Immediate Applications of Biotech in Tree Breeding

Immediate Applications of Biotech in Tree Breeding? ? 061201 Dag Lindgren

Background • • • Harvesting time, now Swedish Forest Tree Breeding wants to harvest the knowledge generated by the investment in the research school. The structured courses are now over. Of course you cannot have made much research, but you should be aware of things like molecular breeding immediatly available done somewhere in the world and you should have learnt to think. You should new more about the detailed and what happened recently than the senior scientists! Now Swedish Forest Tree Breeding want to benefit from that. I do not mean that you should comment now, but in the coming months, preferable to me or Bengt. I am asked to do this and not a specialist on some of what I say or talk about a chosen subject or my speciality. But there is a process initiated to draw out what is available now for operations. This presentation is partly a start up of a long discussion, not only a statement of facts. Part of it I have never presented before. Some sentences are generalisations meant to stimulate thought. Skog. Forsk has initiated a discussion about ”immediate” biotech applications the coming year as a part of a revision of the Swedish Tree Breeding plan. Bengt and I will think particularly on biotech. This presentation can (among other things) be seen as a preparation for that. The other presentations Dec 1 will give a good and updated introduction, and we may revise with the new knowledge they bring. I ask Rosario and Pär to send Bengt and me any material relevant to (almost) immediate applications. A part of the discussion is off course what is immediate, but this is not the right place and time for that. GMO – half the time - will be about GMO and is almost identical to a presentation in Umeå 06 -03 -08. However, I may rush it through if times get short in the end… Is GMO immediate? Skogen Oktober 2006: ”Om mindre än ett decennium kan det finnas genetiskt modifierade lövplantor på marknaden” ”menar Ove Nilsson”. Therefore it is a need of reviewing this subject in the context immediate applications. There are many present now, which were not present at the earlier presentation. http: //www. genfys. slu. se/staff/dagl/Documentations/GMO_presentation_06. ppt GMO presentation is available on the web with the URL given, the first presentation I feel will rather soon be improved so I do not want to share it too widely yet. But if you ask me for it on email you will get it as a reply (when I get your email also).

Earlier history not encouraging! Triploidi! • The initiating event in plant breeding was that the plant breeder Nilsson-Ehle detected 1935 that an aspen with big leaves was triploid, thus a direct link between genetic constitution and production. The picture is a ramet of that tree. Triploids was the first idea tried, but never become important.

Mutations • I have a BSc in physics and was employed at the Royal College of Forestry 1967 to run dosimetry for mutation research. The professor at that time (Åke Gustafsson) thought that mutations could become useful. I soon realised that this was a dead end forest trees and started to think on something else.

1995 Skog. Forsk förädlingsutredning initiated work with • Early tests • Faster cycling (early flowering) • Early tests (mostly in phytotron) did not work. • Flowering stimulation not evaluated yet Still I think people are keen to go on with the same desires… so most of the questions will off course remain evergreens. .

What do I think is useful? System studies, our forest tree breeding has become more efficient by a factor of 2 (? ) the last 3 decades. Modern biotech and better understanding of genes (even the simple Mendelian) has not been very helpful for operative tree breeding till now. Some reasons for faster improvement: -Less emphasise on each plus tree, but instead more plus trees. -Less unproductive waiting for controlled crosses in seed orchards. -Clonal testing instead of progeny testing for Norway spruce. -Better electronic data management and better calculation methods (BLUP, group merit selection) -More streamlined plant production and field trials. -Single tree plots. ”Modern” biotech has been less important

Applications of molecular methods: germplasm use (from IPRI) Comment Suggested immediate practical implications for long term breeding Gene mapping and identification Basic science Hardly Detecting somaclonal variation As clones are much used, this should be investigated Encourage studies Evaluating germplasm for useful genes Basic science, but useful tools may appear Likely useful diagnostic tests will appear Pedigree analysis Immediate appliable but Probably, think more! complicates things and is it worth it? Hybrid identification Immediate appliable, but relevant only for orchards No

Marker aided selection • • • Important genes may not be identifyable and reproducible, it may just appear so. Negative genes (inbreeding) may seem important, thus MAS effort may result in weeding of unimportant genes rather than selection for important Important gene in one genetic and physical environment may not be important in another (thus a lot of possible interactions not considered in simple experiments). Field performance is the added action of all important genes, MAS gives just a small part of them. Higher selection effort for one thing means lower for all others. The breeding stock is often a limited resource. A large added administrative trouble, costs for sampling and preparation, costs for analyses, costs for compiling data from different sources. If one start select for good DNA-pieces, it is an evident risk of faster accumulation of coancestry and subsequent inbreeding. Growth in mature age may be the most important, and probably to composite and complex to locate on the gene map. Etc

