IC4R007-Phenomics-2015-26104586

Project Title

• Field phenomics for response of a rice diversity panel to ten environments in Senegal and Madagascar. 2. Chilling-induced spikelet sterility

The Background of This Project

• Rice spikelet sterility caused by chilling during microspore stage of panicle development is a major cause of the rareness of the indica sub-species in cool environments. A diversity panel of 200 indica accessions including traditional and improved accessions forming 4 genetic sub-groups (I1 to I4), along with 22 accessions representing other genetic groups of Oryza sativa L., was field phenotyped for spikelet sterility under flooded management. Environments were six seasonal climatic situations (sowing dates) in Senegal, ranging from cool to hot; and at two altitudes (857 and 1497 m asl) and two years in Madagascar.

Plant Culture & Treatment

• The panel was a sub-sample of the ORYTAGE species-wide (O.sativa L.) diversity panel of Cirad (http://ricephenonetwork.irri.org/ diversity-panels/orytage-diversity-panels). It was composed of 200 indica accessions, augmented for comparative purposes with 22 accessions representing other genetic groups (3 aus, 3 temperate japonica, 14 tropical japonica, 2 aromatic). The indica population covered improved and traditional varieties from all tropical regions but had large sub-populations from Madagascar (36) and W-Africa(48, thereof 31 improved lines from AfricaRice bred in Senegal) to capture adaptations to the climatic constraints at the experimental sites. Twenty-one improved varieties and lines were from IRRI(Philippines). A complete list of accessions’ geographic origin and seed sources is presented in Table S1. The 203 accessions of the indica sub-species (200 indica and 3 aus) were genotyped with 825 single nucleotide polymorphism (SNP) markers well distributed in the genome and submitted to a genetic structure analysis using the software Structure v 2. 3. 1 (Pritchard et al., 2000), following a methodology similar to that described in detail in Courtois et al.(2013) for another rice panel. Accessions that had more than 66% of their genome coming from a given sub-group were assigned to this sub-group. The assignment of the accessions to the four sub-groups numbered I1 to I4 that were identified in the panel is indicated in Table S1. The other accessions, which were intermediate between sub-groups, were considered as admixed (Im).

Research Findings

'Fig. 1. Seasonal dynamics of spikelet sterility and duration from sowing to flowering for four check varieties in Senegal. Means of six replications and SEM are presented.'

• Sowing dates were in February (07/02/2009, Date (1)March (07/03/2009, Date (2), April (07/04/2009, Date (3), July(17/07/2009, Date (4), September (17/09/2009, Date (5) and October (19/10/2009, Date (6). Dates 1–3 fell into the hot-dry season, Date 4 corresponded to the recommended main (wet) season,Date 5 was a late wet season sowing date frequently associated with cool nights during reproductive development, and Date6 was a date generally associated with chilling stress causing spikelet sterility, associated with very low air humidity (Fig. 1)

• The cold tolerant check Chomrong was not tested in Madagascar.For the remaining three checks, duration to flowering was almost constantly 23 days longer at high altitude (1497 m asl) than at mid altitude (857 m asl) (Fig. 2).

'Fig. 2. Incidence of spikelet sterility (A) and duration from sowing to flowering (B)for three check varieties observed at two altitudes in Madagascar. Error bars indicate SEM of means observed in two years.'

• Sterility levels observed at given sowing dates or locations among check varieties are difficult to compare because different phenology causes de-synchronization of the cold sensitive booting stage (coinciding with microspore stage). The researchers thus regressed the observed sterility against the estimated mean minimal water temperature during the presumed sensitive phase for each observation (Fig. 3), excluding data potentially affected by heat induced sterility.

'Fig. 3. Response of fraction spikelet sterility of three check varieties to estimated minimum water temperature (Tw) at booting to heading stage; in Madagascar (two altitudes confounded, filled symbols) and Senegal (6 sowing dates confounded, open symbols). Cases of temperatures >23 ◦C were removed to exclude incidence of heat induced sterility. Lines indicate linear regression and 95%-confidence interval.'

• To illustrate the diversity of observed responses, 25 selected accessions from different genetic groups and having contrasting thermal response of spikelet sterility are presented in Fig. 4. The same type of linear regression analysis was performed as for the checks in Fig. 3, but data from Senegal and Madagascar were combined in a single regression analysis (but represented with different symbols in Fig. 4)

'Fig. 4. Examples of contrasting genotypic responses of fraction spikelet sterility to estimated minimum water temperature (Tw) between booting and heading stages. (A1–E1) Highly sensitive indica accessions. (A2–E2) Tolerant indica types. (A3–E3) Improved rice varieties with varying tolerance. (A4–B4) Tolerant temperate japonica types. (C4–A5)Tropical japonica types. (B5–C5) Aromatic types. D5: Aus type. E5: all observations confounded. Linear regression lines are provided with 95% confidence interval. Filled symbols, Madagascar (two altitudes and years confounded); open symbols, Senegal (6 sowing dates confounded but cases of temperature >23 ◦C removed to exclude potential incidence of heat sterility).'

Labs working on this Project

• Cirad, Umr AGAP (Department BIOS) and Upr AIDA (Department ES), F-34398 Montpellier, France
• IRRI, CESD Division, DAPO Box 7777, Metro Manila, Philippines
• Université d’Antananarivo, Département de Biologie et Ecologie Végétales, BP 906, Antananarivo 101, Madagascar
• SRR FOFIFA, BP 230, Antsirabe 110, Madagascar
• Africa Rice Center, Sahel Station, P.B. 96, St. Louis, Senegal

Corresponding Author

• M. Dingkuhn:m.dingkuhn@irri.org