Role of rice PPS in late vegetative and reproductive growth

Abstract

The rice peter pan syndrome-1 (pps-1) mutant shows a prolonged juvenile phase and early flowering. Although the early vegetative phase and flowering time of pps-1 have been closely examined, the phenotypes in the late vegetative and reproductive phases are not yet well understood. In the ninth leaf blade of pps-1, the relative length of the midrib was comparable to the sixth leaf blade of wild-type. Moreover, pps-1 had a small inflorescence meristem and small panicles. These phenotypes indicate that in pps-1 the juvenile phase coexists with the late vegetative phase, resulting in small panicles. Gibberellin is known to promote the juvenile-adult phase transition. d18-k is dwarf and has a prolonged juvenile phase. Double mutant (d18-k pps-1) showed the same phenotype as the pps-1, indicating that PPS is upstream of GA biosynthetic genes.

There are three main developmental phases in the life cycle of higher plants: embryogenesis, the vegetative phase, and the reproductive phase. The vegetative phase is further divided into two sub-phases, juvenile and adult.^1,2 Since plants can initiate reproductive growth under appropriate environmental conditions only during the adult phase, the juvenile-adult transition is essential for plant development. Although this has been studied in maize, Arabidopsis and rice,^3-5 the mechanism remains poorly understood. Previously,⁶ we reported that the unique rice mutant peter pan syndrome-1 (pps-1) had a long juvenile phase and early flowering. However, studies of pps-1 have mainly been confined to the young seedling stage and heading date. In this report, we examined the late vegetative phase and panicle production in pps-1.

The plant hormone gibberellin (GA) is known to promote the juvenile to adult transition,^7,8 and PPS positively regulates GA biosynthesis.⁶ To examine whether PPS functions upstream of GA metabolic genes, we constructed a d18-k pps-1 double mutant.

Phenotypes of pps-1 in the Late Vegetative and Reproductive Stages

Even in the sixth leaf blade of pps-1, the midrib length was comparable to that of the third leaf blade in wild type. This suggests that adult higher leaves have more juvenile characters in pps-1 than in wild-type plants. To examine whether pps-1 enters a normal adult phase, we observed pps-1 midrib formation in the late vegetative phase. In the ninth leaf blade of wild-type plants, the midrib was more than 80% developed, whereas in pps-1 the midrib was only ~60% developed (Fig. 1A and B), comparable to the fourth leaf in wild-type plants (Fig. 2B). Thus, pps-1 continued to show incomplete midrib formation even in the late vegetative phase.

Figure 1. Phenotypes of pps-1 in late vegetative and reproductive phases. (A) Cross sections of wild type (left) and pps-1 (right) 9th leaf blade cut at 55% from the base. Bar = 100μm. (B) Relative length of midrib in 9th leaf blages of wild type and pps-1. (C) Mature pps-1 plant. Bar = 5cm. (D) Panicles of wild type (left) and pps-1 (right). Bar = 5cm. (E) Cross sections of spikelets in wild type (left) and pps-1 (right). Bars = 100μm. (F) Numbers of primary branches, secondary branches and spiklets per panicle in wild type and pps-1. (G) Median longitudinal sections of young panicle at primary rachis branch differentiation stage in wild type (left) and pps-1 (right). Enlarged views of inflorescence meristems are shown in the lower panels. Bars = 100μm.

Figure 2. Phenotypes of d18-k and d18-k pps-1 double mutant. (A) 12-d-old seedlings from left to right, wild-type, d18-k, pps-1 and d18-k pps-1. Bar = 5cm. (B) Relative length of midrib in 2nd, 3rd and 4th leaf blades of wild-type, d18-k, pps-1, and d18-k pps-1. (C) Cross sections of wild-type, d18-k, pps-1 and d18k pps-1 4th leaf blades cut at 10% and 50% from the base. Bars = 100μm.

In the reproductive phase, the pps-1 plants produced dark green and extremely small panicles (Fig. 1C and D), whereas the flower was similar in size to that of wild type (Fig. 1E). We therefore counted the number of primary and secondary branches and spikelets per panicle. In pps-1, these were all obviously smaller than in wild type (Fig. 1F). It is thought that panicle size is positively correlated with inflorescence meristem size, as found in the rice aberrant panicle organization 1 (apo1) mutant, which has small panicles and a small inflorescence meristem,⁹ and that enhanced APO1 expression leads to a larger inflorescence meristem and increased numbers of branches and spikelets.¹⁰ To examine whether the small panicles of pps-1 are caused by a small inflorescence meristem, we measured the size of the inflorescence meristem and found that the pps-1 inflorescence meristem was significantly smaller than that of wild type (Fig. 1G). The shoot apical meristem (SAM) of wild-type rice gradually enlarges toward reproductive conversion; however, the pps-1 SAM was smaller than the wild-type SAM in the vegetative phase, and pps-1 flowered earlier than wild type.⁶ Accordingly, as the SAM of pps-1 remained small during reproductive development, small panicles were produced.

The incomplete midrib formation in the late vegetative phase and small inflorescence meristem in pps-1 suggest that a juvenile program coexists with the adult program in this mutant during the late vegetative phase.

Analysis of the d18k pps-1 Double Mutant

Previously, we showed that the amount of active GA₁ was significantly reduced in pps-1.⁶ This reduction in GA₁ content in pps-1 was caused by reduced expression of GA3ox2 and GA20ox2, which encode enzymes involved in GA biosynthesis, and enhanced expression of GA2ox4, which encodes a GA-catabolizing enzyme.⁶ To genetically confirm the relation between PPS and GA metabolic genes, we constructed a d18-k pps-1 double mutant. d18-k is a loss-of-function allele of D18, which encodes GA3ox2,¹¹ and the d18-k single mutant has a dwarf phenotype (Fig. 2A). To estimate the juvenility of d18-k, we examined the development of the midrib. Midribs were rarely observed in the second leaves of both wild type and d18-k. In wild-type third and fourth leaf blades, the midrib was approximately 40 and 60% of the total leaf blade length, respectively. In d18-k, midrib formation was suppressed to about 25 and 50% of the total length in the third and fourth leaf blades, respectively (Fig. 2B and C). These results suggest that the juvenile-adult transition is retarded in d18-k, although to a lesser extent than that in pps-1, where midrib formation was 0 and ~10% in the third and fourth leaf blades, respectively (Fig. 2C).

We then crossed a d18-k homozygous plant with a pps-1 heterozygous plant to construct a double mutant. The double mutant showed the same seedling features as pps-1 (Fig. 2A). In pps-1, the second and third leaf blades had almost no midrib, and midrib formation in the fourth leaf blade was less than 10% of the total length (Fig. 2B and C). In the double mutant, the second and third leaf blades had almost no midrib, as found in pps-1. The relative midrib length in the fourth leaf blade of the double mutant was comparable to that in the fourth leaf blade of pps-1 (Fig. 2B and C). Therefore, the mutation of d18-k in a pps-1 background does not enhance incomplete midrib development. This confirms that PPS functions upstream of GA3ox2/D18.

Incomplete midrib formation in the late vegetative phase of pps-1 suggests that the juvenile program is operating in the pps-1 ‘adult’ phase. In addition, the phenotype of the d18-k pps-1 double mutant indicates that PPS functions upstream of GA biosynthetic genes.

Disclosure of Potential Conflicts of Interest

No potential conflicts of interest were disclosed.