Previous PageTable Of ContentsNext Page

Increased stability of kernel weight under drought through selection of a reduced-tillering gene in wheat

Jaquie Mitchell1,2, Scott Chapman2, Greg Rebetzke3 and Shu Fukai1

1 The University of Queensland, School of Land, Crop and Food Sciences, Brisbane 4072, Qld, Australia.
Email jaquie.mitchell@uq.edu.au
2
CSIRO Plant Industry, QBP, 306 Carmody Rd, St Lucia 4067 Brisbane, Qld, Australia.
3
CSIRO Plant Industry, PO. Box 1600 Canberra, ACT, 2601, Australia.

Abstract

Australian wheat production environments are typically water-limited and characterised by terminal drought. Post-anthesis drought contributes to a high incidence of small or shriveled wheat kernels (screenings) that reduce crop value. It was hypothesised that the incorporation of the tiller inhibition (tin) gene into wheat germplasm may contribute to the maintenance of large kernel weight (KW) and reduction in screenings (SCR), particularly in terminal water stress environments. One irrigated and six dryland field experiments (2005 & 2006) and two rainout shelter experiments with an irrigated control (2007) were utilized to investigate the value of the tin gene in maintenance of KW under variable water supply in Queensland. Twenty two (2005 & 2006) and six (2007) Silverstar near-isogenic lines (NILs) contrasting for reduced tillering (tin1) and wild type (Tin1) alleles were utilised. Averaged across experiments, the KW of tin lines was 11% greater than free-tillering lines. Reduced-tillering lines produced up to 53% fewer SCR than free-tillering lines. The KW advantage was largely associated with reduced spike number/m2 (r2=0.27 to 0.79), with kernel number per spike being less important. On average grain yield (GY), which ranged from 2.0-6.1 t/ha, tended to be lower (-9%) in tin lines, particularly in high-yielding environments with high spike density. However, at moderate spike density (≈300 spike/m2) in all environments, particular tin lines could be identified which achieved high KW, low SCR and relatively high GY. There exists potential for selection among tin progeny to achieve simultaneous GY and KW improvement and reduction in SCR for terminal stress environments. Consequently, considerable increases in crop value (eg $45-313/ha) can be achieved with the use of tin lines in terminal stress environments.

Keywords

grain weight, tin gene, water stress, dryland agriculture, northern region, cereal

Introduction

It is well documented that increased yield of modern wheat varieties, as with other cereals, has largely been achieved through selection for increased kernel number per unit area ( Fischer 2008; Sayre et al. 1997). In contrast, KW is considered a relatively stable yield component across years and locations with small genotypic variation (Fischer and Hillerislambers 1978). However, following many years of selection, especially under higher input conditions, there has been a tendency for modern wheat varieties to produce a substantial proportion of low weight kernels (Acreche and Slafer 2006) even under non water-limiting conditions. Furthermore, when lines are exposed to a water deficit, particularly a post-anthesis terminal water stress, as often experienced in the northern and western growing regions of the Australian wheat belt, small or ‘pinched and shriveled’ kernels result. Growers wish to minimize screenings, as the value of the crop is reduced when the percentage of screenings is above 5%.

Selection for the reduced-tillering tin gene has been shown to result in larger stems, higher harvest index and increased kernel weight (Duggan et al. 2005; Richards 1988) with minimal reduction in GY in southern environments. It may be in the northern region, where the target plant density is relatively low and crops are reliant on stored soil moisture with terminal water stress being frequent (Chenu et al. 2009), that the greatest benefit of tin germplasm may exist. Thus, a series of field experiments were conducted in the northern region to evaluate the performance of reduced and free-tillering Silverstar sister lines. It was hypothesised that the incorporation of the tin gene into wheat germplasm may contribute to the maintenance of KW and reduction in SCR.

