Heat stress in crop plants: its nature, impacts and integrated breeding strategies to improve heat tolerance

Abstract

Increasing severity of high temperature worldwide presents an alarming threat to the humankind. As evident by massive yield losses in various food crops, the escalating adverse impacts of heat stress (HS) are putting the global food as well as nutritional security at great risk. Intrinsically, plants respond to high temperature stress by triggering a cascade of events and adapt by switching on numerous stress-responsive genes. However, the complex and poorly understood mechanism of heat tolerance (HT), limited access to the precise phenotyping techniques, and above all, the substantial G × E effects offer major bottlenecks to the progress of breeding for improving HT. Therefore, focus should be given to assess the crop diversity, and targeting the adaptive/morpho-physiological traits while making selections. Equally important is the rapid and precise introgression of the HT-related gene(s)/QTLs to the heat-susceptible cultivars to recover the genotypes with enhanced HT. Therefore, the progressive tailoring of the heat-tolerant genotypes demands a rational integration of molecular breeding, functional genomics and transgenic technologies reinforced with the next-generation phenomics facilities.

The global climate will witness an increase of 2–4°C temperature at the end of 21st century (IPCC 2007). More importantly, the predictions based on global climate model analysis suggest that the tropical and subtropical regions of the world will be the worst sufferer from the forthcoming disaster of heat stress (HS) (Battisti and Naylor 2009). Lobell and Gourdji (2012) have reported linear trends in increase in average maximum and minimum temperatures per decade (1980–2011) with average of 0.3°C T_max and 0.2°C T_min, respectively. As a consequence of increase in temperature, alterations in plant phenologies such as spring and autumn phenologies were noticed across different plant species and also within plant species (Ibáñez et al. 2010, Li et al. 2014a). In particular, the earliest flowering species in spring are the most sensitive to high temperature (Wolkovich et al. 2012). Therefore, increasing HS in various crop species is emerging as an alarming issue especially in concern to the global food security. In response to HS, plants are endowed with different mechanisms and regulatory networks, viz. regulating vital genes, managing numerous physiological and biochemical adaptations and so forth. The complex genetic structure of heat tolerance (HT) offers a great challenge to breeders (Blum 1986), which is further exacerbated by the presence of large magnitude of G × E and epistatic interactions (Cossani and Reynolds 2012). This article highlights the enlarging adversities of HS and their substantial impacts on crops that have been seen in the recent past. Further, an attempt has been made to provide an overview on underlying mechanism that includes heat signal perception crosstalk, and role of heat-shock proteins (HSPs) and transcription factors (TFs). Finally, we discuss optimizing breeding strategies that focus on efficient utilization of germplasm resources and existing genetic variability together with exploring the future scope for integrating a wide range of emerging forward and reverse genetic techniques, and the phenomics platforms.

Crop Yield Losses Due to HS: Retrospect and Future Projections

Region-wise overview on crop losses

The unprecedented rise in temperature extremities in the last decades revealed a broad range of anomalies associated with HS. Its severe impacts on various crops have been demonstrably evident in different parts of the world. In Yangtze River Valley (YRV), the pivotal rice-growing area of China repeatedly (six times) faced HS, causing severe loss of mid-season rice (Tian et al. 2009). The prolonged HS resulted in poor seed set (up to 10% only) in heat-susceptible rice genotypes and hybrids in Hubei and Sichuan provinces of China, leading to severe loss in rice production (Tian et al. 2009). Similarly, by analysing data from 1979 to 2000, You et al. (2009) noted up to 10% reduction in wheat yield, which could be credited to 1°C temperature increase during wheat-growing season in China. Taken recent trends into consideration, it is envisaged that India will lose 0.45 tons/ha, while China will encounter a 4–7% reduction in rainfed wheat yields, which would be attributable to a rise of 0.5°C in average temperature by 2050 (Easterling et al. 2007). Manigbas and Sebastian (2007) have reported that the high temperature (38.0–39.9°C) during mid-April to early May in major rice-growing areas in Philippines (particularly central and northern parts of Luzon) renders these regions more vulnerable to HS.

Several research groups (Chatrath et al. 2007, Joshi et al. 2007a, Singh et al. 2007a) have highlighted the increasing vulnerability of the major wheat-growing areas to HS. The likely vulnerable regions include South Asia that covers Eastern Gangetic Plains (EGP), central and peninsular India and Bangladesh. By 2050, Indo Gangetic Plain (IGP) of mega environment-1 (ME1), which accounts for almost 15% of the global wheat production, will be redefined as heat-stressed, irrigated and short-season production mega environment (Ortiz et al. 2008). Moreover, Gupta et al. (2012) pointed out that the five MEs (ME4, ME5, ME6, ME9 and ME12) under wheat cultivation are inflicted due to HS and shortage of water. Similarly, climatic model analysis delineates central and eastern Asia, central North America and northern part of Indian subcontinent as the major HS-prone regions for growing wheat, maize, rice and soybean (Teixeira et al. 2013). Further, it is reported by the Working Group II of IPCC that India would likely to suffer from 10 to 40% loss in crop production due to HS by 2080–2100 (see Aggarwal 2008).

As a result of abnormal rise in July temperatures (6°C above the long-term means) coupled with the occurrence of scanty rainfall, Europe faced the hottest summer in 2003 and witnessed record losses in crop yield (Stott et al. 2004, Ciais et al. 2005). Likewise, adverse effects of increasing temperature on crop yield are likely to reflect in Pannonian zone encompassing Hungary, Serbia, Bulgaria and Romania (Olesen et al. 2011). Recently, noticeable reduction in maize yield was recorded in France due to maximum temperature rising beyond 32°C each day (Hawkins et al. 2013). Similarly, dry and hot summer causing shrinkage of maturity period in European wheat-growing regions are also experiencing marked deterioration in yield (Semenov et al. 2014). Furthermore, given the ever-expanding severity of HS, whole Europe will likely to face substantial crop losses not primarily due to drought but because of increasing incidence of HS (Semenov and Shewry 2011).

In Africa, analysis of historical data covering more than 20 000 maize trials suggested 1% and 1.7% yield losses under rainfed and drought conditions, respectively, due to rise of each one degree temperature beyond 30°C (Lobell et al. 2011a). Similarly, reclassification of new maize MEs is anticipated, and more importantly, maize MEs within sub-Saharan Africa region are envisioned to encounter severe threat of temperature rising by 2.1°C (Cairns et al. 2012, 2013a).

Future yield reductions of 13% and 16% in corn and soybean, respectively, were indicated by analysing the patterns of temperature change in USA during 1976–2006 (Kucharik and Serbin 2008). Likewise, Southern plains and south-west regions of the United States suffered sizeable agricultural loss estimating US$5 billion owing to combined drought and HS (NCDC 2011). Lobell et al. (2013) suggested that the rise in temperature beyond 30°C would exert noteworthy negative impacts in rainfed maize in USA and Africa.

Crop-wise impact of HS worldwide

Substantial reductions in yield were documented in various important food crops worldwide (Lobell and Field 2007, Lobell et al. 2008). For instance, massive yield loss of 5.18 million tons was reported from 3 million ha area in China in rice due to the disastrous HS episode of 2003 (Li et al. 2004, Tian et al. 2009). Similarly, Lobell et al. (2008) recorded 4–14% yield loss in rice due to 1°C increase in temperature in South-East Asia.

In addition to rice, Reynolds et al. (2001) asserted that seven million hectares in developing countries as well as 36 million ha in temperate environments representing the prominent wheat-growing areas are likely to be severely hit by HS. Similarly, Lobell and Field (2007) reported an annual global yield reduction of 19 million tons in wheat, costing $2.6 billion, due to unusual warming during 1981–2002. According to a report published by USDA (2010), nearly 30% of the total wheat production was reduced in Russia, which was attributable to drought and HS. Worldwide, temperature trend analysis (1980–2008) has shown about 5.5% reduction in wheat production (Lobell et al. 2011a). Recently, based on 9 years' satellite data on wheat growth in northern India, Lobell et al. (2012a) suggested that temperature above 34°C increases the rate of senescence, thereby causing significant loss in grain yield.

In case of maize, warming between 1981 and 2002 has accounted losses up to 12 million tons per year comparable to a monetary loss of $1.2 billion (Lobell and Field 2007). In a similar study, up to 10% reduction in maize yield was recorded due to increase in temperature (6°C) during the grain-filling stage (Thomson 1966). More severe maize yield reductions have been projected in USA by Kucharik and Serbin (2008) based on the patterns of temperature change that were observed in USA during 1976–2006. By 2100, yield loss of 30% in corn was predicted in USA using the nonlinear and asymmetric temperature and yield relationship analysis (Schlenker and Roberts 2009). A 3.8% reduction in maize production was shown by global temperature trend analysis using historical yield data from 1980 to 2008 (Lobell et al. 2011a). Similarly, Lobell et al. (2013) observed that rise of temperature beyond 30°C will cause negative impacts in rainfed maize in USA and Africa. Additionally, Deryng et al. (2014) have suggested that extreme HS at anthesis could potentially decrease 45% of the global maize yield by 2080s as compared to 1980s.

In soybean, nonlinear and asymmetric temperature and yield relationship analysis has predicted almost 46% yield loss in USA before year 2100 (Schlenker and Roberts 2009). Future yield reduction of 16% has been indicated in soybean based on the patterns of temperature change prevailed in USA during 1976–2006 (Kucharik and Serbin 2008). Similarly in barley, warming from 1981 to 2002 caused loss of eight million tons per year, costing around $1.0 billion (Lobell and Field 2007). Therefore, given the burgeoning agricultural adversities and increasing evidences of negative impact of HS on crops worldwide, an immediate attention is needed to transform breeding strategies and to reconstruct crop ideotypes followed by an in silico testing using simulation models (Gupta et al. 2012, Semenov and Stratonvitch 2013).

