The Plant Cell, Vol. 13, 521–534, March 2001, www.plantcell.org © 2001 American Society of Plant Physiologists

Isolation and Characterization of Rice Phytochrome A Mutants

Makoto Takano,

a,1

Hiromi Kanegae,

b

Tomoko Shinomura,

c

Akio Miyao,

b

Hirohiko Hirochika,

b

and Masaki Furuya

c

a

Department of Plant Physiology, National Institute of Agrobiological Resources, Tsukuba, Ibaraki 305-8602, Japan

b

Department of Molecular Genetics, National Institute of Agrobiological Resources, Tsukuba, Ibaraki 305-8602, Japan

c

Hitachi Advanced Research Laboratory, Hatoyama, Saitama 350-0395, Japan

To elucidate phytochrome A (phyA) function in rice, we screened a large population of retrotransposon (

Tos17

) inser-tional mutants by polymerase chain reaction and isolated three independent

phyA

mutant lines. Sequencing of the

Tos17

insertion sites confirmed that the

Tos17

s interrupted exons of

PHYA

genes in these mutant lines. Moreover, thephyA polypeptides were not immunochemically detectable in these

phyA

mutants. The seedlings of

phyA

mutantsgrown in continuous far-red light showed essentially the same phenotype as dark-grown seedlings, indicating the in-sensitivity of

phyA

mutants to far-red light. The etiolated seedlings of

phyA

mutants also were insensitive to a pulse offar-red light or very low fluence red light. In contrast,

phyA

mutants were morphologically indistinguishable from wildtype under continuous red light. Therefore, rice phyA controls photomorphogenesis in two distinct modes of photoper-ception—far-red light–dependent high irradiance response and very low fluence response—and such function seemsto be unique and restricted to the deetiolation process. Interestingly, continuous far-red light induced the expression of

CAB

and

RBCS

genes in rice

phyA

seedlings, suggesting the existence of a photoreceptor(s) other than phyA that canperceive continuous far-red light in the etiolated seedlings.

INTRODUCTION

Light is one of the most important environmental stimuli thatplays a pivotal role in the regulation of plant growth, devel-opment, and metabolic activities. The perception of environ-mental light by plants is achieved by a family of plantphotoreceptors that includes the phytochromes (Neff et al.,2000), cryptochromes, and several others (Briggs andHuala, 1999), which are capable of detecting a wide spec-trum of light wavelengths ranging from UV to far-red light.

Physiological and photochemical studies in recent de-cades have shown that presumed phytochrome-mediatedresponses in plants can be classified into three different re-sponse modes: red (R)/far-red (FR) reversible responsesdesignated the low fluence response (LFR) and two otherresponses named the very low fluence response (VLFR) andthe high irradiance response (HIR) according to their energyrequirements (Mohr, 1962; Briggs et al., 1984). In a typicalLFR, plants respond to 1 to 1000

m

mol m

2

2

of R light,whereas the VLFR requires only 0.1 to 1

m

mol m

2

2

of R light.The HIR requires prolonged exposure to light of relativelyhigh photon flux and shows fluence rate dependence. Ac-

tion spectra for VLFR, LFR (Shinomura et al., 1996), and HIR(Shinomura et al., 2000) using phytochrome-deficient mu-tants of Arabidopsis showed that phyA is the principal pho-toreceptor involved in the VLFRs, such as induction of seedgermination (Shinomura et al., 1996) and

CAB

gene expres-sion (Hamazato et al., 1997), and the FR light–induced HIRs,including inhibition of hypocotyl elongation (Quail et al.,1995; Shinomura et al., 2000). On the other hand, phyB con-trols LFR in a R/FR reversible mode.

Phytochromes in higher plants are encoded by a smallgene family that is composed of five members (

PHYA

to

PHYE

) in Arabidopsis (Sharrock and Quail, 1989; Clack etal., 1994). Within the

PHY

gene family, phytochrome A(phyA) has been studied most extensively for its molecularproperties and physiological function (Furuya, 1993; Quail etal., 1995) to identify its specific role(s) in a number of phyto-chrome-mediated events leading to plant growth and devel-opment. The mode of photoperception by phyA seems tobe common among higher plants, but the function of phyAin vivo could be somewhat distinctive in different plant spe-cies at different developmental stages. For example, pea

phyA

mutants grown under long day photoperiods show ahighly pleiotropic phenotype in mature plants, includingshort internodes, thickened stems, delayed flowering andsenescence, longer peduncles, and higher seed yield(Weller et al., 1997), whereas the Arabidopsis

phyA

mutants

1

To whom correspondence should be addressed. E-mail mtakano@abr.affrc.go.jp; fax 81-298-38-8384.

Dow

nloaded from https://academ

ic.oup.com/plcell/article/13/3/521/6009486 by guest on 18 July 2021

522 The Plant Cell

grown under white light are morphologically indistinguish-able from wild-type plants (Nagatani et al., 1993; Whitelamet al., 1993).

Although phytochromes are the most extensively character-ized photoreceptors in plants, their diverse functions in theregulation of plant development have been characterizedmainly in dicots, and little information in this regard is avail-able in monocots. Phytochrome genes were isolated in mono-cots long ago (oat

PHYA

, Hershey et al., 1985; maize

PHYA

,Christensen and Quail, 1989; rice

PHYA

, Kay et al., 1989; rice

PHYB

, Dehesh et al., 1991), but the in vivo functions have notbeen elucidated, mainly due to the unavailability of phyto-chrome mutants. To date, the only phytochrome mutant foundin monocots was a

phyB

mutant of sorghum that was origi-nally identified as a

ma

3

R

early-flowering allele (Childs et al.,1997). Very recently, a phytochrome chromophore-deficientmutant was identified in rice (Izawa et al., 2000).

Rice has served as a model monocot plant system inwhich to explore complex genetic and physiological phe-nomena because a wealth of information is available on itsgenomic structure and functionality (Shimamoto, 1995). Thetask of designing and executing complex genetic studies inrice has been further facilitated by the availability of discreteresearch materials and tools, such as the high density ge-netic linkage map and thousands of expressed sequencetag clones made available by the Rice Genome ResearchProgram (Yamamoto and Sasaki, 1997). A recent addition tothese tools is the development of a gene knockout systemusing a rice retrotransposon named

Tos17

(Hirochika et al.,1996; Hirochika, 1997, 1999), which makes it possible toisolate mutants for genes of interest. In this system, the ac-tive

Tos17

can be used to easily generate a large number of

Tos17

-tagged mutant lines, and the mutants of a specificgene can be identified from the large mutant population(mutant panels) by polymerase chain reaction (PCR) using

Tos17

- and gene-specific primers.The generation and characterization of phytochrome mu-

tants of rice can broaden the existing view of phytochromefunction in plants, especially in terms of the differences inphytochrome functionality in monocots and dicots. In thisarticle, we report the isolation and characterization of a

phyA

mutant from a monocot (rice) plant. Because the mu-tant phenotype of rice deficient in phyA was unknown, weadapted the reverse genetics strategy to discover the rice

phyA

mutant. Physiological analyses of the

phyA

mutants atthe young seedling stage revealed that phyA is responsiblefor the inhibition of coleoptile elongation that is produced byirradiation with a pulse of very low fluence R (VLF-R) light.Moreover, phyA also is involved in the inhibition of mesocotylelongation and the induction of the gravitropic response incrown roots under continuous FR light. In contrast, phyAdeficiency did not impart any noticeable morphologicalchanges in plants grown under natural light conditions. Thefunction of phyA in rice appears to be unique and restrictedto the deetiolation process.

RESULTS

Isolation of Rice

phyA

Mutants from Mutant Panels

We screened a large population of

Tos17

insertional mu-tants of rice by PCR to isolate rice

phyA

mutants. Three mu-

Figure 1. The Structure of the PHYA Gene and Insertion Sites of Tos17 in the PHYA Gene of Three Mutant Lines.

The structure of the PHYA gene is derived from sequence information available in the database (GenBank accession number X14172). The exonand intron regions of PHYA are presented as closed and open bars, respectively. Insertion sites of Tos17 sequences in the PHYA gene of threemutant lines are indicated by vertical bars labeled with the names of mutant lines and the range of nucleotides at the sites of insertion. The inte-grated Tos17 sequences are shown by shaded bars with insertion orientation. The locations of primers designed for PCR screening areindicated by arrows. PstI sites used for DNA gel blot analysis also are indicated.

