NYC4, the rice ortholog of Arabidopsis THF1, is involved in the degradation of chlorophyll – protein complexes during leaf senescence


For correspondence (e-mail


Yellowing/chlorophyll breakdown is a prominent phenomenon in leaf senescence, and is associated with the degradation of chlorophyll – protein complexes. From a rice mutant population generated by ionizing radiation, we isolated nyc4-1, a stay-green mutant with a defect in chlorophyll breakdown during leaf senescence. Using gene mapping, nyc4-1 was found to be linked to two chromosomal regions. We extracted Os07g0558500 as a candidate for NYC4 via gene expression microarray analysis, and concluded from further evidence that disruption of the gene by a translocation-related event causes the nyc4 phenotype. Os07g0558500 is thought to be the ortholog of THF1 in Arabidopsis thaliana. The thf1 mutant leaves show variegation in a light intensity-dependent manner. Surprisingly, the Fv/Fm value remained high in nyc4-1 during the dark incubation, suggesting that photosystem II retained its function. Western blot analysis revealed that, in nyc4-1, the PSII core subunits D1 and D2 were significantly retained during leaf senescence in comparison with wild-type and other non-functional stay-green mutants, including sgr-2, a mutant of the key regulator of chlorophyll degradation SGR. The role of NYC4 in degradation of chlorophyll and chlorophyll – protein complexes during leaf senescence is discussed.


Chlorophyll is an essential molecule for harvesting light energy during photosynthesis. Higher plants contain two species of chlorophyll, chlorophyll a (Chl a) and chlorophyll b (Chl b). Chl a is found in all chlorophyll – protein complexes in the thylakoid membrane, including photosystem I (PSI), photosystem II (PSII) and the cytochrome b6f complex. PSI and PSII are multi-subunit protein complexes that play a central role in the absorption and transfer of light energy. Chl b is present only in PSI-associated light-harvesting complex I (LHCI) and PSII-associated light-harvesting complex II (LHCII), which are localized in thylakoid membranes and have a role in energy harvesting. LHCI and LHCII apoproteins are encoded by Lhca and Lhcb genes, respectively.

During leaf senescence, chlorophyll and these chlorophyll – protein complexes are degraded. Yellowing is caused by the degradation of chlorophyll and by unmasking of pre-existing carotenoids (Matile, 2000). The first step in degradation of the major chlorophyll, Chl a, was believed to be removal of a phytol residue by chlorophyllase; however, a recent study suggested that there is another pathway for chlorophyll breakdown involving the α/β hydrolase-fold family protein PHEOPHYTIN PHEOPHORBIDE HYDROLASE (PPH)/NON-YELLOW COLORING3 (NYC3) (Tanaka et al., 2011). PPH removes phytol residues from pheophytin a (Phein a), generating pheophorbide a (Pheide a) in vitro (Schelbert et al., 2009). This pathway suggests that the first step of Chl a degradation is de-chelation of Mg2+. Pheide a may be generated by the chlorophyllase pathway. Pheide a is converted into a red chlorophyll catabolite by pheophorbide a oxygenase (Pružinská et al., 2003; Tanaka et al., 2003). The first step of Chl b degradation is reduction of Chl b into 7-hydroxymethyl chlorophyll a by NON-YELLOW COLORING1 (NYC1) and NYC1-like (NOL) (Kusaba et al., 2007; Sato et al., 2009). 7-hydroxymethyl chlorophyll a is further reduced to Chl a by 7-hydroxymethyl Chl a reductase (Meguro et al., 2011), i.e. Chl b is converted into Chl a and degraded via the Chl a degradation pathway described above.

Stay-green mutants, which retain green during senescence, are useful tools for the analysis of leaf senescence. They are classified into functional and non-functional mutants. In functional stay-green mutants, the progression of senescence is retarded under senescence-inductive conditions. In non-functional stay-green mutants, the progression of senescence is not retarded for most senescence parameters, except chlorophyll breakdown. A number of chlorophyll catabolic enzyme genes have been identified as the non-functional stay-green genes. So far, STAYGREEN (SGR)/NONYELLOWING1 (NYE1) is the only non-functional stay-green gene that does not encode a chlorophyll catabolic enzyme (Jiang et al., 2007; Park et al., 2007; Ren et al., 2007). Recently, SGR was found to recruit chlorophyll catabolic enzymes and promote chlorophyll breakdown (Sakuraba et al., 2012). Loss-of-function of SGR results in inhibition of chlorophyll breakdown, while its over-expression causes premature yellowing, suggesting that the transcript level of SGR is a critical regulatory factor of chlorophyll breakdown. The majority of non-functional stay-green mutants retain not only chlorophylls but also chlorophyll-binding proteins, suggesting a close relationship between chlorophyll breakdown and the stability of chlorophyll – protein complexes. For example, degradation of LHCII during leaf senescence requires the breakdown of Chl b, while degradation of LHCII is necessary for the breakdown of Chl a and some carotenoids contained in LHCII (Kusaba et al., 2007; Sato et al., 2009).

