dom- parison of the light-exaggerated phenotype indiiced by the
inant negative mutations) have been identified by this CaMV 35s promoter to those induced by two other pro-
method (Cherry et al., 1992, 1993; Stockhaus et al., 1992; moters show that this spatial distribution of ectopic phy-
Boylan et al., 1994).Almost a11 work to date has utilized the tochrome is important for eliciting the light-exaggerated
CaMV 355 promoter and not the native PHY promoters to phenotype.
drive expression of exogenous phytochromes. As a result,
MATERIALS A N D METHODS
it has been argued that the light-exaggerated phenotype is
induced by artificially high levels of phytochrome in tis- Plant Material
sues that normally lack the native chromoprotein and thus
Tobacco lines (Nicotiana tabacum, cv Xanthi) expressing
may not reflect phytochrome functioning in its normal
full-length oat phytochrome A under the control of the
capacity. However, a similar phenotype was recently ob-
CaMV 35s promoter were those described by Keller et al.
served when an Arabidopsis phytochrome B was expressed
(1989) and Cherry et al. (1992). The Arabidopsis UBQZ pro-
in Arabidopsis under its own promoter. This supports the
moter was as described by Callis et al. (19913) and the
possibility that the light-exaggerated phenotype represents
CAB140 promoter was obtained courtesy of Dr. Anthony
the photoreceptor functioning in its correct context and
Cashmore (Leutwiler et al., 1986).A 1.4-kb EcoR[/PstI frag-
thus is relevant to normal phytochrome action (Wester et
ment containing the entire CAB promoter was cloned from
al., 1994).
pAB14016LBS into pBluescript KS+ (Stratagene, La Jolla,
Important clues concerning phytochrome function may CA) containing a 40-bp EcoRI/BamHI linker. The linker
be uncovered by examining the physiological basis of the (5-GAATTCGGAGCTGCAGATGTCT - TCCTCP.AGGCCT-
light-exaggerated phenotype. Its pleiotropic nature sug- GGATCC-3)provided codons for the first five amino acids
gests that elevated levels of phytochrome alter hormone of oat phytochrome A ending in a StuI site and introduced
balances and/or sensitivities (Cherry and Vierstra, 1994). a PstI site immediately 5 to the initiation codon. An EcoRI/
Severa1 lines of evidence have specifically implicated GAs. StuI fragment containing the CAB promoter and linker was
For example, transgenic tomatoes expressing high levels of attached in a tripartite ligation to a 323-bp Stu1,KpnI frag-
oat phytochrome A (Boylan and Quail, 1989) are remark- ment encoding amino acids 6 to 113 of oat phytxhrome A
ably similar in phenotype to tomato mutants defective in derived from pBinphyt (Cherry et al., 1992) m d pBlue-
GA biosynthesis (Koornneef et al., 1990). Both types of script KS+ digested with EcoRI/KpnI. The CAB prometer/
plants display a shortened stature, curled leaves, and in- phytochrome-coding region fusion was then cloned as a
creased leaf and fruit pigmentation. In contrast, the Brassica 1.7-kb EcoRIIKpnI fragment into the 14.3-kb IlcoRI/KpnI
ein and the Sorgkum ma,R mutants, which have both unde- fragment of pBinphyt containing the rest of the phyto-
tectable levels of phytochrome B (Childs et al., 1992; Devlin chrome-coding region.
et al., 1992) and elevated levels of GAs (Rood et al., 1990; The UBQZ promoter consisted of a 2.5-kb HindIII/
Beall et al., 1991) have an opposite phenotype character- MZuI fragment (Callis et al., 1990). This region was ap-
ized by substantially elongated stems and hypocotyls. pended to the phytochrome A-coding region using a 93-
Elongated internodes are also observed in Pisum mutants bp SaZI/XkoI oligonucleotide linker. The 1inker was
that have putative defects in the phytochrome signal trans- constructed by annealing and extending two o verlapping
duction chain (Zv [Nagatani et al., 19901and Zipl [Frances et oligonucleotides of 61 and 62 bp (5-GGGGTCGACACGCG-
al., 19921) or that contain aberrantly high levels of GAs (sln TACATTGACATATATAAACCCGCCTCCTCCTTGTTTA-
[Reid et al., 1992a1).Taken together, these data suggest that GGGTTTCTACT-3 and 5-GGGCTCGAGGAC4TCTTTT-
phytochrome responsiveness and GAs are inversely re- GTGTTTC-GTCTTCTCTTAAGTAGAAACCCTf1AACAA-
lated; elevated Pfr levels repress GA biosynthesis and/or GGAGGA-3). The resulting double-stranded linker
sensitivity, which in turn retards stem elongation, whereas contained an interna1 MluI site and codons for the first
reduced Pfr levels elevate GA biosynthesis and/or sensi- three amino acids of oat phytochrome A, which mded with
tivity, which then increases stem elongation. Although a XkoI site. The linker was digested with Sal1 anli XkoI and
some data implicate phytochromes in altering the sensitiv- inserted into the SaZI/XkoI sites of pBluescript KS-, and
ity of plants to GAs ( e g . Pisum), most data suggest that the UBQl promoter was added following digestion of the
phytochromes act by altering GA biosynthesis (Weller et plasmid with HindIII and MZuI. The coding region for oat
al., 1994). phytochrome A was obtained from a 5 reconstruction of
Phytochrome A Overexpression in Tobacco 799

