A MAGIC population as an approach to the conservation and development of genetic diversity of winter barley for breeding purposes by on-farm management

Eine MAGIC-Population als ein Ansatz zur Erhaltung und Entwicklung züchterisch nutzbarer genetischer Diversität von Wintergerste durch on-farm Management

Journal für Kulturpflanzen, 71 (11). S. 286–298, 2019, ISSN 1867-0911, DOI: 10.5073/JfK.2019.11.02, Verlag Eugen Ulmer KG, Stuttgart

Correspondence

Dr. Lorenz Bülow, Julius Kühn-Institut (JKI), Federal Research Centre for Cultivated Plants, Institute for Breeding Research on Agricultural Crops, Erwin-Baur-Str. 27, 06484 Quedlinburg, Germany, E-mail: lorenz.buelow@julius-kuehn.de, phone: +49–3946–47705

Accepted

17 October 2019

This is an Open Access article distributed under the terms of the Creative Commons Attribution 4.0 International License (https://creativecommons.org/licenses/by/4.0/deed.en).
Dies ist ein Open-Access-Artikel, der unter den Bedingungen der Creative Commons Namensnennung 4.0 International Lizenz (CC BY 4.0) zur Verfügung gestellt wird (https://creativecommons.org/licenses/by/4.0/deed.de).

Thirty-two founder lines were identified representing the genetic marker diversity within this set of 58 varieties (Léon et al., 2008) and showing variation in terms of yield, stability, winter hardiness and resistance to leaf rust, net blotch, powdery mildew, BaYMV and BaMMV (Table 2). In the period from 2008 to 2015, the selected winter barley varieties were crossed (Frese et al., 2012) according to the Multi-parent Advanced Generation Inter-Cross scheme (Cavanagh et al., 2008) to produce a heterogeneous multiparent winter barley population.

Based on a principal component analysis (PCA), 32 diverse founder lines representing the genetic diversity of the 58 analysed varieties were selected and assorted to four groups of eight parents each. The initial crosses were carried out according to this grouping (Table 2). Group I of the parental generation G0 consisted of the pairwise initial crossings (Sorna × Gaulois), (Vulcan × Cervoise), (Frances × Peragis), and (Birgit × Gilberta), group II of (Pict × Alissa), (Alinghi × Merlot), (Lomerit × Leibniz), and (Fridericus × Naomie), group III of (Eiszapfen × Alpaca), (Banteng × Ludmilla), (Mammut × Venus), and (Engelens II × Catinka), and group IV of (Elbany × Andrea), (Engelens-6zlg × Hord. Hex. Giganteum), (Vogelsanger Gold × Plana), and (Erfurt × Traminer). Within each of the four groups, the resulting sixteen G1 lines were intercrossed with offspring resulting from crosses of the same group yielding lines for the next generation G2. To exclude accidental selfings, heterozygosity of G1 and G2 offspring was verified by using selected polymorphic SSR markers. Due to a limited analysis capacity, this validation was restricted to G1 and G2 offspring only. Generally, for each intercrossing generation at least 20 independent plants were used in order to avoid losing alleles. The MAGIC crossing scheme (Cavanagh et al., 2008) was accordingly continued until the production of G6 seeds, each one harboring parts of all genomes from the 32 parental founder lines. Aliquot amounts of seeds from each of the resulting 324 G6 lines were combined in 2015 to form a highly heterogeneous and heterozygous population that was subsequently propagated at the Julius Kühn Institute in Quedlinburg.

SSR analyses were carried out with selected polymorphic SSR markers (Ramsay et al., 2000). The PCR was performed in a 20 μl reaction volume using the PeqSTAR 96 HPL Thermocycler (PeqLab, Erlangen, Germany). Regarding the primers, two different PCR mixes were used: a) a universal fluorescent-labeled M13-primer (0.07 μM) in combination with an M13-tailed marker- specific forward primer (0.1 μM) and a marker-specific reverse primer (0.17 μM) as described by Schuelke (2000) or b) a Cy5 labeled marker-specific forward primer (0.5 μM) was added to the marker-specific reverse primer (0.5 μM). PCR mix a) was preferred, but in case of unsuccessful amplification, PCR mix b) was employed for the labeling reaction. Generally, a touchdown PCR profile with decreasing annealing temperature by 0.5°C per cycle in the course of 12 cycles from 62°C to 56°C was used that was followed by 35 cycles. However, adapted PCR profiles were applied depending on the specific primer pair employed. The sequence of the forward and reverse primers, the repeat motif and the PCR profiles are given by Ramsay et al. (2000). PCR products were separated and detected on a GenomeLab™ GeXP platform (AB Sciex GmbH, Germany) applying multiplexing.

