DEPARTMENT for ENVIRONMENT, FOOD and RURAL AFFAIRS CSG 15

Research and Development

Final Project Report
(Not to be used for LINK projects)

Two hard copies of this form should be returned to:

Research Policy and International Division, Final Reports Unit
DEFRA, Area 301
Cromwell House, Dean Stanley Street, London, SW1P 3JH.
An electronic version should be e-mailed to resreports@[Link]

Project title Conventional and biotechnological genetic improvement of hardy nursery

stock
     

DEFRA project code HH1026SHN

Contractor organisation Horticulture Research International

and location East Malling, West Malling
Kent, ME19 6BJ

Total DEFRA project costs £ 380,279

Project start date 01/04/00 Project end date 31/03/03

Executive summary (maximum 2 sides A4)

Objectives

1. To propagate Acer platanoides selections identified as resistant to Verticillium wilt in HH1012SHN,

and now as part of a new EC Shared Cost Project, so that those best for resistance and
propagation can proceed to industry-funded trial; stocks of wilt-resistant Acer clones will be
available for industry exploitation and a seed orchard of resistant clones will be established.
2. To undertake genetic studies, principally in Buddleia, Ribes, Sambucus and Sorbus, to underpin
the complementary HDC-funded programme HNS18b, e.g. by allowing design of crosses to
pyramid useful characters; papers will be published on genetics of horticulturally important
characters in these genera and the information will allow the HDC-funded programme to proceed
more effectively.
3. To raise and identify selections of Rhododendron combining tolerance to lime with horticultural
merit, and to identify genetic markers for lime tolerance; promising selections will be passed to the
industry for assessment of commercial merit.
4. To raise tetraploid forms of rootstocks that would benefit from greater dwarfing ability and of sterile
interspecific hybrids that would be useful for further breeding; tetraploid selections will be made
available to the industry for evaluation as dwarfing rootstocks and fertile tetraploids will be used
as parents in the HDC-funded breeding programme.
5. To introduce a dwarfing gene into one or more HNS subjects by transformation; details of passing
transgenic material to the industry are uncertain at this stage.
6. To disseminate information to the industry and scientific community via publications and open
days and transfer breeding lines to industry; this technology transfer will be as outlined in the five
previous objectives.
Methods and Findings
CSG 15 (Rev. 6/02) 1
 Project Error: Reference source not found DEFRA Error:
title Error: Reference source not found project code Reference
source not
found

Acer

Ten clones of Acer platanoides that have been consistently rated as resistant to Verticillium wilt were
supplied to a commercial nursery and seed merchant to establish a seed orchard of wilt resistant
clones. Twenty more have been propagated for delivery and five more shortlisted. The seed orchard
will provide planting stock having, it is hoped, enhanced resistance to this disease. Individual clones
could be tried as clonal rootstocks.

Genetics

The elderberry cultivar Sambucus nigra 'Black Beauty' that was developed by collaboration between
the previous Defra project and a complementary HDC project is achieving UK sales of about 50,000
p.a. and received EU Plant Breeders Rights in 2003. A second cultivar 'Black Lace' - with very dark
purple laciniate foliage - was entered for Rights in 2002 and will be launched at the Chelsea Show in
May 2003. Various crosses were undertaken to clarify the inheritance of a dwarf gene in some 'Black
Beauty' siblings. Seedlings of mountain ash Sorbus aucuparia 'Sheerwater Seedling' x S.
cashmireana (pink flowers) and of whitebeam S. aria x S. chamaemespilus (pink flowers) were
assessed in the field. All had pink flowers, indicating that this character is dominant in both donors
and seedlings were identified for backcrossing to S. aucuparia and S. aria to introgress pink flower
colour into these particular species. Other progenies of S. aucuparia, from crosses of 'Dirkenii'
(yellow leaves) with 'Pendula' (weeping habit) and 'Fastigiata' (columnar habit) segregated for leaf
colour - but none of the seedlings had weeping or columnar habit. This indicates that these
characters are recessive, which may be confirmed by appropriate backcrossing. Progenies of the
flowering currant Ribes sanguineum, 'Brocklebankii' (yellow leaves) x 'White Icicle' (white flowers)
were assessed. Seedlings segregated for leaf colour, but all had pink flowers, indicating that white
flower colour is recessive. A small collection of hardy and/or scented jasmines was established. The
polyploidy of the accessions was determined cytometrically to avoid making crosses between ploidy
levels. Flower morph, pin versus thrum was also recorded, and test pollinations were made to
demonstrate that crosses within morphs were usually sterile whereas those between morphs were
fertile. This information will be useful when planning a crossing programme.