pCV35phyt (Keller et al., 1989) in which a translationally 2 mM PMSF prior to use (Cherry et al., 1993). Samples for
silent XhoI site was introduced 8 bp downstream from the spectrophotometric assays were made 0.1% polyethylenei-
initiator ATG (Cherry et al., 1992; J.R. Cherry and R.D. mine and clarified by centrifugation for 10 min at 48,OOOg.
Vierstra, unpublished data). In a tripartite ligation, a 2.6-kb Phytochrome was concentrated by ammonium sulfate (0.26
HindIII/XhoI fragment containing the UBQZ promoter and g/mL extract) precipitation and resuspended in one-tenth
the N-terminal codons of oat phytochrome A, a 5.0-kb to one-twentieth of the original volume in 25% ethylene
XhoI/SalI encoding the remainder of oat phytochrome A, glycol, 50 mM Tris-HC1 (pH 7.8,4"C), 5 mM Na,EDTA, 14
and pBinl9 digested with SalI/HindIII were ligated to- mM 2-mercaptoethanol, and 2 mM PMSF. Protein content in
gether to create an oat phytochrome A gene under the the clarified extracts was determined by the Bradford
control of the UBQZ promoter. For a11 genes used, intro- method (Bradford, 1976). Spectrophotometric analyses
duced changes in nucleotide sequence were verified by were conducted using a Shimadzu (Kyoto, Japan) UV3000
DNA sequence analysis of the affected regions using the dual-wavelength spectrophotometer. Phytochrome content
dideoxy chain termination method. (A[AA]) was measured as (AA,,, - AA,,,) following satu-
Genes were introduced into tobacco (cv Xanthi) by rating irradiation with red and far-red light and converted
Agrobacterium-mediated transformation (Cherry et al., to (AA665 - AA,,o) using the equivalence factor 2.18 (Cher-
1992). Kanamycin-resistant plants were screened for oat ry et al., 1991).
phytochrome A expression by immunoblot analysis with
polyclonal antibodies against oat phytochrome A (Shanklin
et al., 1987; Cherry et al., 1992). Plants expressing oat lmmunological Techniques
phytochrome A under the control of the CaMV 355, CAB, Oat phytochrome A content was determined in triplicate
and the UBQZ promoters were designated 35S-PHYA, by sandwich ELISA (Cherry, et al., 1992). Affinity-purified
CAB-PHYA, and UBQ-PHYA, respectively. Three indepen- rabbit polyclonal antibodies against oat phytochrome A
dent lines of each were used as the starting material in this (Shanklin et al., 1987) were used as the capture antibody
study. Progenies of initial transformants homozygous for and the murine monoclonal antibody Oat-22, specific for
the inserted genes were identified by their ability to com- oat phytochrome A (Cordonnier et al., 19851, was used as
pletely retain oat phytochrome A expression in subsequent the detection antibody. In this case, the initial crude ex-
generations. Heterozygous plants used in the dose-re- tracts were clarified twice at 48,0008 for 10 min and used
sponse analysis were progenies of the initial transformants. directly. Purified oat phytochrome A, quantitated by dual-
A11 plants were grown in a greenhouse under natural wavelength difference spectroscopy (AtAA,,, - AA,,,]),
diurna1 light cycles. The effects of GA were tested by foliar was used as the standard.
application of various concentrations of GA, (0-100 p~ Tissue printing involved gently pressing 1-mm tissue
solutions containing 100 pL/L of Triton X-100). Spraying slices onto CaC1,-treated nitrocellulose as described by Ye
began 40 d after sowing when the wild-type plants were et al. (1992). Prints were incubated with either polyclonal
approximately 5 cm tall and the 35s-PHYA plants were rabbit antibodies against oat phytochrome A at 0.5 pg/mL
approximately 1 cm tall and continued at 3- to 4-d inter- or preimmune serum diluted 1:lOOO. Primary antibodies
vals. Plant height was measured every 3 to 4 d until the were detected with alkaline phosphatase-conjugated goat
onset of flowering, as determined by the appearance of anti-rabbit immunoglobulins (0.25 pg/mL) and the sub-
flower primordia. For the dose-response analysis of phy- strates 5-bromo-4-chloro-3-indolyl phosphate and ni-
tochrome content, a11 tissue from the apex to the first troblue tetrazolium (Cherry et al., 1992).
13-cm-long leaf (measured from the base of the leaf to the
tip along the midvein) of mature 74-d-old plants was used.
For analysis of the spatial distribution of phytochrome, Quantification of CAs
tissue was harvested at seven different positions: position
1, the apex to leaves 5 to 6 cm in length; position 2, leaves For quantitative analysis of GAs, apical portions from
6 to 7 cm in length; position 3, leaves 10 to 11 cm in length the shoot tip to the first mature leaf were collected from 50
(youngest expanded leaves); position 4, leaves 16 to 17 cm wild-type plants and 50 homozygous 35s-PHYA plants
in length; position 5, leaves 19 to 20 cm in length; position expressing high levels of oat phytochrome A (Cherry et al.,
6, leaves 22 to 23 cm in length (old leaves); and position 7, 1991). The plants were grown for 63 d (following sowing)
stem section between positions 3 and 4. in a greenhouse. GA levels were quantified as described by
Talon et al. (1991). Briefly, the tissue was lyophilized,
weighed, and extracted in methanol. The GAs, [17,17-
Phytochrome Extractions
2H,]GA,, (90% enrichment), [20-2HlGA,, (95.7% enrich-
Phytochrome extractions and analyses were performed ment), [17,17-2H2]GA,, (94.2% enrichment), [17,17-
under a dim green safelight unless otherwise noted. Frozen 2H,]GA,, (99.3% enrichment), and [17,17-2H,]GA, (99.2%
tissue was pulverized at liquid nitrogen temperatures in a enrichment) were added as interna1 standards. GAs were
mortar and pestle. The powder was homogenized at a ratio purified by chromatography via charcoal, silicic acid, and
of 2 mL/g fresh weight in 37.5% ethylene glycol, 75 mM QAE Sephadex, followed by reverse-phase HPLC. The
Tris-HC1 (pH 8.3,4"C), 105 mM ammonium sulfate, and 7.5 samples were methylated and trimethylsilylated, and the
mM Na,EDTA and made 20 mM sodium metabisulfite and derivatized material was quantitated by GC-selected ion
800 Jordan et al. Plant Physiol. Vol. 107, 1995

