Changes in Grain Yield and Yield Attributes Due to Cultivar Development in Indica Inbred Rice in China

Abstract

Inbred rice has been grown more and more widely, while the planting area of hybrid rice has decreased by approximately 25% in China since 1995. This study aimed to assess the changes in grain yield and yield attributes due to cultivar development in indica (Oryza sativa ssp. indica) inbred rice in China. Field experiments were conducted in 2019 and 2020 to determine the performance of grain yield and yield attributes of an indica super inbred rice cultivar Jinnongsimiao (JNSM) released in 2010 by comparing it with an indica high-yielding inbred rice cultivar Guichao 2 (GC2) released in 1978 and an indica super hybrid rice cultivar Y-liangyou 900 (YLY900) released in 2016. Results showed that JNSM produced 18% higher grain yield than GC2 but 6% lower grain yield than YLY900. Compared with GC2, JNSM had higher spikelets per panicle, spikelet-filling percentage, and harvest index by 67%, 4%, and 11%, respectively. Compared with YLY900, JNSM had 14% lower grain weight and 19% lower biomass production during the pre-heading period. The difference in biomass production during the pre-heading period between JNSM and YLY900 was explained more by crop growth rate than growth duration. This study suggests that (1) the recently released indica super inbred rice cultivar JNSM outyields the old indica high-yielding inbred rice cultivar GC2 as a result of increasing panicle size, spikelet-filling percentage, and harvest index, and (2) further improvement in grain yield in indica inbred rice can be achieved by improving biomass production through promoting pre-heading crop growth.

1. Introduction

Rice is the main source of calorie intake for almost 70% of the population in China [1]. Increasing rice yield is significant for ensuring national food security in China with an arable land per capita that is only one-third of the world average [2]. Rice yield experienced two quantum leaps in China as a result of the development of semi-dwarf cultivars in the 1950s and hybrid cultivars in the 1970s [3]. To further increase rice yield, a super rice breeding program was established in China in 1996 to develop new high-yielding rice cultivars based on the ideotype concept [4]. Up to 2021, 100 hybrid and 35 inbred rice cultivars were approved as super rice cultivars by the Ministry of Agriculture and Rural Affairs of China (MARA) (https://www.ricedata.cn/variety/superice.htm, assessed on 8 August 2022). It has been well-documented that the development of super hybrid cultivars has increased rice yield potential by more than 10% [5,6].

Although the development of super hybrid rice cultivars provides an opportunity to increase rice yield in China, the planting area of super hybrid rice cultivars is less than 8% of the national total rice planting area [7]. What is worse, the planting area of hybrid rice has decreased by approximately 25% in China since 1995 [2]. Fortunately, however, the decrease in hybrid rice planting area did not cause a decrease in rice yield in China, but on the contrary the rice yield in China increased by nearly 15% during the period when the hybrid rice planting area decreased [2]. This suggests that grain yield has actually been improved in China not only for hybrid rice but also for inbred rice. However, there is limited information available on the yield improvement in inbred rice in China.

In this study, two-year field experiments were performed to determine the performance of grain yield and yield attributes of a recently released indica (Oryza sativa ssp. indica) super inbred rice cultivar by comparing it with an old indica high-yielding inbred rice cultivar and a recently released indica super hybrid rice cultivar. The objective of this study was to assess the changes in grain yield and yield attributes due to cultivar development in indica inbred rice in China.

2. Materials and Methods

2.1. Experimental Details

Field experiments were conducted at the research farm of Hengyang Academy of Agricultural Sciences (26°53′ N, 112°29′ E), Hunan Province, China in 2019 and 2020. The experimental site has a subtropical monsoon humid climate, and no low- or high-temperature stress occurred during the rice-growing season in 2019 and 2020 (Figure 1). The soil of the experimental field had the following chemical properties at the upper 20 layer before transplanting in 2019: pH 6.24, 18.4 g kg⁻¹ organic carbon, 2.66 g kg⁻¹ total N, 0.73 g kg⁻¹ total P, and 6.55 g kg⁻¹ total K.