A small collection of roses reported to be resistant to blackspot was established and we discussed
with commercial rose breeders how genetic research could help meet their needs.

Rhododendron

Some 270 Rhododendron seedlings raised in the previous project were planted out in soil of relatively
high pH and were monitored for horticultural merit and resistance to lime-induced chlorosis. Three
selections, all from the cross of 'Exbury' x 'Cunningham's White', were shortlisted for industry funded
trial. A replicated trial of 10 reportedly lime tolerant accessions was planted in the field and
monitored. R. caucasicum 3707, R. williamsianum 1774, R. ciliatum 3056 and R. rubiginosum 1768
were the least chlorotic and could be worth using as parents. Several microsatellite primers were
developed and optimised. A progeny from the cross R. hirsutum PA54/1 (tolerant) x R. ferrugineum
3379 (susceptible) was assessed for resistance to lime induced chlorosis in vitro. The pH of the
medium was progressively increased from 5.0 to 6.4 over several weeks and growth and chlorosis
was measured. Three genotypes were identified as tolerant. However the progeny size of 23 was
too small to undertake QTL mapping. Steps were taken to increase progeny size, but this has proved
difficult.

Tetraploidy

CSG 15 (Rev. 6/02) 2

 Project Error: Reference source not found DEFRA Error:
title Error: Reference source not found project code Reference
source not
found

Over 100 seedlings were raised from the intercrossing in 2000 of two yellow flowered selections from
the progeny resulting from crossing our tetraploid form of Buddleia globosa (yellow flowers) x the
naturally tetraploid B. davidii 'Nanhoensis Alba' (dwarf habit, white flower, large inflorescences). They
were planted in the field and should flower in 2003. Tetraploid forms of Sorbus aucuparia 'Biessneri'
were grown for evaluation as rootstocks and tetraploid forms of pear x apple selection PyMa2 were
identified and grown on for use in crossing. Tetraploids of the lilac hybrids Syringa vulgaris 'Vestale' x
S. pinnatifolia 104-18 and 104-32 were induced with colchicine treatment in vitro. Some tetraploid
clones of 104-18 and 104-25 that had been induced previously flowered. Some had substantially
larger flowers than the diploid and, importantly, were fertile which will allow further crossing.
Tetraploids were successfully induced in the pear x quince hybrid x Pyronia veitchii. When they
flower they could be valuable for breeding rootstocks for pear.

Transformation

Preliminary experiments were undertaken to develop routinely reliable regeneration methods for
Buddleia davidii and to determine appropriate levels of kanamycin and paromomycin for selection of
transformants. Then a series of experiments were undertaken to transform B. davidii. However,
although putative transformants that survived on the kanamycin selection medium were obtained, it is
thought that these were chimaeric and it was not possible to confirm the presence of a small amount
of transformed tissue by PCR.