rnonitoring with magnetic field switching. Yields of inter-


na1 standards were used to quantitate endogenous GAs.
Each analysis was performed in duplicate. The results in
Table I are averages of two replicates that showed similar
results. 100 c
I
WT ~ W M /
A

RESULTS
In this study, we explored the possibility that the light-
exaggerated phenotype induced by ectopic expression of
phytochrome in tobacco is caused, in part, by reduced
levels of GAs. As a simple first test, wild-type plants and
plants expressing high concentrations of oat phytochrome
A driven by the CaMV 35s promoter (35s-PHYA) were
treated with foliar applications of GA, at 3- to 4-d intervals,
and their growth rates were measured. (Although it is
probable that GA, is not an active GA in tobacco, it is likely
converted to the active formIs1 once in the plants.) Typi-
cally, high levels of oat phytochrome A reduces concomi- 40 50 60 70
tantly the growth rate and the final height of mature to-
bacco (cv Xanthi) by nearly 4-fold (Fig. 1; Cherry et al.,
1991, 1992). As can be seen in Figure 1, GA, reversed this
B Time (d)

dwarf phenotype. After a lag of approximately 10 d fol-


lowing the start of GA, application, the growth rate of
35s-PHYA plants increased 386% (from 1.1 to 4.25 cm/d)
when treated with 60 p~ GA, as compared to untreated
35s-PHYA plants. As a result, treated plants eventually
grew to heights comparable to those of untreated wild-type
plants (Fig. 1A). The 35s-PHYA plants still retained ele-
vated levels of oat phytochrome A after GA, treatment,
thus eliminating the possibility that GA, application

--
blocked ectopic phytochrome accumulation. We could de-
tect significant growth enhancement at GA, levels 2 3 p ~ ;
at higher concentrations of GA,, the growth rate asymp-
totically approached that of GA,-treated wild-type plants
(Fig. 1B). For comparison, GA, application had little effect
o L
O 1 10 100
on wild-type plants; plants treated with 510 p~ showed
no significant affect on growth, and the highest leve1 ap- GA3 [pMI
plied (100 FM) increased growth rate by only 23% (from 4.3
to 5.15 cm/d; Fig. 1B). Chl content was also affected by GA, Figure 1. Effect of CA, on the growth rate of wild-type tobacco (WT)
application; levels within 35s-PHYA plants treated with 50 and tobacco expressing oat phytochrome A under the dirlxtion of the
CaMV 35s promoter ( 3 5 5 ) .CA, was sprayed onto the le,wes every 3
p~ GA, declined to those of wild-type plants (data not
to 4 d beginning 40 d after sowing. A, Height of wild-type (O,) . and
shown). Not a11 effects associated with the light-exagger-
35s-PHYA (O, O) tobacco treated with either water Aone (open
ated phenotype were repressed by GA, application. Most symbols) or 60 ~ L GA,
M (filled symbols). Each point represents data
notably, the decrease in apical dominance (as detected by averaged from 10 plants. B, Growth rate of wild-type (0) and 35S-
the development of axillary shoots), a delay in senescence, PHYA (O) tobacco treated with various concentrations of CA,. An
and the reduction in leaf width were still evident in the average growth rate was determined during a 1O-d period (from d 55
GA,-treated 35s-PHYA tobacco even at the highest concen- to d 65) for tobacco treated with each concentration of GA, (see A).
trations used (100 p ~data ; not shown).
Suppression of the dwarf phenotype by GA, suggested dards (Table I). The GA profile in wild-type plants was
that oat phytochrome A represses GA biosynthesis and not indicative of a GA biosynthetic pathway involving conver-
the sensitivity of the plants to the hormone. This possibility sion of the inactive precursors, GA,, and GA,,, into the
was supported by comparing the endogenous GA levels biologically active GAs, GA, and GA,, respectik ely (Table
within wild-type and 35s-PHYA plants. Both wild-type I). A similar GA profile was detected in 35s-PHYA plants
and 35s-PHYA plants were grown simultaneously for 63 d but the total levels of GAs were significantly reduced (ap-
and harvested prior to the development of detectable proximately 40% of wild-type levels). Importantl y, the lev-
flower primordia. At that time, the heights of wild-type els of the two biologically active GAs, GA, and GA4, were
and 35s-PHYA plants were 49.5 5 6.7 and 12.3 ? 3.5 cm, reduced by nearly 4-fold (Table I).
respectively. The profile of GAs from apical tissue was then To help identify the cellular site(s) where ciat phyto-
resolved by GC/MS and quantitated using interna1 stan- chrome A affects tobacco growth and possibly GA levels,
Phytochrome A Overexpression in Tobacco 801