In this study, we isolated the stay-green mutant non-yellow coloring 4 (nyc4) in rice (Oryza sativa). In nyc4, the Fv/Fm value, which reflects PSII activity, remains at a high level during senescence. This is a very different characteristic from those observed in the non-functional stay-green mutants sgr and nyc3 (Sato et al., 2007; Morita et al., 2009). Consistent with this observation, the PSII core subunits were stable during senescence in nyc4. The nyc4-1 mutation was induced by a translocation-related chromosomal rearrangement. Map-based cloning of genes disrupted by a translocation has rarely been reported, but we successfully identified NYC4 using a combination of coarse mapping and microarray analysis. NYC4 encodes an ortholog of THYLAKOID FORMATION1 (THF1) in Arabidopsis thaliana (Wang et al., 2004). THF1 is a multi-function protein that is involved in acclimation to high light, sugar sensing and disease resistance (Keren et al., 2005; Huang et al., 2006; Zhang et al., 2009; Wangdi et al., 2010). We discuss a possible role for NYC4 in the degradation of chlorophyll and chlorophyll – protein complexes.


Isolation and characterization of nyc4

During screening of rice M2 plants mutagenized using carbon ion beams, we isolated non-yellow coloring 4-1 (nyc4-1), the detached leaves of which retain green at 7 days after dark incubation (DAD) at 28ºC (Figure 1a). Before dark incubation, the levels of cholorophyll were 3.09 nmol mg−1 FW (Chl a) and 0.85 nmol mg−1 FW (Chl b) in the wild-type cultivar Nipponbare, and 3.61 nmol mg−1 FW (Chl a) and 0.87 nmol mg−1 FW (Chl b) in nyc4-1 (Figure 1b). The Chl a content in nyc4-1 was slightly higher than that in the wild-type (Student's t test, < 0.05). During the 8 days of dark incubation, the Chl a and Chl b contents decreased drastically in Nipponbare, but 3.6- and 2.3-fold higher levels of Chl a and Chl b, respectively, were observed in nyc4-1. The Chl a/b ratio was largely unchanged in nyc4-1, being 4.1 before dark incubation and 3.9 at 8 DAD. This is in contrast to nyc1 and nol, in which the Chl a/b ratio becomes approximately 1 at the late stage of leaf senescence (Kusaba et al., 2007; Sato et al., 2009). Membrane ion leakage, a senescence parameter reflecting cell death, increased during dark incubation similarly in Nipponbare and nyc4-1, suggesting decreased leaf functionality in nyc4-1 despite the retention of green; thus, nyc4-1 is a non-functional stay-green mutant (Figure 1c). Surprisingly, the Fv/Fm value, which reflects PSII activity, was maintained at a high level of approximately 0.8, even during the late stage of senescence in nyc4-1, but decreased to <0.6 in Nipponbare (Figure 1d). This feature of nyc4 is unique among non-functional stay-green mutants of rice, including nyc1, sgr, nol and nyc3 (Kusaba et al., 2007; Sato et al., 2007, 2009; Morita et al., 2009). In nyc4-1, expression of the senescence-inducible genes involved in chlorophyll degradation (SGR, NYC3, NYC1, NOL, RCCR1 and OsPAO) and other senescence-inducible genes [Os07g0683200 encoding a NAC transcription factor (Sato et al., 2011b) and Osh36 (Lee et al., 2001)] was induced at 5 and 8 DAD, suggesting that senescence proceeded in nyc4-1 (Figure 1e). This observation confirms that nyc4-1 is a non-functional stay-green mutant, although its Fv/Fm value is maintained during senescence. nyc4-1 plants had slightly smaller plant height and a lower tiller number than wild-type plants did, and also showed a stay-green phenotype under natural growth conditions, although the lower leaves turned white as described below (Figure S1a,b).

Figure 1.

Physiological characterization of nyc4-1.

(a) Dark-induced leaf senescence in nyc4-1. The same leaves are shown at various incubation times (0, 4 and 7 DAD). nyc4-1 leaves did not yellow even at 7 DAD, when wild-type leaves were completely yellow.

(b) Change of chlorophyll content over time in nyc4-1 during dark incubation. Solid line, wild-type; dotted line, nyc4-1; closed circles, Chl a; open circles, Chl b. The bars indicate standard errors (= 3).

(c) Change of membrane ion leakage over time in nyc4-1 during dark incubation. Values relative to the total ions in the boiled samples are shown (%). Solid line, wild-type; dotted line, nyc4-1. The bars indicate standard errors (= 4).

(d) Change of Fv/Fm value over time in nyc4-1 during dark incubation. Solid line, wild-type; dotted line, nyc4-1. The bars indicate standard errors (= 4).

(e) Quantitative RT-PCR analysis of senescence-inducible genes in nyc4-1. mRNA was extracted from the detached leaves of Nipponbare or nyc4-1 incubated in the dark. Actin2 was used as a reference gene.