HS in Crops: Influencing Plants from Germination to Maturity

A series of in-depth reviews were published in recent years describing the detrimental impacts of HS on crop growth and development (Wahid et al. 2007, Bita and Gerats 2013, Hasanuzzaman et al. 2013). High temperature stress impairs various vital physiological processes in plants, including photosynthesis, respiration and transpiration through impeding carbon assimilation (Stone 2001). The stress-led alterations in plant's physiology hinder the overall reproductive processes and eventually result in substantial yield loss (Barnabás et al. 2008, Hedhly et al. 2008, Zinn et al. 2010).

Impacts of HS on warm-season crops

Considering impacts on warm-season crops, rice is among the major food crops that often encounter the challenges of increasing HS in tropical and subtropical regions, especially South and South-East Asia (Jagadish et al. 2012, Manigbas et al. 2014). Escalating HS has raised a serious concern in major rice-growing areas of Asia (Catherine and Gemma 2012). In rice, HS impedes various stages starting from the emergence to the harvesting (Krishnan et al. 2011, Shah et al. 2011). Temperature beyond 40°C strictly restricts the emergence of rice seedlings (Yoshida et al. 1981, Akman 2009). Simulation studies have shown a 10% reduction in rice yield with rise of each 1°C minimum temperature during growing periods (Peng et al. 2004). In a similar way, Lyman et al. (2013) have also observed a 6.7% reduction in rice yield, which is attributable to an average 1°C rise in temperature during the growing period. Several attributes such as plant height, tiller number and total dry weight were known to be affected negatively by temperature rising beyond 32°C/25°C (Yoshida et al. 1981).

Concerning differential susceptibility of various growth phases to HS, flowering stage and microsporogenesis were reported to be the most vulnerable in rice (Satake and Yoshida 1978, Nakagawa et al. 2002). Subjecting rice to 41°C temperature for 4 h at flowering stage resulted in the complete sterility (IRRI 1976). Likewise, spikelet sterility was noted at temperatures rising beyond 33°C, and notably, most of the grains showed emptiness above 35°C (Matsui et al. 1997, 2000, Jagadish et al. 2007). In case of maize, impaired protein synthesis in embryo led to the inhibition of germination at temperatures beyond 37°C (Riley et al. 1981). The coleoptile growth is completely halted in maize at 45°C (Weaich et al. 1996, Akman 2009). Similarly, HS leads to an enhancement in respiration and reduction in photosynthesis in maize (Crafts-Brander and Salvucci 2002), which often causes pollen sterility and inflicts kernel development (Schoper et al. 1987a,b, Cheikh and Jones 1994).

Instances are reported in soybean, where prevalence of HS during seed filling stage caused lowering in germination and seedling vigour (Egli et al. 2005). Under tropical humid conditions, increase in temperature above 30°C impoverished seed production in soybean (Lindsey and Thomson 2012). Negative impact of high temperature on growth and photosynthesis in different legumes was also documented by McDonald and Paulsen (1997). For example, temperature above 30°C hampers microsporogenesis in common bean (Porch and Jahn 2001, Rainey and Griffiths 2005, Porch 2006). A wide range of reproductive traits from pollen viability to fertilization process including pod and seed setting were investigated in common bean (Anthony et al. 1980, Weaver and Timm 1988, Gross and Kigel 1994) and groundnut (Prasad et al. 1998, 1999a,b, 2000, 2001). Apart from legumes, demonstrably reliable evidences were also gained in other important crops such as sorghum (Eastin et al. 1983, Prasad et al. 2008) and cotton (Singh et al. 2007a). In sorghum, failure of germination and epicotyl emergence was reported due to soil seed zone temperature exceeding 45°C (Peacock et al. 1993). Likewise, cessation of seedling emergence was noticed beyond 30°C in tomato (Camejo et al. 2005). Additionally, HS diminishes net photosynthetic rate (Camejo et al. 2005), pollen grains and their viability (Pressman et al. 2002) and ultimately impairs the fruit development (Peet et al. 1998, Adams et al. 2001).

Impact of HS on cool-season crops

With regard to cool-season crop, more than 13.5 million ha of wheat-growing area in India is under HS (Joshi et al. 2007a,b). Given increasing severity during heading to maturity, HS is becoming a major concern in cooler northern wheat-growing regions (Liu et al. 2014). Joshi et al. (2007a) and Singh et al. (2007a) reported Eastern Gangetic Plains (EGP), central and peninsular India and Bangladesh as the most vulnerable regions to HS in South Asia for growing wheat. Severe damages due to HS were also noted in late-sowing wheat, in particular during grain-filling stage (Rane et al. 2007). Joshi et al. (2007a) and Rane et al. (2007) reported that curtailing ‘cool period’ in wheat also resulted in yield reduction. Likewise, Asseng et al. (2011) speculated that temperature variation of ±2 °C during growing season could potentially reduce 50% yield in Australia. HS impairs seed germination in wheat by disrupting the function of enzymes that are associated with starch breakdown (Essemine et al. 2010).

Concerning flowering period, key determining stages involving grain size and grain number development, and ultimately the yield represent the most vulnerable stages in wheat (Semenov 2009, Farooq et al. 2011). Various abnormalities were observed in wheat that could be ascribed to HS, which include unusual ovary development (Saini et al. 1983), tapetum degradation during microspore meiosis and pollen sterility (Saini et al. 1984, Sakata et al. 2000, Zinn et al. 2010), disruption in starch synthesis during early grain-filling period and lowering in biomass production (Reynolds et al. 2007). In case of cool-season legumes, germination rates of chickpea and lentil seeds are retarded owing to increase in temperature above 33 and 24.4°C, respectively (Covell et al.1986), while anomalies were manifested with soil temperature above 32°C in lettuce (Gray 1975). Several investigations describing growth retardation, and adverse impact on photosynthesis in Brassica juncea L. (Hayat et al. 2009) and inflicted seed production due to hindrance in micro- and mega-gametophyte fertility in B. napus were noted under HS (Young et al. 2004).

Molecular Mechanism Underlying HS Acclimation

Stimuli of abiotic stress perceived by the plant generate an integrated signalling cascade that is generally triggered by receptors embedded in plasma membrane of the cell (Vigh et al. 1998, Sangwan et al. 2002, Los and Murata 2004). Secondary messengers such as Ca²⁺ ion and calmodulins (Liu et al. 2003, Wu and Jinn 2010), calcium-dependent protein kinases (CDPKs), calcineurin B-like (CBL) (Das and Pandey 2010) and CBL-interacting protein kinases (CIPKs) (Kolukisaoglu et al. 2004) act as calcium sensor. This event switches the mitogen-activated protein (MAP) kinases on, followed by the activation of transcription factors and subsequently the concerned heat-shock protein (HSP) genes (Sangwan et al. 2002, Suri and Dhindsa 2008). The complex pathway underlying HS perception and signal transduction are described in detail elsewhere (Larkindale and Knight 2002, Sung et al. 2003, Kaur and Gupta 2005, Saidi et al. 2010).

While confronting high temperature stress and alleviation from damage of cellular protein structure essential for survival in stressed conditions, plant triggers a novel class of protein called HSPs. These HSPs serve as molecular chaperons to maintain conformational protein functions as well as cellular protein refolding, thereby protecting plants under HS conditions (Baniwal et al. 2003, Wang et al. 2004).

Based on molecular weight, function and amino acid sequence, the HSPs can be categorized into five major types, viz. Hsp100, Hsp90, Hsp70, Hsp60 and sHSPs (Schlesinger 1990, Schoffl et al. 1998, Kotak et al. 2007, Gupta et al. 2010a,b, Al-Whaibi 2011). Higher plants generate HSPs as and when these are subject to HS conditions (Vierling 1991). In response to extreme environmental stresses, plant expresses stress-responsive genes that are regulated by numerous TFs via binding to the cis-acting promoter of the concerned gene (Nakashima et al. 2009, Zhang et al. 2009a, Dubos et al. 2010, Chen et al. 2012a, Mizoi et al. 2012).