Options for further work

A CSG7 has recently been agreed with Defra for a five-year continuation project. The new project
will build on the success of HH1026 and will concentrate on developing breeding lines and tools. It
has the following objectives:

 To clarify genetics of horticulturally important characters in Sambucus, Buddleia, Sorbus and

Jasminum to inform breeding programmes.
 To identify sources of resistance to blackspot, and perhaps aphids, in Rosa, and transfer breeding
lines to a commercial breeding programme, with a view to subsequently developing a LINK
research programme.
 To demonstrate that tetraploidy has restored fertility to interspecific hybrids of Syringa and Pyrus x
Malus and check for self compatibility and mode of inheritance, and complete backcrossing
programme of tetraploid Buddleia globosa to B. davidii to introgress yellow flower colour.
 To determine if tetraploidy in the climbers Clematis and honeysuckle (Lonicera), and shrubs such
as Physocarpus induces more compact habit, useful for small gardens.
 To develop transformation methods for hardy ornamentals, assessing the xylose isomerase and
MAT vector systems in Buddleia davidii, and establishing regeneration methods for two additional
subjects, for use in subsequent transformation experiments and technology transfer to the
industry.

Relevance to Defra policy

This work on improvement of woody ornamentals accords with the ROAME A that recognises that
genetic improvement and novelty help the industry meet the quality demands of customers and retain
and extend market share by increasing competitiveness.

CSG 15 (Rev. 6/02) 3

 Project Error: Reference source not found DEFRA Error:
title Error: Reference source not found project code Reference
source not
found

CSG 15 (Rev. 6/02) 4

 Project Error: Reference source not found DEFRA Error:
title Error: Reference source not found project code Reference
source not
found

Scientific report (maximum 20 sides A4)

Background

The project sought to use genetic improvement techniques, conventional and biotechnological, to
develop novel lines of woody ornamentals for the benefit of the industry. The Hardy Nursery Stock
(HNS) trade, which supplies trees, shrubs and herbaceous perennials to the gardening public and for
amenity planting, has a farm-gate value of some £300 million. The trade regards new improved
cultivars as its life blood, stimulating public interest and overcoming such problems as difficult
propagation, poor adaptation to certain soil types, excessive vigour and disease susceptibility, or
mediocre appearance. With the exception of a few genera such as Rosa and Camellia, where
industry or amateurs have conducted breeding programmes, little systematic improvement work has
been undertaken.

The work undertaken capitalised on success in developing techniques and material in HH1012SHN.
It focused on Acer, Buddleia, Rhododendron, Ribes, Sorbus and Syringa - all important subjects in
the HNS trade. It followed several strategies as appropriate - selection, breeding and
biotechnological.

Objectives

1. To propagate Acer platanoides selections identified as resistant to Verticillium wilt in HH1012SHN,

and now as part of a new EC Shared Cost Project, so that those best for resistance and
propagation can proceed to industry-funded trial; stocks of wilt-resistant Acer clones will be
available for industry exploitation and a seed orchard of resistant clones will be established.
2. To undertake genetic studies, principally in Buddleia, Ribes, Sambucus and Sorbus, to underpin
the complementary HDC-funded programme HNS18b, e.g. by allowing design of crosses to
pyramid useful characters; papers will be published on genetics of horticulturally important
characters in these genera and the information will allow the HDC-funded programme to proceed
more effectively.
3. To raise and identify selections of Rhododendron combining tolerance to lime with horticultural
merit, and to identify genetic markers for lime tolerance; promising selections will be passed to the
industry for assessment of commercial merit.
4. To raise tetraploid forms of rootstocks that would benefit from greater dwarfing ability and of sterile
interspecific hybrids that would be useful for further breeding; tetraploid selections will be made
available to the industry for evaluation as dwarfing rootstocks and fertile tetraploids will be used
as parents in the HDC-funded breeding programme.
5. To introduce a dwarfing gene into one or more HNS subjects by transformation; details of passing
transgenic material to the industry are uncertain at this stage.
6. To disseminate information to the industry and scientific community via publications and open
days and transfer breeding lines to industry; this technology transfer will be as outlined in the five
previous objectives.

The project was in direct accord with ROAME A of MAFF/Defra which recognises that genetic
improvement and novelty helps the industry meet the quality demands of the customers and retain
and extend market share by improving competitiveness.