Positional cloning of NYC4

To isolate NYC4 by positional cloning, we crossed nyc4-1 with the indica wild-type cultivar ‘Kasalath.’ The resulting F1 plants showed the wild-type phenotype for dark-induced senescence, but with reduced fertility, although the nyc4-1 homozygous line did not show an obvious reduction of fertility. From the F2 plants, we selected nine plants showing a reliable stay-green phenotype under dark incubation. Unexpectedly, all the plants with the stay-green phenotype were homozygous for the japonica (mutant) allele in the chromosome 3 markers RM3872 (33.6 cm), RM3545 (36.9 cm) and MRG0338 (40.3 cm), and the chromosome 7 markers RM1973 (70.3 cm), RM3826 (73.2 cm), RM3799 (75.6 cm), C53905 (81.4 cm) and MRG0330 (96.1 cm), suggesting that nyc4-1 maps to two chromosomal regions (Figure 2a and Table S1). Given the low fertility of the F1 plants and the mapping results, we hypothesize that nyc4-1 was generated by a chromosomal rearrangement involving a reciprocal translocation. This hypothesis raises the possibility that the translocation eliminates transcription of the 3′ part of NYC4 due to the loss of its promoter region.

Figure 2.

Map-based cloning of NYC4.

(a) The two mapped regions of nyc4-1. The closed circles indicate markers that show complete linkage to the stay-green phenotype. The open circles indicate markers that show linkage with some recombinants. The arrows indicate the positions of Os03g0260100 and Os07g0558500.

(b) Microarray analysis of gene expression within the candidate region of chromosome 7 (Os07g0480800–Os07g0661400) in nyc4-1. The ratios of expression in nyc4-1 to that in wild-type are shown. Mean values for pre-senescent (0 DAD) and senescent (3, 5 and 8 DAD) leaves were used for analysis. Os07g0558500 is indicated by an arrowhead.

(c) Structures of Os07g0558500 and Os03g0260100 and the fused gene in nyc4-1. Os07g0558500 and Os03g0260100 were fused in nyc4-1. The breakpoint is shown by the vertical dotted lines. The arrows labeled ‘Inv’ indicate the positions of PCR primers used for the inverse PCR. The arrows labeled ‘7gF’, ‘7gR’ and ‘3gR’ are the PCR primers used in (d). The grey and white boxes indicate the coding sequence of Os07g0558500 and Os03g0260100.

(d) Complementation of nyc4-1 with an Os07g0558500 genomic clone. nyc4-1 harboring a 10 kb SmaI fragment containing the entire Os07g0558500 gene showed yellowing at 7 DAD, although nyc4-1 did not exhibit yellowing. nyc4-1 and the transformant harbor the fused gene amplified using primer pair 7gF and 3gR, and the transformant also harbors wild-type Os07g0558500 (amplified using primer pair 7gF and 7gR) and hygromycin phosphotransferase (HPT).

Thus, we compared the expression signature of genes located in the mapping regions between wild-type and nyc4-1 using the rice 4 × 44K microarray RAP-DB, in which most of the probes are designed based on alignment of the 3′ untranslated region (UTR) of each gene. A comparison between Nipponbare and nyc4-1 mRNAs in pre-senescent (0 DAD) and senescent (3, 5 and 8 DAD) leaves revealed that expression of Os07g0558500, which is located within the mapped region of chromosome 7, was drastically reduced in nyc4-1 (Figure 2b). To test the hypothesis that reduced expression of this gene was caused by a translocation, we further investigated the breakpoint at Os07g0558500. First, we tried to amplify several fragments of Os07g0558500 by PCR, and found that a particular pair of primers (7gF and 7gR) did not amplify the predicted DNA fragment in nyc4-1 (Figure 2c,d). On the basis of the assumption that the translocation breakpoint is between these primers, we performed an inverse PCR experiment using the DNA sequence of this region (Figure 2c and Figure S2). The amplified DNA fragment was found to contain part of Os03g0260100 following Os07g0558500, suggesting that the translocation occurred between Os07g0558500 and Os03g0260100 (Figure 2c). Consistent with this, Os03g0260100 was located within the other mapped region of nyc4-1 (Figure 2a). The rearrangement is thought to have occurred between the third intron of Os07g0558500 and the first intron of Os03g0260100. This rearrangement produced a fusion gene between Os07g055850 and Os03g0260100, with a frameshift in the fusion gene. Taken together with the fact that, nyc4-1 is recessive, nyc4-1 is not a gain-of-function mutation. Thus both Os07g0558500 and Os03g0260100 may be candidates for NYC4.

Os03g0260100 encodes a membrane protein with an unknown function. Os07g0558500 encodes a rice protein that is similar to THYLAKOID FORMATION1 (THF1), a chloroplast-targeted protein in A. thaliana (Wang et al., 2004; Keren et al., 2005). We introduced a genomic fragment harboring the entire Os07g0558500 gene into nyc4-1 to complement the stay-green phenotype. The resultant transformant showed yellowing but nyc4-1 retained green at 7 DAD, suggesting that Os07g0558500 is the NYC4 gene (Figure 2d).