Methods

Near-isogenic lines (NILs) varying in tiller number were developed by twice back-crossing the tin gene into the free-tillering, high-screenings variety, Silverstar (Rebetzke pers.comm.). Lines containing the tin1 allele, as identified by molecular marker Xgwm136 (Spielmeyer and Richards 2004), were classed as reduced tillering (tin), whereas those with the alternative wild-type allele were free-tillering (W). Seven field experiments were conducted at Gatton (-27.55,152.33), Kingsthorpe (-27.47,151.83) and Emerald (-23.47,148.15), QLD in 2005 and 2006. Each experiment evaluated 22 Silverstar NILs (11 Silverstar tin and 11 W lines) for which variation in height and days required to reach anthesis date were minimized, to reduce the confounding effect of these two major traits. A subset of 6 lines was used in 2007 Rainout shelter experiments. All plots were machine planted and harvested (with trimmed ends and plot lengths measured) averaging 6 m long and 7 rows wide with a row spacing of 0.25 m. Rainout shelter experiments were manually harvested at maturity. Quadrats (0.125-0.5 m2) for biomass and tiller counts were taken at various developmental stages throughout the season. Anthesis dates, grain yield and yield components, and grain screenings based on industry standards (40 shakes on Agtator with 2mm slotted sieve) were determined. For each line, kernel weight was determined for a random sample of 300 grain.

Established plant density ranged from 74 to 157 plants / m2. Apart from 2005 experiments which were augmented design all other experiments were randomised complete block designs with two (2006) and three (2007) replications. Sown into dry soil the Gatton and Kingsthorpe experiments were irrigated (41-63 mm) to ensure good plant establishment. The Gatton irrigated trial received an additional 54mm irrigation between rainfall events (34mm) pre-anthesis, with 107 mm rainfall occurring post-anthesis.

Results

Head Number

The incorporation of the tin gene reduced head number per unit area in Silverstar lines, in all experiments (Table 1). On average, Silverstar free-tillering lines produced between 305 and 542 heads/ m2 depending on experiment; in contrast tin lines produced between 197 and 385 heads/ m2. Therefore, Silverstar tin lines produced between 23% to 50% less heads per unit area than free-tillering lines.

Table 1: Experiment code, year, locations, post-anthesis stress type and head number per unit area of free-tillering (W) and reduced (tin) tillering Silverstar lines grown in 10 Queensland field experiments between 2005-2007.

Post-anthesis

Head number per m2

Experiment code

Year

Location

Stress

W

tin

Reduction due to tin(%)

05GAT

2005

Gatton

None

337 24.8

258 18.7

23

06GATI

2006

Gatton

None

477 15.0

261 21.1

45

06GATS

2006

Gatton

None

521 18.7

383 20.2

27

07C

2007

Gatton

None

542 35.1

385 39.2

29

06EMN

2006

Emerald

mild

363 10.9

252 17.4

30

06EMW

2006

Emerald(WR)

mild

305 12.0

197 12.6

35

07MS

2007

Gatton

mild

485 24.3

315 19.5

35

05KIN

2005

Kingsthorpe

Severe

401 21.5

281 21.8

30

06KIN

2006

Kingsthorpe

Severe

463 13.5

230 28.8

50

07SS

2007

Gatton

Severe

384 23.0

285 24.5

26

Mean all

     

428

285

33

Mean stress experiments

400

260

35

WR indicates wide row spacing (0.5m)

Kernel weight, screenings and grain yield

In the experiments that suffered post-anthesis water stress, the KW advantage of tin lines ranged from 3 to 36% (Table 2). The larger kernel weight of tin lines translated to between 13 and 53% reduction in screenings produced compared with free-tillering lines. A negative genotypic relationship exists between KW and SCR (Mitchell et al. 2008; Sharma and Anderson 2004). Under mild stress conditions there was no difference in average GY of tin and free-tillering Silverstar lines; an observation also made in southern and western regions (Duggan et al. 2005; Whan 1988). However, under severe stress conditions in Kingsthorpe the GY reduction due to tin lines was 23 and 26%. This was due in part to low head number of tin lines relative to free-tillering lines. This GY penalty was not observed in 2007 Gatton severe terminal stress trial where tin lines achieved head number around 300 heads/m2.

Table 2: Kernel weight (mg), screenings (%) and grain yield (t/ha) of free-tillering (W) and reduced (tin) tillering Silverstar lines grown in 10 Queensland field experiments between 2005-2007.