Dow

nloaded from https://academ

ic.oup.com/plcell/article/13/3/521/6009486 by guest on 18 July 2021

Isolation of Rice

phyA

Mutants 523

tant panels were subjected to PCR-based screening usingall combinations of four

PHYA

-specific (Figure 1) and two

Tos17

-specific primers. As a result, we identified three mutantlines,

osphyA-1

,

osphyA-2

, and

osphyA-3

, in which

Tos17

swere inserted into the

PHYA

genes. The amplified DNA frag-ments corresponding to the three identified lines werecloned and sequenced to determine the insertion sites andthe orientation of transposed

Tos17

, and the results aresummarized in Figure 1.

Tos17

was found to be inserted inthe forward orientation near the end of the second exon inline

osphyA-1

. In the two other lines, the insertion of

Tos17

was detected in reverse orientation in the third (

osphyA-2

) orfourth exon (

osphyA-3

). In all three mutant lines, the trans-poson was inserted into the coding sequence of the

PHYA

genes and the predicted amino acid sequences were termi-nated prematurely. The results suggest that

Tos17

-tagged

phyA

alleles are functionally impaired.The insertion of

Tos17

into the

PHYA

gene was furtherconfirmed by DNA gel blot hybridization. The DNA was iso-lated from progeny plants of line

osphyA-1

and hybridizedwith a cDNA clone of the

PHYA

gene to distinguish the ge-notypes of mutant and wild-type alleles. On the basis of therestriction map of

PHYA

(Figure 1), cutting with PstI shouldgive two hybridizing bands (0.8 and 4.3 kb) for the wild-typealleles, whereas three bands (6.6, 1.8, and 0.8 kb) are sup-posed to be hybridized for the mutant alleles because the

insertion of

Tos17

adds one more PstI site. As expected, weobserved three banding patterns in the progeny of the

os-phyA-1

line: (1) homozygous for the insertion (Figure 2,lanes 4 and 5); (2) heterozygous (Figure 2, lane 3); and (3)wild type (Figure 2, lanes 1, 2, and 6). Because the numberof plants showing these genotypes follows the Mendelianratio of 1:2:1, it was demonstrated that the mutant alleleswere stably transmitted to the next generation according toMendel’s first law. We also confirmed the insertion of

Tos17

into the

PHYA

gene and the normal inheritance of the mu-tant alleles in two other mutant lines,

osphyA-2

and

osphyA-3

(data not shown).

Immunochemical Analysis of the PHYA Apoprotein

The level of PHYA apoprotein in the insertional mutants wasdetermined by immunoblotting to examine if the mutant al-leles of

PHYA

were null mutations. Figure 3 shows that thePHYA apoprotein was undetectable in crude protein ex-tracts from etiolated

phyA

mutant progeny (

osphyA-1-2

and

osphyA-2-10

), whereas progeny with wild-type

PHYA

alleles(

osphyA-1-4

and

osphyA-2-9

) had similar levels of PHYA asthe wild-type plants (cvs

Akitakomachi

and

Nipponbare

).Thus, the

PHYA

gene appeared to be completely knockedout in the

osphyA-1

and

osphyA-2

mutant lines.

Effect of Continuous FR Light on Growth Responses in

phyA

Mutants

We examined the effect of continuous irradiation with R andFR light on the growth responses of young wild-type and

osphyA-1

mutant seedlings. The results are presented asoriginal photographs of seedlings (Figure 4) and as mea-surements of mean lengths of coleoptile and mesocotyl

Figure 2. DNA Gel Blot Analysis of Segregating Progeny of osphyA-1Mutant Lines.

The DNA gel blot of PstI-digested genomic DNA was hybridized withthe complete sequence of PHYA cDNA. Typical autoradiogram pro-files of wild type (Akitakomachi) and six progeny of the osphyA-1 lineare shown. Hybridized bands of 4.3 kb are from plants with the wild-type allele, and those of 6.6 and 1.8 kb are from plants with the mu-tant allele. Therefore, lanes 4 and 5 are homozygous (2/2) for theinsertion of Tos17, lane 3 is heterozygous (1/2) for the same insert,and lanes 1, 2, and 6 are homozygous for the wild type (WT; 1/1).

Figure 3. Immunoblot Detection of PHYA Apoproteins in Dark-Grown Seedlings.

Seven-day-old etiolated seedlings of wild type (WT; Akitakomachi[A] and Nipponbare [N]) and segregating progeny of two phyA mu-tant lines (osphyA-1 and osphyA-2) were used for the extraction ofcrude protein. Two progeny (homozygous for the insertion, osphyA-1-2 and osphyA-2-10; and wild type, osphyA-1-4 and osphyA-2-9)were chosen for each mutant line. Each lane was loaded with 50 mgof total protein for the detection of PHYA by the monoclonal anti-ryePHYA antibody (mAR07).

Dow

nloaded from https://academ

ic.oup.com/plcell/article/13/3/521/6009486 by guest on 18 July 2021

524 The Plant Cell

(Figures 5A and 5B). The coleoptiles and mesocotyls of thedark-grown seedlings reached their maximum lengths 6days after the induction of germination without any addi-tional changes for up to 19 days (data not shown).

Both the wild-type and osphyA-1 seedlings exhibited atypical elongation of coleoptiles and mesocotyls whengrown in constant darkness (Figures 4 and 5D). The differentappearance of the two dark-grown osphyA-1 seedlings inFigure 4 is due to a difference in the timing of germination.We did not observe significant difference of growth rate ofthe shoot between wild type and phyA mutants. A com-monly observed inhibition of coleoptile and mesocotyl elon-gation also was apparent when the seedlings of wild typeand osphyA-1 mutants were exposed to continuous R lightfor 11 days. In this experiment, exposure to continuous R

light reduced the growth of coleoptiles to one-sixth of that indark-grown seedlings in both wild-type and osphyA-1 seed-lings (Figures 4 and 5A, cR). Similarly, growth of the meso-cotyl was inhibited completely under continuous R light inboth lines (Figures 4 and 5B, cR). In contrast, growth re-sponses of the osphyA-1 seedlings to continuous FR lightwere completely different from those of the wild type. Whenthe seedlings were exposed to FR light for 11 days, the os-phyA-1 coleoptiles and mesocotyls elongated like those ofdark-grown seedlings (Figures 4B and 5, cFR), whereaswild-type coleoptiles were short (Figures 4A and 5A, cFR)and wild-type mesocotyls were undetectable (Figures 4Aand 5B, cFR).

To determine the irradiation period with FR light neededfor the detection of such growth inhibition, exposure timesto FR light were gradually shortened until growth inhibitionwas no longer observed. Two-day-old etiolated seedlings,which are most sensitive to FR light for the inhibition of me-socotyl elongation, were exposed for various durations ofcontinuous FR light from 1 hr to 5 days. After the irradiation,they were kept in darkness until the measurement of meso-cotyl lengths in the 7th day. Their mesocotyl lengths werecompared with those of seedlings grown in constant dark-ness as a control (Figure 5C). When the wild-type seedlingswere exposed to continuous FR light for 3 hr or shorter, theirmesocotyls were as long as those of seedlings grown in thedark (Figure 5C). However, when they were exposed forlonger than 6 hr, their mesocotyl lengths became shorter asthe duration of light exposure increased (Figure 5C). In con-trast, the mesocotyl elongation of osphyA-1 mutants wasnot affected even by a 5-day exposure to continuous FR(Figure 5C). These results indicate that the inhibition of me-socotyl elongation is detectable in wild-type seedlings whenexposed to a prolonged FR light irradiation (more than 6 hr).Essentially similar responses to continuous R and FR lightwere observed in another phyA mutant, osphyA-2 (data notshown). These results indicate that etiolated seedlings of thephyA mutants are insensitive to FR light for the modulationof coleoptile and mesocotyl length.

Effect of Light Pulse Irradiation on Coleoptile Growth

We examined the effect of R and FR light pulse irradiationwith both low fluence (LF) and very low fluence (VLF) on thecoleoptile growth of wild-type and osphyA-1 seedlings. Eti-olated seedlings with coleoptiles longer than 10 mm wereexposed to a pulse of LF-R, FR, or VLF-R light and werekept in the dark. Their coleoptile lengths were measured 2days after the irradiation. Exposure to a pulse of LF-R light(z1 mmol m22) completely inhibited the growth of coleop-tiles in both the wild type and osphyA-1 (Figure 6, LF-R).Subsequent exposure to a pulse of FR light (z20 mmol m22)partly reversed the effect of LF-R light exposure in the wildtype (Figure 6, LF-R/FR). This LF-R/FR reversible responsewas observed much more clearly in osphyA-1 (Figure 6, LF-R,

Figure 4. Phenotypic Comparison of Wild-Type (WT) and phyA Mu-tant Plants Grown under Different Light Conditions.