MAS • 20 years of QTL research gave us good insight about genome organization, but no meaningful QTL to hang the hat on. • FAO conference 2003 on MAS. From summary: 2. 7 Practical applications of MAS “Although documentation was limited, the current impact seemed small…” • Efforts have been going on for at least a decade forest trees in Sweden, without coming near to an application. • At least one scientist reasearching it in Sweden (Reza Yasdani) got the sack… and no-one of the pioneers in MAS applied to forest tree breeding is with us any more…

Not extreme close linkage • Cis- and trans- different in different families • Different associations in different materials

Associations (populationwide linkage disequilibrium) • Usually not large effects • It will probably take some time (10 years) till the pine and spruce genome will be densily covered. (? ? Pärs föredrag) • Not that mature field, so remains to see what will come

I searched my 50 first hits on the web I also looked in Vicopedia • ”Association mapping” – I found no successful breeding operation • QTL– I found no successful breeding operation (the limit to ”successful” may be debated)

Place in system • MAS research should not be seen as applied and motivated by the possible practical applications. • Instead it should be viewed as basic research. • Guess it is on way down internationally as both basic and applied research. (My feeling only. )

Parentage • It is now possible to decide which (among suspects) are the parents to a tree. That could find immediate applications in operative breeding. • Reliability is high enough. • Costs is probably affordable for some applications and decreases.

• Can more directly estimate ”effective size” (status number), which actually is a requirement in BC. (Similar to Wrights FST)

Error elimination • The breeding population could be checked for errors in pedigree. • Erronously tagged clones occur. Yoshinary suggests 10% on average and 50% in special cases. Johan checked an Umeå seed orchard material and found no error in 36 ramets. A month ago I personally checked a Turkish seed orchard and found at least one error among 75 ramets. • I believe Swedish breeders are extremely clever and careful, so the error-rate in breeding material is probably below 5%. But this figure could be good to verify! • Probably it is not justified to make it a routine, but at least it is something which requires some more discussion

Routine DNA sampling? • Perhaps a good idea to routinely store a DNA sample of every member of the breeding population forever? (”Gene conservation”)

Polycross • Progeny testing and creating of progenies for selection of a new breeding population may be more efficiently combined. • Trees can be mated with mixtures of pollen from good trees for parental ranking, and in the same time creating a new breeding population with known pedigrees by identifying which of the fathers with markers.

Molecular diagnostic tools • The state of plants – not always directly coupled to the genes that control the physiological process – will be easier to record with molecular symptoms. New measurement devices for relevant characters can be foreseen.

Breeding without breeding • Molecular parentage control has the potential to make controlled crosses redundant. Controlled cross (recombination) is one of the bottlenecks in tree breeding and it is a much bigger obstacle than generally accepted. No convincing schemes have been presented yet, but the concept is new and I believe it will be possible to identify procedures which could be practically useful and efficient ”immediatly”. • E. g. Norway spruce are tested in clone trials in operative breeding and based on the results the best clones are later mated in archives. If instead the trial was measured first when cones were emerging; open pollination from the best clones collected and progeny checked for if the father also was among top clones and if so cloned and tested…. It may be a more efficient design than the present

Flower stimulation • Hormons (gibberrelins) are involved an important background for the history of this department and the Umeå plant physiology group (Arne Dunberg), but the history is older than that. • Gibberrelins are used operationally to get crossings of Scots pine, which speeds up operations somewhat. However, ”top working” is regarded as operatively more promising. • Not used operationally in seed orchards and efforts give variable results • Even if it would succeed in seed orchards it is not given it would be allowed (work hazards) • I do not foresee large progress the next years (thus breeding plan should be based on operative methods).

Clones for conifer massmultiplication • • • Cuttings SE Both systems operative for conifers (as well as others) Both systems results in considerable more expensive plants Cutting propagation takes time to multiply a clone To store clones for cutting propagation is expensive and troublesum SE propagation is fast, clone storage is cheap and simple. A slight risk for accumulation of somatic mutations Ease of multiplication varies among genotypes, more so for SE than cuttings.

Economic calculation Lennart Eriksson 050823, http: //www. nordgen. org/nsfp/doc/konferenser/2005 frosta/NSFP 2005_lennart_eriksson. ppt, G 36 spruce, rotation 51 years, 3000 plants/ha, 2% interest, kkr/ha, Dag complement Plantation with Production level Added cost Belarus Provenance Cuttings 100% 0 Present Value 62 Land Value 97 115% 4. 5 78 123 Somatic 120% Embryogenes Tre. O 115% Seed orchard 1. 5 87 136 0. 2 78 123

The message • If calculations are the right, the increase in present value of the stand or in land value can more than pay even for expensive ways to get genetic gain • For some reason forestry does not seem to have as clever economists as SLU, it is difficult to get them to buy this argument! • But at least 5 years ago forestry bought the idea for seed orchard seeds. • But it requires a 50 fold increase in dedication of forestry to accept the clonal idea!