Three indica rice cultivars, Guichao 2 (GC2), Jinnongsimiao (JNSM), and Y-liangyou 900 (YLY900) were arranged in a completely randomized block design with four replications and a plot size of 30 m². The three cultivars were chosen because they were among the best indica cultivars in terms of grain yield when they were released. GC2 is an inbred cultivar developed by the Guangdong Academy of Agricultural Sciences (GAAS), with Guiyangai 49 as the female parent and Chaoyangzao 18 as the male parent. This cultivar was released in 1978 and was widely grown by farmers in China during the 1980s. JNSM is an inbred cultivar developed by the GAAS, with Jinhuaruanzhan as the female parent and Guinongzhan as the male parent. This cultivar was released in 2010 and was approved as a super rice cultivar by the MARA in 2012. YLY900 is a two-line hybrid cultivar developed by the Biocentury Seed Industry Co., Ltd., Shenzhen, China with the R58S as the female parent and R900 as the male parent. This cultivar was released in 2016 and was approved as a super rice cultivar by the MARA in 2017.

Pre-germinated rice seeds were sown in a seedbed on 25 April, and 25-day-old seedlings were manually transplanted at a hill spacing of 20 cm × 20 cm with two seedling per hill. Rice plants received 180 kg N hm^ࢤ2 (50% at 1 day before transplanting, 20% at 7 days after transplanting, and 30% at panicle initiation), 90 kg P₂O₅ hm^ࢤ2 (100% at 1 day before transplanting), and 180 kg K₂O hm^ࢤ2 (50% at 1 day before transplanting and 50% at panicle initiation). All plots were flooded with a water depth of 5–10 cm from transplanting to 7 days before maturity and then drained to prepare for harvesting. Weeds, insects, and pathogens were controlled according to locally recommended plant protection measures and chemicals.

2.2. Sampling and Measurements

Ten hills (0.4 m²) of rice plants were sampled from each plot at heading (HD) and maturity (MA). The rice plants sampled at HD were separated into leaves, stems, and panicles. Each plant organ was oven-dried at 70 °C to a constant weight to determine biomass production from sowing (SO) to HD. The rice plants sampled at MA were hand-threshed after counting the number of panicles. Filled and unfilled spikelets were separated by submerging them in tap water. The number of filled spikelets was counted using a digital automatic seed counter (SLY-C, Zhejiang Top Cloud-Agri Technology Co., Ltd., Hangzhou, China). The number of unfilled spikelets was manually counted. Straw and filled and unfilled spikelets were oven-dried at 70 °C to a constant weight to determine biomass production from SO to MA.

The data of the number of panicles and filled and unfilled spikelets and filled grain dry weight were used to calculate yield components including panicles per m², spikelets per panicle, spikelet-filling percentage, and grain weight. In addition, biomass production from HD to MA was calculated as the difference between the biomass production from SO to MA and the biomass production from SO to HD. Biomass remobilization (i.e., the remobilization of stored reserves into growing grains) was calculated as the difference between the filled grain dry weight and the biomass production from HD to MA. The harvest index was calculated as the ratio of the filled grain weight to the biomass production from SO to MA. Crop growth rate during the period from SO to HD, HD to MA, and SO to MA was calculated by dividing the biomass production from SO to HD, HD to MA, and SO to MA by the growth duration for the corresponding period. Temperature-use efficiency during the period from SO to HD, HD to MA, and SO to MA was calculated by dividing the biomass production from SO to HD, HD to MA, and SO to MA by the cumulative active temperature (>12 °C, the biologically low temperature for indica rice) for the corresponding period.

Grain yield was determined by harvesting rice plants from a 5 m² area in each plot and adjusting to a standard moisture content of 14%. Daily grain yield was calculated by dividing the grain yield by the growth duration from SO to MA.

2.3. Statistical Analysis

Data were analyzed by analysis of variance (ANOVA) using Statistix 8.0 (Analytical Software, Tallahassee, FL, USA). The statistical model of ANOVA included replication, the main effects of variety and year, and the interactive effect of variety and year. Means of varieties were compared by the least significant difference test at the 0.05 probability level.