Kernel weight (mg)

Screenings (%)

Grain yield (t/ha)

Experiment

W

tin

Advantage of tin (%)

W

tin

Advantage of tin (%)

W

tin

Advantage of tin (%)

05GAT

31.0 2.3

30.9 2.5

-3

7.4

8.3

12

3.6 0.6

3.7 0.7

3

06GATI

30.9 1.1

33.8 1.1

10

4.1

4.3

6

6.0 0.7

4.9 0.7

-18

06GATS

29.4 1.0

32.1 1.0

9

6.7

5.1

-24

5.2 0.3

4.6 0.3

-11

07C

31.3 0.5

31.6 0.7

1

3.6

5.7

58

6.1 0.2

5.0 0.2

-18

06EMN

22.8 1.5

23.6 1.4

3

21.7

18.9

-13

2.6 0.2

2.7 0.2

3

06EMW

22.9 1.0

25.9 1.0

13

21.2

17.7

-17

2.1 0.1

2.2 0.1

4

07MS

23.5 0.4

26.4 0.8

12

13.0

8.9

-32

4.1 0.2

4.1 0.2

1

05KIN

18.3 0.9

20.9 0.9

14

16.4

8.9

-46

2.7 0.2

2.0 0.2

-26

06KIN

20.4 1.5

27.6 1.5

36

24.9

11.8

-53

2.7 0.3

2.0 0.3

-23

07SS

21.8 0.5

24.9 1.0

14

14.5

10.6

-27

2.9 0.3

2.7 0.3

-6

                   

Mean all

25.3

27.8

11

13.4

10

-14

3.8

3.4

-9

Mean stress

21.6

24.9

15

18.6

12.8

-31

2.8

2.6

-8

Averaged across the experiments, a negative association (r2=0.41) existed between mean KW and GY, with tin lines tending to maintain higher KW and lower GY compared to free-tillering lines. However, a number of tin lines consistently performed well and maintained a high KW with relatively high GY.

Crop value – cost/benefit analysis

The estimated crop value was determined for the top yielding line from each group (tin and free-tillering) in the 2007 control, mild and severe stress experiments. These experiments were utilised for this purpose because terminal stress developed which resulted in the production of SCR, and GY achieved by tin lines was comparable to free-tillering lines with relatively high established head number (≈300 heads/m2). Crop value was determined using the Base Rates from the AWB Golden Rewards Pool pricing (www.awb.com.au/growers) and the increments /decrements based on SCR, protein and moisture percent (Table 3). Base Rates were moderately high ($448.35/t) from 2007/2008 and relatively low ($216.50/t) from 2004/2005 seasons.

In the high yielding favourable conditions (2007 control), lower SCR (3 vs. 9%) and the 1.0 t/ha yield advantage of free-tillering lines results in a $583 and $332/ha advantage in crop value at high and low base rates respectively. In contrast, under mild stress conditions when there was essentially no GY difference, and SCR generated by free-tillering lines result in a downgrading from Australian Prime Hard (APH) to Australian Utility Hard (AUH), the tin lines produced an $ 87 and $45/ha advantage in crop value at high and low base rates respectively. This crop value advantage of tin lines became even greater, up to $313 and $126/ha, when under severe stress conditions where free-tillering lines are downgraded to feed quality as a result of >15% SCR. When GY potential for a given environment was not realized eg tin lines in 2005 and 2006 Kingsthorpe experiments, due to low head number (257 and 192 heads/m2), free-tillering lines achieved higher crop values (data not shown). In the 2005 and 2006 Kingsthorpe experiments the price reduction due to SCR was not offset by the yield advantage of free-tillering lines. Generally, a GY advantage and therefore economic advantage was also observed for free-tillering lines in favourable environments.

Conclusion

In the Silverstar material there appears to be a distinct advantage of tin in maintenance of KW and subsequent reduction in SCR in terminal stress environments, such as those often experienced in the northern and western Australian production environments. There was evidence to suggest that tin was associated with a GY penalty when less than ≈300 heads per m2 was achieved. However, in all environments tested particular tin lines were found to achieve high KW with the maintenance of high GY. From the results of the simple cost/benefit analysis it would appear that the use of tin lines in the northern region, where terminal water stress is frequently encountered, could be of considerable value to the farmer, provided that similar GY to that of free-tillering lines can be achieved.

