**Abstract**

Rice is an important staple food crop across the world. It is mainly cultivated under irrigated lowland and also rain-fed upland conditions where drought stress is often noticed. Global climate change predicts an intensification of drought stress in future due to uneven rainfall which was witnessed for the last few years. Confronting drought stress can deliver fruitful crop returns in rice and scope for research extents. Drought stress affects the overall plant growth and yield. A prominent improvement has been made during last two decades in our understanding of the mechanisms involved in adaptation and tolerance to drought stress in rice. In order to achieve the marked crop returns from rainfed areas, there is a requisite of drought tolerant rice varieties, and genetic improvement for drought tolerance should be a prime area of concern in the future. A huge rice germplasm is available and good number of the germplasm possess drought tolerance and these genomic regions have been exploited in developing some drought tolerant rice varieties. The application of available genotyping methodologies, the identification of traits of interest, and key genetic regions associated with the drought tolerance have opened new prospects to successfully develop new drought tolerant varieties. This chapter deals with the importance of drought tolerance in rice crop followed by the evolution of molecular markers and breeding techniques in identifying drought tolerant QTL's/ genes and their utilization in the improvement of drought tolerant rice varieties.

**Keywords:** Rice, Drought tolerance, Markers, QTLs, Varieties

## **1. Introduction**

Rice is a staple cereal consumed by more than half of the world's population. It is cultivated in wide agro-ecological conditions including rain-fed conditions where drought stress is often evident due to erratic rainfall. Rice crop consumes around 3000 L water to produce 1 Kg rice. Drought stress is one of the major factors that leads to decreased rice production [1] and it can expand to above 50% of the global arable land by 2050 [2] due to loss of ground water, global climate change leading to decreased water sheds. Drought stress was noted in approximately 42 Mha of riceproducing area [3]. It is of two types, terminal and intermittent [4].

Drought stress is usually a dry condition where the water availability is less than a threshold level which causes damage to plants [5].

Lack of water for a long time leading to the death of plants is terminal drought while lack of water for a short time that leads to improper growth is intermittent drought [6].

Drought stress tolerance varies among the plant species and is defined as the ability of a plant to grow, develop and produce significant yield as well as economic benefit [7]. It is also defined as the ability of plant to survive at minimum water level in the fresh tissue (23%) [6].

Drought stress leads to morphological, physiological and biochemical changes in plants which ultimately lead to decreased yields. In response, plants synthesize reactive oxygen species (ROS), proteins and osmolytes to maintain turgor pressure. This osmotic adaptation provides dehydration tolerance to tissues [8–10]. But, this did not show yield benefit in rice [11].

The first landmark achievement in drought rice tolerance study was the identification of a small region on chromosome 8 homologous to chromosome 7 of wheat [12]. A marker based selection was proposed and this region was worked out to identify QTL's for vegetative leaf rolling and root traits- thickness and root to shoot ratio [13].

Historically, upland-adapted germplasm was of Japonica type while lowlandadapted germplasm was of Indica type [14]. In general, Japonica genotypes were dehydration avoidant while Indica types were dehydration tolerant. Geographically separated evolution and sterility problems have limited hybridization between the two types [15]. If breeding for both osmotic adjustment and rooting capacity is considered desirable, then the linkage between high osmotic adjustment and poor root traits needs to be broken.

RG1 QTL was identified while working with 52 recombinant inbred (RI) lines (F 7), a randomly sampled subset of a population originally developed to study the genetics of resistance to rice blast (*Pyricularia oryzae*) [16].

## **2. First generation markers**

#### **2.1 Restriction fragment length polymorphism (RFLP)**

Rice RFLP maps developed at Cornell University [17, 18] and Japan [19, 20] is the basis of gene mapping research. In the 1988 wet season at IRRI, Co39 (maternal), a lowland, Indica cultivar developed in India, and Moroberekan, a traditional upland, Japonica cultivar originally developed in Guinea. Moroberekan is considered to be resistant to drought while Co39 is drought susceptible. About 50 F1 seeds were obtained from the cross and only 15 F1 seeds were randomly chosen and grown in a greenhouse to obtain an F2 population. About 300 F2 seeds were randomly selected and planted in the Rapid Generation Advance (RGA) [21] greenhouse from F2 to F6 using single seed descent (SSD). All panicles were bagged at each generation and F7 seeds were used for genotype analysis [16].

DNA was extracted from the leaves of the two parents and digested with restriction enzymes DraI, EcoRI, EcoRV, HindIII and Scal. The digested DNAs were electrophoresed on 0.9% agarose gels and transferred to Hybond N+ membranes (Amersham Corp., Chicago) according to the manufacturer's instructions. 280 DNA clones distributed throughout the 12 chromosomes of rice (184 rice genomic clones, coded RG; 62 rice cDNAs, coded Rz; and 28 oat cDNAs, coded CDO) were linearized and labeled with 32PdCTP by the random hexamer method [22]. Hybridized filters were washed once in 1.5 X SSPE and once in 0.5 X SSPE at 65° for 15–20 min. Filters were exposed to X-ray film −80° at with one intensifying screen for 1–4 days. 127 informative probes were used for segregation analysis of the RI lines using the procedures outlined above. Mapmaker [23] and Map Manager [24] were used to establish the RFLP map.

Since the *indica* × *japonica* crosses suffer from sterility, distorted segregation, etc., [16], breeders chose to work on the accessions of *indica or japonica* sub-species, RFLP was considered laborious over RAPD which gained popularity among researchers [25].