Seedlings of wild type (A) (Akitakomachi) and osphyA-1 mutants (B)were grown under complete darkness (D), continuous R light (cR), orcontinuous FR light (cFR) for 11 days. Two seedlings are shown foreach treatment. White arrowheads indicate the tips of coleoptiles.Bars 5 1 cm.

Dow

nloaded from https://academ

ic.oup.com/plcell/article/13/3/521/6009486 by guest on 18 July 2021

Isolation of Rice phyA Mutants 525

LF-R/FR, and LF-R/FR/LF-R). These results indicate thatphyA is not involved in this LF-R/FR reversible control ofgrowth inhibition in the rice coleoptile.

Next, we examined the effect of VLF-R and FR light andfound that a pulse of either FR light (20 mmol m22) or VLF-Rlight (z1 mmol m22) was able to inhibit growth of the coleop-tile in the wild type but not in osphyA-1 (Figure 6, VLF-R andFR). These results indicate that the inhibition of coleoptileelongation by a pulse irradiation of VLF-R or FR light in eti-olated rice seedlings is mediated primarily by phyA.

Root Gravitropic Response

We found another phenotypic difference in the root growthbetween wild-type and osphyA-1 seedlings. Growth orienta-tions of crown roots under different light conditions are pre-sented in Figure 7. In the dark, the crown roots of both wild-type and osphyA-1 seedlings grew mostly horizontally (Fig-ures 7B and 7D). In other words, the crown roots of dark-grown rice seedlings are plagiotropic (growing parallel ornearly parallel to the surface). Under continuous R light, thegravitropic response of crown roots was induced and theroots grew with a peak frequency distribution at 150 to1808 in both wild-type and osphyA-1 seedlings (Figure 7B,cR). Wild-type seedlings grown in continuous FR lightshowed a similar geotropic response as seedlings grown incontinuous R light, although the peak of the distribution incontinuous FR light shifted slightly toward horizontal (Figure7B, cFR). Crown roots of osphyA-1 were insensitive to con-tinuous FR light, because their orientation was similar to thatobserved in the dark-grown seedlings (Figure 7B, cFR).These results indicate that the gravitropic response of ricecrown roots is modulated by continuous irradiation withboth R and FR light and that the phyA mutation caused theloss of response to continuous FR light.

We examined crown roots to find the duration of prolongedirradiation with FR light needed for the induction of the gravi-tropic response. Five crown roots emerge sequentially fromthe nodal portion of the rice coleoptile (Hoshikawa, 1989). Weexamined only the first of these five to emerge (Figure 7A,white arrowheads). Four-day-old etiolated seedlings, whenthe length of the first crown roots was z9 6 2 mm (most sen-sitive stage to FR light for detection of this response), wereexposed to various durations of continuous FR light from 15min to 48 hr. They were grown in the dark until growth mea-surement at the 7-day-old stage. The growth angles of thefirst crown roots were measured and compared with those ofseedlings grown in the dark or continuous FR light as a con-trol (Figure 7C). When the wild-type seedlings were exposedto continuous FR light for 1 hr or less, the growth orientationof the crown root was equivalent to that of seedlings grown inthe dark (Figure 7C). In one z24-hr treatment, longer expo-sure times produced a larger growth angle in the first crownroots in wild-type seedlings (Figure 7C). The FR light irradia-tion for 48 hr caused the equivalent effect on this responseobtained by continuous FR light irradiation (Figure 7C). Incontrast, osphyA-1 mutants treated with various durations ofcontinuous FR light showed growth orientation angles aslarge as those of seedlings grown in the dark (Figure 7C). Irra-diation with VLF-R light (z1 mmol m22) did not induce thegravitropic response of the first growing crown root in wild-type seedlings (data not shown). Essentially, similar effects ofcontinuous FR light and VLF-R light on the second growingcrown roots were observed (data not shown). These resultsindicate that gravitropic growth orientation was induced onlywhen the wild-type seedlings were exposed to a prolongedFR light irradiation (more than 3 hr).

Figure 5. Coleoptile and Mesocotyl Lengths of Wild Type and phyAMutants Grown under Different Light Conditions.

(A) and (B) Wild-type (Akitakomachi; open bars) and osphyA-1(hatched bars) seedlings were grown under continuous R light (cR),under continuous FR light (cFR), or in complete darkness (D) for 11days. The lengths of coleoptiles (A) and mesocotyls (B) of theseseedlings were measured individually.(C) Effect of continuous FR light on the inhibition of the mesocotylelongation of wild-type (Akitakomachi) and osphyA-1 seedlings.Two-day-old etiolated seedlings were irradiated with continuous FRlight for intervals ranging from 1 hr to 5 days (d) and kept in darknessuntil 7th day. Their mesocotyl lengths were measured and the rela-tive lengths compared with the dark control were shown. Error barsrepresent SE; n 5 30. N. indicates that the corresponding tissue wasnot detected.

Dow

nloaded from https://academ

ic.oup.com/plcell/article/13/3/521/6009486 by guest on 18 July 2021

526 The Plant Cell

The seminal roots, in contrast to crown roots, grew at anangle between 170 and 1908 below horizontal, irrespectiveof the presence or absence of light in both wild-type andphyA mutant seedlings. Thus, gravitropic growth orientationin rice seminal roots is light independent.

Phenotypes of Mature Plants of Rice phyA Mutants

To compare the phenotypic peculiarities exhibited by the ma-ture plants, both wild-type and phyA mutant plants wereraised in the paddy field, and plant shape, plant height, andheading day were observed. However, we could not detectany significant differences for the measured phenotypic traits.Because the variety Nipponbare is strongly photoperiod sen-sitive, we carefully measured the heading date of osphyA-2

mutant lines. But no significant difference was observed be-tween wild-type and osphyA-2 mutant plants. The meandays-to-heading value for wild-type plants (n 5 21) was112.7 6 1.2 days and that for osphyA-2-10 plants (n 5 70) was112.5 6 2.5 days. We also raised the wild-type and mutantplants under short day (10.5 hr of light/13.5 hr of dark) condi-tions, but still no significant difference was observed for daysto heading in either genotype (wild-type, 50.0 6 1.5 days [n 532]; osphyA-2-9 with wild-type allele, 51.7 6 1.4 days [n 5 13];osphyA-2-10 with mutant allele, 51.6 6 1.7 days [n 5 14]).Therefore, rice phyA mutants were morphologically indistin-guishable from wild-type plants when grown under natural lightconditions.

Expression Patterns of Light-Regulated Genes

We determined the effect of R and FR irradiation on theabundance of transcripts for CAB and RBCS genes in thewild type and phyA mutants. Mutant and wild-type plantswere grown in complete darkness for 7 days and exposedto continuous R or FR light. Control seedlings were kept inconstant darkness during the experiment. The seedlingswere harvested 0, 4, 12, and 24 hr after the irradiation andthe accumulation of CAB and RBCS transcripts was investi-gated by RNA gel blot analysis. As shown in Figure 8A,RBCS transcripts were present at a low level even in thedark, and the amount of these transcripts was increasedequally by both R and FR light in wild-type plants. In phyAmutant plants, R light exerted the same effect as in wild-type plants, as expected. Unexpectedly, however, RBCSgene expression was induced in phyA mutants by continu-ous FR light to almost the same level as in wild-type plants.In the case of CAB gene expression (Figure 8B), we werenot able to detect any signal in the dark (column D or row 0hr) samples. Both R and FR light induced an accumulationof CAB transcripts in wild-type plants, which showed a fluc-tuation due to circadian rhythms. In phyA mutant plants, Rlight induced the same CAB gene expression pattern as inwild-type plants. FR light treatment also resulted in a similarfluctuation of CAB gene expression, although this expres-sion level was much lower than that of wild-type plants.These results show that phyA seems to play a minor role inthe induction of RBCS gene expression by FR light, whereasfor the induction of CAB gene expression by FR light, phyAclearly play a more prominent role, although other photore-ceptor(s) must also be involved.

DISCUSSION

Isolation of Rice phyA Mutants from Mutant Panels

In this article, we report the isolation and characterization ofphyA mutants in a monocot (rice) plant. Identification and

Figure 6. Effects of a Pulse of Light Irradiation on Coleoptile Growthin Young Seedlings of Rice phyA Mutants.