Each parent variety was genotyped to detect accidental self-pollination and to ensure that only offspring from cross-pollination are used to produce the next generation. Before starting the next crossing step, G1 and G2 offspring was verified with polymorphic SSR markers by means of parent-specific alleles (Table 3). The composite population was finally established by pooling aliquot amounts of 300 seeds of each of the G6 progenies. The bulked seed sample was subsequently propagated in 2015/2016 in Quedlinburg.

Fifty-one different soil-climate-areas have been defined for Germany that are derived from meteorological and soil data and result in a consent map of soil-climate-areas released by German agricultural federal and state authorities (Roßberg et al., 2007). Based on these different areas, 12 eco-geographically contrasting locations were selected to cultivate sub-populations in order to promote the development of differently adapted germplasm (Table 4). These locations constitute the network for the dynamic management of winter barley genetic resources and involve three locations under organic management practices, too. In addition, a further location in Southern Sweden near Alnarp was included (Table 4). Since 2016, sub-populations of the material have been continuously cultivated under high and low agricultural input conditions including crop protection measures at each one of these locations. Adaptation of winter barley sub-populations to different climatic, soil and agricultural input conditions will take place over a period of at least 6 to 8 years. It is however expected to maintain the network of sites as a conservation infrastructure for dynamic management of plant genetic resources over a much longer period. In order to maximise adaptation to the different regional conditions, locally usual cultivations practices like tillage, seed bed preparation and seed density are applied at each of the locations. Based on samples taken in each year, changes in the allele frequencies within and between locations will be monitored at the DNA level.

The winter barley sub-populations are expected to adapt to the specific local conditions. The climatic conditions in Germany have been recorded continuously by meteorological stations in the past. To characterise differences between the locations, 30-year average data on temperature, precipitation, radiation, sunshine, and frost were collected for the period from 1981 to 2010 for each one of the German locations from the ClimateImpacts­Online portal (http://www.climateimpactsonline.com/) at administrative district level. However, climate is steadily changing and the locations may be subjected to different degrees of climate change resulting in the need for local climate change predictions. When designing such a long-lasting project, potential changes of climatic conditions have also to be taken into account. As possible future scenarios, four Representative Concentration Pathways (RCPs) for greenhouse gases were adopted by the United Nations Intergovernmental Panel on Climate Change (IPCC) in its Fifth Assessment Report (IPCC, 2014). These are RCP2.6, RCP4.5, RCP6, and RCP8.5 named according to the radiative forcing in W/m² in the year 2100 compared to pre-industrial values. RCP8.5 represents the scenario with the highest greenhouse gas emissions and was considered by the IPCC as an extreme case. Nevertheless, current greenhouse gas emission rates match or even exceed the RCP8.5 scenario (Hayhoe et al., 2017), and therefore, this scenario fits the actual development most closely. The Potsdam Institute for Climate Impact Research (PIK) calculated regional climate projections for the next 60 to 100 years by employing the STARS model (STatistical Analogue Resampling Scheme) (Orlowsky et al., 2008) for all German administrative districts. As a result of changing conditions, the climatic water balance will decrease, especially during the main growing season of crops from May to October (Gerstengarbe et al., 2013). This and other meteorological parameters will have implications for future agricultural economy. At the country level, Germany like Northern European countries is supposed to economically benefit from climate change since its current temperature is below the economical optimum (Ricke et al., 2018). At the regional level, PIK used an Integrated Regional Model Assessment to determine future yields of agricultural crops in Germany. Yield decreases are expected for summer-grown crops in these areas of Germany where water availability is limited. On the other hand, winter cereals will be less influenced and may even benefit from the increased precipitation during the winter season (Gerstengarbe et al., 2013). Meteorological and agronomic predictions based on the RCP8.5 scenario and median temperature increase were collected from the ClimateImpactsOnline portal for the period from 2011 to 2040 for each one of the 12 German locations to track and assess future developments of the sub-populations. There are no explicit data on winter barley yields, but winter wheat data can be accessed to infer that most of the German locations may benefit from climate change while some might expect decreased winter cereal yields. The degree of prospected yield changes is remarkable and ranges from 30.9% decrease to 26.7% increase. Details on past and prospected winter wheat yields for the German locations are given in Table 5.