CSG 15 (Rev. 6/02) 5

 Project Error: Reference source not found DEFRA Error:
title Error: Reference source not found project code Reference
source not
found
Results and findings

Acer

The work on propagation and resistance testing of Acer platanoides was a relatively small part of the
project. Ten clones of A. platanoides that have been consistently rated as resistant to Verticillium wilt
were supplied to the commercial nursery and seed merchant, Forestart, to establish a seed orchard
of wilt resistant clones. Twenty more clones were propagated for delivery in 2003. A licensing
agreement is being drawn up between HRI and Forestart.

In three batches of selections propagated by cuttings were tested for field resistance to wilt by placing
straw coated with mycelium of V. dahliae underneath the roots when planting in the inoculation plot.
In May 2000, one, two or three replicates of 113 selections that had been propagated by cuttings in
1998 and planted in the inoculation plot in 1999 were scored for leaf symptoms, vascular staining and
apical die back. Fifty showed few, if any, symptoms. Likewise in May 2001 and 2002, two more
batches, of 113 and 114 selections, were assessed, of which about 50 and 35 were classed as
resistant. In all 142 selections were tested, of which 24 appear to be consistently resistant.

To establish a well replicated trial, 34 selections were propagated in 2000. Cuttings were collected in
July for rooting under fog. Rooting varied from 5% for clone AP9 to 90% for AP5. Thirty clones
propagated well enough to include in the trial, which was planted in 2001. Wilt resistance was
assessed in 2002. None of the 5 clones of UK origin (AP clones) was reliably resistant. Of the
remaining 25, five were confirmed as resistant and will be offered to Forestart.

CSG 15 (Rev. 6/02) 6

 Project Error: Reference source not found DEFRA Error:
title Error: Reference source not found project code Reference
source not
found
Genetics

Our elderberry cultivar, Sambucus nigra 'Black Beauty' (formerly selection 528-180), that was
developed by collaboration between a previous MAFF/Defra project and a complementary HDC
project, has proved to be a popular plant for gardeners. It has UK sales of about 50,000 plants per
annum and has started to earn useful royalties from overseas. EU Plant Breeders Rights were
granted in Feb 2003 under the name 'Gerda'. Preparations were made for a second S. nigra
selection from Defra/HDC collaboration, 595-35, to be launched at the Chelsea Show in May 2003.
This was entered for EU Plant Breeders Rights in October 2002 under the name 'Eva' and will be
released as 'Black Lace'. It is homozygous for a semidominant gene for purple leaf colour and
homozygous for a recessive gene for laciniate leaf form.

The progeny from which 'Black Beauty' was selected included a small proportion of dwarf seedlings.
Some of these, 528-14, -17, -112, -175 and -157, were intercrossed in 2001 and 2002 to clarify the
genetics of this character, which may be under the control of two loci. The dwarf habit is a desirable
character for modern gardens. In 2002 some were also crossed with 'Black Lace' with a view to
combining the foliage of 'Black Lace' with dwarf habit. A few seedlings were raised from the 2001
crosses and are to be planted out this year, but germination was poor. Therefore a different
stratification regime is being used for the seed from the 2002 crosses. Also, where sufficient seed
was available, a comparison of seed treatments was made on seed of several of the crosses:
compared to pre-soaking with water for two days, some seed was also pre-treated with sulphuric acid
then soaked in water, and some was pre-treated with both sulphuric acid, soaked in water and then in
a gibberellic acid solution for approximately 16 hours.

Various progenies of Sorbus (the popular mountain ashes and whitebeams) which had been planted
in the field during the previous project were recorded for flower and habit characteristics. In progeny
631, from the cross of S. aucuparia 'Sheerwater Seedling' x S. cashmireana (pink flowers), all
seedlings inherited the pink flowers indicating that pink flower colour is a dominant character and that
S. cashmireana is homozygous for this character. Seedling 631-1 was particularly floriferous and
compact and was earmarked for crossing back to S. aucuparia, to introgress pink flower into
mountain ash. In progeny 633, from the cross S. aria 'Chrysophylla' x S. chamaemespilus (pink
flowers), all seedlings were likewise pink flowered and seedling 633-3 was chosen for backcrossing
to S. aria to produce a pink flowered whitebeam.