Harnessing Crop Germplasm Repertoire for Breeding Against HS

Keeping ‘yield’ as the principal criterion, serious breeding efforts were made to develop high-yielding cultivars in most of the important food crops. The conventional crop breeding schemes relying solely on selection and intermating have unintentionally resulted in paucity in the genetic variation especially for economically important traits that underwent domestication/selection (Tanksley and McCouch 1997, Eyre-Walker et al. 1998, Lee 1998, Gur and Zamir 2004, McCouch 2004). Therefore, accelerating crop improvement demands an extensive search for genetic variability in cultivated as well as in wild species. In the context, heat-tolerant gene(s)/QTLs and the component traits conferring HT must be explored thoroughly within the entire gene pool, especially targeting the non-adapted and underutilized crop wild relatives (CWRs) and the landraces (Lee 1998, Fernie et al. 2006). Impressive accomplishments were achieved in harnessing the natural genetic variation for HT, and additional efforts are underway to introduce the heat-tolerant QTLs/genes into different genetic backgrounds (Jagadish et al. 2008, Jiang-lin et al. 2011). The presence of HT was examined in cultivars and wild species of wheat (Rawson 1986, Edhaie and Waines 1992, Reynolds et al. 1994a). Similarly, a rice wild relative (Oryza meridionalis) exhibited higher growth rate and less affected photosynthesis at 45°C compared to O. sativa ssp. japonica (Scafaro et al. 2010). Two rice genotypes, viz. ‘Dular’ and ‘Todorokiwase’, offered higher HS tolerance at booting stage at 39°C, while ‘Milyang 23’ showed tolerance at flowering stage at 38°C (Tonorio et al. 2013). On the other hand, tolerance was manifested at both the above-mentioned stages by Giza 178. Jagadish et al. (2008) reported advantages offered by the genotype CG14 (O. glaberrima) that reached peak anthesis stage earlier than O. sativa under both controlled and HS conditions. In addition, they also noted that the cultivar ‘N22’ possessed marked HS tolerance (Jagadish et al. 2010a,b, Madan et al. 2012) exhibiting 64–86% spikelet fertility at 38°C temperature compared to the susceptible cultivars, viz. ‘Azucena’ and ‘Moroberekan’, that had meagre fertility (up to 8%) (Jagadish et al. 2008). Later, higher pollen viability and spike fertility under HS were confirmed in rice genotypes, viz. ‘N22’ and ‘NH219’ (Poli et al. 2013). In recent past, introgression breeding in rice facilitated transfer of HT from ‘N22’ to ‘Xieqingzao B’ line by developing BC₁F₈ lines (Jiang-lin et al. 2011). Additionally, advanced line derived from Gayabyeo/N22 cross has offered HS tolerance and high yield (Manigbas et al. 2014). Given the importance of anther dehiscence in imparting HS tolerance, a study was undertaken in rice to evaluate anther characteristic, especially closure of locules under high temperature (Matsui and Omasa 2002). Consequently, japonica rice cultivars ‘Nipponbare’ and ‘Akitakomachi’ exhibited higher fertility under 37.5°C/26°C temperature during flowering (Matsui and Omasa 2002).

In wheat, wide variation was revealed in the wild progenitor Aegilops tauschii Coss. in comparison with the tolerant cultivar ‘C273’ for important HT-related traits such as cell membrane stability and ‘TTC’ (2, 3, 5-triphenyl tetrazolium chloride)-based cell viability (Gupta et al. 2010a). Recently, A. tauschii was successfully used as a donor for incorporating HT-relevant component traits such as cell membrane stability and chlorophyll retention into cultivar ‘PBW550’ through backcrossing (Sehgal et al. 2011). Likewise, identification of A. speltoides Tausch and A. geniculata Roth species offering HS tolerance at reproductive stage opens up new opportunities for incorporation of HT genes in hexaploid wheat in near future (Pradhan et al. 2012a). By screening wheat against stressed conditions involving both drought and HS, ALTAR 84/AO'S’ and ALTAR 84/A. tauschii genotypes remained least affected, notably at two critical stages: (i) from emergence to anthesis and (ii) from emergence to postanthesis (another 21 days after anthesis) (Pradhan et al. 2012a). More recently, attempts were made in wheat aiming at introgression of wheat-Leymus racemosus chromosome to ‘Chinese spring’ cultivar to enhance HT and better adaptation under HS (Mohammed et al. 2014). In search of some novel sources for HT, Hede et al. (1999) explored potential of wheat landraces, and consequently, three Mexican landraces were identified that carried superior canopy temperature depression trait. Further, concerning germination-related traits, two genotypes, viz. ‘Moomal-2000’ and ‘Mehran-89’, performed better at 20–30 °C (Buriro et al. 2011). Recently, while investigating photosynthetic activity in flag leaves during grain-filling stage, Feng et al. (2014) identified cultivar ‘Jimai22’ having 6% less reduction in grain yield under HS. Additionally, this cultivar also offered benefits including stability of PSII and carboxylation activity under HS. Evaluation at terminal growth stage focusing stay green trait resulted in the discovery of three promising genotypes on the basis of maximum grain development and higher survival under stressed conditions (Rehman et al. 2009). More recently, promising genotypes ‘WH1021’ and ‘WH730’ showing enhanced yield under HS were discovered in wheat (Dhanda and Munjal 2012). In the same way, three synthetic wheat lines, viz. SYN 11, SYN36 and SYN44, were declared as highly tolerant to heat based on cluster analysis of morphological attributes and ISSR markers (Sharma et al. 2014a).

In case of tropical grain legumes, direct selection targeting yield led to the recovery of a tolerant genotype ‘California Blackeye No. 27 (CB27)’ characterized by the tolerance to HS at reproductive stage (Ehlers et al. 2000). In parallel, two additional genotypes ‘B89-200’ and ‘TN88-63’ also showed higher yield under hot short days (Ehlers and Hall 1998). Likewise, four common bean genotypes, viz. ‘SRC-1-12-1-182’, ‘SRC-1-12-1-48’, ‘98020-3-1-7-2’ and ‘98012-3-1-2-1’, were found to be tolerant to HS on the basis of stress indices such as heat tolerance index (HTI), heat susceptibility index (HSI) and geometric mean (GM) analyses (Porch 2006). In chickpea, HS-tolerant genotype ‘ICCV 92944’ was identified while screening diverse genotypes under field conditions (Gaur et al. 2012). Several other chickpea genotypes with enhanced HT are listed recently by Jha et al. (2014). Pollen-based screening of genotypes delivered various tolerant accessions in different crop species. Examples include ‘DG 5630RR’ in soybean (Salem et al. 2007), ‘AZ100’ in maize (Petolino et al. 1992) and ‘ICC1205’ and ‘ICC15614’ in chickpea (Devasirvatham et al. 2012, 2013). Recently, ‘La Posta Sequia C7-F64-2-6-2-2’ and ‘DTPYC9-F46-1-2-1-2’ were identified in maize carrying tolerance to both stresses, that is, drought and heat (Cairns et al. 2013a).

Apart from the genotypes described above, several hybrids including ‘YH-1898’, ‘KJ. Surabhi’, ‘FH-793 ND-6339’ and ‘NK -64017’ provided stable yield under HS (Rahman et al. 2013). Important cultivars of tomato ‘Fresh market 9’, ‘Saladette’, ‘Processor 40’ and ‘Solar set’ have been reported to set fruit under HS (Abdul-Baki 1991).

‘Heat escape’ means an alternative mechanism through which plant completes its life cycle before the onset of HS. This property has been exploited in developing early maturing genotypes in durum wheat such as ‘Waha-1’, ‘Omrabi-5’ and ‘Massara-1’ (Al-Karaki 2011). By using temperature induction response (TIR) technique, Senthil-Kumar et al. (2003) observed adequate genetic variability for thermotolerance among the parental lines of the hybrid ‘KBSH-1’, viz. ‘CMS234A’, ‘CMS234B’ and ‘6D-1’. In a similar manner, application of chlorophyll fluorescence measurement techniques in USDA upland cotton collection resulted in the identification of nine wild accessions with high vegetative HT (Wu et al. 2013). The availability of potential donors for HT would encourage plant breeders not only to deploy these novel sources directly in breeding schemes but also to excavate the resilient alleles that underlie tolerance.

Physiological and Adaptive Trait Breeding for HS Tolerance

Concerning yield improvement under stressed conditions, several factors such as low heritability and a complex network of major and minor QTLs limit the efficacy of direct selection method (Hittalmani et al. 2002, Leung 2008, Manavalan et al. 2009). Compounding the problem, environmental factors present additional barriers to breeding for high-yielding lines with enhanced HT. Furthermore, poor understanding of genetic inheritance of heat coupled with less availability of validated QTLs/cloned gene(s) for HS tolerance in plants limits the progress of crop improvement (Cossani and Reynolds 2012). Given the context, genetic tailoring of the physiological traits could be a promising approach for incorporating gene(s)/QTLs that determine complex abiotic stress tolerance in crop plant. Physiological trait-based breeding strategy provides advantage over conventional breeding approach (Reynolds and Trethown 2007). Additionally, physiological approach offers the benefit of maximizing the probability to harness the more relevant additive gene actions (Reynolds and Trethown 2007). Extensive efforts were undertaken for physiological trait breeding in wheat at International Maize and Wheat Improvement Center (CIMMYT), which primarily relied on three basic steps: i) characterization of the parental genotypes in various crossing schemes; ii) appropriate mating scheme designed to target traits; and iii) selection of superior progeny from the early generations (Reynolds et al. 2009).

Central to the HS acclimation, breeding should be aimed at physiological traits that are related to canopy structure, delayed senescence, photosynthesis efficiency, less respiration rates, reproductive traits and the harvest index (Cossani and Reynolds 2012, Gupta et al. 2012). As evident from the above description, emphasis should be given to capture the genetic variation in plant phenologies that confers better adaptation under stressed conditions (Evans 1993). In parallel, selection for morpho-physiological traits that are involved in drought acclimation and indirectly associated with yield constitutes an alternative approach for enhancing drought tolerance in crop plants (Richards 1996, Tuberosa et al. 2002). The potentiality of this approach for enhancing HT was explored in wheat (Cossani and Reynolds 2012). Substantial genetic variability for photosynthetic rate under HS was detected in wheat (Blum 1986, Delgado et al. 1994). Similarly, the genetic variability was evaluated by subjecting 16 wheat genotypes to HS at CIMMYT (Reynolds et al. 2000). Likewise, high level of photosynthetic rate in leaves was demonstrated to serve as a potential indicator of HS tolerance in rice (Restrepo-Diaz and Garces-Varon 2013). The genetic variability resulted from loss of chlorophyll content and premature leaf senescence (Al-Khatib and Paulsen 1984, Harding et al. 1990). In a likely manner, three HS-tolerant landraces were identified in wheat by evaluating leaf chlorophyll content (LCC), canopy temperature depression (CTD) and thousand-kernel weight (KWT) (Hede et al. 1999). While screening over 1000 wheat genotypes, the chlorophyll fluorescence was established as an important physiological parameter (Sharma et al. 2012). Apparently, CTD could be considered as the potential mechanism underlying heat escape in cotton (Cornish et al. 1991), and it is also demonstrated to be an effective selection parameter against HS in wheat (Amani et al. 1996, Fischer et al. 1998).