In autumn of 2016, all locations were provided with seeds from the propagated winter barley population ERG16–1 and asked to continuously cultivate the population on two plots of approximately 100 m² applying two different fertilizer and pesticide intensities. Given that the locations correspond to different conditions regarding soil and climate, usual local cultivation practices should be applied in respect to crop rotation, sowing date, sowing density, fertilisation, use of pesticides, and harvest. The intensive or high input variant at each location stands for these usual local cultivation conditions whereas the extensive or low input variant represents reduced fertiliser and pesticide intensities. Low input fertiliser application rate should amount to approximately one third of the high input rate. Regarding the low input variant, plant growth regulators, fungicides and insecticides should be completely avoided and only be used in case of probable total loss of the harvest. Use of pesticides only applies to the conventionally managed locations while locations under organic managing practices completely reject pesticides. Therefore, variants of the organically managed locations only differ in the fertiliser application rate. When needed, special local measures are applied, for example the urban location Dahlem situated in the city of Berlin covers the plots with nets to prevent bird damage after sowing and before harvest. All local treatments and cultivation conditions are being continuously documented. The harvested seeds at each location are saved for sowing the next generation of the population. To avoid possible edge effects, only the harvest of the central part of each plot is kept as seed. The different locations constitute the network of sites since 2016 when sub-populations were first cultivated, and from then, the populations have been grown continuously at the locations. It is intended to continue cultivation and seed saving for 6 to 8 years. In order to ensure this long-term cooperation between the managing institutions of the locations, a memorandum of understanding was signed to promote the continuity of the network. The Julius Kühn Institute was appointed to currently take over the coordination of the network and is in charge of the network management. This includes services like documentation, data management and long-term sample storage.

Samples of approximately one kg of each harvest from all locations are sent every year to the Julius Kühn Institute in Quedlinburg for central storage and later evaluation. For better comparison, evaluation will take place after the period of site-specific adaptation at one location with all the collected samples. Since the adaptation period lasts 6 to 8 years, standard genebank long-term storage measures (Rao et al., 2006) are applied to the seed samples, which includes drying to a low moisture content of about 5% (Ellis and Roberts, 1980) before storage at low temperature (Dickie et al., 1990). Purified and dried seed samples from the locations are stored in air-tight 1000 ml wide-neck transparent PVC containers with PP screw-on caps and PE-foam seals (Carl Roth GmbH + Co. KG, article no. H318.1) at –20°C in a chest freezer. In addition to the samples from all locations and all years, seed samples from the initial evolutionary bulk (ERG16–1) that was sent out to all locations as well as from the 32 parent varieties are treated and stored analogously.

Fig. 1. Pedigree representation by ROBUSTUM for barley line ERG11–66/5 (highlighted) from MAGIC cros­sing generation G2. The pedigree displays the line’s parentage up to its four parental founder varieties as well as all descendants derived from the line down to generation G5. The number of levels to be shown can be adjusted (not shown) and the interactive functionality enables retrieval of additional details on the breeding material, its pedigree in XML format and the corresponding breeding actions.

An effective and efficient approach for a dynamic management system with wheat was described by Goldringer et al. (2001) and is being developed by the INRA at Le Moulon (France) towards an on-farm management system involving farmers (Enjalbert et al., 2011). This French approach is being accompanied and supported by a Seed History and Network Management System, an information system called SHiNeMas (Rivière et al., 2013). Farmers in Germany have already expressed interest in the present winter barley population. The network may therefore be broadened towards a future participatory breeding network. ROBUSTUM will take over the documentation required by such a network. Recently, the process to register the winter barley population as a heterogeneous population variety at the German Federal Plant Variety Office has been initiated to allow for easy distribution to interested farmers. The registration of heterogeneous population varieties is also possible in Germany since the publication of the regulation on the placing into the market of seed of populations of the species oat, barley, wheat and corn (BGB, 2015). The regulation refers to the implementing decision of the Euro­pean Commission on the organisation of a temporary experiment providing for certain derogation rules for the marketing of populations of the plant species wheat, barley, oats and maize (EC, 2014). This experiment is part of the agricultural policy trying to increase inter- und intra specific genetic diversity in crop production as well as on grassland (BMELV, 2007). It is expected that a more diversity-friendly agricultural policy will not only promote sustainable farming activities but also help maintaining and enhancing good ecological infrastructures, and promote the conservation plant genetic resources (Paracchini and Bertaglia, 2018).

