Transcript 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.
SkogForsk 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/GMOs/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 for
forest trees and started to think on
something else.
1995 SkogForsk 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 for
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 and 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/2005frosta/NSFP2005_lennart_eriksson.ppt,G36 spruce,
rotation 51 years, 3000 plants/ha, 2% interest, kkr/ha,
Dag complement
Plantation with
Production
level
Added
cost
100%
0
Present
Value
62
115%
4.5
78
123
Somatic
120%
Embryogenes
TreO
115%
Seed orchard
1.5
87
136
0.2
78
123
Belarus
Provenance
Cuttings
Land
Value
97
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,
totalt 156 000 frön
Avvecklat, för dyrt!
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
2004
1*
=1
0.6*
=0.6
STT
Crosses of good
tested parents
2006
2016?
2
2
SE
Bulk
1995
Use of clones over the World
Eucalyptus
-total of 15 000 000 ha
P mariana east Canada
-1500 ha annually
P sitchensis Ireland and Scotland
- 2500 ha annually
Eucalyptus spp.
-total of 1 217 000 ha
Acacia spp.
-total of over 200 000 ha
Populus spp.
-total of 1 567 000 ha
-in Europe 949 000 ha
Salix spp.
-in Europe 20 000 ha
Cryptomeria japonica (sugi)
-in Japan 5 000 000 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)
Norway Spruce:
10,000 ha non-tested;
1,000 ha tested
Hybrid Aspen: 400 ha;
10 years
Poplar: 1,000 ha since
many years
Norway Spruce: 250 ha
non-tested; 20 ha tested
Hybrid Aspen: <1500 ha;
since 1997
Birch: Curly (10years) and
Silver (13 years) 1,400 ha
Aspen: small area;
6 years
Poplar: 2,000 ha;
12 years
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)
H. Aspen - 1000 ha/yr
A. lasiocarpa?
Poplar 500ha/yr
Norway Spruce
- 100ha/yr
Birch - 400 ha/yr
NS - 500 to 1000 ha/yr
SP - gearing up
No change to
Aspen
Sitka Spruce 500ha/yr
Questionnaire 2002
Nordic Group for The Management of
Genetic resources of Trees;
Summarised by Steve Lee
H. Aspen - 300ha/yr
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 progeny-testing
•Phenotypic testing better
than progeny-testing at low
budget
Danusevicius and Lindgren 2002
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