NYC4 is the functional ortholog of THF1

According to phylogenetic analysis, Os07g0558500 is thought to be an ortholog of THF1 (Wang et al., 2004). The molecular function of THF1 is poorly understood, except for its chloroplast localization and interaction with the α-subunit of G proteins (Huang et al., 2006). To confirm that NYC4 is the functional ortholog of THF1, we first examined the chloroplast localization of NYC4 using an NYC4–GFP fusion protein (Figure 3a). The plasmid harboring NYC4–GFP was introduced transiently together with a plastid-localizing S9–RFP plasmid into onion epidermal cells via particle bombardment. NYC4–GFP co-localized with S9–RFP, suggesting that NYC4 is a chloroplast protein, similar to THF1.

Figure 3.

NYC4 is the functional ortholog of THF1.

(a) Intracellular localization of NYC4. The GFP and RFP constructs were transiently co-introduced into onion epidermal cells by particle bombardment. NYC4–GFP, GFP fused with full-length NYC4 at the N-terminal end; S9–RFP, plastid-targeted RFP protein; merged, merged images of the GFP and RFP images.

(b) The leaves of nyc4-1 just before heading. The flag leaf and the first, second and third leaves below the flag leaf of wild-type and nyc4-1 are shown.

(c) Light sensitivity of nyc4-1. Fv/Fm values for plants acclimatized to shade conditions (approximately 50 μmol m−2 sec−1; open bars) and plants exposed to high light for 30 min (approximately 2000 μmol mol m−2 sec−1; gray bars). The bars indicate standard errors.

The thf1 mutant of A. thaliana is a variegated leaf mutant, but the phenotype is not observed under low light conditions, suggesting that this phenotype is induced by normal/high light conditions (Keren et al., 2005). In nyc4-1, variegation was not observed even in the field, i.e. high light conditions, but the mutant plants turn white rather than yellow just before heading, which is similar to bleaching (Figure 3b). To determine whether the production of reactive oxygen species (ROS) in the nyc4-1 leaves was involved in the ‘bleaching’, we examined the ROS level in nyc4-1 and wild-type using diaminobenzidine tetrahydrochloride for hydrogen peroxide and nitroblue tetrazolium staining for superoxide (Figure S2). However, we did not observe significant differences in hydrogen peroxide or superoxide production between wild-type and nyc4-1. Next we examined the sensitivity to high light of nyc4, because thf1 is reportedly sensitive to high light (Figure 3c). Nipponbare and nyc4-1 plants placed under shaded conditions (approximately 50 μmol m−2 sec−1) for 3 h showed comparable Fv/Fm values of 0.82 and 0.81, respectively. At 30 min after transfer to field conditions (approximately 2000 μmolm−2 sec−1), the Fv/Fm value was slightly reduced in Nipponbare (0.79) but significantly reduced in nyc4-1 (0.475), suggesting that nyc4-1 is sensitive to light, which is similar to the observation for thf1 in A. thaliana. These observations confirm that NYC4 is an ortholog of THF1.

Expression of NYC4

By referring to RiceXPro, a rice gene expression profile database (Sato et al., 2011a,b), we found that NYC4 is expressed in various tissues, but is expressed more strongly in green tissues such as the leaf, stem, lemma and palea (Figure S3). We then examined the light/senescence regulation of NYC4 expression. Expression of NYC4 was down-regulated within 24 h of dark incubation, and then gradually increased during the extended dark incubation, and finally reached the level similar to that observed before dark incubation (Figure 4). This expression pattern is consistent with the function of NYC4: acclimation to high light and promotion of chlorophyll degradation in senescence.

Figure 4.

Expression of NYC4 during dark incubation.

The expression of NYC4 was analyzed by quantitative RT-PCR. mRNA was extracted from detached leaves of Nipponbare incubated in the dark. Actin2 was used as a reference gene.

Analysis of photosynthetic proteins

As mentioned above, nyc4 has unique characteristics among non-functional stay-green mutants, such as maintenance of the Fv/Fm value during senescence. To characterize nyc4 in relation to other non-functional stay-green mutants, nyc4-1 was compared with sgr-2 and nyc3-2 in terms of various parameters related to leaf senescence.

First we examined photosynthetic proteins by Western blot analysis (Figure 5). In Nipponbare, the LHCI apoproteins Lhca1 and Lhca2 were degraded completely at 7 DAD. nyc4-1, sgr-2 and nyc3-2 showed significant amounts of Lhca1 and Lhca2 at 7 DAD, but retention in sgr-2 was greater. Similarly, the PSI subunits PsaF and CP1 were degraded completely in Nipponbare at 7 DAD, but were retained in sgr-2. The major LHCII apoprotein Lhcb1 and the minor LHCII apoprotein Lhcb4 were more stable in nyc4-1, sgr-2 and nyc3-2 than in Nipponbare at 7 DAD, with particularly high levels in sgr-2. The PSII core subunits D1 and D2 were completely degraded in Nipponbare at 7 DAD but retained in nyc4-1; only a residual amount was detected in sgr-2 and nyc3-2 at 7 DAD. In addition, other PSII subunits (CP43 and CP47) were also more stable in nyc4-1 than in sgr-2 or nyc3-2 during senescence. In blue-native PAGE analysis, bands containing the PSII monomer and dimer were present in nyc4-1 at 7 DAD, consistent with these observations (Figure S4). These results suggest that nyc4 has a major effect on PSII stability during leaf senescence, while sgr has a stronger effect on the stability of PSI-related proteins, including PSI subunits and LHCI apoproteins. This idea is consistent with the observation that a high Fv/Fm value (i.e. PSII activity) was maintained during senescence in nyc4-1 only. No obvious difference in the stability of the Rubisco large subunit was observed between nyc4-1 and Nipponbare, consistent with the observation that nyc4-1 is a non-functional stay-green mutant.