Table 1. GA levels in tobacco expressing oat phytochrome CAB


Wild type, Nontransformed tobacco. 35S, Tobacco overexpressing I
oat Dhvtochrome A under the control of the CaMV 35s Dromoter.
CA Wild Type 35s
ng/g dry wt
35SWild Type-
%
400
300
t n
1.4 0.3 21
0.2 0.2 1 00 200
14.5 7.3 50
17.1 4.4 26 100
21.7 5.7 26
34.3 17.3 n
50
0.6 0.3 50 " 1 2 3 4 5 6 7
9.4 2.7 29
3.r( UBQl
the tissue distribution of the oat chromoprotein in light-
Q 200 1
grown plants was determined. Gross analysis involved 3i4 100
quantitation of phytochrome levels in various locations by
sandwich ELISA using the monoclonal antibody Oat-22, s
% o
1 2 3 4 5 6 7
which is specific for oat phytochrome A (Cordonnier et al.,
1985; Cherry et al., 1992). As can be seen in Figure 2, oat
CaMV 35s

g0
phytochrome A was detected immunologically in a11 tis-
sues examined from 35S-PHYA plants. Immature leaves
contained high levels that declined as the leaves aged (cf.
positions 1 and 6, Fig. 2). Phytochrome A was especially 800
abundant in stem tissue, having nearly twice the levels
found in leaves when calculated relative to total protein
content (position 7).
More precise localization of oat phytochrome A used
tissue printing of light-grown plants (Reid et al., 1992b).
Because we were unable to obtain suitable tissue prints
from leaves and meristems, the study was confined to the
more rigid stem and petiole sections (Fig. 3; data not 4001
shown). Immunodetection used polyclonal antibodies
against oat phytochrome A, which weakly cross-reacted 300
with tobacco phytochromes (Cherry et al., 1991). In the
absence of this primary antibody or when preimmune se-
200
rum was used as a substitute, only a faint ring of brown
100
staining was observed in the tissue prints that was coinci-
dent with the ring of vascular tissue (Fig. 3, A and F). This n
color was distinct from the blue staining derived from the w 1 2 3 4 5 6 7
histochemical reagents used and likely resulted from POSITION
brown phenolic compounds leaching from the sections. In
wild-type plants, weak but immunospecific blue staining Figure 2. Spatial distribution of oat phytochrome A expressed in
of vascular tissue and the epidermis was reproducibly seen tobacco under the control of either Arabidopsis CAB, Arabidopsis
(Fig. 3D). Even though this labeling was not observed ubiquitin (UBQl),or CaMV 35s (35s) promoters. For each promoter,
when wild-type stem sections were probed with preim- three independently transformed lines were selected that expressed a
mune sera (Fig. 3A), it was uncertain whether this pattern range of phytochrome levels, and two plants from each line were
analyzed for phytochrome content (represented by the individual
actually reflected the distribution of endogenous tobacco
bars in the graphs) at various positions within the plant. Horizontal
phytochromes or was artifactual, given the weak affinity of
lines at each position represent the average expression levels of all
the antibody to tobacco phytochromes. However, when six plants in each group. Plants were grown in a greenhouse and
35S-PHYA plants were analyzed, intense blue staining of tissue was harvested 72 d after sowing. Levels of oat phytochrome A
vascular tissue was evident just inside and outside the were determined by sandwich ELISA using monoclonal antibody
vascular ring (Fig. 3C). This ring was coincident with Oat-22, which is specific for the oat phytochrome A polypeptide, and
phloem and the surrounding companion cells (Fig. 3F). As plotted relative to total sample protein. No phytochrome was de-
with wild-type plants, slight immunostaining was ob- tected in wild-type plants by this assay (data not shown). Position 1,
served in the epidermis, but little staining was present in Apex to leaves 5 to 6 cm in length; position 2, leaves 6 to 7 cm in
the cortex, xylem, and pith. Analogous results were ob- length; position 3, leaves 1 O to 1 1 cm in length (youngest expanded
tained with severa1 different transgenic lines expressing leaves); position 4, leaves 16 to 17 cm in length; position 5, leaves 19
to 20 cm in length; position 6, leaves 22 to 23 cm in length (older
35S-PHYA (data not shown).
leaves); and position 7, stem sections between positions 3 and 4.
802 Jordan et al. Plant Physiol. Vol. 107, 1995