Some younger progenies of Sorbus aucuparia from crosses of 'Dirkenii' (yellow leaves), 'Pendula'
(weeping habit) and 'Fastigiata' (columnar habit) were planted in the field and assessed. These
segregate for green versus yellow leaves, as mentioned in the previous report, indicating a single
dominant nuclear gene for which 'Dirkenii' is heterozygous and a Short Communication on the
subject has been drafted. However none of the seedlings appears to have weeping or columnar
habit, indicating that these characters are recessive. Several yellow leaved seedlings from progeny
774 ('Dirkenii' x 'Fastigiata') were noted as being resistant to the scorching often caused by bright
sunlight. When these flower one will be chosen for backcrossing to 'Fastigiata' to confirm the
recessive nature of columnar habit and to combine it with yellow leaf. A couple of yellow-leaved
seedlings from progeny 778 ('Dirkenii' x 'Pendula') have likewise been earmarked for backcrossing to
'Pendula' in due course.

CSG 15 (Rev. 6/02) 7

 Project Error: Reference source not found DEFRA Error:
title Error: Reference source not found project code Reference
source not
found
In the case of flowering currant, Ribes sanguineum, there is little to report. Two progenies of
'Brocklebankii' (yellow leaves) x 'White Icicle' (white flowers) and the reciprocal were assessed in the
field. As previously reported, these both segregated for green versus yellow leaves. In the field the
yellow leaved seedlings proved to be low growing, and most of them scorched badly. All seedling
had pink flowers, indicating that white flower colour is recessive. Seedling 730-20 ('White Icicle' x
'Brocklebankii') could be worth backcrossing to 'White Icicle' to confirm the recessive nature of white
flower colour, but whether a low growing yellow leaved, white flowered Ribes sangineum would be
commercially useful is debatable.

A small collection of hardy and/or scented species of the popular climbing subject Jasminum was
established. The ploidy of the accessions was determined by cytometric analysis of young leaves. It
appeared to range from diploid to hexaploid. This knowledge can help us avoid making crosses that
would give rise to sterile triploids or pentaploids. Flower type, pin versus thrum, was also recorded.
Intra- and inter-specific crosses of J. beesianum (red flowers) and J. officinale (large scented white
flowers) were made in the glasshouse, within and between the pin and thrum morphs. Intraspecific
and even interspecific pollinations between morphs succeeded, and hybrid seed was set. In 2002
seed of these crosses was sown. Three seedlings germinated from the cross J. beesianum 4010 (P)
x J. officinale 4330 (T), one of which survives. Three seedlings germinated and two survive from a
cross J. beesianum 4331 (T) x J. officinale 'Affine' 4013. Three seedlings germinated and one
survives from a within morph pollination, J. officinale 'Affine' 4013 (P) x J. beesianum 4010 (P). It
would be worth repeating the crosses that set seed to confirm the success of the within morph
hybridisation and to increase the population size for selection of promising seedlings.

A small collection of Rosa (rose) was established. This includes various species reported to be
resistant to blackspot. We met with commercial rose breeders to discuss how genetic research could
help meet industry needs and some ideas were incorporated into our CSG7 for the follow-on project.