The wild type (Akitakomachi; open bars) and the osphyA-1 mutants(hatched bars) were germinated and grown in the dark for 4 days (d)as described in Methods and then exposed to the following lightregimens: Dark, kept in the dark; LF-R, a pulse of LF-R; LF-R/FR, apulse of LF-R followed by a FR pulse; LF-R/FR/LF-R, a pulse of LF-Rfollowed by FR and then LF-R pulses; VLF-R, a pulse of VLF-R light;FR, a pulse of FR. After the light irradiation, seedlings were kept inthe dark for 48 hr, and coleoptile lengths were measured from 30seedlings. Error bars represent SE.

Dow

nloaded from https://academ

ic.oup.com/plcell/article/13/3/521/6009486 by guest on 18 July 2021

Isolation of Rice phyA Mutants 527

characterization of mutants in living organisms is indispens-able to efforts to understand and elucidate the function ofparticular genes. Mutants for phytochrome genes have beenisolated in Arabidopsis and several other dicot species(Koornneef and Kendrick, 1994; Fankhauser and Chory,1997). We used the gene knockout system (Hirochika, 1997,1999) to isolate the rice phyA mutants, because isolation ofmutants in rice by phenotypic selection was not as easy asin Arabidopsis.

Site-directed manipulation of chromosomal genes is a

powerful tool with which to investigate the in situ function ofisolated genes. A gene knockout system has been estab-lished in mice to produce phenotypes resulting from theelimination of functional genes or the loss of gene function.However, in plants, gene targeting has not become widelyused because of the low frequency of homologous recombi-nation (Thykjaer et al., 1997; Gallego et al., 1999; Jelesko etal., 1999). The transposon-tagging system provides an alter-native method in plants to knock out specific genes. In thissystem, instead of targeting a specific gene, large populations

Figure 7. Effect of Light on the Growth Orientation of Crown Roots of Wild Type (WT) and phyA Mutants.

(A) Representative photographs of crown roots of wild-type (Akitakomachi) and osphyA-1 seedlings grown under continuous FR light for 11days. White arrowheads indicate first growing crown roots. Bars 5 0.5 cm.(B) Histograms of growth angles of crown roots. Wild-type (Akitakomachi) and osphyA-1 seedlings were maintained in complete darkness (D) orwere grown under continuous R light (cR) or continuous FR light (cFR) for 11 days. Angles of crown roots were measured from the horizontalaxis and divided into 108 intervals. The numbers of crown roots of wild-type and osphyA-1 seedlings are indicated by light blue and magentabars, respectively.(C) Effects of continuous FR light on the gravitropic growth orientation of the first growing crown roots (indicated by white arrowheads in [A]).Four-day-old etiolated seedlings of wild type (Akitakomachi; light blue bars) and osphyA-1 mutants (magenta bars) were irradiated with continuousFR light for intervals ranging from 15 min to 48 hr and kept in darkness until the 7th day. The growth angles of the first-grown-crown roots weremeasured. As controls, those of the seedlings grown in the dark or under the continuous FR light are shown. Error bars represent SE; n 5 20.

Dow

nloaded from https://academ

ic.oup.com/plcell/article/13/3/521/6009486 by guest on 18 July 2021

528 The Plant Cell

of mutants (mutant panels) are generated by random inser-tion of retrotransposons and the mutants for a gene of inter-est are isolated by PCR-based screening of mutant panels.We used this system to isolate several independent lines inwhich the PHYA gene was knocked out.

In each of the phyA mutant lines, an exon of the PHYAgene was interrupted by the insertion of Tos17 sequence(Figure 1) and an in-frame stop codon was produced justdownstream of the insertion site. These sequences have thepotential to produce truncated PHYA molecules. However,the aberrant transcripts corresponding to the truncatedPHYA molecules seem not to be translated, because immu-nodetectable bands were absent from two mutant lines (os-phyA-1-2 and osphyA-2-10; Figure 3). The transcript level ofthe PHYA gene also was determined in the etiolated seed-

lings by RNA gel blot analysis. An expected, the transcriptlength of PHYA (4 kb) was not detected in either osphyA-1-2or osphyA-2-10. Therefore, we categorized these mutantlines as null mutants for the phyA function.

Each mutant line in the mutant panels that we used has 5to 10 copies of transposed Tos17. Thus, it is important toeliminate the possible effects of the insertion of Tos17s intogenes other than PHYA. Fortunately, we were able to isolatethree independent mutant lines (osphyA-1, osphyA-2, andosphyA-3) and found the same phenotypes from the twomutant lines we checked (osphyA-1 and osphyA-2), whichproved that the phenotypes observed in the present studywere caused by the loss of function of PHYA.

Role of phyA in the Photomorphogenesis ofRice Seedlings

Screening of the mutant panels resulted in the isolation ofthree allelic phyA mutants that show a dramatically reducedmorphological response to continuous FR light, including in-hibition of coleoptile and mesocotyl elongation and of thegravitropic response of crown roots (Figures 4, 5, and 7).Together with the sequence analysis (Figure 1) and immu-nochemical data (Figure 2), these results indicate that the al-tered photomorphogenesis in the phyA mutants results froma deficiency in functional phyA and that phyA is the predom-inant phytochrome in the dark-grown seedlings that medi-ates deetiolation responses induced by continuous FR lightin rice. On the other hand, when grown under continuous Rlight (Figures 4, 5, and 7) or white light (data not shown), thephenotypic properties of phyA seedlings were essentiallythe same as those of the wild type, suggesting that phyA ismostly dispensable for the photomorphogenesis of riceseedlings under these conditions.

A pulse irradiation of LF-R light is known to inhibit theelongation of coleoptiles of rice seedlings when the coleop-tile is longer than 10 mm (Pjon and Furuya, 1967). This re-sponse was partially reversed by subsequent irradiationwith a pulse of FR light, indicating that it was the R/FR re-versible response (Pjon and Furuya, 1967). In the presentstudy, we repeated the same experiment for the phyA mu-tants and found that they showed significantly clearer re-versibility of R/FR response compared with the wild type(Figure 6, LF-R, LF-R/FR, and LF-R/FR/LFR). We think thatin the wild type, the FR pulse–induced reverse responsemediated mainly by phyB was almost canceled by phyA-mediated growth inhibition induced by the FR pulse. The re-sults presented here clearly show that the coleoptile elonga-tion in wild-type plants is inhibited by a FR light pulse andare consistent with the results using a single short exposure(Pjon and Furuya, 1967) or hourly pulses of FR light expo-sure (Casal et al., 1996). In addition, we have demonstratedthat a pulse of VLF-R or FR light that inhibited the coleoptileelongation in wild-type seedlings was not effective for theinhibition of coleoptile elongation in phyA mutants (Figure 6,

Figure 8. Induction of CAB and RBCS Genes by R and FR Light inWild-Type (WT) and phyA Mutant Seedlings.

Seven-day-old etiolated seedlings of wild type (Nipponbare [N]) andtwo independent phyA mutants (osphyA-1and osphyA-2) were ex-posed to continuous R light (cR) or continuous FR light (cFR) for 0, 4,12, and 24 hr (h) and harvested for RNA isolation. Control seedlingswere kept continuously in the dark (D) and harvested at the sametime point as cR- or cFR-treated seedlings. Samplings at the timepoint of 12 hr were skipped. Five micrograms of total RNA was blot-ted and hybridized with rice CAB and RBCS probes. rRNA wasstained with methylene blue.

Dow

nloaded from https://academ

ic.oup.com/plcell/article/13/3/521/6009486 by guest on 18 July 2021

Isolation of Rice phyA Mutants 529

VLF-R and FR). These results indicate that the growth inhibi-tion of coleoptiles in rice seedlings is mediated by phyA inVLFR mode as well as by other phytochromes (probablyphyB) in LFR mode of photoperception. It is interesting tonote that the light-induced inhibition of the coleoptile andmesocotyl elongation is mediated by the different light per-ception modes. A pulse of VLF-R or FR light was able to in-duce the inhibition of the coleoptile elongation, whereas aprolonged FR light irradiation was required to inhibit the me-socotyl elongation and VLF-R light was not effective for theinhibition. Therefore, as far as phyA is concerned, the light-induced inhibition of the coleoptile is mediated by VLFRmode and that of the mesocotyl by HIR mode. Moreover,we used the phyA mutants to present direct evidence thatphyA mediates FR light–induced gravitropic growth in thecrown roots of rice (Figure 7). The effects of light on gravi-tropic responses have been reported in many organs of var-ious plant species, including coleoptiles of oat (Blaauw, 1961)and maize (Wilkins and Goldsmith, 1964) and roots of maize(Mandoli et al., 1984; Feldman and Briggs, 1987). However,the reported effects varied greatly depending on the plantspecies and the organs, and the variation could be attributedin large part to the existence and involvement of multiple pho-toreceptors. Arabidopsis was the only plant species in whichan assignment of phytochrome molecular species involvedin the gravitropic responses was determined using phyto-chrome-deficient mutants (Robson et al., 1996). According tothat report, hypocotyls of Arabidopsis exhibit negative gravi-tropism (growing against the gravity vector) in the dark, andeither phyA or phyB is capable of inducing randomization ofthe direction of hypocotyl growth in R light.