in UBQ-PHYA plants were substantially lower (data not


shown). Both ELISA and tissue printing confirmed the
expected location of the oat phytochrome A driven by the
CAB and UBQ1 promoters and showed each to express the
protein in distributions different from that of the CaMV
35S promoter (Figs. 2 and 3). For CAB-PHYA plants, oat
phytochrome A was detectable in all tissues. The majority
was found in the apex and young leaves (Fig. 2), consistent
with the increased activity of the CAB promoter during
chloroplast biogenesis (Leutwiler et al., 1986). In contrast to
35S-PHYA plants, only moderate levels were present in
stem tissue. Phytochrome concentrations declined in more
mature and senescing leaves in accord with their reduced
photosynthetic capacity. As expected, the UBQ-PHYA
plants expressed the chromoprotein most strongly in the
apex (position 1: apex to leaves 5-6 cm in length; Fig. 2).
Accumulation was apparent in stems, although little was
detected in mature or senescing tissues. Tissue printing of
CAB-PHYA and UBQ-PHYA plants expressing the highest
levels of oat phytochrome A showed that relatively little
oat phytochrome A accumulated in vascular tissue (Fig. 3,
B and E). In contrast to 35S-PHYA plants, neither the
CAB-PHYA nor UBQ-PHYA plants displayed enhanced
Figure 3. Distribution of phytochrome in stems of tobacco express- immunostaining of vascular tissue (or any other stem tis-
ing oat phytochrome A. Stem cross-sections were tissue printed onto sue) beyond that observed with wild-type plants.
nitrocellulose membranes and probed with preimmune serum (A) or
To assess the importance of phytochrome location in
with polyclonal antibodies against oat phytochrome A (B-E). Plants
examined include wild-type tobacco (D) and tobacco expressing oat
eliciting the light-exaggerated phenotype, the biological
phytochrome A under the direction of either the CaMV 35S promoter activity of oat phytochrome A under the control of the CAB
(A and C), Arabidopsis CAB promoter (B), or the Arabidopsis UBQ1 and UBQ1 promoters was determined and compared to
promoter (E). F, Stem section from 35S-PHYA plants stained for that using the CaMV 35S promoter. To accomplish this, a
protein with toluidine blue. phytochrome dose-phenotype response curve was gener-
ated for all three, in which plant height at maturity was
High concentrations of oat phytochrome A in vascular correlated with the amount of phytochrome present in the
tissue (possibly phloem) suggested to us that the dwarf apical portions of the mature plants (Cherry and Vierstra,
response of 35S-PHYA plants may be induced by artifi- 1994). To maximize the range of oat phytochrome A that
cially high levels of photoreceptor in this tissue type. To accumulated from each construct, homozygous and het-
test this possibility, oat phytochrome A was also expressed erozygous plants, derived from three independently trans-
under the control of two other promoters, Arabidopsis CAB formed lines, were used. The levels of phytochrome were
and UBQ1, that were expected to generate a different tissue determined either by red-minus-far-red difference spec-
distribution of the chromoprotein. The tissue distribution troscopy (Fig. 4, left), which detects both oat phytochrome
of the oat phytochrome A protein under the direction of A and the endogenous tobacco phytochromes that bear
each promoter was then determined and correlated with chromophores, or by sandwich ELISA (Fig. 4, right), using
the ability of each to induce the light-exaggerated pheno- Oat-22, which is specific for the oat phytochrome A
type. Given the involvement of the CAB gene product in polypeptide (Cherry et al., 1992, 1993). In all cases, the
photosynthesis, its promoter was expected to function pri- increased levels of spectrophotometrically detectable phy-
marily in photosynthetic tissue, mainly leaf parenchyma tochrome was matched by a similar accumulation of the oat
(Leutwiler et al., 1986). The CAB promoter is also induced phytochrome A polypeptide and not by elevated levels of
by Pfr, thus creating a potential feedback loop where the the endogenous tobacco phytochromes. This result sup-
activity of the CAB-PHYA gene would be enhanced further ports previous data showing that a majority of the oat
by increasing levels of oat phytochrome A. The UBQ1 phytochrome A polypeptides associates with chromophore
promoter was chosen because it directs expression in mer- and assembles into a spectrally active chromoprotein in
istematic tissue (a predicted site of phytochrome action tobacco (Cherry et al., 1991, 1992, 1993).
[Pratt, 1994]) and expresses to some extent in most cell As can be seen in Figure 4, plants expressing a wide
types, including epidermis, mesophyll, cortex, pith, and range of oat phytochrome A levels under the direction of
pollen (Callis et al., 1990). the CAB promoter were generated. Some contained spec-
Transgenic CAB-PHYA and UBQ-PHYA tobacco were trophotometrically detectable levels of total phytochrome
successfully generated; whereas the average levels of oat that were 16 times greater than in wild-type plants, levels
phytochrome A in the CAB-PHYA plants were nearly that even exceeded that synthesized in 35S-PHYA plants
equivalent to those in 35S-PHYA plants, the average levels (Fig. 4, top). However, the growth of CAB-PHYA plants
Phytochrome A Overexpression in Tobacco 803