3. Results

3.1. Growth Duration and Cumulative Active Temperature

Growth duration from SO to HD was 1–2 d and 7 d shorter in JNSM than in GC2 and YLY900, respectively (

Table 1. Growth duration and cumulative active temperature of three rice cultivars grown in 2019 and 2020.

Cultivar †	Growth Duration (d) ††			Cumulative Active Temperature (°C)
	SO–HD	HD–MA	SO–MA	SO–HD	HD–MA	SO–MA
2019
GC2	92	32	124	1145	594	1739
JNSM	90	26	116	1107	487	1594
YLY900	97	42	139	1241	736	1977
2020
GC2	89	35	124	1409	637	2046
JNSM	88	24	112	1390	443	1833
YLY900	95	40	135	1518	700	2218

^† GC2, YLY900, and JNSM are Guichao 2, Y-liangyou 900, and Jinnongsimiao, respectively. ^†† SO, HD, and MA represent sowing, heading, and maturity stages, respectively.

). JNSM had 6–11 d and 16 d shorter growth duration from HD to MA compared to GC2 and YLY900, respectively. Growth duration from SO to MA in JNSM was 8–11 d and 23 d shorter than that in GC2 and YLY900, respectively.

JNSM had 19–38 °C and 128–134 °C lower cumulative active temperature during the period from SO to HD compared to GC2 and YLY900, respectively (Table 1). Cumulative active temperature during the period from HD to MA in JNSM was 107–194 °C and 249–257 °C lower than that in GC2 and YLY900, respectively. Cumulative active temperature during the period from SO to MA was 145–213 °C and 383–385 °C lower in JNSM than in GC2 and YLY900, respectively.

3.2. Grain Yield and Yield Components

Grain yield and daily grain yield significantly differed among three cultivars (

Table 2. Grain yield and daily grain yield of three rice cultivars grown in 2019 and 2020.

Cultivar †	Grain Yield (t ha−1)			Daily Grain Yield (kg ha−1 d−1)
	2019	2020	Mean ††	2019	2020	Mean ††
GC2	7.23	7.67	7.45 c	58.3	61.9	60.1 c
JNSM	8.54	9.00	8.77 b	73.6	80.4	77.0 a
YLY900	9.44	9.20	9.32 a	67.9	68.1	68.0 b
Analysis of variance (F-value)
Cultivar (C)	37.27 **			32.06 **
Year (Y)	3.01 NS			8.29 *
C × Y	2.57 NS			2.93 NS

* and ** denote significance at the 0.05 and 0.01 probability level, respectively. ^NS denotes non-significance at the 0.05 probability level. ^† GC2, YLY900, and JNSM are Guichao 2, Y-liangyou 900, and Jinnongsimiao, respectively. ^†† Means not sharing the same letters are significantly different at the 0.05 probability level.

). Averaged across two years, JNSM produced 18% higher grain yield than GC2 but 6% lower grain yield than YLY900. Daily grain yield was 28% and 13% higher in JNSM than in GC2 and YLY900, respectively. Grain yield did not significantly differ between 2019 and 2020, whereas daily grain yield significantly differed between the two years. Averaged across three cultivars, daily grain yield was 5% higher in 2020 than in 2019. There was no significant interaction effect of cultivar and year on either grain yield or daily grain yield. Therefore, combined data across two years were presented in the subsequent tables.

The difference in panicles per m² was not significant among three cultivars (

Table 3. Yield components of three rice cultivars across two years (2019 and 2020).

Cultivar †	Panicles m−2	Spikelets Panicle−1	Spikelet Filling (%)	Grain Weight (mg)
GC2	222 a	134 b	85.3 b	32.8 a
JNSM	218 a	224 a	88.4 a	21.9 c
YLY900	205 a	242 a	79.4 c	25.6 b

Within a column, means not sharing the same letters are significantly different at the 0.05 probability level. ^† GC2, YLY900, and JNSM are Guichao 2, Y-liangyou 900, and Jinnongsimiao, respectively.