In terms of root gravitropism, it has been suggested thatroots of many plant species grow in diagravitropic orienta-tion in the dark and that light induces their gravitropic growth(Feldman, 1984). For example, in maize (var Merit), light-induced gravitropic response in etiolated primary roots ismediated by both LFR, showing R/FR reversibility (Mandoliet al., 1984), and VLFR, showing that FR light could not re-verse the R light induction (Feldman and Briggs, 1987).These observations suggested the involvement of multiplephytochromes in the induction of gravitropic responses.However, we have presented here conclusive evidence thatphyA induces gravitropic growth in the crown roots of ricethrough continuous FR light, and photoreceptor(s) otherthan phyA seem to induce the same response in LF-R light(Figure 7).

It is interesting to note that in rice, such light-inducedgravitropic bending was observed only in the crown rootsand not in the seminal (primary) roots, which grow straightdown irrespective of the light treatment (Figure 7A). Inmaize, the primary roots show the light-induced gravitro-pism (Feldman and Briggs, 1987). Therefore, there seem tobe variations among plant species in the light-induced grav-itropic response. In rice, crown roots emerge from nodalportions and tend to develop horizontally. Because thecrown roots, which come out from higher nodes, start de-

veloping horizontally into air or water, they need to benddownward to reach the soil. Therefore, light-inducedchanges in root gravitropism could be an adaptive mecha-nism, and immunochemically detectable PHYA that is highlyaccumulated in the root cap (Hisada et al., 2000) may playsome role in light-induced gravitropism.

Phenotypes of Mature phyA Mutants of Rice

Only a limited number of phyA mutants have been reportedto date, including the loss-of-phyA-function mutants of Ara-bidopsis (Nagatani et al., 1993; Parks and Quail, 1993;Whitelam et al., 1993), tomato (van Tuinen et al., 1995), andpea (Weller et al., 1997). In Arabidopsis and tomato, phyAdeficiency was reported to have little effect on the morpho-logical appearance of mature plants grown under whitelight. In contrast, phyA-deficient pea mutants (fun1-1)showed a highly pleiotropic phenotype when they weregrown to maturity under long day conditions (Weller et al.,1997), indicating that phyA is involved in the detection ofdaylength in pea.

In rice, we did not detect a significant difference betweenmutant and wild-type plants grown under natural light con-ditions for any of the measured phenotypic traits. Clough etal. (1995) overexpressed the oat PHYA gene in rice, but thetransgenic rice plants were not phenotypically different fromwild-type plants when grown under high fluence white light.These observations indicate that phyA has little effect in ma-ture rice plants grown under natural light conditions.

Rice cv Nipponbare is strongly photoperiod sensitive, anda delay in flowering is noted under long-day conditions. Inshort-day conditions (10.5 hr of light/13.5 hr of dark), bothwild-type and osphyA-2 plants started heading z50 daysafter sowing, whereas in natural light conditions (long-dayconditions), the days to heading was prolonged to z113days in both osphyA-2 and the wild type. These observa-tions suggest that phyA does not play an important role inphotoperiod sensing for flowering in rice. Recently, Izawa etal. (2000) showed that the photoperiodic sensitivity 5 (se5)mutant, which is completely deficient in photoperiodic re-sponse (Yokoo and Okuno, 1993), lacks the phytochromechromophore and therefore has no functional phytochromes.In their work, se5 flowered very early even under long-dayconditions (14 hr of light/10 hr of dark), which indicates thatsome phytochrome species detect the long-day photope-riod and mediate the delay in flowering. Because our phyAmutants responded to the long-day photoperiod, we can saythat the delay in flowering observed by Izawa et al. (2000) wasdue to some phytochrome species other than phyA.

Expression of CAB and RBCS Genes in phyA Mutants

Phytochromes are known to mediate the induction of ex-pression for genes involved in photosynthesis (Kuno et al.,

Dow

nloaded from https://academ

ic.oup.com/plcell/article/13/3/521/6009486 by guest on 18 July 2021

530 The Plant Cell

2000). In particular, CAB and RBCS genes are well-charac-terized nuclear genes whose expression is regulated byphytochromes at the transcriptional level. The loss of phyAfunction in the phyA mutants should alter the expressionpattern of CAB and RBCS genes. In Arabidopsis, it was re-ported that irradiation with both VLF-R and FR light couldinduce the expression of the CAB gene in phyB mutants,whereas in the phyA mutants, CAB gene expression couldbe induced by LF-R light but not by FR light (Hamazato etal., 1997). These results suggest that phyA is the sole photo-receptor in Arabidopsis that can induce CAB gene expres-sion upon irradiation with FR light in dark-grown seedlings.

In rice phyA mutants, however, FR light induced RBCSand CAB gene expression in dark-grown seedlings at differ-ent levels. The induction of RBCS gene expression was al-most to the same level as in wild-type plants (Figure 8A),whereas the induction level of CAB gene expression wasmuch less in phyA mutants than in wild-type plants (Figure8B). These results show that phyA contributes to the accu-mulation of RBCS and CAB transcripts differently under FRlight and also suggest that photoreceptor(s) other than phyAcan mediate the expression of these genes in response to ir-radiation with FR light in etiolated rice seedlings. Only threephytochrome genes (PHYA, PHYB, and PHYC) have beenreported to date in rice (Tahir et al., 1998), and phyB is as-sumed to be a sensor for classic R/FR photoreversible re-sponses (Shinomura et al., 1996). Therefore, phyC could bea candidate photoreceptor for such photoresponse. In Ara-bidopsis, phyC is thought to be a type II phytochromebased on its photostable nature and its small amount inboth etiolated and green plants, but its photoregulatoryroles have not been fully elucidated because monogenicmutants of phyC have yet to be isolated (Quail, 1998). Thus,we cannot exclude the possibility that phyC mediates theCAB gene expression induced by FR light. Regardless ofthe facts to be established in the future, the present studydemonstrates that a photoreceptor(s) other than phyA per-ceives FR light for the induction of CAB gene in the etiolatedseedlings.

Modes of phyA Action in Rice

The present study has demonstrated that the photoinhibi-tion of coleoptile elongation in rice induced by a pulse oflight irradiation is mediated by both phyA in VLFR mode andprobably phyB in LFR mode. In contrast, the photoinhibitionof mesocotyl elongation and the light-dependent gravitropicresponse of crown roots are HIRs and are mediated byphyA under continuous FR light and by some other phyto-chrome(s) under continuous R light.

The results presented here clearly show that rice phyAcontrols photomorphogenesis in two distinct modes of pho-toperception—VLFR and HIR—and are consistent with theresults obtained for phyA mutants in Arabidopsis and tomato(Nagatani et al., 1993; Parks and Quail, 1993; Whitelam et

al., 1993; Reed et al., 1994; van Tuinen et al., 1995). In Ara-bidopsis, phyA is known to control development and growth(Chory et al., 1996) in several different modes of action, in-cluding seed germination in VLFR mode (Shinomura et al.,1996) and inhibition of hypocotyl elongation in FR light–induced HIR mode (Nagatani et al., 1993; Parks and Quail,1993; Whitelam et al., 1993). Therefore, phyA seems to playa similar role in the deetiolation of monocot and dicot plants,in spite of the fact that FR light induces the inhibition of co-leoptile elongation in monocots (i.e., rice) and hypocotylelongation in dicots (i.e., Arabidopsis). Although elongationinhibition of both coleoptiles and hypocotyls seems to resultfrom the inhibition of cell elongation, the modes of photo-perception by phyA in those responses are quite different,the former being via VLFR and the latter being via HIR(Shinomura et al., 2000).