amounts of oat phytochrome A were compared, 35s-PHYA


CAB-PHYA
plants were shorter in almost a11 cases.
I
DI SC USSI ON
Here we provide evidence that the light-exaggerated
phenotype induced by ectopic expression of phytochromes
is elicited, at least in part, by alterations in GA biosynthe-
sis. The involvement of GAs is implicated by (a) the ability
of foliar applications of GA, to suppress the dwarf pheno-
I I
type in 35s-PHYA plants and (b) a nearly 4-fold reduction
O .O2 .O4 .O6 .O8 O 25 50 75 100
in the active GAs, GA, and GA,, from apical tissue in
35s-PHYA plants versus wild-type plants. The response of
UBQ-PHYA 35s-PHYA plants to exogenous GA, discounts the possi-
bility that Pfr represses sensitivity to active GAs. The in-
volvement of GAs is consistent with the phenotypic simi-
larity of various plant species altered in phytochrome
levels or responsiveness to those defective in GA metabo-
lism (see above). The dwarf response observed in 35S-
PHYA plants results from a decrease in cell expansion and
not from a decline of cell division (Nagatani et al., 1991):an
effect that agrees well with the specific role of GAs in
" promoting cell elongation (Jones, 1982).
O .o2 .O4 .O6 O 25 50 75 100 The major GAs in wild-type tobacco belong to the early
PHYTOCHROME PHYTOCHROME 13-hydroxylation pathway (Reid, 1993). The presence of
(A(AAlg protein) (udgpmtein) GA, and GA, suggests that the non-3,13-hydroxylation
Figure 4. Phenotypic response of tobacco to increasing levels of oat pathway also operates in this species. The mechanism by
phytochrome A expressed under the direction of either the CaMV which excess Pfr affects the production of active GAs, GA,
35s promoter (O), CAB promoter (O),or t h e UBQ7 promoter (+). and GA,, is unclear. Since none of the GAs analyzed were
Wild-type (A) and transgenic plants expressing oat phytochrome A individually increased in 35s-PHYA plants, it suggests that
under the direction of the three promoters were grown concurrently
the block in GA biosynthesis occurred prior to the synthe-
under natural diurna1 light cycles. When flower primordia were
detected (74 d after sowing),apical leaf tissue was harvested, and the
sis of GA,,. The inability of GA, application to suppress a11
phytochrome was extracted and concentrated by ammonium sulfate effects induced by oat phytochrome A overexpression (e.g.
precipitation. Plant height was plotted relative to phytochrome con- partia1 release of apical dominance and a delay in senes-
tent as determined either by red-minus-far-red difference spectro- cence) indicates that other hormonal factors may partici-
scopy (AA665 - AA,,,) (left), which detects both endogenous to- pate. Chory et al. (1994) recently showed that cytokinin
bacco phytochromes and oat phytochrome A that bear application can phenocopy light-regulated development in
photoreversible chromophores, or by sandwich ELISA (right) using dark-grown Arabidopsis. We have preliminary data that
the monoclonal antibody Oat-22 that is specific for the oat phyto- wild-type and 35s-PHYA plants have indistinguishable cy-
chrome A polypeptide. Top, Expression directed by the CAB pro- tokinin levels, which suggests that excess phytochrome A
moter compared to the 3 5 s promoter. Bottom, Expression directed by does not alter cytokinin biosynthesis (M. Brenner, J.
the USQl promoter compared to the 35s promoter.
Cherry, and R.D. Vierstra, unpublished data).
This study and others (Boylan and Quail, 1989, Kay et al.,
was never inhibited to the same extent as that of 35s-PHYA 1989; Cherry et al., 1992, 1993; Boylan et al., 1994) clearly
plants. 35s-PHYA plants containing high levels of oat phy- establish that ectopic phytochrome A can accumulate to
tochrome A grew only to an average height of 24 cm (or high levels in transgenic plants, even in light-grown plants
28% the height of wild-type plants) with substantial dwarf- in which high rates of Pfr degradation occur (Vierstra,
ing evident even with modest increases in spectrally de- 1994). Moreover, when appropriate promoters are used, it
tectable chromoprotein (2- to 3-fold). In contrast, CAB- is possible to create plants that accumulate the photorecep-
PHYA plants grew to an average height of 54 cm (or 63% tor with different tissue distributions. The fact that high
the height of wild-type plants) even at the highest levels of levels can be achieved by the CAB and the CaMV 35s
oat phytochrome A achieved. Similar results were obtained promoters suggests that many cell types have the capacity
with UBQ-PHYA plants (Fig. 4, bottom). However, the to convert the apoprotein into a photoreversible chro-
analysis was limited because of the relatively low levels of moprotein. However, the failure of CAB-PHYA and UBQ-
chromoprotein obtained with this promoter; the maximal PHYA plants to respond as effectively as 35s-PHYA plants
leve1 of total spectrally detectable phytochrome attained to elevated levels of oat phytochrome A indicates that not
was only 3.5 times that present in wild-type plants. Despite a11 cells/ tissues are equally responsive. This is especially
this limitation, it was evident that UBQ-PHYA plants never evident for tobacco leaves that failed to elicit the maximal
realized the same extent of dwarfing as that attained with dwarf response even when they contained high concentra-
35s-PHYA plants. When plants accumulating equal tions of oat phytochrome A. Thus, the biological activity of
804 Jordan et al. Plant Physiol. Vol. 107, 1995

