SHIFTING TRENDS OF BREAD WHEAT BREEDING IN ITALY ASSESSED BY THE STUDY OF A TEMPORAL-BASED CORE COLLECTION

Patrizia Vaccino1*, Leonardo Ormoli1, Corrado Costa2, Stefano Negri1, Maurizio Perenzin1

Consiglio per la Ricerca e la sperimentazione in Agricoltura –1CRA-SCV, via R. Forlani 3, 26866 Sant’Angelo Lodigiano (LO); 2CRA-ING, via della Pascolare, 16, 00015 Monterotondo Scalo (RM) – Italy;

E-mail: patrizia.vaccino@entecra.it

ACKNOWLEDGEMENTS The research was partly sponsored by the Italian Ministry of Agriculture in the framework of the Project RGV-FAO (DM 13763, 24/06/11; DM 1123, 12/10/12).

INTRODUCTION Germplasm collections represent valuable genetic resources for breeding programmes which aim to the synthesis of improved genotypes, better suited to the technological and nutritional needs of the market. One of the more recent ways to exploit the natural genetic diversity of germplasm collections is the so-called association analysis, or association mapping (AM), whose basic objective is to detect correlations between genotypes and phenotypes on the basis of linkage disequilibrium (LD). The sheer size of many germplasm collections often hinders their best exploitation; hence the need to identify core collections, i.e. representative samples of manageable size, where to perform the initial assessments, before re-exploring broader ranging materials. Given these assumptions, within the huge Triticum aestivum collection available at CRA-SCV, a sub-collection of 157 lines was identified and an association mapping project was started .

MATERIALS AND METHODS

RESULTS AND DISCUSSION

A set of 157 bread wheat varieties representative of wheat breeding in Italy from the beginning of XX century up to date was assembled. The lines were classified into nine groups on the basis of their period of cultivation. Group 1 consists of local landraces, group 2 includes selections from local populations, group 3 comprises the first cultivars obtained by Nazareno Strampelli by crossing Italian and foreign germplasm. Each of the remaining groups consists of cultivars released in the following (consecutive) period of twenty years, and having at least one parent in the previous group. The lines were cropped (single rows 1.0 m long and 0.60 m apart) for two growing seasons (2010-11 and 2011-12) in two different locations in Italy; heading, growth habit, plant height, lodging and susceptibility to rusts (Puccinia spp), septoriosis (Septoria), fusarium head blight (Fusarium spp), powdery mildew (Erysiphe graminis) were recorded. After harvest, the genotypes were characterized for spike shape, length, colour, density and awnedness, glumes hairiness, seed colour, size and vitreousness, according to the IBPGR (1985) descriptors. The number of spikelets and seeds per spike, and the thousand-kernel weight (TKW) were also determined. The lines were subsequently ground to wholemeal with a 1-mm-sieve Cyclotec mill (Foss Tecator AB, Höganäs, Sweden). Protein content (PC) (N × 5.7, dry weight, AACC 39-10), and hardness (AACC 39-70 [AACC International, 2000]) were determined by a NIR System Model 6500 (FOSS NIRSystems, Laurel, MD). The SDS sedimentation volume was according to Preston et al. (1982). In parallel, DNA was extracted and analysed by the 90K wheat Infinium array (Illumina). The phenotypic traits were analyzed by univariate (i.e., ANOVA) and multivariate (i.e., Correspondence analysis; CA) approaches. For CA the quantitative variables were converted to multiple categories and then binarised together with the qualitative ones (Burt table).