Figure 5.

Stability of photosynthetic proteins during senescence in nyc4 and other non-functional stay-green mutants.

Western blot analysis was performed using various antibodies against photosynthetic proteins. Proteins were extracted from pre-senescent (0 DAD) and fully senescent (7 DAD) leaves of Nipponbare, nyc4-1, sgr-2 and nyc3-2. The Rubisco large subunit (Rubisco L) was visualized by Coomassie brilliant blue G-250 staining of the SDS–PAGE gel.

HPLC analysis of photosynthetic pigments

Next we examined photosynthetic pigments using HPLC analysis (Figure 6a). Higher levels of Chl a and Chl b were found in the leaves of nyc4-1 than those of Nipponbare at 6 DAD (Figure 1c). In addition, repression of carotenoid degradation and the conversion of lutein into lutein 3-acetate were apparent in nyc4-1 (Kusaba et al., 2009). Interestingly, a significant amount of Phein a was observed in nyc4-1, but Phein a was fully degraded in Nipponbare at 6 DAD (Figure 6b). Phein a is not only a component of D1, which is involved in electron transport, but is also an intermediate compound in the chlorophyll breakdown pathway. In the nyc3-2 mutant of the Phein a catabolic enzyme gene, Phein a content increased 3.5-fold during senescence, suggesting that the accumulated Phein a was generated from Chl a during its degradation in addition to the pre-existing Phein a contained in D1 and D2. In nyc4-1 and sgr-2, Phein a levels were largely unchanged during senescence, in contrast to nyc3-2.

Figure 6.

HPLC analysis of photosynthetic pigments in nyc4-1 and non-functional stay-green mutants.

(a) HPLC profiles of photosynthetic pigments extracted from pre-senescent (0 DAD) and senescent (6 DAD) leaves of Nipponbare and nyc4-1. Absorption was measured at 410 nm. N, neoxanthin; V, violaxanthin; L, lutein; La, lutein 3-acetate; b, Chl b; a, Chl a; P, Phein a; C, β-carotene.

(b) Accumulation of Phein a in senescent leaves of Nipponbare, nyc4, sgr and nyc3. Phein a content was measured by HPLC analysis.


Identification of the gene disrupted by a chromosome-level mutation

Ionizing radiation often induces chromosomal rearrangements such as inversions and translocations, which cause gene disruptions or, in some cases, gene fusions. In most cases, these mutations are loss-of-function and recessive; however, both loss-of-function and gain-of-function mutations are possible for gene fusions. Theoretically, these mutations may involve two gene mutations associated with two chromosomal breakpoints. Reports of map-based cloning of mutant genes caused by such mutations are rare, probably because of the difficulty in applying map-based cloning procedures to such a situation. For example, for an inversion, it is very difficult to narrow down the candidate region because recombination is suppressed in the inverted region. In this paper, we report map-based cloning of a mutant gene caused by a chromosomal rearrangement involving translocation. In general, F1 plants generated by crossing a wild-type plant with a mutant with a translocation show reduced fertility, making it difficult to obtain a sufficient number of homozygous mutant plants for fine mapping and gene cloning. We selected a limited number of mutant plants from the F2 population and identified two candidate regions for NYC4. To isolate NYC4, we performed microarray analysis because expression of the 3′ end of the disrupted gene may be reduced in the mutant due to loss of its promoter. We successfully identified a gene that was expressed at a very low level in the mutant within the mapped regions, which was identified as NYC4 in further analysis. Our strategy may be a useful one for identification of genes mutated by chromosome-level mutations, including large inversions. It is able to identify the mutated gene directly by very coarse mapping. Fluorescence in situ hybridization (FISH) mapping is another strategy to isolate mutated genes caused by chromosome-level mutations (Haas et al., 1993). FISH mapping identifies the bacterial artificial chromosome (BAC) clone that paints two chromosomal regions in a FISH image, suggesting that the BAC clone covers the breakpoint. To identify the mutated gene, FISH mapping requires finer mapping than in our approach. If the candidate region is large an enormous number of in situ hybridization experiments using BAC clones covering the candidate region would be required. Furthermore, gene cloning by FISH mapping requires identification of the mutated gene in the genomic clone. The expression of Os03g0260100, which is the other mutated gene in nyc4-1, was not reduced (instead it was rather up-regulated; Figure S5), suggesting that our microarray-based approach may not be always effective, but it is worth considering as a potential strategy. Alternatively, next-generation sequencing may be a useful approach for identification of translocation breakpoints (Chen et al., 2008).