Table 3: Grain yield; grain protein; and screenings (and based on these); pool pay grade with base rate; and increment adjustments (based on protein, screenings and moisture); Providing estimated pool return per tonne; and overall crop value, for the highest yielding tin and W Silverstar lines in 2007 control, mild and severe terminal stress experiments.

Experiment

Control

Mild stress

Severe stress

Group

tin

W

tin

W

tin

W

Line

SsrT65

SsrW35

SsrT65

SsrW35

SsrT17

SsrW35

Grain yield (t/ha)

5.7

6.7

4.9

5

4.1

4.4

Grain protein (%)

11.4

11.5

13

12.5

13.4

14

Screenings (%)

9

3

8

14

10

15

             

Pool Pay Grade#

APH

APH

APH

AUH

APH

FEED

Base rate (2007/2008 Season)

448.35

448.35

448.35

432.85

448.35

352.4

+/- Protein, Moisture & Screening Increment

-24.5

-9.5

-6.5

-19

-6

0

Estimated Pool Return ($/t)

423.85

438.85

441.85

413.85

442.35

352.4

Crop Value ($/ha)

2711

3293

2400

2313

2041

1729

             

Base rate (2004/2005 Season)

216.5

216.5

216.5

204

216.5

170

+/- Protein, Moisture & Screening Increment

-19

-4

-6.5

-8

-8.5

0

Estimated Pool Return ($/t)

197.5

212.5

210

196

208

170

Crop Value ($/ha)

1263

1595

1141

1095

960

834

# APH, Australian Prime Hard; AUH, Australian Utility Hard; FEED, feed quality

References

Acreche MM and Slafer GA 2006. Grain weight response to increases in number of grains in wheat in a mediterranean area. Field Crops Research 98, 52-9.

Chenu K, Hammer GL, Dreccer F and Chapman SC 2009. Environment characterisation as an aid to wheat improvement in north east Australia. 14th Australasian Plant Breeding & 11th SABRAO Conference, Cairns, Australia, 10-13 August 2009.

Duggan BL, Richard, RA, van Herwaarden AF and Fettell NA 2005. Agronomic evaluation of a tiller inhibition gene (tin) in wheat. I. Effect on yield, yield components, and grain protein. Australian Journal of Agricultural Research 56, 169-78.

Fischer RA 2008. The importance of grain or kernel number in wheat: A reply to sinclair and jamieson', Field Crops Research 105, 15-21.

Fischer RA and Hillerislambers D 1978. Effect of environment and cultivar on source limitation to grain weight in wheat', Australian Journal of Agricultural Research 29, 443-58.

Mitchell JH, Chapman S, Rebetzke GJ and Fukai S 2008. Increasing grain size and reducing screenings in wheat using a tiller inhibition gene - investigating grain morphology by image analysis. In Global Issues: Paddock Action. Proceedings of the 14th ASA Conference, Adelaide, South Australia, 21-25 September 2008.

Richard, RA 1988. A tiller inhibitor gene in wheat and its effect on plant-growth. Australian Journal of Agricultural Research 39, 749-57.

Sayre KD, Rajaram S and Fischer RA 1997. Yield potential progress in short bread wheats in northwest mexico', Crop Science 37, 36-42.

Sharma DL and Anderson WK 2004. Small grain screenings in wheat: Interactions of cultivars with season, site, and management practices. Australian Journal of Agricultural Research 55, 797-809.

Spielmeyer W and Richards RA 2004. Comparative mapping of wheat chromosome 1as which contains the tiller inhibition gene (tin) with rice chromosome 5s. Theoretical and Applied Genetics 109, 1303-10.

Whan BR, Delane R. and Gilmour R 1988. The potential of reduced tillering wheats in dry environments. Seventh International Wheat Genetics Symposium, Cambridge, England, 13-19 July.

Previous PageTop Of PageNext Page