Considerable extent of phenotypic variation (PV) was noticed in wheat (Pierre et al. 2010), and meaningful conclusions were generated suggesting correlation of CTD value with HT in wheat (Balota et al. 2008, Pradhan et al. 2012a). Similarly, the cooler canopy temperature (CT) caused higher yield under HS in wheat (Kumari et al. 2012, Mondal et al. 2013). Reynolds et al. (1994a) experimentally demonstrated that CTD in conjunction with flag leaf stomatal conductance and photosynthetic rate is positively correlated with yield in wheat under HS. Blum and Ebercon (1981) surveyed the genetic variation for membrane thermostability (MT) in different crops and advocated that MT under HS should be considered as a vital component for measuring HT. Practicing selection under HS for MT during anthesis stage delivered significant yield improvements in wheat (Shanahan et al. 1990). Various physiological traits and their relative contributions to HT in wheat are discussed in detail by Gupta et al. (2012).

Screening against HS based on parameters such as electrolyte leakage from cell membrane and chlorophyll fluorescence revealed negative association of membrane injury with specific leaf weight in some legume species including groundnut and soybean (Srinivasan et al. 1996). Combination of the two selection parameters, viz. high chlorophyll content and MT, was implicated to carry out selections in Brassica and wheat (Ristic et al. 2007, Kumar et al. 2013). Further, higher pollen grain fertility under HS may serve as an important criterion for measuring HT (Rodriguez-Garay and Barrow 1988). Targeting pollen selection under HS in cotton, heat-tolerant genes were transferred from a donor line ‘7456’ (G. barbadense L.) to a heat-sensitive genotype ‘Paymaster 404’ through backcrossing (Rodriguez-Garay and Barrow 1988). In addition to pollen selection, relative cell injury level (RCIL) under HS could also be taken as a reliable index in determining HS tolerance in cotton (Sullivan 1972, Khan et al. 2008). Similarly, Petolino et al. (1992) proposed gametic selection as a viable option for addressing HS in maize.

With regard to root traits, adaptation of root respiratory carbon metabolism can offer tolerance to soil temperature by managing the ion uptake load as investigated in Agrostis species (Rachmilevitch et al. 2006). Furthermore, Huang et al. (2012) deduced that the efficient carbon and protein metabolism conferred higher thermotolerance to roots at 45°C in Agrostis scabra (a C₃ perennial grass species). From phenology point of view, selection of early flowering and maturity could enable us to escape HS in spring-sown chickpea in Mediterranean region and south India (Toker et al. 2007, Berger et al. 2011). Likewise, early-maturity-led escape mechanism enabled addressing HS in wheat in Eastern Gangetic Plains and various South Asian locations (Joshi et al. 2007a, Mondal et al. 2013). Stay green or delayed senescence imparts yield improvement under abiotic stress (Thomas and Smart 1993, Borrell et al. 2000, Thomas and Howarth 2000, Harris et al. 2007). Screening of a total of 963 diverse wheat accessions at various sowing time suggested that stay-green trait associated with CTD could be a strong indicator of HT (Kumari et al. 2007). However, stay-green trait is less important in the context of yield on account of disability in translocation of stem reserves to grain under HS (Blum 1998). While under conditions encompassing HS alone as well as HS and drought, stay-green trait measured as normalized difference vegetation index (NDVI) at physiological maturity exhibited a positive correlation with yield (Lopes and Reynolds 2012). Therefore, physiological trait-based breeding remains a promising improvement strategy to develop heat-tolerant genotypes without causing yield penalty.

Intervention of Molecular Markers for HT Breeding

QTL identification and trait mapping by molecular markers

To date, conventional breeding schemes have been extensively deployed for uncovering the HT-relevant gene(s) and their inheritance patterns, thus illuminating the causal molecular mechanism (Wahid et al. 2007, Farooq et al. 2011). To this end, advances in DNA marker discovery and genotyping assays have permitted the accurate determination of chromosomal position of the QTLs responsible for HT in different crops (Jagadish et al. 2010a, Pinto et al. 2010, Paliwal et al. 2012, Bonneau et al. 2013). An updated list of QTLs associated with HT in various crops is shown in Table 1. Further, information related to important QTLs controlling tolerance against various abiotic stresses in different crops can be accessed from PLANTSTRESS site (http://www.plantstress.com/biotech/index.asp?Flag=1).