Phein a metabolism and chlorophyll breakdown

Although Phein a is known to be accumulated in the pph mutant in Arabidopsis and the nyc3 mutant in rice, the Phein a content in other stay-green mutants has not been reported (Morita et al., 2009; Schelbert et al., 2009). In this study, we examined the Phein a content in nyc4-1 and two other stay-green mutants. Phein a accumulation increased in nyc3-2 during senescence, but was largely unchanged in nyc4-1 and sgr-2 during senescence. Because PPH/NYC3 encodes Phein a hydrolase, the increase in the level of Phein a in nyc3-2 is thought to be caused by blockage of the metabolism of Phein a. Only a small amount of Phein a was detected in the nyc3-2 mutant at 6 DAD compared to Chl a and Chl b (Phein a, 0.15 nmol mg−1 FW; Chl a, 1.96 nmol mg−1 FW: Chl b, 0.83 nmol mg−1 FW), indicating that the main green pigments observed in senescent nyc3 leaves are Chl a and Chl b. This suggests that the blockage of Phein a metabolism inhibits Chl a and Chl b breakdown, probably by a feedback regulation system via accumulation of Phein a. Phein a is thought to be retained in nyc4 due to the stabilization of D1/D2 during senescence. Consistent with this idea, the Fv/Fm values remained at high levels during senescence in nyc4. On the other hand, the majority of D1/D2 was degraded in sgr, suggesting that the accumulated Phein a existed as a free form. Phein a may not be metabolized into Pheide a due to impairment of PPH/NYC3 function in sgr. It is unclear whether the accumulated free Phein a in sgr has an additional effect on inhibition of chlorophyll breakdown.

The roles of NYC4 in degradation of chlorophyll and chlorophyll – protein complexes

Loss of SGR/NYE1 function results in the stay-green phenotype, while over-expression of SGR/NYE1 causes premature yellowing, suggesting that the expression level of SGR/NYE1 is a key regulatory factor for progression of chlorophyll breakdown (Jiang et al., 2007; Park et al., 2007; Ren et al., 2007). Furthermore, it has been reported that SGR is a protein that promotes chlorophyll breakdown by recruiting chlorophyll catabolic enzymes (Sakuraba et al., 2012). This suggests that SGR is a regulator of chlorophyll-degrading activity in vivo. Degradation of LHCII during senescence requires the function of the Chl b reductases NYC1 and NOL, suggesting that Chl b breakdown is necessary for LHCII degradation (Kusaba et al., 2007; Sato et al., 2009). Similarly, PSI and Lhca proteins were retained at high levels in sgr-2, suggesting that degradation of PSI and Lhca proteins depends on breakdown of chlorophyll, probably Chl a, because both PSI and Lhca proteins are degraded in Chl b reductase mutants during senescence. (Kusaba et al., 2007; Sato et al., 2009). However, the PSII core proteins D1, D2, CP43, and CP47 were degraded in sgr-2, suggesting that SGR function and chlorophyll breakdown are not very important for PSII degradation during senescence.

Both nyc4-1 and sgr-2 showed a stay-green phenotype, but the retention pattern of chlorophyll – protein complexes during senescence in nyc4-1 was different from that in sgr-2, particularly for D1 and D2, suggesting that NYC4 has a function distinct from SGR in degradation of chlorophyll – protein complexes. In view of the hypothesis that SGR is the key regulator of chlorophyll degradation (Jiang et al., 2007; Park et al., 2007; Ren et al., 2007; Sakuraba et al., 2012), it is probable that the primary function of NYC4 is not the regulation of chlorophyll breakdown; rather, NYC4 may be necessary for degradation of chlorophyll – protein complexes, particularly D1 and D2. This possibility is consistent with the fact that THF1 was originally identified as the PSII-associated Psb29 protein (Kashino et al., 2002).

Arabidopsis THF1 is reportedly involved in PSII activity, particularly under photo-inhibitory light conditions (Zhang et al., 2009). In fact, light-dependent D1 degradation, which is mediated by FtsH protease in the PSII repair cycle, was shown to be impaired in the thf1 mutant. Degradation of photo-damaged D1 involves several proteases, among which FtsH plays a central role as a processive enzyme (Kato and Sakamoto, 2009). Interestingly, lack of THF1 leads to a concomitant decrease in FtsH in chloroplasts, suggesting connectivity between THF1 and FtsH (Keren et al., 2005; Zhang et al., 2009). Although physical interaction of the two proteins was not reported, these data, together with the fact that THF1 is associated with PSII, suggest that THF1 is involved in PSII maintenance through protein degradation. Little is known about the factors and proteases involved in degradation of D1/PSII during leaf senescence. In nyc4, LHCII and PSI are partially retained in addition to PSII during leaf senescence. One possibility to explain this phenomenon is that defective PSII degradation indirectly affects the stability of LHCII (and PSI) during senescence. Alternatively, NYC4 may be partially involved in chlorophyll breakdown in addition to PSII degradation. In nyc4, SGR is induced normally but chlorophyll is retained during senescence (Figure 1e). NYC4 may be partially required for SGR to promote chlorophyll breakdown. While confirmation of these possibilities requires further experimentation, our characterization of nyc4 in rice demonstrates an interesting type of stay-green phenomenon that has not been reported previously.