This difference raises the question of whether the twomodes of responses via phyA—VLFR and HIR—are com-pletely different or share some overlapping pathways.Recent findings in phytochrome biochemistry, including nu-clear localization and kinase activity, might lead us to aplausible explanation for these physiological responses(Reed, 1999; Neff et al., 2000). Important work in this regardincludes the following: (1) intracellular localization studies ofphyA (Kircher et al., 1999; Hisada et al., 2000); (2) the dem-onstration of light-dependent kinase activity of oat phyAs(Wong et al., 1989; Biermann et al., 1994; Yeh and Lagarias,1998); and (3) the identification of a protein that can bephosphorylated by oat phyA in a light-regulated manner(Fankhauser et al., 1999). On the basis of such information,it can be hypothesized that phyA signaling for photomor-phogenesis responses may depend on combinations of thevariables, including nuclear versus cytoplasmic location,phosphorylated versus unphosphorylated, and dark-synthe-sized Pr, photoconverted Pfr, and photocycled Pr. It re-mains to be determined in the near future which proteinsinteract with each species of phyA described above andhow these various control mechanisms interact with eachother.

Perspective

In rice, a minimum of three phytochrome genes (PHYA,PHYB, and PHYC) are expressed (Tahir et al., 1998). Recentevidence in Arabidopsis indicates that individual membersof the phytochrome family show different manners of photo-perception (Shinomura et al., 1996) and have specializedfunctions (Reed et al., 1994; Quail, 1998; Whitelam et al.,1998). To characterize the functions of all known phyto-chromes in rice, we have also been attempting to isolatephyB and phyC mutants from the mutant panels of rice, but nocandidates have been obtained to date. On the other hand,we recently identified three additional mutant alleles of thePHYA gene from the mutant panels. Previously, Hirochika etal. (1996) analyzed the border sequences of eight random

Dow

nloaded from https://academ

ic.oup.com/plcell/article/13/3/521/6009486 by guest on 18 July 2021

Isolation of Rice phyA Mutants 531

Tos17 insertion sites and found one of them to be a PHYAgene. Therefore, the chromosomal region where PHYA re-sides appears to be a focus for the transposition of Tos17.In fact, the findings of Hirochika’s group inspired us to at-tempt the isolation of Tos17 insertional mutants of phyto-chrome genes in rice. We have screened 11,600 mutantlines so far, but our failure to obtain the phyB and phyC mu-tants indicates that Tos17-induced mutations must be satu-rated to make this system applicable to any gene of interest.We would need to produce a large number of mutagenizedlines to achieve this saturation level, and further screening ofthe upcoming panels is expected to reveal the phyB andphyC mutants.

METHODS

Isolation of phyA Mutants from Mutant Panels

Large populations of rice (Oryza sativa) mutants generated by meansof Tos17-mediated mutagenesis were made available by the Labora-tory of Gene Function at the National Institute of Agrobiological Re-sources (Tsukuba, Japan). Details of mutagenesis with Tos17 havebeen described (Hirochika, 1999). To identify phyA mutants from alarge mutant population, the mutant lines were divided into severalgroups, and within each group the lines were aligned as two- orthree-dimensional matrices. A particular group of mutant lines wastermed a “mutant panel.” For example, a mutant panel of 625 linesconsisted of a matrix in which all of the mutant lines were aligned in25 rows and 25 columns. Plant samples were collected and mixed tomake a pool for DNA extraction from all of the individual mutants in aparticular row or column. The pooled DNA was subjected to poly-merase chain reaction (PCR) analysis so that a maximum of 50 (25 125) PCR reactions were enough to survey 625 mutant lines. A mutantline with the insertion of Tos17 in a specific gene could be identified byusing primers specific to the gene of interest and Tos17. If a desiredmutant exists in the panel, DNA pools from one row and one columnmust give amplified DNA fragments of the same size, and the mutantline can be addressed on the matrix by the row/column number.

Screening for the phyA Mutant by PCR Amplification

We screened three mutant panels (NA0, NB0, and NC0) for the isola-tion of phyA mutants. NA0 was a two-dimensional (23 3 24) mutantpanel, whereas NB0 and NC0 were three-dimensional (8 3 10 3 12)panels. Table 1 lists the details of two Tos17-specific primers (LTR1and LTR4) and four PHYA-specific primers (AF, BF, CF, and ER) thatwere used in all possible combinations for PCR amplification of 300ng of genomic DNA from each pooled DNA sample. To eliminatenonspecific amplifications, we designed additional PHYA-specific(AF1, BF1, CF1, and ER1) and Tos17-specific (LTR2 and LTR5) prim-ers just upstream of each primer for nested PCR (Table 1). The am-plified DNA fragments were cloned and sequenced to confirm theinsertion of Tos17.

Genomic DNA Isolation and DNA Gel Blot Hybridization

Genomic DNA was isolated from young leaves using the cetyl-tri-methyl-ammonium bromide method (Murray and Thompson, 1980).The genotype analysis was performed using 2 mg of genomic DNAdigested with PstI, fractionated on an 0.8% agarose gel, and trans-ferred to Hybond N1 membranes (Amersham Pharmacia Biotech). Afull-length clone of PHYA cDNA was random labeled and used as aprobe for the genotype analysis.

Immunochemical Analysis

Crude protein extracts were prepared from 7-day-old etiolated seed-lings of wild type and phyA mutants for the immunochemical detec-tion of PHYA. Approximately 0.5 g of seedlings was homogenized inthe presence of 1 mL of phytochrome extraction buffer (100 mM Tris-HCl, pH 8.3, 5 mM EDTA, 0.2% 2-mercaptoethanol, and 100 mMphenylmethylsulfonyl fluoride). The supernatant was recovered aftercentrifugation at 10,000g for 15 min twice, and 0.66 volume of satu-rated ammonium sulfate was added. The resultant precipitate wasdissolved in 100 mL of phytochrome extraction buffer, and the pro-tein concentration was determined. Fifty micrograms of each proteinextract was separated by SDS-PAGE (10% acrylamide gel) and elec-troblotted to a nitrocellulose membrane. The monoclonal antibodyused in this study was the anti-rye PHYA antibody (mAR07) from the

Table 1. Nucleotide Sequences of PHYA- and Tos17-Specific Primers

PHYA-SpecificPrimer Namea PHYA-Specific Primer Sequence Primer Positionb

Tos17-SpecificPrimer Namec Tos17-Specific Primer Sequence

AF GAAGCATGGAGGTGCTATGC 5278–5297 LTR1 TTGGATCTTGTATCTTGTATATACAF1 TATGCAATACGGTGGTCAAGG 5293–5313 LTR2 GCTAATACTATTGTTAGGTTGCAABF CCAACATCATGGACCTTGTG 5944–5963 LTR4 CTGGACATGGGCCAACTATACAGTBF1 ACGTGATATTGCCTTCTGGC 6038–6057 LTR5 CCAATGGACTGGACATCCGATGGGCF CCTCATCCATCGAGTTCTAGC 4327–4347CF1 ATCTTGTTCAGCAGGAGTGG 4558–4577ER AGGATGAAGGTTGACATGCC 8583–8564ER1 GACATCGCCATTCATGAGC 8545–8527

a AF1, BF1, CF1, and ER1 are the nested PCR primers of AF, BF, CF, and ER, respectively.b Position of the first and last nucleotides in phy18 (PHYA) sequence (GenBank accession number X14172).c LTR2 and LTR5 are the nested PCR primers of LTR1 and LTR4, respectively.

Dow

nloaded from https://academ

ic.oup.com/plcell/article/13/3/521/6009486 by guest on 18 July 2021

532 The Plant Cell

monoclonal antibody stock at the Hitachi Advanced Research Labora-tory (Hatoyama, Japan). Immunochemical detection was performedusing an immunoblot kit (alkaline phosphatase system; Bio-Rad).