ectopic phytochromes appears d e p e n d e n t not only on the Callis JA, Raasch JA, Vierstra RD (1990) Ubiquitin extension
concentration of the photoreceptor but also on its spatial proteins in Arubidopsis thuliunu: structure, localizati(m, and ex-
distribution. pression of their promoters in transgenic tobacco. J Biol Chem
265: 12486-12493
Obviously, to completely understand the function(s) of Cheny JR, Hershey HP, Vierstra RD (1991) Charactt2rization of
phytochromes, information regarding their tissue distribu- tobacco expressing functional oat phytochrome. Plant Physiol
tion will be required. Whereas the distribution of phyto- 96: 775-785
chromes i n dark-grown plants is k n o w n (at least for phy- Cheny JR, Hondred D, Walker JM, Keller JM, H8:rshey HP,
tochrome A), the distribution in green plants is unknown, Vierstra RD (1993) Carboxy-terminal deletion analysis of oat
phytochrome A reveals the presence of separate clomains re-
mainly because of the low levels of phytochromes present quired for structure and biological activity. Plant Cell5: 565-575
(Pratt, 1994). Ectopic expression of o a t phytochrome A Cherry JR, Hondred D, Walker JM, Vierstra RD (1'292) Phyto-
u n d e r the direction of the CaMV 35s promoter results i n chrome requires the 6-kDa N-terminal domain for full biological
high phytochrome A concentrations in stem and petiole activity. Proc Natl Acad Sci USA 8 9 5039-5043
vascular tissue. This distribution is indistinguishable from Cherry JR, Vierstra RD (1994) Use of transgenic plantli to analyze
phytochrome structure/function. In RE Kendrick, GHM Kro-
that observed when the CaMV 35s promoter is used to nenberg, eds, Photomorphogenesis in Plants, Ed :!. Martinus
drive expression of the histochemical marker, GUS (Jeffer- Nijhoff Publishers, Dordrecht, The Netherlands, pp 271-297
son et al., 1987).Thus, the expression patterns in 35s-PHYA Childs KL, Cordonnier-Pratt M-M, Pratt LH, Morgan PW (1992)
plants seem dependent solely on the promoter and not on Genetic regulation of development in Sorghum bicolor. VIL mu,R
the capacity of vascular tissue to assemble the holoprotein flowering mutant lacks a phytochrome that predominates in
green tissue. Plant Physiol 99: 765-770
or on its failure to degrade Pfr. With respect to stem Chory J, Reinecke D, Sim S, Washburn T, Brenner M (1994) A
elongation, the enhanced phenotypic sensitivity of 35S- role for cytokinins in de-etiolation in Arubidopsis. tlet mutants
PHYA plants, compared w i t h CAB-PHYA and UBQ-PHYA have an altered response to cytokinins. Plant Physiol 104
plants, implicates the high concentrations of phytochrome 339-347
Cordonnier M-M, Greppin H, Pratt LH (1985) Moncclonal anti-
A in vascular tissue. If we assume that phytochromes reg-
bodies with differing affinities to the red-absorbing iind far-red-
ulate GA levels, this vascular tissue Pfr may i n turn affect absorbing forms of phytochrome. Biochemistry 2 4 3246-3253
GA synthesis, but whether this happens by altering GA Devlin PF, Rood SB, Somers DE, Quail PH, Whitelani GC (1992)
synthesis within vascular tissue directly or within other Photophysiology of the elongated internode (ein) mutant of
tissue indirectly is unknown. Clearly, identification of the Brussicu rupu. Plant Physiol 100 1442-1447
Edgerton MD, Santos MO, Jones AM (1993) In vivo supression of
cellular site(s) of GA biosynthesis is needed to resolve such
phytochrome aggregation by GroE chaperonin in Escherichiu coli.
possibilities. Plant Mo1 Biol 21: 1191-1194
Frances S, White MJ, Edgerton MD, Jones AM, Elliot RC,
Thompson WF (1992) Initial characterization of a pea mutant
ACKNOWLEDCMENTS with light independent photomorphogenesis. Plimt Cell 4:
1519-1530
We thank Drs. L.H. Pratt and M.-M. Cordonnier for the gift of Furuya M, Song P-S (1994) Assembly and properties 2f holophy-
monoclonal antibody Oat-22, Dr. Howard Hershey for original oat tochrome. Molecular properties and biogenesis of pk.ytochrome.
phytochrome A gene and oligonucleotides used to create the var- In RE Kendrick, GHM Kronenberg, eds, Photomorphogenesis in
ious promoter fusions, and Dr. Joel Cherry for generating the Plants, Ed 2. Martinus Nijhoff Publishers, Dordrechi, The Neth-
erlands, pp 105-140
35s-PHYA plants used in this study. We also thank Joseph Walker
Jefferson RA, Kavanagh TA, Bevan MW (1987) GUS fusions:
for providing technical assistance. P-glucuronidase as a sensitive and versatile gene fu!;ion marker
in higher plants. EMBO J 6: 3901-3907
Received September 19, 1994; accepted November 14, 1994. Jones RL (1982) Gibberellins control of cell elongation. In PE
Copyright Clearance Center: 0032-0889/95/107/0797/09. Wareing, ed, Plant Growth Substances 1982. Academic Press,
New York, pp 121-130
Kay SA, Nagatani A, Keith B, Deak M, Furuya M, Chua N-H
LITERATURE ClTED
(1989) Rice phytochrome is biologically active in transgenic
tobacco. Plant Cell 1: 775-782
Beall FD, Morgan PW, Mander LN, Miller FR, Babb KH (1991) Keller JM, Shanklin J, Vierstra RD, Hershey HP (19(39).Expres-
Genetic regulation of development in Sorghum bicolor. V. The sion of a functional monocotyledonous phytochrone in trans-
mu,R allele results in gibberellin enrichment. Plant Physiol 9 5 genic tobacco. EMBO J 8: 1005-1012
116-125 Kendrick RE, Kronenberg GHM (1994) Photomoi.phogenesis
Bowler C, Neuhaus G, Yamagata H, Chua N-H (1994) Cyclic GMP in Plants, Ed 2. Martinus Nijhoff Publishers, Dordrecht, The
and calcium mediate phytochrome phototransduction. Cell 77: Netherlands
73-81 Koomneef M, Bosma TDG, Hanhart CJ, van der Vcen JH, Ze-
Boylan M, Douglas N, Quail PH (1994) Dominant negative sup- evaart JAD (1990) Isolation and characterization of C;A-deficient
pression of Arubidopsis photoresponses by mutant phytochrome mutants in tomato. Theor Appl Genet 8 0 852-857
A sequences identifies spatially discrete regulatory domains in Leutwiler LS, Meyerowitz EM, Tobin EM (1986) Stiucture and
the photoreceptor. Plant Cell 6 449460 expression of three light-harvesting chlorophyll a / b binding
Boylan MT, Quail PH (1989) Oat phytochrome is biologically protein genes in Arubidopsis thuliunu. Nucleic Aci d s Res 14:
active in transgenic tomatoes. Plant Cell 1: 765-773 4051-4064
Boylan MT, Quail PH (1991) Phytochrome A overexpression in- Liscum R, Hangarter RP (1993) Genetic evidence that the red-
hibits hypocotyl elongation in transgenic Arubidopsis. Proc Natl absorbing form of phytochrome B modulates gravitropism in
Acad Sci USA 88: 10806-10810 Arubidopsis thuliunu. Plant Physiol 103: 15-19
Bradford M (1976) A rapid and sensitive method for the quanti- Nagatani A, Kay SA, Deak M, Chua N-H, Furuya M (1991) Rice
tation of microgram quantities of protein utilizing the principle type I phytochrome regulates hypocotyl elongatictn in trans-
of protein-dye binding. Ana1 Biochem 7 2 248-254 genic tobacco seedlings. Proc Natl Acad Sci USA 88: 5207-5211
Phytochrome A Overexpression in Tobacco 805