In summary, we isolated a stay-green mutant gene in rice, NYC4, using a unique microarray-based approach. NYC4 encodes the ortholog of Arabidopsis THF1. nyc4 retained chlorophyll during senescence, although changes in other senescence parameters indicated that senescence proceeded, suggesting that nyc4 is a non-functional stay-green mutant. However, nyc4 retains some functionality, such as PSII activity (Fv/Fm), in contrast to other non-functional stay-green mutants. This phenomenon is consistent with the observation that PSII subunits are more stable than in other non-functional stay-green mutants. THF1 is a multi-functional protein that is involved in sugar signaling, disease resistance, and acclimation to high light (Keren et al., 2005; Huang et al., 2006; Zhang et al., 2009; Wangdi et al., 2010). This study suggests that THF1 has another function; the regulation of chlorophyll – protein complex degradation during leaf senescence, although it has not been confirmed that thf1 shows a stay-green phenotype. The G protein α-subunit is reportedly involved in THF1 function in both sugar signaling and acclimation to high light (Huang et al., 2006; Zhang et al., 2009). It will be interesting to examine the involvement of G protein signaling in degradation of chlorophyll/chlorophyll – protein complexes during leaf senescence in rice and Arabidopsis.

Experimental Procedures

Plant materials

nyc4-1 was obtained from a rice (Oryza sativa L. cv. Nipponbare) M2 population derived from seeds irradiated with carbon ion beams (200 MeV). For dark-induced senescence, detached leaves were incubated in water at 28ºC.

Analysis of photosynthetic pigments, membrane ion leakage, and photochemical efficiency

For pigment extraction, plant tissue was ground in a mortar with liquid nitrogen and extracted using 80% acetone. Chl a and Chl b levels were determined as described by Porra et al. (1989). The HPLC apparatus was equipped with a Symmetry C8 column (Waters, and a photodiode-array detector (SPD-M10A; Shimadzu, as described by Zapata et al. (2000) and Tanaka et al. (2003). The Fv/Fm value was measured using a JUNIOR-PAM fluorometer (Walz, according to the manufacturer's instructions. To measure membrane ion leakage, leaves were floated on 500 μl deionized water and incubated in the dark at 28ºC. Conductivity was measured using a Twin Cond B-173 conductivity meter (Horiba, Relative membrane ion leakage (%) was defined as follows: 100 × (electrolyte leakage before boiling/electrolyte leakage after boiling).

Quantitative RT-PCR

Total RNA was isolated from leaves of wild-type and nyc4-1 plants using a total RNA extraction kit (RBC Bioscience, First-strand cDNA was synthesized from 1 μg total RNA using ReverTra ACE-α (TOYOBO, The transcript level was determined by quantitative RT-PCR using a KAPA SYBR FAST qPCR kit (KAPA Biosystems, and a Rotor-Gene Q real-time PCR cycler (Qiagen, The primers used for amplification are listed in Table S2.

Protein analysis

We extracted proteins from leaf samples (100 mg fresh weight) using 400 μl of 2 × SDS buffer (0.125 m Tris, pH 6.8, 4% SDS, 4% mercaptoethanol, 1% bromophenol blue (BPB), 20% glycerol). The extracted proteins were diluted tenfold using 1 × SDS buffer (62.5 mm Tris, pH 6.8, 2% SDS, 2% mercaptoethanol, 0.5% BPB, 10% glycerol) and subjected to SDS–PAGE with or without boiling. Proteins were detected by Western blot analysis. Antibodies against Lhca1, Lhca2, Lhcb1, Lhcb4, CP47 and D2 were purchased from Agrisera (, while the anti-PsaF antibody was provided by Y. Takahashi (Graduate School of Natural Science and Technology, Okayama University, Japan). The antibody against D1 was described previously by Kato et al. (2012), and those against CP1 and CP43 were described previously by Tanaka et al. (1991). Detection of each protein was performed using an ECL Prime Western blotting detection system (GE Healthcare, and an ODYSSEY Fc imaging system (LI-COR, The Rubisco large subunit was visualized by Coomassie Brilliant Blue G-250 staining of the SDS–PAGE gel. Blue-native PAGE analysis was performed as described by Zhang et al. (2012).

Map-based cloning

Nine F2 plants from a cross between nyc4-1 and the indica rice cv. Kasalath that exhibited the stay-green phenotype during dark incubation were used for coarse mapping. Detached leaves were incubated at 27°C in the dark for 8 days, and green was judged by eye. The translocation breakpoints in nyc4-1 were identified by inverse PCR. After double digestion with EcoRI and StuI, and self-circularization using Ligation High (TOYOBO), the circularized DNA was initially amplified by PCR using primers NYC4 inverse F1 (5′-CGAAGACCCTGAGCAATACA-3′) and NYC4 inverse R1 (5′-AAAGTTGACAGGAAAGGGGG-3′), and then with the nested PCR primers NYC4 inverse F2 (5′-TTCACAGTGATGTTAGCATCTT-3′) and NYC4 inverse R2 (5′-GCCACAATTCAACAGAAAAGC-3′). The PCR products were separated in agarose gels, purified using a QIAquick gel extraction kit (Qiagen), and sequenced directly.