). JNSM had 67% higher spikelets per panicle than GC2, while the difference in spikelets per panicle was not significant between JNSM and YLY900. Spikelet-filling percentage was 4% and 11% higher in JNSM compared to GC2 and YLY900, respectively. Grain weight was lower in JNSM than in GC2 and YLY900 by 33% and 14%, respectively.

3.3. Biomass Production and Translocation

There were no significant differences in biomass production from SO to HD, from HD to MA, and from SO to MA between JNSM and GC2 (

Table 4. Biomass production and translocation of three rice cultivars across two years (2019 and 2020).

Cultivar †	Biomass Production (g m−2) ††			Biomass Remobilization (g m−2)	Harvest Index
	SO–HD	HD–MA	SO–MA
GC2	943 b	658 a	1601 b	53 c	0.445 b
JNSM	971 b	672 a	1643 b	135 b	0.493 a
YLY900	1195 a	621 a	1816 a	243 a	0.478 a

Within a column, means not sharing the same letters are significantly different at the 0.05 probability level. ^† GC2, YLY900, and JNSM are Guichao 2, Y-liangyou 900, and Jinnongsimiao, respectively. ^†† SO, HD, and MA represent sowing, heading, and maturity stages, respectively.

). JNSM had 19% lower biomass production from SO to HD than YLY900. The difference in biomass production from HD to MA was not significant between JNSM and YLY900. Biomass production from SO to MA was 10% lower in JNSM compared to YLY900.

Biomass remobilization in JNSM was 155% higher than that in GC2 but was 44% lower than that in YLY900 (Table 4). JNSM had 11% higher harvest index than GC2. The difference in harvest index was not significant between JNSM and YLY900.

3.4. Crop Growth Rate and Temperature-Use Efficiency

The difference in crop growth rate during the period from SO to HD was not significant between JNSM and GC2 (

Table 5. Crop growth rate and temperature-use efficiency of three rice cultivars across two years (2019 and 2020).

Cultivar †	Crop Growth Rate (g m−2 d−1) ††			Temperature-Use Efficiency (g °C−1)
	SO–HD	HD–MA	SO–MA	SO–HD	HD–MA	SO–MA
GC2	10.4 b	19.6 b	12.9 b	0.74 b	1.07 b	0.85 b
JNSM	10.9 b	26.9 a	14.4 a	0.78 b	1.45 a	0.96 a
YLY900	12.4 a	15.1 c	13.3 b	0.87 a	0.86 b	0.87 b

Within a column, means not sharing the same letters are significantly different at the 0.05 probability level. ^† GC2, YLY900, and JNSM are Guichao 2, Y-liangyou 900, and Jinnongsimiao, respectively. ^†† SO, HD, and MA represent sowing, heading, and maturity stages, respectively.

). JNSM had 37% and 12% higher crop growth rate than GC2 during the periods from HD to MA and from SO to MA, respectively. Crop growth rate was 12% lower in JNSM than in YLY900 during the period from SO to HD, whereas it was higher in JNSM compared to YLY900 by 78% during the period from HD to MA and by 8% during the period from SO to MA.

There was no significant difference in temperature-use efficiency during the period from SO to HD between JNSM and GC2 (Table 5). Temperature-use efficiency was 36% and 13% higher in JNSM compared to GC2 during the periods from HD to MA and from SO to MA, respectively. JNSM had 10% lower temperature-use efficiency during the period from SO to HD. Temperature-use efficiency was higher in JNSM than in GC2 during the periods from HD to MA and from SO to MA by 69% and 10%, respectively.