The production of and access to food and feed in sufficient quantity and quality today and tomorrow is pivotal to the implementation of the sustainability concept and an enormous challenge to agriculture facing numerous production limits. The challenge to agriculture research today is to investigate and develop land utilisation systems that (i) contribute to the maintenance and improvement of ecosystem function, (ii) treat the non-renewable resources sustainably, (iii) can buffer production systems against negative effects of weather extremes and other hazards, and (iv) can produce food, fodder and other agricultural commodities in sufficient quantities and qualities (Baeumer, 1997; Gregorius, 2001; Fuhrer, 2003). How inter-specific diversity can be deployed to implement resilient production systems is currently being inves­tigated (Lin, 2011; Raseduzzaman and Jensen, 2017). Research projects concern inter alia the effects of intercropping systems on weed control in pea and barley (Corre-Hellou et al., 2011), wheat and subclover (Radicetti et al., 2018), canola and legume mulch (Lorin et al., 2015) or the recovery of fertiliser phosphorus (Xia et al., 2013). Inter-varietal diversity, i.e. specific fractions of the intra-specific diversity combined in varieties of a crop species, offers benefits in yield response, yield stability and disease control (Newton, 2016; Newton and Guy, 2009) in barley. The yield increase or disease index decrease of mixtures compared to the average of their components in pure stands depends on the specific features of the components and how the components interact in the mixture under local production conditions. High intra-specific diversity thus has the potential for environmental resilience, i.e. buffering the effects of environmental and climate variability (Döring et al., 2010). In addition, evolutionary processes contribute to a durable conservation of biodiversity.

Bulk populations constitute a significant contribution to the conservation and development of genetic diversity that can be used in breeding and contribute to the replenishing of the barley gene pool. MAGIC populations compared to classical biparental populations highly benefit from the inter-crossing process due to a faster decay in LD and breakup of initial linkage blocks. The high number of recombination events is also a prerequisite for new allele combinations and influences the genetic diversity of the population. In the case of self-pollinating species, the adaptation and enrichment of favourable alleles can be measured after 6 to 10 generations (Goldringer et al., 2001). Heterozygosity will substantially decrease in the first years after the last crossing and only thereafter, a continuous effect of the selection forces that is not masked by heterozygosity will be detectable. Because of this, in the present barley MAGIC approach cultivation and adaptation within a network of sites shall take place over a long period until 2023. Thereafter, the evaluation of the adapted and differentiated MAGIC sub-populations will be carried out to assess the impact of local adaptation on the allele constitution and to analyse the temporal and local dynamics of biodiversity and the genetic causes of the adaptation processes at molecular level. Specific traits of interest may be identified through marker-trait association studies or high-resolution mapping of quantitative traits by genotyping by sequencing (GBS).

In the future, the MAGIC population may continuously be improved by introducing modern varieties originating from classical plant breeding and use of a male sterility trait to increase recombination rates. This will enable breeders to use the population as a new source for complex traits. The evolutionary plant breeding concept translates the breeding of landraces by farmers and the functions of traditional seed supply systems of the past into science-supported crop diversity management practices of today. Generally, diversification is a usual measure to increase the stability of a system, and even financial service providers diversify their portfolios in particular in periods of increasing instability to minimise the financial risks (Figge, 2001). The present approach addresses the increasing pressure on finite natural resources caused by a growing human population and changing consumption patterns (Steffen et al., 2015) and increasing insecurity in agricultural production systems. Evolutionary plant breeding approaches will also play a very prominent role in solving problems related to climate change (Schimmelpfennig et al., 2018). After successful adaptation, a better fitness of the population against boitic and abiotic stress factors is expected. Thus, the cultivation of evolutionary bulks provides the plant breeders with a continuously filled resource of locally adapted variability that will serve to improve cultivated plants in the future. The creation and cultivation of populations appears to be a suitable measure for successful in-situ conservation of plant genetic resources under on-farm conditions. Bulk populations can be considered as course of action in plant breeding and contribution to a genetic diversification of agricultural production systems.

The winter barley MAGIC parents were kindly provided by the department of Plant Breeding of the University of Giessen, INRES Plant Breeding of the University of Bonn and German breeding companies. Thanks to the members of the network for cultivating and harvesting the subpopulations in the past and following years.

Many thanks to Peter Wehling and two anonymous reviewers for revising the manuscript.