The 10 kb SmaI fragment of PAC clone P0567H04 (accession number AP005195) obtained from the National Institute of Agrobiological Sciences (NIAS) DNA bank, which contains the entire Os07g0558500 gene, was cloned into the SmaI site of pZH2B, a binary vector derived from pPZP202 (Hajdukiewicz et al., 1994). nyc4-1 calli were transformed with this construct by Agrobacterium-mediated transformation as described by Fukuoka et al. (2000). The primers used for genomic PCR were 7gF (5′-TAAGGGGTTGGCTTGTGTCT-3′), 7gR (5′-TGAACTATTTGGCTGGGCTC-3′), 3gR (5′-CAGCTTTCATGTATGTCGCC-3′), HmF1 (5′- CGGTCAATACACTACATGGC-3′) and HmR1 (5′-CCGTCAACCAAGCTCTGATA-3′).

Microarray analysis

Detached leaves from wild-type and nyc4-1 were incubated in distilled water at 27°C under dark conditions, and collected at 0, 3, 5 and 8 days after treatment. Total RNA was isolated from each sample using an RNeasy plant mini kit (Qiagen). The extracted RNA was quantified using an ND-1000 UV-VIS spectrophotometer (NanoDrop,, and checked for quality using an Agilent 2100 bioanalyzer (Agilent Technologies, Labeling was performed using a Quick Amp Labeling Kit, Two-Color (Agilent Technologies) in the presence of Cy3-labeled CTP for wild-type and Cy5-labeled CTP for nyc4-1, according to the manufacturer's instructions. For microarray hybridization, a mixture of 825 ng Cy3-labeled cRNA and 825 ng Cy5-labeled cRNA was fragmented and hybridized to the rice 4 × 44K microarray RAP-DB (Agilent Technologies) at 65°C for 17 h. Hybridization and washing of the hybridized slides were performed according to the manufacturer's instructions. The slides were scanned on an Agilent G2505B DNA microarray scanner, and background correction of the raw signals was performed using Agilent Feature Extraction software (version

Construction and visualization of the GFP fusion protein

The plastid-localized RFP construct was an RFP derivative of OsPRS9TP–GFP, containing the transit peptide from rice ribosomal protein S9 (Arimura et al., 1999). The DNA fragment for the full-length coding region of NYC4 was amplified by PCR using PrimeSTAR GXL polymerase (TaKaRa, and primers NYC4-GFP-F1 (5′-TTACGTCGACTCTAGAATGGCGGCCATATCTTCGCT-3′) and NYC4-GFP-R1 (5′-TGCTCACCATGGATCCATGCCTCATGGAATTGAGACT-3′). Amplified DNA was cloned into the XbaI–BamHI site of pJ4-GFP (Igarashi et al., 2001) using an In-fusion HD cloning kit (TaKaRa). Particle bombardment was performed using a PDS-1000/He particle gun (Bio-Rad, We introduced 1.6 μg of plasmid precipitated onto 1.0 μm gold beads into onion (Allium cepa) epidermal cells. Protein expression was observed 24 h after bombardment.

Detection of ROS

Detection of ROS was performed as described by Kato et al. (2009) with slight modifications. In situ detection of hydrogen peroxide was performed by diaminobenzidine tetrahydrochloride staining. Leaves were vacuum-infiltrated with 0.1% diaminobenzidine tetrahydrochloride in 60 mm Tris (pH 7.5) for 1 min and incubated at room temperature under light. Chlorophyll was removed by boiling in 90% ethanol solution. In situ detection of superoxide was performed by nitroblue tetrazolium staining. Leaves were vacuum-infiltrated with 10 mm NaN3 in 10 mm potassium phosphate buffer (pH 7.8) for 1 min, and then incubated in 0.1% nitroblue tetrazolium solution at room temperature under light. Chlorophyll was removed by boiling in a solution composed of acetic acid, glycerol and ethanol (1:1:3).

Accession numbers

The Rice Annotation Project Database accession numbers for the sequences referred to in this paper are shown in parentheses: NYC4 (Os07g0558500), NYC3 (Os06g0354700), SGR (Os09g0532000), OsPAO (Os03g0146400), RCCR1 (Os03g0146400), NYC1 (Os01g0227100), NOL (Os03g0654600), Osh36 (Os05g0475400) and Actin2 (Os03g0654600).


We thank Junko Kishimoto and Yumi Nagashima for their technical assistance, Yuichiro Takahashi (Graduate School of Natural Science and Technology, Okayama University, Japan) for providing the anti-PsaF antibody, Hiroyasu Yamaguchi (National Agriculture and Food Research Organization), Yoshihiro Hase, Naoya Shikazono and Atsushi Tanaka (Japan Atomic Energy Agency) for irradiating rice seeds with carbon ions, Yohsuke Takahashi (Hiroshima University) for help with particle bombardment, and Masaharu Kuroda (National Agriculture and Food Research Organization) for providing the pZH2B binary vector. This work was supported by Core Research for Evolutional Science and Technology to M.K., Grant-in-Aid for Scientific Research (B) (21380007) to M.K., and Joint Research Program implemented at the Institute of Plant Science and Resources, Okayama University, to M.K.