Plant-Growth Measurement and Light Treatment

For measurements of coleoptile and mesocotyl length, dehuskedrice seed were surface sterilized and placed on 0.6% agar in glasstubes. The glass tubes were kept in a light-tight box at 48C for 3days. Seed were placed on the agar so that embryos had an upwardorientation. The seed in the glass tubes were transferred subse-quently to 258C in the dark for 24 hr before the light treatment. In theexperiment with continuous light irradiation, seedlings were exposedto continuous red (R) light (photon fluence of 50 mmol m22 sec21) orcontinuous far-red (FR) light (photon fluence of 90 mmol m22 sec21)for 11 days. For the experiments with different intervals of continu-ous FR light treatments on mesocotyl growth, 2-day-old etiolatedseedlings were irradiated with continuous FR light for intervals rang-ing from1 hr to 5 days and kept in darkness until the measurement onthe 7th day. For irradiation with pulses of light, 4-day-old etiolatedseedlings with coleoptile lengths of 13.2 6 1.5 mm in the wild typeand 10.9 6 1.4 mm in osphyA-1 were exposed to a pulse of R light(17.8 mmol m22 sec21 for 56 sec for low fluence R [LF-R], 142 nmolm22 sec21 for 7 sec for very low fluence [VLF-R]) and/or FR light (28.4mmol m22 sec21 for 703 sec) and then kept in the dark for 2 days untilthe measurement. The lengths of coleoptiles and mesocotyls weremeasured using a digimatic caliper (CD-15C; Mitsutoyo, Tokyo, Ja-pan). The R and FR light were obtained using broad-band filters ac-cording to Shinomura et al. (1994). For measurement of the growthorientation of crown roots and seminal roots, seedlings were grownin the dark or under continuous R or FR light for 7 days. The orienta-tion of root growth was measured using a protractor as the angle be-tween the horizontal plane (08) and the orientation of the proximal 10-mm part of crown roots of 10 to 30 mm in length (or of the seminalroot of 20 to 80 mm in length). Orthogravitropic orientation was mea-sured as a positive angle. All experiments were conducted threetimes with 30 to 50 plants per data point. To examine the effect ofdifferent intervals of continuous FR light treatments on gravitropicgrowth orientation of the first growing crown roots, 4-day-old etiolatedseedlings were irradiated with continuous FR light for intervals rang-ing from 15 min to 48 hr and kept in darkness until the measurementon the 7th day.

Plant Materials and Growth Conditions

Seed of Nipponbare and the mutant line osphyA-2-10 were sown onApril 21, 1999, and seedlings were transplanted on May 26 in an irri-gated rice field. Plants were grown under natural daylength condi-tions, and the heading date was monitored for the appearance of thefirst panicle. Nipponbare and mutant lines osphyA-2-10 and osphyA-2-9 also were grown in a growth chamber in short-day conditions(288C; light cycle, 10.5 hr of light/13.5 hr of dark).

RNA Isolation

For RNA isolation, dehusked and sterilized seed were sown on 0.8%agar and kept at 48C for 1 day. The rice seedlings were grown at 288Cfor 7 days in complete darkness and then transferred to continuous Rlight (2.0 mmol m22 sec21) or FR light (29 mmol m22 sec21). The seed-

lings were harvested at 4 time points (0, 4, 12, and 24 hr) after the ir-radiation. Light sources were white fluorescent tubes filtered throughtwo layers of R film (red No. 22; Tokyo Butai Shomei, Tokyo, Japan)for R light and special FR fluorescent tubes (FL20S.FR-74; Toshiba,Tokyo, Japan) filtered through red and blue film (red No. 22 and blueNo. 72; Tokyo Butai Shomei) for FR light. The seedlings to be used ascontrols were kept in the dark. Total RNA for the light-induced geneexpression experiments was isolated from leaf parts of seedlingswith the RNeasy Plant Mini Kit (Qiagen, Valencia, CA).

ACKNOWLEDGMENTS

We thank H. Hanzawa for providing monoclonal antibody, W. Kuriharafor immunochemical analysis, K. Ikeda for physiological experi-ments, and Prof. T. Yokota for discussion and support. We aregrateful to Drs. M. Tahir, J.M. Reed, and C. Scutt for critical readingof the manuscript. We also thank K. Yagi and Y. Iguchi for their tech-nical assistance. This work was partly supported by grants from theProgram for the Promotion of Basic Research Activities for Innova-tive Biosciences, Japan, to M.T., H.H., and M.F. and from the HitachiAdvanced Research Laboratory (Grant B2023) to M.F.

Received September 25, 2000; accepted January 23, 2001.

REFERENCES

Biermann, B.J., Pao, L.I., and Feldman, L.J. (1994). Pr-specificphytochrome phosphorylation in vitro by a protein kinase presentin anti-phytochrome maize immunoprecipitates. Plant Physiol.105, 243–251.

Blaauw, O.H. (1961). The influence of blue, red and far-red light ongeotropism and growth of the Avena seedling. Acta Bot. Neerl.10, 397–450.

Briggs, W.R., and Huala, E. (1999). Blue-light photoreceptors inhigher plants. Annu. Rev. Cell Dev. Biol. 15, 33–62.

Briggs, W.R., Mandoli, D.F., Shinkle, J.R., Kaufman, L.S., Watson,J.C., and Thompson, W.F. (1984). Phytochrome regulation ofplant development at the whole plant, physiological, and molecu-lar levels. In Sensory Perception and Transduction in AneuralOrganisms, G. Colombetti, F. Lenci, and P.-S. Song, eds (NewYork: Plenum Press), pp. 265–280.

Casal, J.J., Clough, R.C., and Vierstra, R.D. (1996). High-irradi-ence response induced by far-red light in grass seedlings of thewild type or overexpressing phytochrome A. Planta 200, 132–137.

Childs, K.L., Miller, F.R., Pratt, M.M.C., Pratt, L.H., Morgan, P.W.,and Mullet, J.E. (1997). The sorghum photoperiod sensitivity gene,Ma3, encodes a phytochrome B. Plant Physiol. 113, 611–619.

Chory, J., Chatterjee, M., Cook, R.K., Elich, T., Fankhauser, C.,Li, J., Nagpal, P., Neff, M., Pepper, A., Poole, D., Reed, J., andVitart, V. (1996). From seed germination to flowering, light con-trols plant development via the pigment phytochrome. Proc. Natl.Acad. Sci. USA 93, 12066–12071.

Christensen, A.H., and Quail, P.H. (1989). Structure and expres-sion of a maize phytochrome-encoding gene. Gene 85, 381–390.

Dow

nloaded from https://academ

ic.oup.com/plcell/article/13/3/521/6009486 by guest on 18 July 2021

Isolation of Rice phyA Mutants 533

Clack, T., Mathews, S., and Sharrock, R.A. (1994). The phyto-chrome apoprotein family in Arabidopsis is encoded by fivegenes: The sequences and expression of PHYD and PHYE. PlantMol. Biol. 25, 413–427.

Clough, R.C., Casal, J.J., Jordan, E.T., Christou, P., and Vierstra,R.D. (1995). Expression of functional oat phytochrome A in trans-genic rice. Plant Physiol. 109, 1039–1045.

Dehesh, K., Tepperman, J., Christensen, A.H., and Quail, P.H.(1991). phyB is evolutionarily conserved and constitutively ex-pressed in rice seedling shoots. Mol. Gen. Genet. 225, 305–313.

Fankhauser, C., and Chory, J. (1997). Light control of plant devel-opment. Annu. Rev. Cell Dev. Biol. 13, 203–229.

Fankhauser, C., Yeh, K.C., Lagarias, J.C., Zhang, H., Elich, T.D.,and Chory, J. (1999). PKS1, a substrate phosphorylated by phy-tochrome that modulates light signaling in Arabidopsis. Science284, 1539–1541.

Feldman, L.J. (1984). Regulation of root development. Annu. Rev.Plant Physiol. 35, 223–242.

Feldman, L.J., and Briggs, W.R. (1987). Light-regulated gravitro-pism in seedling roots of maize. Plant Physiol. 83, 241–243.

Furuya, M. (1993). Phytochromes: Their molecular species, genefamilies, and functions. Annu. Rev. Plant Physiol. Plant Mol. Biol.44, 617–645.

Gallego, M.E., Sirand-Pugnet, P., and White, C.I. (1999). Positive-negative selection and T-DNA stability in Arabidopsis transforma-tion. Plant Mol. Biol. 39, 83–93.

Hamazato, F., Shinomura, T., Hanzawa, H., Chory, J., andFuruya, M. (1997). Fluence and wavelength requirements for Ara-bidopsis CAB gene induction by different phytochromes. PlantPhysiol. 115, 1533–1540.

Hershey, H.P., Barker, R.F., Idler, K.B., Lissemore, J.L., andQuail, P.H. (1985). Analysis of cloned cDNA and genomicsequences for phytochrome: Complete amino acid sequences fortwo gene products expressed in etiolated Avena. Nucleic AcidsRes. 13, 8543–8559.