Nagatani A, Reid JB, Ross JJ, Dunnewijk A, Furuya M (1990) gion of phytochrome A result in an increase in biological activ-
Internode length in Pisum. The response to light quality and ity. Genes Dev 6: 2364-2372
phytochrome type I and I1 levels in Iv plants. J Plant Physioll35: Talon M, Zeevaart JAD, Gage DA (1991) Identification of gib-
667-674 berellins in spinach and effects of light and darkness on their
Pratt LH (1994) Distribution and localization of phytochrome levels. Plant Physiol 97: 1521-1526
within the plant. In RE Kendrick, GHM Kronenberg, eds, Pho- Vierstra RD (1993) Illuminating phytochrome functions. There is
tomorphogenesis in Plants, Ed 2. Martinus Nijhoff Publishers, light at the end of the tunnel. Plant Physiol 103: 679-684
Dordrecht, The Netherlands, pp 163-185 Vierstra RD (1994) Phytochrome Degradation. In RE Kendrick,
Quail PH (1991) Phytochrome: a light-activated molecular switch GHM Kronenberg, eds, Photomorphogenesis in Plants, Ed 2.
that regulates plant gene expression. Annd Rev Genet 2 5 Martinus Nijhoff Publishers, Dordrecht, The Netherlands, pp
389-409 141-1 62
Reid JB (1993) Plant hormone mutants. J Plant Growth Regul 12: Wagner D, Tepperman JM, Quail PH (1991) Overexpression of
207-226 phytochrome B induces a short hypocotyl phenotype in trans-
Reid JB, Ross JJ, Swain SM (1992a) Internode length in Pisum: a genic Arubidopsis. Plant Cell 3: 1275-1288
new, slender mutant with elevated levels of C,, gibberellins. Wahleithner JA, Li L, Lagarias JC (1991)Expression and assembly
Planta 1 8 8 462-467 of spectrally active recombinant holophytochrome. Proc Natl
Reid P, Pont-Lezica R, de1 Campillo E, Taylor R (1992b) Tissue Acad Sci USA 8 8 10387-10391
Printing: Tools for the Study of Anatomy, Histochemistry, and
Weller JL, Ross JJ, Reid JB (1994) Gibberellins and phytochrome
Gene Expression. Academic Press, New York
regulation of stem elongation in pea. Planta 192: 489-496
Rood SB, Williams PH, Pearce D, Murofushi N, Mander LN,
Pharis R (1990) A mutant gene that increases gibberellin pro- Wester L, Sommers DE, Clack T, Sharrock RA (1994) Transgenic
duction in Brussica. Plant Physiol 93: 1168-1174 complementation of the hy3 phytochrome B mutation and re-
Shanklin J, Jabben M, Vierstra RD (1987) Red light-induced sponse to PHYB copy number in Arubidopsis. Plant J 5 261-272
formation of ubiquitin-phytochrome conjugates: identification Ye Z-H, Song Y-R, Vamer JE (1992) Insolubilization of cell wall
of possible intermediates of phytochrome degradation. Proc proteins in developing soybean stems. In P Reid, R Pont-Lezica,
Natl Acad Sci USA 84: 359-363 E de1 Campillo, R Taylor, eds, Tissue Printing: Tools for the
Stockhaus J, Nagatani A, Halfter U, Kay S , Furuya M, Chua N-H Study of Anatomy, Histochemistry, and Gene Expression. Aca-
(1992) Serine-to-alanine substitutions at the amino-terminal re- demic Press, New York, pp 51-54