The lines were clustered into nine groups, from group 1, including ancient populations, to group 9, including cultivars released from 2000 to 2006. Significant differences for all the parameters were observed (ANOVA not reported). A clear trend from old to new varieties was observed, leading towards earliness and reduction of plant size (Fig. 1). The n° spikelets/spike did not shown extensive variation among groups, while an increase of the n° seeds/spike, along with kernel weigh reduction, was observed going from group 1 to 9 (Fig. 2). For qualitative parameters, a strong increase of kernel hardness was observed starting from the cultivars released in the ‘80s (group 7; fig. 3A). The eldest lines had higher protein content than the recent ones (Fig. 3B); however, in the recent lines is evident a gradual increase of SDS sedimentation volume, a good indicator of gluten quality, with the highest values in groups 7, 8 and 9 (Fig. 3C). This reflects the breeding trend in the last thirty years: besides yield improvement, a strong selection for qualitative traits was pursued.

REFERENCES AACC International (2000) Approved methods of the American Association of Cereal Chemists. 10 th ed. Am. Assoc. of Cereal Chem. St. Paul, MN. International Board for Plant Genetic Resources (IBPGR); Commission of the European Communities (CEC) (1985). Revised descriptor list for wheat (Triticum ssp.) AGPG: IBPGR/85/210. Preston K.R., March P.R., Tipples K.H., 1982. An assessment of SDS-sedimentation test for prediction of Canadian bread wheat quality. Can. J. Plant Sci. 62: 545-553.

0

20

40

60

80

100

120

140

1 2 3 4 5 6 7 8 9

Plant height (cm)

0

5

10

15

20

25

30

35

40

45

1 2 3 4 5 6 7 8 9

Heading (days from 01/04) A

B

Fig. 1 – Average heading time (A) and plant height (B) in the groups identified in the bread wheat core collection. Standard errors (SE) are represented by bars.

30,0

32,0

34,0

36,0

38,0

40,0

42,0

44,0

46,0

48,0

50,0

1 2 3 4 5 6 7 8 9

Thousand kernels weight (g)

10

12

14

16

18

20

1 2 3 4 5 6 7 8 9

N° spikelets/spike

0

10

20

30

40

50

60

70

1 2 3 4 5 6 7 8 9

N° seeds/spike

A

B

C

Fig. 2 – Average n° spikelets/spike (A), seeds/spike (B), 1000 kernels weight (C) in the groups identified in the bread wheat core collection. SE are represented by bars.

10,0

11,0

12,0

13,0

14,0

15,0

16,0

17,0

18,0

1 2 3 4 5 6 7 8 9

Protein content (% db)

20

25

30

35

40

45

50

1 2 3 4 5 6 7 8 9

SSV (mL)

0

10

20

30

40

50

60

70

80

1 2 3 4 5 6 7 8 9

Hardness

Fig. 3 – Average kernel hardness (A), protein content (B) and SDS sedimentation volume (C) in the groups identified in the bread wheat core collection. SE are represented by bars.

A B C

The first axis of multivariate CA (Fig. 4) mainly discriminated between ancient (groups 1-3, left side of figure 4A) and more recent groups. Being the CA an analysis where observations and variables are positioned in the same space, it is possible to select the categorized variables which follow the same trend observed for the groups. In particular Fig 4B shows how (from ancient to newer – i.e. from left to right):

• Plant height tends to decrease • Thousand kernels weight tends to decrease • Protein content tends to decrease • Awnedness follows this trend: from very long awn to awned and awnless • Kernel size tends to decrease

The statistical analysis for the definition of allele diversity, the phenomic analyses (i.e., coupling genotypes and phenotypes information) population structure and LD investigations are currently under way.

-1

-0.5

0

0.5

1

1.5

2

2.5

3

3.5

-1.5 -1 -0.5 0 0.5 1

Axis1 (3.8%)

Axis

2 (

3.4

%)

Group1 Group2 Group3

Group4 Group5 Group6

Group7 Group8 Group9

-1

0

1

2

3

4

5

-2,5 -2 -1,5 -1 -0,5 0 0,5 1

Axis

2 (3

.4%

)

Axis1 (3.8%)

Plant height

Thousand kernels weight

Protein content

awnedness

Kernel size

Fig. 4 – Output of the multivariate Correspondence Analysis (observations and variables are plotted in different graphs); (A) plot of the observations, (B) plot of the variables which show trends on the axis 1.

A B