QTL study of chlorophyll content as a g

QTL study of chlorophyll content as a genetic parameter of drought tolerance in barley



D. This, C. Borries, I. Souyris & B. Teulat

UMR 1096, plant breeding department, ENSAM-INRA, 2 place P. Viala 34060 Montpellier cedex. France.
Correspondence: Dominique This. Fax: 33 467045415;

email: thisd@ensam.inra.fr



Barley is considered as relatively adapted to drought conditions in the Mediterranean basin although unpredictable environment may lead to dramatic decreases of yield in this area. Barley genotypes, in particular landraces and wild species, represents a large source of variation for specific morpho-physiological parameters that may contribute to yield and yield stability under drought conditions (Forster et al. 99), and that could be introgressed into improved varieties. An extensive study of QTL controlling the genetic variation of such traits may provide new breeding methodologies with molecular markers assisting selection. In our research group, most effort has been given to the study of the genetic control of osmotic adjustment parameters variation, using a Recombinant Inbred Line progeny derived from the cross Tadmor x ER/APM (Teulat et al., 1998; This and Teulat-Merah, 1999). However, this analysis has to be combined with the evaluation of other determinants of plant behaviour under different environmental conditions, such as growth parameters (Teulat et al. 1997), but also photosynthesis or light interception efficiency.

Some specific traits may lead to the variation of photosynthesis efficiency under drought condition: The pale leaf colour observed at the end of the growth cycle in barley landraces from the Middle East and well adapted to harsh Mediterranean conditions such as Tadmor (tolerant parent of the studied progeny), seemed related to their drought tolerance capacity (van Oosterom and Acevedo 1992). Watanabe et al. (1995) attributed this colour to a decrease of chlorophyll content per unit area compared to dark-green genotypes. This could result from a passive adaptation to high solar radiation reducing the light absorption, leading to a specific adaptation to drought conditions in the Mediterranean area (Tardy et al. 1998). Thus, the measurement of the total chlorophyll content could be considered as a simple quantitative trait related to drought tolerance in barley crosses from the Middle East, related to the resistance of photosystem II to high temperatures, which is an important factor of drought stress in this environment.

Quantitative trait loci (QTL) controlling parameters related to drought tolerance variation in barley were identified, using a random block incomplete design (9 blocks) conducted in a growth chamber at the Genetic and Plant Breeding laboratory of ENSA-INRA-Montpellier (France), at an early stage of growth. The experiment was described and detailed in Teulat et al. (1998). The water stress was imposed at the 4-leaf stage by stopping the irrigation for the first set of plants while the other set was maintained well watered. After 12 days, the relative soil moisture content was 14% of the field capacity (FC) for the stressed-plants and 100% FC for the irrigated plants (pots were weighed and watered daily). Total chlorophyll content (CC) was measured at 100% and 14% of FC together with other water status parameters on 105 recombinant inbred lines (RILs) in each block, on the last fully expanded leaf, with a portable meter (Minolta Camera, SPAD-502). QTL have been identified with adjusted means (block and pot within block effects fixed) and strengthened by examination of the QTL positions from each block separately (data not shown).

For all the data sets, phenotypes of a subset of 167 RILs were analyzed using 118 molecular markers. The map was presented in This and Teulat-Merah (1999). We obtained 15 linkage groups assigned to the seven barley chromosomes. Without considering the main gaps, the new map represented a minimum of 1101 cM. The QTL data were analyzed by interval mapping method with MAPMAKER/QTL version 1 (Lander and Bostein 1989) and verified by single-factor analysis of variance with QGeneTM version 2.30 (Nelson 1997). The putative QTL were declared significant when the LODscore $ 2.0 and at a P<0.005.

A large variation was observed from the phenotypic data for chlorophyll content (CC) in the two water treatments. High significant block and pot within block effects have been detected for all the traits measured including CC (Teulat et al. 1998). Indeed, a sort of chronological decrease between the mean values from each of the 9 blocks was noticed for CC (Teulat 1997, phD thesis). Thus, the experiment showed that even in controlled conditions, testing a great number of genotypes is difficult without excluding environmental and technical effects. However, significant RIL effects were detected for CC at 14% FC and at 100% FC. The experimental design and the adjustment of the data by fixing the block and pot within block effects (Teulat et al. 1998) allowed to reduce or limit some part of the environmental effects. With the adjusted means, the large variation observed from phenotypic means was found again and an important water treatment effect was also obtained for CC (P<0.0001).

When detecting QTL for total chlorophyll content (CC) with adjusted data, on either the whole experiment, or on the four first blocks or the five last blocks separately, the same chromosomal regions were generally detected as with phenotypic means but with a higher significance. Results with adjusted data are presented in table 1. Some QTL regions are detected specifically with the water stress treatment: on BCD1066 and 4 cM from WG380 (chromosome 7H), 13 cM from CDO669 (chromosome 4H), and on BCD348B marker (chromosome 6H). The chromosome 2H region between BCD1069 and CDO366 and a chromosome 4H region 4 to12 cM from SB734A are involved in CC variation in both water-stressed and irrigated conditions. On the contrary a QTL was detected 22 cM from dhn1 on chromosome 5H specifically in irrigated conditions. For all QTL detected, Tadmor provided negative effects on CC values, in accordance with Watanabe results (1998) on this landrace.