Table 1. List of QTLs/gene(s) conditioning HT in different crops

Crop	Markers linked to the QTLs	Name/No. of the QTL/loci	Chromosomal location/Linkage group (LG)	Mapping population	Range of PV (%)	References
Arabidopsis	SNP	5 THERM QTLS (THERM1e, THERM3.e, THER4.1.e, THERM4.2.e and THERM5.e)	1, 3, 4, 4 and 5	-	7	Li et al. (2014a,b),
Azuki bean	-	HQTL1 and HQTL2 pollen viability under HS		BC (Vigna riukiuensis × V. angularis)	-	Kaga et al. (2003), Vaughan et al. (2005), Tomooka et al. (2011),
Barley	SSR	34 putative QTLs	-	BC2DH (Scarlett × ISR42-8)	-	Mohammed (2004)
Brassica	AFLP, RAPD	5 QTLs	Three different LG		-	Shuancang et al. (2003)
Cowpea	SNP	Five regions of 9% of the cowpea genome, Cht-3		RIL (CB27 IT × 82E-18)	11.5–18.1	Lucas et al. (2013)
	SNP	Hbs1, Hbs2 and Hbs3		IT93K-503-1 × CB46 IT84S-2246 × TVu14676	6.2–77.3	Pottorff et al. (2014)
Maize	RFLP	6 QTLs (cellular membrane stability)		RIL (T232 × CM37)	-	Ottaviano et al. (1991)
	RFLP	3–8 QTLs, heat-shock protein (HSP) expression	-	RIL (T232 × CM37)	-	Frova and Sari-Gorla (1993),
	RFLP	5 QTLs for IPGG and six QTLs for IPTG.	-	RIL (T232 × CM37)	-	Frova and Sari-Gorla (1994)
Potato	AFLP, SSR	QTLs resistance to internal heat necrosis (IHN)	LG 4, 5, 7 and 10	RIL (Atlantic × B1829-5)	-	McCord et al. (2011)
Rice	-	qhr1, qhr3-1, qhr4-3, qhr8-1, qhr11-1 and qhr11-2	1, 3, 4, 8 and 11	DH (IR64 × Azucena)	1.3–22.8	Cao et al. (2003)
	RFLP	3 QTLs	1, 4 and 7	BIL Nipponbare/Kasalath//Nipponbare	8.94–17.25	Zhu et al. (2005)
	-	9 QTLs thermotolerance for amylose content and gel consistency	6 and 8	Nipponbare/Kasalath//Nipponbare	-	Zhu et al. (2006a)
	SSR	WBK–qWK1-1, qWK1-2, qWK2 and qWK8	1, 2 and 8	RIL (Chiyonishiki × Koshijiwase)	8.8–15	Tabata et al. (2007)
	SSR	2 putative QTLs associated with white-back kernels	4, 6	(Hana-echizen × Niigata-wase)	15.2–59.6	Kobayashi et al. (2007)
	SSR	qHt3, qHt9a	3 and 9	RIL (T219 × T226)	7.6–11.4	Chen et al. (2008)
	SSR	3 QTLs (qhts-2, qhts-3 and qhts-5)	LG 2, 3 and 5	RIL (Zhongyouzao No. 8 × Fengjin)	6.59–10.72	Zhang et al. (2008)
	SSR	RM3735 and RM3586 loci	4 and 3	F2 (996 × 4628)	3–17	Gui-lian et al. (2009)
	-	qtl_2.3,qtl_4.1, qtl_1.1,qtl_2.2,qtl_8.2,qtl_1.1,qtl_8.1, qtl_10.1,qtl_1.1,qtl_3.4,qtl_8.3,qtl_10.1, qtl_1.1,qtl_11.1,qtl_10.1,qtl_1.1,qtl_10.1 and qtl_11.1	1, 2, 3, 4, 8, 10, 11	(Bala × Azucena)	7–17.6	Jagadish et al. (2010a)
	SSR	2 QTLs	4 and 10	RIL (996 × 4628)	21.3–25.8 11.5–11.6	Xiao et al. (2011)
	SSR	qPF4, qPF6	4 and 6	RIL (996 × 4628)	15.1–9.31	Ying-hui et al. (2011)
	SNP	qHTSF1.1 qHTSF4.1	1 4	BC1F1, BC2F2 and F2 (IR64 × N22)	12.6–17.6	Ye et al. (2012)
	SNP	OsHTAS locus	9	F1 and F2 (HT54 × HT13)	-	Wei et al. (2013)
	SSR	qHTS1-1	1	Introgressed line YIL106	6.83–14.63	Lei et al. (2013)
	SSR	qHTS1-2, qHTS2, qHTS3 and qHTS8	1, 2, 3 and 8	(Teqing × O. rufipogon)
	SSR	(qWB3, qWB4, qWB6 and qWB9) QTLs for WBK	3, 4, 6 and 9 for WBK	(Hana-echizen × Niigata-wase)	(31.5–36.8) WBK	Kobayashi et al. (2013)
		(qKW3-1, qKW3-2, qKW6, qKW7 and KW10) QTLs for KW	3, 3, 6, 7 and 10 for KW		(8.4–12.1) DTH
		(qDH1, qDH3 and qDH6) QTLs for DTH	1, 3 and 6 for DTH
	SSR SNP	QTL for white-back grains	6	RIL (Tohoku 168 × Kokoromachi)	-	Shirasawa et al. (2013)
	SSR	9 QTLs	3, 4, 6, 8, 10 and 11	BC2F2 (OM5930x N22)	17.1–36.2	Buu et al. (2014)
Sorghum	RFLP	Stg 1, Stg 2, Stg 3 Stg 1, Stg 2 responsible for heat-shock protein expression	LGA, LGD and LGJ	RIL	46	Xu et al. (2000)
Tomato	fAFLP	QTLs related to fruit set under HS	1, 3, 4, 6, 7 and 9	F2 (Jab-95 × Caribe)	32.8	Grilli et al. (2007)
	RAPD and SSR	2 QTLs	LG3 and LG7	F2 (01137 × CLN2001A)	-	Xiang-yang et al. (2008)
	-	29 putative QTLs, 2 QTLs contributed in viability of pollen under HS		RIL (CLN1621 L × CA4)	-	Schafleitner (2014)
Wheat	SSR	1.4 genes		(Ventnor × Karl 92)	-	Yang et al. (2002)
	SSR	2 QTLs		(Ventnor × Karl92)	11–12.	Yang et al. (2002)
	AFLP and SSR	3 QTLs	LG1B, 5B and 7B	RIL (Kauz × MTRWA116)	27.3–44.3	Mohammadi et al. (2008)
	SSR	Common QTL for drought and heat stress traits	1B-a, 2B-a, 3B-b, 4A-a,	RIL (Seri × Babax)	17 (yield QTL)	Pinto et al. (2010)
			4B-b, and 7A-a		28 (CT)
	AFLP, SSR	(Q75%Gh.ksu-2A, Q75%Gh.ksu-2A,	2A, 6B, 3A, and 7A	RILs (Ventnor × Karl 92)	53 (75%Q),	Vijayalakshmi et al. (2010)
		Q75%Gh.ksu- 3B) 75%G,			28 (25%G),
		(Q25%Gh.ksu-2A, Q25%Gh.ksu-2A) 25%G,			63 (50%G),
		(Q50%Gh.ksu-2A, Q50%Gh.ksu-6A) 50%G, (QMrsh.ksu-2A)			40 (MRS),
		(QTmrsh.ksu-2A, QTmrsh.ksu-6A, QTmrsh.ksu-6B) TMRS,			55 (TMRS),
		(QPgmsh.ksu-3A, QPgmsh.ksu-6B) PGMS,			36.4 (PGMS),
		(QFv/Fmh.ksu-7A) Fv/Fm			11.2 (Fv/Fm)
	SSR	5 QTLs	1A, 2A, 2B, and 3B associated with HIS	RIL (Halberd × Cutter)		Mason et al. (2010)
	SSR	3 QTLs (QSg. bhu-1A, QSg.bhu-3B and QSg.bhu-7D)	1AS, 3BS and 7DS	RIL (Chirya3 x Sonalika)	38.7	Kumar et al. (2010)
	SSR	14 QTLs for heat susceptibility index (HIS)	1B, 3B, 4A, 5A,	F2:6 RIL (Halberd × Karl 92)	Individual QTL	Mason et al. (2011)
		7 QTLs colocalized for HIS and TD trait	5B and 6D		(4.5–19.3)
	SSR	Xgwm132-linked QTL, Xgwm577-linked QTL and Xgwm617-linked QTL	6B, 7B and 6A	F1, F2 (Debra × Yecora Rojo)	3–25.	Barakat et al. (2011)
	SSR	12 QTLs	1A, 7A, (3B, 3A, 5B), 2D, 1D, (2A, 2B, 2D),	F2 (Ksu106 × Yecora Rojo)	22–64	Barakat et al. (2012)
	SSR	QHthsitgw.bhu-2B, QHthsitgw.bhu-7B, QHthsitgw.bhu-7D, (TGW), QHthsiYLD.bhu-7B, QlsYLD.bhu-7B, (YLD), QHthsigfd.bhu-2B (GFD), QHtctd.bhu-7B (CTD), Qls-dm.bhu-7D (DM)	1B, 7D, (4A, 5D), (7A, 7B, 7D), 5A and 3B, BL, 7BL, 7DS (TGW), 7BL (YLD) 2B1 (GFD), 7BL (CTD) and 7DS (DM)	RIL (NW1014 × HUW468)	9.78-20.34 (TGW) 13.21 (YLD), 20.15 (GFD), 19.81 (CTD) and 7.42 (DM)	Paliwal et al. (2012)
	DArT and SSR	2 QTLs, (Q.Yld.aww-3B-2 and Q.Yld.aww-3D)	3B	DH, RIL (RAC875 ×  Kukri)	22	Bennett et al. (2012)
	SSR	4 QTLs (Qsdscon.tam-1B, Qsdsheat.tam-1D, Qsdscon.tam-4A and Qsdssta.tam-7A)	LG 1B, 1D, 4A, and 7A	RILs (Halberd × Cutter)	18–30	Beecher et al. (2012)
	SSR	Marker locus gwm299	3BL	DH, RIL (RAC875 × Kukri)	-	Bonneau et al. (2013)
	AFLP, SSR, DArT	TKW QTLs linked or pleiotropic to DH and DM	4B and 7D-b	RIL (Barbax × Seri)	39 (TKW)	Lopes et al. (2013)
	AFLP, SSR, DArT	Few QTLs associated with ECG	7D-a and 7D-b	RIL (Barbax × Seri)	17.4 (ECG)	Lopes et al. (2013)
	AFLP, SSR, DArT	Consistent QTLs were detected for CTvg and CTgf	4A	RIL (Barbax × Seri)	16 (Ctgf)	Lopes et al. (2013)
	SSR	QChlc.tamu-1B (chlorophyll content), QFlt.tamu-2B (FLTD), QIkw.tamu-5A (IKW)	-	19 families consisting of 384 individuals developed from 3-way cross	-	Ali et al. (2013)
	SSR	7 QTLs associated with HSI traits, GFD, TGW, GY and CT 9 QTLs associated with GFD, TGW, GY and CT under HS	1D, 6B, 2D and 7A	DH (Berkut × Krichauff)	-	Tiwari et al. (2013)
		QTL for leaf and spike temperature depression and leaf wax	1B and 5A	RIL (Halberd × Karl 92)	8–12	Mondal et al. (2014)

• BC, backcross; BILs, Backcross inbred lines; TKW, thousand-kernel wt; DH, Double Haploid; DM, days to maturity; HS, heat stress; HSI, Heat susceptibility index; GY, Grain yield; TGW, Thousand grain weight; GFD, Grain-Filling duration; CT, Canopy temperature; TD, Temperature depression; CTD, canopy temperature depression; PV, phenotypic variation; IPGG, pollen germinability; IPTG, pollen tube growth; WBK, white-back kernels; KW, 1000 kernel weight; DTH, days to heading; EGC, Early ground cover; CTvg, Canopy temperature under vegetative growth; CTgf, Canopy temperature under grain-filling stage; 75%G, 75% green; 25%G, 25% green; 50%G, 50% green; MRS, maximum rate of senescence; TMRS, time to maximum rate of senescence; PGMS, percentage greenness at maximum senescence; Fv/Fm, chlorophyll fluorescence.

By applying 245 restriction fragment length polymorphism (RFLP) markers in 98 backcross inbred lines (BILs) derived from the cross (Nipponbare × Kasalath) × Nipponbare, three QTLs for HT were detected each on chromosome 1, 4 and 7 (Zhu et al. 2005). These three QTLs explained 8.94%, 17.25% and 13.50%, respectively, of the entire PV during grain-filling stage (Zhu et al. 2005). In addition to QTL analysis, application of BSA using a set of SSR markers in 279 F₂ (996 × 4628) individuals resulted in identification of two loci associated with HT, viz. RM3735 (chromosome 4) and RM3586 (chromosome 3), that controlled 17% and 3% proportions, respectively, of the total PV (Gui-lian et al. 2009). Similarly, eight QTLs governing spike fertility under high temperature were mapped on different chromosomes, such as 1, 2, 3, 8, 10 and 11 (Jagadish et al. 2010a). Recently, a major dominant locus OsHTAS (Oryza sativa heat tolerance at seedling stage) was identified from the genotype HT54, which contributed high temperature tolerance at 48°C especially during seedling and grain-filling stages (Wei et al. 2013). Likewise, two major QTLs for HT, namely qHTSF1.1 (R² = 12.6%) and qHTSF4.1 (R² = 17.6%), were detected on chromosome 1 and 4, respectively, in BC₁F₁ and F₂ progeny generated from the cross IR64 × N22 (Ye et al. 2012). Recently, a QTL study involving 90 introgression lines provided five QTLs explaining PVs in the range of 6.83–14.63% (Lei et al. 2013). An introgression line Y106 carrying two QTLs for HS tolerance (qHTS1-1 and qHTS3) was identified while transferring genes from the wild rice (O. rufipogon Griff.) (Lei et al. 2013).