4. Discussion

By comparing two representative indica high-yielding inbred rice cultivars released 32 years apart, i.e., an old high-yielding inbred cultivar GC2 released in 1978 and a super inbred cultivar JNSM released in 2010, it was found that the recently released cultivar JNSM has a yield advantage of nearly 20% over the old cultivar GC2. Furthermore, it was also found that larger panicle size (more spikelets per panicle), higher spikelet-filling percentage, and higher harvest index are responsible for the higher grain yield in JNSM than in GC2. It has been well-documented that the improvement in panicle size is also an important reason for higher grain yields in super indica hybrid rice cultivars than in old indica hybrid rice cultivars [5,6]. However, spikelet-filling percentage and harvest index do not explain the yield advantage of super indica hybrid rice cultivars compared to old indica hybrid rice cultivars. This indicates that the path for genetic improvement of grain yield may be not entirely the same between indica inbred and hybrid rice.

Spikelet-filling percentage and harvest index are two positively related traits in rice [8], and the harvest index is determined by the potential sink size, the biomass production during the post-heading period, and the biomass remobilization [9]. In this study, the higher harvest index in JNSM was mainly attributable to larger sink size (more spikelets per panicle) and higher biomass remobilization than GC2, because the difference in biomass production during the post-heading period was comparable between the two cultivars.

It is generally reported that improved indica inbred rice cultivars have a yield disadvantage of 10–20% than indica hybrid rice cultivars [10,11,12]. In this study, the magnitude of yield disadvantage in the super indica inbred rice cultivar JNSM compared to the super indica hybrid rice cultivar YLY900 was only 6%, which is lower than that generally reported. This also partly supports that grain yield has been genetically improved in indica inbred rice.

Tillering capacity and panicle size are generally lower in indica inbred rice cultivars than in indica hybrid rice cultivars [13]. However, this was not responsible for the difference in grain yield between JNSM and YLY900, because there were no significant differences in panicles per m² and spikelets per panicle between these two cultivars. In this study, the lower grain yield in JNSM than in YLY900 was partially attributable to lower grain weight. As far as we know, the low grain weight in JNSM is associated with the increased demand for slender-grain rice with superior quality that has encouraged breeders to develop cultivars with low grain weight [14]. In addition, our recent study showed that decreasing grain size is also a feasible way to reduce the occurrence of chalkiness in rice [15]. This indicates that increasing grain weight may be not a practical way for further improving grain yield in indica inbred rice.

Lower biomass production during the pre-heading period was another reason for the lower grain yield in JNSM than in YLY900. The lower biomass production during pre-heading period in JNSM was attributable to both shorter growth duration and lower crop growth rate than in YLY900. The growth duration and crop growth rate explained approximately 40% and 60% of the difference in biomass production during pre-heading period between JNSM and YLY900, respectively. This suggests that more attention should be paid to promoting crop growth during the pre-heading period to further improve grain yield in indica inbred rice. This strategy is also useful for improving the lodging resistance in rice plants [16,17].

In addition, this study showed that JNSM had higher crop growth rate and temperature-use efficiency during the post-heading period than both GC2 and YLY900, indicating that JNSM is a typical late-stage vigor cultivar. It has been reported that late-stage vigor is related to the improvements in morphological traits of root systems and photosynthetic function of leaves in super hybrid rice cultivars [18,19,20]. However, there is limited information available on the mechanisms underlying the late-stage vigor in super inbred rice cultivars. This highlights the need for a fundamental understanding of the morphological and physiological characteristics regarding the late-stage vigor in JNSM.

There is a limitation that must be acknowledged in the present study. Up to 2021, the MARA has approved 35 super inbred rice cultivars and 100 super hybrid rice cultivars with high yield potential. However, only one super inbred cultivar and one super hybrid cultivar were used in this study. Further studies involving more super inbred and hybrid rice cultivars are required to generate more conclusive results.

5. Conclusions

The recently released indica super inbred rice cultivar JNSM outyields the old indica high-yielding inbred rice cultivar GC2 by nearly 20% as a result of increasing panicle size, spikelet-filling percentage, and harvest index. Further improvement in grain yield in indica inbred rice can be achieved by improving biomass production through promoting pre-heading crop growth. In addition, this study also shows that (1) the yield advantage in hybrid rice over inbred rice observed in this study is lower than that reported previously and (2) JNSM is a typical late-stage vigor cultivar with high crop growth rate and high temperature-use efficiency during the post-heading period.