Analyzing separately 4 or 5 blocks allowed most of the time the detection of different QTLs, indicating a variation in the genetic control of CC during the experiment. However, some QTL regions have been also detected in field conditions, using adjusted means of three years measurements, at anthesis, for the same population (detailed experimental field conditions will be presented elsewhere, QTL results are also presented in table 1). For example chromosome 7H region near BCD1066 and QTL regions detected on chromosome 6H and chromosome 5H have been already detected when considering the adjusted mean of three years experiment conducted in Montpellier95, Mauguio97 (South of France) and Granada96 (Spain). A QTL was also found in field conditions on chromosome 2H near CDO64, however this regions is quite genetically distant from the previous one identified. Two QTL have been also detected by Baum et al (1997) in field conditions for a barley population involving Tadmor as a parental line on chromosome 2H near MWG882 and on chromosome 5H near MWG533. Those markers are not mapped in our population, but the consistency of those QTL regions may be increased by this comparison.

In addition, on chromosomes 7HS, 2H and 4HL, QTL found for CC genetic variation were in regions where QTL controlling several other traits related to water-stress response were revealed (This and Teulat 99, Teulat et al. in preparation). Such QTL clusters should be analyzed more in detail to distinguish pleiotropic effects from a clustering of genes potentially involved in water stress adaptive response.

The detection of QTL for chlorophyll content associated with the detection of candidate genes within them could help to understand the resistance to high temperature. The potential of chlorophyll content in term of drought tolerance character will have to be however considered with care, since this may represents a very specific adaptation to some geographic conditions but may be antagonist in other conditions less submitted to heat stress with photosynthesis efficiency. In a marker assisted selection, one should identify clearly the environmental target and no unique response can certainly be given world-wide for barley breeding.



Table 1. QTL detected for chlorophyll content by interval mapping, from adjusted of the total experiment, *the 4-first blocks, **the 5-last blocks for 14% FC (water stress) and 100% FC (irrigated) and from adjusted means of three field experiments (Montpellier 1995, Granada 1996 and Mauguio 1997). %var: individual variance explained by the QTL.

Treatment Chromosome Left marker (Distance to peak LOD in cM) Magnitude of peak LOD Estimated additive effects# %var
Water stress treatment

(controlled conditions)



7H

7H

2H

4H

4H

6H

BCD1066 (0) and**

WG380 (4)*

McaaEaccD (18) and *

CDO669 (13) and *

SB734A (12) and *

BCD348B (0)**

2.10

2.27

2.21

2.77

2.83

3.79

-0.98

-2.48

-2.00

-1.29

-1.28

-1.68

5.7

15.1

8.3

9.7

9.6

18.7

Irrigated treatment

(controlled conditions)



2H

4H

5H

McaaEaccD (0)

SB734A (4) and **

dhn1** (22)

3.13

4.25

2.31

-1.47

-1.89

-3.00

8.3

13.6

16.8

Montpellier 1995, Granada 1996 and Mauguio 1997

(field conditions)

2H

6H

5H

CDO64 (0)

McaaEaccC (0)

WG908 (4)

3.15

2.11

2.45

-1.40

-1.28

-1.35

8.4

7

7.8

#The estimated genetic effect is expressed in the SPAD unit and correspond to the contribution of Tadmor's alleles compare to Er/Apm.



References

Baum M., Sayed H., Araus J.L., Grando S., Ceccarelli S. (1996) QTL analysis of agronomic important characters for dryland in barley by using molecular markers. In Proceedings of the V International Oat Conference and the VII International Barley Symposium. 1, pp. 241-243. Saskatoon, Saskatchewan, Canada.

Forster, B.P., Ellis, R.P., Thomas, W.T.B., Newton, A.C., Robinson, D., Tuberosa, R., This, D., El Gamal, A.S., Bahri, M.H., Ben Salem, M. 1999. Molecular markers for abiotic stress tolerance in barley. Journal of Experimental Botany (in press).

Lander E.S., Botstein D. (1989) Mapping Mendelian factors underlying quantitative traits using RFLP linkage maps. Genetics 121, 185-199.

Nelson J.C (1997) QGENE: software for marker-based genomic analysis and breeding. Molecular breeding 3: 239-245.

Tardy F., Créach A., Havaux M. (1998). Photosynthetic pigment concentration, organization and interconversions in a pale green Syrian landrace of barley (Hordeum vulgare L., Tadmor) adapted to harsh climatic conditions. Plant, Cell and Environment 21, 479-489.

Teulat B., Monneveux P., Wery J., Borries C., Souyris I., Charrier A. & This D. (1997a) Relationships between relative water content and growth parameters under water stress in barley: a QTL study. The New Phytologist 137, 99-107.

Teulat B., This D., Khairallah M., Borries C., Ragot C., Sourdille P., Leroy P., Monneveux P., Charrier A. (1998) Several QTLs involved in osmotic adjustment trait variation in barley (Hordeum vulgare L.). Theoretical and Applied Genetics 96, 688-698.

This D., Teulat-Merah B. (1999) Towards a comparative genomics of drought tolerance in cereals: lessons from a QTL analysis in barley. International workshop-CIMMYT Mexico june 21-25 1999 (http://www.cimmyt.mx/Research/ABC/).

Van Oosterom E.J., Acevedo E. (1992) Adaptation of barley (Hordeum vulgare L.) to harsh Mediterranean environments. Euphytica 62, 1-14.

Watanabe N., Naruse J., Austin R.B., Morgan C.L. (1995) Variation in thylakoid proteins and photosynthesis in Syrian landraces of barley. Euphytica 82, 213-220.