Under terminal HS, various physiological and agronomic traits contributing to grain yield improvement or yield components directly or indirectly in wheat have been discussed using multienvironment analysis (Mason et al. 2011, Bennett et al. 2012, Paliwal et al. 2012, Bonneau et al. 2013). In wheat, evaluation of senescence-related traits revealed nine QTLs across different chromosomes (2A, 6A, 6B, 3A, 3B and 7A) (Vijayalakshmi et al. 2010). A heat susceptibility index (HSI) was constructed using parameters such as spike yield and temperature depression (TD) of spike, and subsequently, a total of 14 QTLs associated with HSI were discovered (Mason et al. 2011). Following sodium dodecyl sulphate sedimentation (SDSS) test, four QTLs for improved baking quality under HS conditions were found on wheat chromosomes 1B, 1D, 4A and 7A (Beecher et al. 2012). Three major QTLs associated with HT were mapped on chromosomes 2B, 7B and 7D in 148 RILs (NW1014 × HUW468), accounting up to 20% PV (Paliwal et al. 2012).

Considering grain-filling rate (GFR) as a pivotal trait governing grain yield in wheat, BSA was employed using SSR markers in an F₂ population of Ksu106 × Yecora Rojo (Barakat et al. 2012). As a result, 12 SSR markers were found to be in close association with GFR in wheat. Similarly, Kirigwi et al. (2007) mapped QTLs for GFR on 4A chromosome. Under HS, a major QTL experiencing 17% variation on yield was detected on chromosome 4A in wheat (Pinto et al. 2010). Given amenability of single nucleotide polymorphism (SNP) markers to automated genotyping assays, one candidate SNP marker was discovered recently in wheat that markedly distinguished heat-tolerant (K7903) and heat-sensitive (RAJ4014) cultivars (Garg et al. 2012). More recently, genome-wide as well as candidate-gene-based association mapping approaches using SNP and diversity array technology (DArT) were applied in chickpea to establish marker–trait associations for HT (Thudi et al. 2014).

Importantly, most desirable yield-relevant physiological traits and their respective chromosomal locations are discussed in wheat (Cossani and Reynolds 2012). In case of maize, 184 RFLP markers were employed in a RIL population, which led to the detection of five QTLs controlling pollen germination, and six QTLs for pollen tube growth under HS (Frova and Sari-Gorla 1994). Earlier, using the same mapping population, six QTLs were identified in maize that controlled cellular membrane stability under HS (Ottaviano et al. 1991). In sorghum, four QTLs (Stg1–4) related to stay green trait were mapped on three LGs, viz. A, D and J. Importantly, co-occurrence of QTLs, namely Stg1 and Stg2, with important photosynthetic genes and HSPs on LG A strongly supported the involvement of LG A in response to heat and drought stress (Xu et al. 2000). In case of adzuki bean, two QTLs, viz.HQTL1 and HQTL2, were identified for enhancing pollen viability under HS (Kaga et al. 2003, Vaughan et al. 2005). Similarly, applying SNP marker in a RIL population (CB27 × IT82E-18), Lucas et al. (2013) reported five important genomic regions that rendered HT in cowpea. In tomato, six QTLs were found to be linked with fruit set under HS (Grilli et al. 2007). Additionally, Xiang-yang et al. (2008) identified two QTLs conferring HT in tomato. Deploying amplified fragment length polymorphism (AFLP) and random amplified polymorphic DNA (RAPD) markers aided in the identification of five QTLs related to HT in Brassica campestris L. ssp. pekinensis (Shuancang et al. 2003). The above-mentioned QTLs identified using molecular markers in different crops provide a way to transfer the causative heat-tolerant gene(s)/QTLs to elite cultivars. In parallel, the fine mapping accompanied by cloning of candidate QTL will help the breeders to commence marker-assisted breeding for incorporating HT in various important crops in near future.

Markers assisted introgression of HT-relevant traits: A faster breeding strategy

Molecular markers enabling recovery of desirable genotypes in a precise and time-saving fashion become imperative while dealing with quantitative traits such as HT (Shirasawa et al. 2013). In rice, a marker-based approach, that is, advanced backcross, was used to develop introgression lines in the background of ‘Teqing’, and later, screening of these lines allowed detection of one heat-sensitive line ‘YIL106’ (Lei et al. 2013). Similarly, heat-tolerant (XN0437T) and heat-sensitive (XN0437S) introgression lines were also recovered from another backcross inbred population derived from the cross (Xieqingzao B × N22)  × Xieqingzao B (Jiang-lin et al. 2011). Near-isogenic lines (NILs) created by introducing qWB6-allele from ‘Hana-echizen’ into the background of ‘Niigata-wase’ showed considerable reduction in the incidence of heat-induced injuries such as white-back kernels (Kobayashi et al. 2013). In a recent study conducted in rice, a 1.5-Mb chromosomal region delimited with markers ktIndel001 and RFT1 was transferred from ‘Kokoromachi’ to ‘Tohoku 168’ using marker-assisted backcrossing. And the resultant NILs had grain quality significantly improved over the susceptible parent ‘Tohoku 168’. The genomic region involved above actually harboured a robust QTL that controlled better grain quality under HS, and governed more than 20% of the entire PV (Shirasawa et al. 2013). A non-exhaustive list of DNA markers related to various HT/component traits available in different crops is presented in Table 2. With such robust markers/candidate gene(s)/QTLs in place, marker-based selection in early generations accompanied by a strategic mating scheme would maximize the genetic gains while breeding for HT.

Table 2. Robust DNA markers to enable faster breeding for HT in various crops

Crop	Trait	Marker	References
Cowpea	Pod set	1_1346, 1_0437 (SNP)	Lucas et al. (2013)
	Seed coat	1_0032, 1_1128 and 1_0640 (SNP)	Pottorff et al. (2014)
Rice	Seedling stage HT	InDel5 and RM7364	Wei et al. (2013)
	Flowering stage and spikelet fertility	SNP17877584, SNP19381891 (SNP); RM570-RM148 (SSR), B1065E10	Ye et al. (2012), Chen et al. (2008), Jagadish et al. (2010a)
	Seed set percentage	M5687 and RM471, RM6132-RM6100 (SSR)	Xiao et al. (2011), Ye et al. (2012)
	Filled grains per panicle and grain yield	RM3586, RM3735 and RM160 (SSR)	Zhang et al. (2009a), Gui-lian et al. (2009), Buu et al. (2014)
	Days to heading	RM151, RM5172 and RM1369 (SSR)	Kobayashi et al. (2013)
	Kernel weight	RM7365 (SSR)	Kobayashi et al. (2013)
	White-back kernel	RM3288 and RM8125 (SSR)	Kobayashi et al. (2007), Kobayashi et al. (2013)
	Yield	RM5749 and RM337 (SSR)	Buu et al. (2014)
Tomato	Fruit set	fAFLP	Grilli et al. (2007)
Wheat	Grain-filling duration (GFD)	Gwm293, gwm11, Xcfd43 (SSR)	Yang et al. (2002), Mohammadi et al. (2008), Sadat et al. (2013)
	Grain-filling duration	Xwmc 407, Xbarc137 (SSR)	Mason et al. (2010, 2011), Sadat et al. (2013)
	Grain-filling rate	Xgwm 577	Barakat et al. (2011), Paliwal et al. (2012)
	Kernel weight	gwm291, Gwm268, WMC527 (SSR)	Mason et al. (2010, 2011), Sadat et al. (2013)
	CTv	wPt-3465, agg/cta-12, acc/ctc-8, aag/cta-7	Pinto et al. (2010)
	CTg	gwm388 (SSR)	Pinto et al. (2010)
	GM	barc065, act/cag-5, wPt-0021	Pinto et al. (2010)
	TGW	aag/cta-5	Pinto et al. (2010)
	75%G, 25%G	Xgwm356/CGT.TGCG-349	Vijayalakshmi et al. (2010)
	MRS	CAG.AGC-101/AGG.CTT-212 (AFLP)	Vijayalakshmi et al. (2010)
	PGMS	Xbarc198/CGT.CTCG-406	Vijayalakshmi et al. (2010)
	TMRS	Xgwm111/Xgwm437 (SSR); Xgwm356/CGT.TGCG-349	Vijayalakshmi et al. (2010)
	HSITGW, HSIGFD and CTD	Xgwm1025–Xgwm745, Xgwm935–Xgwm1273, Xgwm1025–Xgwm745 (SSR)	Paliwal et al. (2012)
	Canopy temperature, early vigour, flag leaf width, peduncle length	RAC875 allele	Bennett et al. (2012)
	Baking quality	barc170, wmc468, wmc707, barc119, gwm337 (SSR)	Beecher et al. (2012)
	Yield	act/cag-5, gwm397	Pinto et al. (2010)
	Grain yield and thousand-grain weight	gwm299 (SSR)	Bonneau et al. (2013)

• G, green; CTv, canopy temperature in the vegetative stage; CTg, canopy temperature in the grain-filling stage; TGW, thousand grain weight; GM, grain number; MRS, maximum rate of senescence; PGMS, percentage greenness at maximum senescence; TMRS, time to maximum rate of senescence.

Progress in Proteomics and Functional Genomics: Expression Profiles and Candidate Genes for HT

Emerging trends in functional genomics facilitate elucidation of the role of candidate genes, genome-wide expression analysis and HS response coupled with regulatory and tolerance mechanism for HS in plant (Sreenivasulu et al. 2007, Vij and Tyagi 2007, Urano et al. 2010).