Cutting…

Avvecklat, för dyrt!

Växthusplantage – ett sätt att producera råmaterial till bulkförökning! Sju växthus kan täcka Sveriges behov I växthuset får det plats 400 plantageträd Varje träd ger 390 matade frön per år, Avvecklat, för dyrt! totalt 156 000 frön 80 % av fröna ger plantor användbara för sticklingproduktion Man producerar 200 sticklingar av varje fröplanta Totalt ger då ett växthus odlingsmaterial för 25 milj. sticklingar, och sju växthus material till 175 milj. sticklingar Kuvattu maaliskuussa, 1995 Mats Eriksson & Curt Almqvist, Skogforsk PLANTaktuellt Nr 2 2003

Vegetative propagation programs with Norway spruce in Sweden added Cost/plant SEK Operator Method Start genotypes Start End Clone Hilleshög Cutting Nursery selections of good provenances 1972 1995 +0. 80 Mellansvenska Cutting Good parents 1995 STT Crosses of good 2006 tested parents SE Bulk 1995 2004 1* =1 0. 6* =0. 6 2016? 2 2

Use of clones over the World Eucalyptus -total of 15 000 ha P mariana east Canada Eucalyptus spp. -total of 1 217 000 ha Acacia spp. -total of over 200 000 ha -1500 ha annually Populus spp. P sitchensis Ireland Scotland -in Europe 949 000 ha - 2500 ha annually -total of 1 567 000 ha Salix spp. -in Europe 20 000 ha Cryptomeria japonica (sugi) -in Japan 5 000 ha Cupressus / Chamaecyparis spp. -total of 40 000 ha Pinus spp. -total of 30 000 ha Estimated from Sonesson et al 2003 According to B. Kellison, IUFRO 2004

Clonal forestry in Nordic countries (total) Based on Questionnaire 2002 Nordic Group for The Management of Genetic resources of Trees; Summarised by Steve Lee Hybrid Aspen – >500 ha since 1999

Clonal forestry in Nordic countries – vision for year 2012 (annual) (of total annual planting ≈ 300 000? ha) Questionnaire 2002 Nordic Group for The Management of Genetic resources of Trees; Summarised by Steve Lee

A million SE plants. Plum Creek Nursery in Georgia

Plantation with SE trees Sept 06 in Georgia Photo Dag Lingren Sept 06

Transplantation maching, SE miniplugs are transplanted to open land. Nursery in Georgia Photo Dag Lingren Sept 06

Forestry with clones • It is technically and biologically possible to propagate spruce by clones for practical forestry. • One or a few percent of the plant market may be penetrated with expensive clonal propagation for curiosity or foreseen development. The rest will not be willing to pay the price. • My guess is that the vision of clonal forestry will remain, but that it does not justify much extra effort in breeding.

Science with clones Potential better science • Replications • Reproducibility

Clone is a tool for a more effective breeding! • • • Used for seed production Gains faster realized Clonal test means testing the sum of genes deployed, progeny testing often are confounded by paternal genes just contributes to noise A seedling is genetically unique; clones can be optimally replicated. Clonal test gives in practice much information about ability to transfer gene to progeny More efficient use of the variation occurring after sexual propagation

Clone is a tool for a more effective breeding! • The efficiency of clonal testing depends on costs • Collaborators instead of competitors (non -egoistic clones) • Test on many environments and choose for wider adaptation • Test in the field and cross in the archive • Combine wood in the field and reproduction in archive • Biotechnical breeding - like transgenetic trees - becomes more feasible

16 14 Clonal selection Breeding value 12 10 Phenotypic selection 8 6 Comparison (at the same dimensioning) of clonal or seedling based testing for the Swedish Norway spruce long term breeding program. Clonal testing adds around 30% to gain. (Rosvall 1999) 4 2 0 0 500 1000 1500 Test size (plants) 2000 2500

Multigenerational comparison of testing strategies in Swedish conifer breeding • Clonal testing is much superior to progenytesting • Phenotypic testing better than progenytesting at low budget Danusevicius and Lindgren

Clone is a tool for a more effective breeding! • Clone testing with cuttings has operatively replaced progeny testing for Norway spruce in Sweden • It may be possible to do the same for Scots pine. Development is going on and continuing small scaled practical applications in part of breeding is recommended for pines (lodgepole and Scots) • Although in the end I guess it will be found usable but not cost efficient for Scots pine. The donour plants are to large and expensive and the success rate will vary too much among clones.

End