Hirochika, H. (1997). Retrotransposons of rice: Their regulation anduse for genome analysis. Plant Mol. Biol. 35, 231–240.

Hirochika, H. (1999). Retrotransposons of rice as a tool for forwardand reverse genetics. In Molecular Biology of Rice, K. Shimamoto,ed (Tokyo: Springer-Verlag), pp. 43–58.

Hirochika, H., Sugimoto, K., Otsuki, Y., Tsugawa, H., and Kanda,M. (1996). Retrotransposons of rice involved in mutations inducedby tissue culture. Proc. Natl. Acad. Sci. USA 93, 7783–7788.

Hisada, A., Hanzawa, H., Weller, J.L., Nagatani, A., Reid, J.B.,and Furuya, M. (2000). Light-induced nuclear translocation ofendogenous pea phytochrome A visualized by immunocytochem-ical procedures. Plant Cell 12, 1063–1078.

Hoshikawa, K. (1989). The Growing Rice Plant. (Tokyo: Nobunkyo).

Izawa, T., Oikawa, T., Tokutomi, S., Okuno, K., and Shimamoto,K. (2000). Phytochromes confer the photoperiodic control of flow-ering in rice (a short day plant). Plant J. 22, 391–399.

Jelesko, J.G., Harper, R., Furuya, M., and Gruissem, W. (1999).Rare germinal unequal crossing-over leading to recombinant geneformation and gene duplication in Arabidopsis thaliana. Proc. Natl.Acad. Sci. USA 96, 10302–10307.

Kay, S.A., Keith, B., Shinozaki, K., and Chua, N.H. (1989). Thesequence of the rice phytochrome gene. Nucleic Acids Res. 17,2865–2866.

Kircher, S., Kozma-Bognar, L., Kim, L., Adam, E., Harter, K.,Schafer, E., and Nagy, F. (1999). Light quality–dependent nuclearimport of the plant photoreceptors phytochrome A and B. PlantCell 11, 1445–1456.

Koornneef, M., and Kendrick, R.E. (1994). Photomorphogenicmutants of higher plants. In Photomorphogenesis in Plants, 2nded, R.E. Kendrick and G.M.H. Kronenberg, eds (Dordrecht, TheNetherlands: Kluwer Academic Publishers), pp. 601–628.

Kuno, N., Muramatsu, T., Hamazato, F., and Furuya, M. (2000).Identification by large-scale screening of phytochrome-regulatedgenes in etiolated seedlings of Arabidopsis using a fluorescentdifferential display technique. Plant Physiol. 122, 15–24.

Mandoli, D.F., Tepperman, J., Huala, E., and Briggs, W.R. (1984).Photobiology of diagravitropic Maize roots. Plant Physiol. 75,359–363.

Mohr, H. (1962). Primary effects of light on growth. Annu. Rev. PlantPhysiol. 13, 465–488.

Murray, M.G., and Thompson, W.F. (1980). Rapid isolation of highmolecular weight plant DNA. Nucleic Acids Res. 8, 4321–4325.

Nagatani, A., Reed, J.W., and Chory, J. (1993). Isolation and initialcharacterization of Arabidopsis mutants that are deficient in phy-tochrome A. Plant Physiol. 102, 269–277.

Neff, M.M., Fankhauser, C., and Chory, J. (2000). Light: An indica-tor of time and place. Genes Dev. 14, 257–271.

Parks, B.M., and Quail, P.H. (1993). hy8, a new class of Arabidop-sis long hypocotyl mutants deficient in functional phytochrome A.Plant Cell 5, 39–48.

Pjon, C.-J., and Furuya, M. (1967). Phytochrome function in Oryzasativa L. I. Growth responses of etiolated coleoptiles to red, far-red and blue light. Plant Cell Physiol. 8, 709–718.

Quail, P.H. (1998). The phytochrome family: Dissection of functionalroles and signaling pathways among family members. Philos.Trans. R. Soc. Lond. B Biol. Sci. 353, 1399–1403.

Quail, P.H., Boylan, M.T., Parks, B.M., Short, T.W., Xu, Y., andWagner, D. (1995). Phytochromes: Photosensory perception andsignal transduction. Science 268, 675–680.

Reed, J.W. (1999). Phytochromes are Pr-ipatetic kinases. Curr.Opin. Plant Biol. 2, 393–397.

Reed, J.W., Nagatani, A., Elich, T.D., Fagan, M., and Chory, J.(1994). Phytochrome A and phytochrome B have overlapping butdistinct functions in Arabidopsis development. Plant Physiol. 104,1139–1149.

Robson, P.R., McCormac, A.C., Irvine, A.S., and Smith, H. (1996).Genetic engineering of harvest index in tobacco through overex-pression of a phytochrome gene. Nat. Biotechnol. 14, 995–998.

Sharrock, R.A., and Quail, P.H. (1989). Novel phytochromesequences in Arabidopsis thaliana: Structure, evolution and differ-ential expression of a plant regulatory photoreceptor family.Genes Dev. 3, 1745–1757.

Shimamoto, K. (1995). The molecular biology of rice. Science 270,1772–1773.

Dow

nloaded from https://academ

ic.oup.com/plcell/article/13/3/521/6009486 by guest on 18 July 2021

534 The Plant Cell

Shinomura, T., Nagatani, A., Chory, J., and Furuya, M. (1994). Theinduction of seed germination in Arabidopsis thaliana is regulatedprincipally by phytochrome B and secondarily by phytochrome A.Plant Physiol. 104, 363–371.

Shinomura, T., Nagatani, A., Hanzawa, H., Kubota, M.,Watanabe, M., and Furuya, M. (1996). Action spectra for phyto-chrome A- and B-specific photoinduction of seed germination inArabidopsis thaliana. Proc. Natl. Acad. Sci. USA 93, 8129–8133.

Shinomura, T., Uchida, K., and Furuya, M. (2000). Elementary pro-cesses of photoperception by phytochrome A for high-irradianceresponse of hypocotyl elongation in Arabidopsis. Plant Physiol.122, 147–156.

Tahir, M., Kanegae, H., and Takano, M. (1998). Phytochrome C(PHYC) gene in rice: Isolation and characterization of a completecoding sequence. Plant Physiol. 118, 1535.

Thykjaer, T., Finnemann, J., Schauser, L., Christensen, L.,Poulsen, C., and Stougaard, J. (1997). Gene targetingapproaches using positive-negative selection and large flankingregions. Plant Mol. Biol. 35, 523–530.

van Tuinen, A., Kerckhoffs, L.H., Nagatani, A., Kendrick, R.E.,and Koornneef, M. (1995). Far-red light-insensitive, phytochromeA–deficient mutants of tomato. Mol. Gen. Genet. 246, 133–141.

Weller, J.L., Murfet, I.C., and Reid, J.B. (1997). Pea mutants with

reduced sensitivity to far-red light define an important role forphytochrome A in day-length detection. Plant Physiol. 114, 1225–1236.

Whitelam, G.C., Johnson, E., Peng, J., Carol, P., Anderson, M.L.,Cowl, J.S., and Harberd, N.P. (1993). Phytochrome A nullmutants of Arabidopsis display a wild-type phenotype in whitelight. Plant Cell 5, 757–768.

Whitelam, G.C., Patel, S., and Devlin, P.F. (1998). Phytochromesand photomorphogenesis in Arabidopsis. Philos. Trans. R. Soc.Lond. B Biol. Sci. 353, 1445–1453.

Wilkins, M.B., and Goldsmith, M.H.M. (1964). The effect of red, far-red and blue light on the geotropic response of coleoptiles of Zeamays. J. Exp. Bot. 15, 600–615.

Wong, Y.-S., McMichael, R.J., and Lagarias, J. (1989). Propertiesof a polycation-stimulated protein kinase associated with purifiedAvena phytochrome. Plant Physiol. 91, 709–718.

Yamamoto, K., and Sasaki, T. (1997). Large-scale EST sequencingin rice. Plant Mol. Biol. 35, 135–144.

Yeh, K.C., and Lagarias, J.C. (1998). Eukaryotic phytochromes:Light-regulated serine/threonine protein kinases with histidinekinase ancestry. Proc. Natl. Acad. Sci. USA 95, 13976–13981.

Yokoo, M., and Okuno, K. (1993). Genetic analysis of earliness muta-tions induced in the rice cultivar Norin 8. Jpn. J. Breed. 43, 1–11.

Dow

nloaded from https://academ

ic.oup.com/plcell/article/13/3/521/6009486 by guest on 18 July 2021