Following proteomics analysis, a total of 54 proteins involved in carbohydrate metabolism, protein synthesis and stress responses during caryopsis were identified in rice under HS condition (Lin et al. 2005). By conducting a similar proteomic study, thermotolerance was noticed in the roots of Agrostis scabra (heat tolerant) at different temperature levels, and the study indicated an up-regulation of proteins in comparison with sensitive species of A. stolonifera (Xu and Huang 2009). Likewise, new insights were gained related to HSPs and regulatory proteins, energy metabolism and redox homeostasis by applying MALDI-TOF MS in rice (Lee et al. 2007). Leaf proteome analysis of wild rice O. meridionalis (under HS at 45°C) manifested differential response of heat-responsive genes and encouraged enzymes associated with Calvin cycle and thiamine biosynthesis (Scafaro et al. 2010). Thirteen differentially expressed protein spots identified by MASS/MALDI-TOF analysis under HS during anthesis shed new light on the mechanism of HT in rice (Jagadish et al. 2010a). Further, differential expressions of phosphoproteins under HS were also examined in rice leaves (Chen et al. 2011). Similarly in maize, MALDI-TOF mass spectrometry allowed detection of sHSP17.4, sHSP17.2 and sHSP26 proteins under combined drought and HS (Hu et al. 2010). MALDI-TOF/TOF analysis also facilitated identification of 81 differentially expressed proteins in alfalfa, thereby furnishing new clues about the molecular basis of HS response (Li et al. 2013c). By applying novel proteomic approach, that is, multidimensional protein identification technology (MudPIT), the role of stomatal proteins was examined and consequently, up-regulation and down-regulation of genes under HS condition in Agave americana was experimentally demonstrated (Shakeel et al. 2013). High levels of six leaf proteins and nine root proteins were obtained in transgenic A. stolonifera, harbouring isopentenyl transferase (ipt) gene that manages HT via regulating cytokinin synthesis (Xu et al. 2010).

Relating to gene expression profiling under high temperature, 22-K DNA microarray, differential hybridization and reverse transcription-PCR were undertaken to examine the regulation of different starch synthesis genes, viz. granule-bound starch synthase I (GBSSI) and branching enzymes (BEIIb) associated with grain filling in rice (Yamakawa et al. 2007). Microarray-based expression analysis of rice pollens that were exposed to high temperature revealed down-regulation of some important tapetum-specific genes under HS, however other genes like Osc6, OsRAFTIN and TDR genes remained unaffected during high temperature (Endo et al. 2009). In addition, by combining microarray and cDNA-AFLP techniques, Bita et al. (2011) demonstrated the transcriptomic response of meiotic anthers in heat-tolerant as well as in heat-sensitive tomato genotypes. As a result, this study provided a set of candidate genes conferring HS tolerance in tomato.

Transcriptome analysis using Affymetrix Grape genome oligonucleotide microarray elucidated stress-responsive genes, and the transcription factors involved in HS. Moreover, the role of HSPs, ascorbate peroxidase and galactinol synthase in imparting HT was also elucidated (Liu et al. 2012). Similarly, application of Affymetrix 22K Barley 1 Gene Chip microarray illustrated the expression patterns of 958 induced genes and 1122 repressed genes during caryopsis development in barley observed under early HS (Mangelsen et al. 2011). Further, this investigation also indicated that the embryo and endosperm are the prime locations for HS response (Mangelsen et al. 2011). Involvement of different HSPs with reactive oxygen species (ROS), hormones and sugars was tested in tomato by affymetrix tomato genome array and cDNA-AFLP-based transcriptome profiling (Frank et al. 2009). Likewise, genome-wide expression patterns of genes were analysed in sensitive and tolerant wheat lines using GeneChip wheat genome array, thereby enabling access to the HS-responsive genes (Qin et al. 2008). A similar transcriptome analysis in potato under high soil temperature uncovered a suite of stress-related genes that encode for HSPs in periderm (Ginzberg et al. 2011).

A comprehensive genome-wide translational analysis clearly depicted a decrease in translation process under HS in Arabidopsis (Yángüez et al. 2013). Given the role of non-coding RNAs in abiotic stress tolerance (Chinnusamy et al. 2007), response of chloroplast small RNA (csRNA) under HS and their prospects in engineering HT in Chinese cabbage were discussed (Wang et al. 2011). Recently, deep sequencing of RNA produced a novel class of conserved RNA and miRNA concerned to heat response in B. rapa, notably the bra-miR398a and bra-miR398b (heat inhibitive) and bra-miR156 h and bra-miR156 g (Yu et al. 2012). Similarly, several heat-responsive miRNA and genes were identified in various plants by applying transcriptome sequencing (Table 3). More recently, a comparative transcriptomic analysis investigated the role of 16 common genes in protein refolding process in switchgrass, rice, wheat and maize (Li et al. 2013b). Genome-wide transcriptional response of 10 ecotypes of Arabidopsis thaliana under HS was investigated by employing Arabidopsis Nimble Gen ATH6 microarrays (Barah et al. 2013). Further, in silico transcript regulatory network model in Arabidopsis presented 35 TFs showing ecotype-specific response to HS (Barah et al. 2013). Taken the above description into account, it is expected that the proteomics and functional genomics can greatly expedite the progress of discovery and functional characterization of the heat-tolerant genes/QTLs. A comprehensive approach involving multiple disciplines to combat HS is given in Fig. 1.

Table 3. Application of deep transcriptome sequencing for identification of heat-responsive gene(s) and regulatory RNA molecules in various plant species

Crop	Genotype	Tissue used for cDNA library construction	Platform used	Candidate genes/regulatory RNA	Reference
Apium graveolens L.	Ventura and Jinnan Shiqin	Leave	Illumina HiSeq 2000	celery-89505, celery-28253, celery-20717 and celery-18591 celery-10247 and celery-75186 genes	Li et al. (2014a)
Arabidopsis thaliana	-	Leave	Solexa	sRNA-encoding genes	Baev et al. (2013)
Arabidopsis thaliana	-	-	Illumina GAIIx sequencer	SR45a mRNA	Gulledge et al. (2012)
Brachypodium distachyon	-	Diverse array of tissue	Illumina	1088 heat-responsive genes	Priest et al. (2014)
Brassica rapa ssp. Chinensis	Wu11	Seedlings	Illumina Genome Analyzer	Heat-responsive csRNA	Wang et al. (2011)
Brassica rapa	Wu11	Above ground part of seedling	Illumina GAII sequencer	BracCSD1 and BracSPL2 genes	Yu et al. (2012)
B. rapa ssp. pekinensis and B. rapa ssp. chinensis	Bre and Wut	Seedlings	Illumina GAII	728 novel cis-natural antisense transcripts (cis-NATs) Heat cis-NATs-derived small interfering RNAs	Yu et al. (2013)
Cicer arietinum	ICC4958	Embryo, leaves, apical meristem, shoots, roots, buds, flowers and pods	Illumina's Genome Analyzer I	DNAJ heat-shock protein Heat-shock proteins (HSP 70, HSP 91)	Hiremath et al. (2011)
Gossypium hirsutum L. G. barbadense L.	Guazuncho 2 VH8-4602	Fibres	454 GSFLXTM Titanium	28 heat-shock protein-annotated genes	Lacape et al. (2012)
Menihot esculenta Crantz	TAI16	Root and leave	Illumina Genome analyser	134 target genes for conserved miRNA sequence and 1002 genes for non-conserved miRNA	Ballén-Taborda et al. (2013)
Oryza sativa	-	Shoot and root	Illumina	39 known miRNAs and 173 miRNAs	Mangrauthia et al. (2013)
Populus tomentosa	-		Solexa sequencing	52 miRNAs	Chen et al. (2012a)
Panex. ginseng C.A. Meyer	Shizhu	Leaf, stem, flower and root	Illumina sequencing	10 heat-responsive miRNA	Wu et al. (2012)
Rhazya stricta	-	Leave	Illumina HiSeq 2000	-	Yates et al. (2014)
Solanum lycopersicum	-	Root	Illumina GAIIX	3 class I heat-shock protein genes	Gupta et al. (2013)
Triticum aestivum L.	Chinese spring and TAM 107	Leave	Solexa	TaGAMYB1 and TaGAMYB2	Xin et al. (2010)
Triticum aestivum L.	Chinese spring and TAM 107	Seed	Solexa	Long npcRNAs (TahlnRNA12, TahlnRNA23 and TahlnRNA29)	Xin et al. (2011)
Triticum aestivum L.	PBW343	Seed	Illumina GAIIx	tae_7, tae_10, tae_15, tae_19, tae_22, tae_45	Pandey et al. (2014)
Triticum aestivum L.	HD2985 and NIAW	Seed	Illumina Hiseq 2000	HSF3, HSFA4a, HSP17, HSP70 and superoxide dismutase (SOD)	Goswami et al. (2014)
Youngia japonica	-	Leave	Illumina Genome Analyzer	T1 Unigene BMK.37824	Peng et al. (2014)

• csRNA, Chloroplast small RNA; npcRNAs, non-protein-coding RNA.

Figure 1

An integrated breeding strategy for improving heat tolerance in crops

Genetic Manipulation of HT Using Transgenic Technologies

With increasingly refined transformation and regeneration protocols, transgenic techniques are becoming attractive tool for designing both biotic and abiotic stress-tolerant crops via manipulating native genes or introducing gene (s) that lie beyond the crop gene pools (Ashraf et al. 2008). In the context of HT, genetic engineering has focused so far primarily on engineering genes that encode TFs, HSPs, chaperones, organic osmolytes, antioxidants and plant growth regulators (Ashraf 2010, Grover et al. 2013). The detailed list of important transgenes contributing to various abiotic stresses including HT is provided in PLANTSTRESS site (http://www.plantstress.com/biotech/index.asp?Flag=1).

Recently, Grover et al. (2013) reviewed genetic manipulation of HSF and HSP genes and the related change in expression levels under HS. Over-expression of HSFs is successfully demonstrated in tomato (Mishra et al. 2002, Giorno et al. 2010), Arabidopsis (Li et al. 2013b, Zhang et al. 2013), tobacco, pepper (Dang et al. 2013) and soybean (Chen et al. 2006, Zhu et al. 2006a). Similarly, over-expression of HSP and related genes was also reported in rice (Katiyar-Agarwal et al. 2003), maize (Nieto-Sotelo et al. 2002), carrot (Malik et al. 1999), Arabidopsis (Prändl et al. 1998, Panchuk et al. 2002, Rhoads et al. 2005, Jiang et al. 2009, Khurana et al. 2013, Li et al. 2013b, Zhang et al. 2013) and tobacco (Sanmiya et al. 2004, Sun et al. 2012). The regulatory role of DREB gene family [a class of Apetala 2 (AP2) transcription factor] in generating abiotic stress response in various crops under stressed conditions has been thoroughly reviewed (Sakuma et al. 2006, Qin et al. 2007, Shinozaki and Yamaguchi-Shinozaki 2007, Lata and Prasad 2011). Higher expression of OsDREB2B gene in rice (Matsukura et al. 2010) and greater response of ZmDREB2A gene in maize (Qin et al. 2007) may cause induction of heat-responsive genes providing adaptation and survival under HS. Similarly, Mizoi et al. (2013) have reported activation of GmDREB2A in soybean under HS. This gene family is known to be associated with HT in various plant species such as chrysanthemum (Hong et al. 2009) and Arabidopsis (Lim et al. 2007, Qin et al. 2007, Matsukura et al. 2010). Likewise, expression of CAP2 gene (another class of AP2 transcription factor) caused high germination efficiency in transgenic tobacco under HS (Shukla et al. 2009). Genetic engineering of membrane lipid causing increase in accumulation of saturated fatty acid offers high temperature tolerance in plants (Grover et al. 2013). This property has been well exploited by silencing chloroplast omega-3 fatty acid desaturase gene, which led to lowering in trienoic fatty acids in tobacco, thereby imparting adaptation to high temperature (Murakami et al. 2000). The omega-3 fatty acid desaturase gene was also found to be involved in high temperature tolerance in transgenic tomato, which had reduced activity of endoplasmic reticulum (Wang et al. 2010).

Concerning osmolyte compounds that contribute to HS tolerance, transformation of Arabidopsis harbouring cod gene from Arthrobacter globiformis resulted in substantial accumulation of glycinebetaine, leading to acclimation against high temperature (Alia et al. 1998). Biosynthesis of glycinebetaine offers high temperature tolerance by protecting PSII from inhibition owing to high ROS in transgenic tobacco (Yang et al. 2007); similarly, the presence of NADP (H) in leaf chloroplast pacifies the ROS activity under HS in transgenic tobacco (Wang et al. 2006). Over-expression of maize acetylcholine esterase AChE gene in transgenic tobacco evidenced its role in HS tolerance (Yamamoto et al. 2011, Yamamoto and Momonoki 2012). In cotton, engineering of AtSAP5-encoding proteins having A20/AN1 zinc finger domains prohibited damage in photosystem (PS) II complex under HS (Hozain et al. 2012). The involvement of isoprene in providing HT was documented in transgenic Arabidopsis expressing isoprene synthase gene from Populus alba (Sasaki et al. 2007) and was also very well characterized in transgenic tobacco (Vickers et al. 2009). Enhanced HT was also credited to increased expression levels of cytokinin oxidase/dehydrogenase (CKX) gene in transgenic tobacco (Macková et al. 2013) and sedoheptulose-1, 7-bisphosphatase (SBPase) gene in transgenic rice (Feng et al. 2007).

Taken the role of antioxidants into account, HvAPX1 gene was introduced into Arabidopsis from barley, which later produced ascorbate, thus imparting HT (Shi et al. 2001). The loss-of-function mutants in O. sativa glycogen synthase kinase3-like gene 1 (OsGSK1) gene created by T-DNA insertion conferred HT in rice (Koh et al. 2007). The cross-species gene transfer offering HS tolerance is well exemplified in tobacco by transferring DnaK1 gene from a halo-tolerant cyanobacterium (Aphanothece halophytica) (Ono et al. 2001). Likewise, transfer of codA gene from Arthrobacter globiformis to tomato evinced tolerance to HS during germination by enhancing expression of several proteins including the mitochondrial small heat-shock protein (MT-sHSP), heat-shock protein 70 (HSP70) and heat-shock cognate 70 (HSC70) (Li et al. 2011). Expression of rice gene ZFP177 and A20/AN1-type zinc finger gene provided HT in tobacco (Huang et al. 2008). Recently, over-expression of chloroplast-targeted DnaJ protein (LeCDJ1) gene also imparted HT in transgenic tomato (Kong et al. 2014). To our knowledge, the major research on transferring the HT genes is invested so far in the model plants such as Arabidopsis, tobacco and rice. Therefore, the next priority should be to extend these technologies to the other agriculturally relevant field crops.

Prospective role of precise and high-throughput phonemics facilities

The continued progress in the new-generation high-throughput DNA sequencing technologies permits generation of enormous sequence information at acceptable prices (Feuillet et al. 2010, Xu et al. 2012, Bevan and Uauy 2013). However, large-scale phenotypic evaluation of crop plants, that is, precise, accurate and high-throughput phenotyping of the traits, remains strenuous (Cobb et al. 2013), in particular the phenotypic assessment of yield/component traits and abiotic stress-tolerance-related traits, which are under the control of complex network involving QTLs, their epistatic interactions and strong G × E interactions (Houle et al. 2010, Sozzani and Benfey 2011, White et al. 2012, Cobb et al. 2013). Taking into account the environment as such (as described by Xu et al. 2012), the accurate assessment of the magnitude of G × E interactions by E-typing will transform the selection procedures from being mere technical (phenotype based) to much more scientific, which would consider all three dimensions (genotype, phenotype and environment), that is, a point-to-space transformation. In this way, by effectively controlling the environmental factors and errors, plant phenomics holds the opportunity to understand the gene function, their response to external environment and subsequently, bridging the gap between genotype and phenotype by facilitating phenotyping of important quantitative traits (Furbank 2009, Furbank and Tester 2010, Sozzani and Benfey 2011, Chen et al. 2014). Since the establishment of Australian Plant Phenomics Facility (APPF) in Australia (Finkel 2009), new avenues are being provided to overcome the phenotyping bottlenecks.

To facilitate more realistic evaluation of the plant responses to environment, field-based phenotyping (FBP) approaches are initiated (White et al. 2012). Some of the remarkable large-scale phenotyping techniques applied recently for screening against abiotic stress are infrared thermography that efficiently captures the genetic variability for stomatal response against water deficit in wheat and barley (Munns et al. 2010) and stomatal behaviour and canopy temperature in potato (Prashar et al. 2013). Infrared thermography, an effective non-invasive tool, has been deployed for phenotyping of HT in wheat and chrysanthemum (Amani et al. 1996, Janke et al. 2013, Prashar and Jones 2014). Additionally, higher value of chlorophyll fluorescence parameter F_v/F_m was observed in wheat, which reflected maximum efficiency of PSII higher photosynthesis rate, higher stomatal conductance and transpiration rate under HS (Sharma et al. 2014a). Further, the above parameter was also used in assessment of genetic variability for HS tolerance in wheat (Sharma et al. 2012, 2014a). Other strategies such as measuring canopy temperature were also implemented for identifying high-yielding HS-tolerant wheat lines with lower canopy temperature (Pinto et al. 2010, Mason and Singh 2014). More recently, high-resolution thermal imaging system was used to precisely measure the leaf temperature (Jones and Sirault 2014). In recent years, various platforms were established for phenotyping, including ‘HTpheno’ for image analysis (Hartmann et al. 2011), ‘PHENOPSIS DB’ to perform image analyses under different environmental conditions in Arabidopsis (Fabre et al. 2011), ‘Gia Roots’ for root architecture analysis (Galkovskyi et al. 2012) and ‘Phenoscope’ for spatial homogeneity (Tisne et al. 2013). Rootscope represents another phenotyping system that is recently used to quantify heat-shock responses in plants (Kast et al. 2013). Excellent reviews were published in recent years describing applications, benefits and challenges of the new-generation phenotyping systems (Sozzani and Benfey 2011, Cobb et al. 2013). In the coming future, the new-generation phenomics platforms would allow cost-effective and user-friendly yet phenome-level screening for HT in crop plants.

Conclusion and prospects

Given the rising demands for global food supply coupled with the severe pressure of population growth and the climate change trajectories, strategies should mainly aim at proper exploration of germplasm and harnessing of novel alleles from wild gene pool. Moreover, the adaptive as well as morpho-physiological traits could be incorporated in high-yielding genotype through physiological trait breeding. In parallel, paradigm shift in molecular marker systems coupled with growing availability of the whole-genome sequences and rapid progress in functional genomics can expedite the discovery of candidate gene conferring HT and their genome-scale expression profiling. Moreover, transgenic approaches hold great promise for transferring HS-tolerant genes across the species. In addition, the hurdle of phenotyping of this complex trait can be overcome by testing genotypes harbouring HT under both stress and non-stressed condition along with multilocation testing. Phenomics platforms raise new hope for difficult-to-measure component traits associated with HT. Additionally, ‘omics’ approaches combined with system biology approach could greatly strengthen the conventional breeding to mitigate the challenges of HS and to streamline the future of sustainable agriculture (Ahuja et al. 2010, Cramer et al. 2011).

Acknowledgements

Authors acknowledge support from Indian Council of Agricultural Research (ICAR), India.

Conflict of interest

The authors declare that there is no conflict of interest.