A visiting worker who was developing seven microsatellite primer pairs for olive tried them on DNA
extracts from a range of related ornamental genera in the Oleaceae. Two of them, EM013 and
EM090, revealed polymorphic loci in Syringa vulgaris and should prove useful for determining the
mode of inheritance, disomic or tetrasomic, in the tetraploid Syringa hybrids described in the section
on tetraploidy.
Rhododendron

Some 250 Rhododendron seedlings from the previous project, germinated in 1996, 1997 and 1998,
were planted in the field on soil of relatively high pH and in 2002 a preliminary assessment was made
of horticultural merit and resistance to lime-induced chlorosis. Three promising selections (96x4-1,
96x4-3, 96x4-51), short-listed for industry-funded trial, came from a cross between R. yakushimanum
'Exbury' x lime tolerant R. 'Cunningham's White'. In addition a batch of Pieris seedlings from the
open pollination of 'Debutante' were planted out and scored and several were noted as promising. In
addition 50 Rhododendron seedlings germinated in 2000 were raised for planting in the field in 2003
for assessment.

A replicated trial of 10 accessions that are reportedly lime tolerant was planted in the field in soil of
relatively high pH and scored for chlorosis. R. caucasicum 3707, R. williamsianum 1774, R. ciliatum
3056 and R. rubiginosum 1768 were among the clones that were least chlorotic and could be worth
using as parents in breeding for resistance. A further 12 reportedly resistant accessions were
propagated for field assessment.

CSG 15 (Rev. 6/02) 8

 Project Error: Reference source not found DEFRA Error:
title Error: Reference source not found project code Reference
source not
found
Ten microsatellite primer pairs designed in the previous project were optimised and tested on DNAs
extracted from a range of rhododendrons including various accessions of R. hirsutum and R.
ferrugineum, among them the parents of the mapping progeny. A further 24 clones were sequenced
from the library and from these and the previously sequenced clones twelve more primer pairs have
been designed, and will be optimised and tested for polymorphism. Four primer pairs from Japan
were found to be informative and another six were obtained for testing.

For assaying lime tolerance in vitro the autotrophic system using suncaps and sorbarods devised in
the previous project was modified to allow long term exposure of seedlings to successively increasing
levels of pH. To minimise problems of salt precipitation at higher pHs, the chelated iron in the
medium was FeEDDHA and some of the microelements were present as acetates. To accommodate
the extra growth, taller culture vessels than the usual honey jars were used. For investigating the
inheritance of lime tolerance only 23 seedlings were available from the cross of R. hirsutum PA54/1
(tolerant) x R. ferrugineum 3379 (susceptible), with three replicates per seedling. The pH of the
medium was progressively increased from 5.0 to 6.4 over several weeks and growth and chlorosis
was measured. This approach made more efficient use of limited replication than a traditional
factorial experiment. The performance of the different genotypes was recorded and in due course
could be compared with microsatellite data. Three genotypes were identified as tolerant. However,
the progeny size of 23 is too small for the detection of quantitative trait loci. To increase the progeny
size, the cross R. hirsutum PA54/1 x R. ferrugineum 3379 was repeated in 2002. The one extant R.
ferrugineum plant began to flower earlier than the R. hirsutum plants and it was cold stored to delay
flower development. Hybridisations were carried out in a warm, humid environment, which reportedly
is ideal for pollen tube development and ovule fertilisation. However, seed resulting from these
crosses failed to develop. Repeat hybridisations are planned for 2003.

A poster was presented for the “Rhodo 02” conference in Edinburgh, presenting our approach and
results and stimulating exchange of ideas with growers and scientists.

Tetraploidy

The tetraploidy studies were undertaken for a range of important genera, Buddleia, Sorbus, Syringa
and Sorbus, as well as two intra-generic hybrids, Pyrus x Malus and x Pyronia.

Over 100 seedlings were raised from the intercrossing in 2000 of two yellow-flowered selections from
the progeny resulting from crossing the tetraploid form of Buddleia globosa (yellow flowers), created
at East Malling, × the naturally tetraploid B. davidii ‘Nanhoensis Alba’ (dwarf habit, white flowers,
large inflorescences). These were planted in the field and, it is expected, will flower in 2003. We
hope some of these seedlings will combine yellow flower colour with large inforescences and the best
may be used for one more generation of crossing.

The tetraploid clone of Sorbus aria ‘Chrysophylla’ reported in 2001 unfortunately died, apparently
weakened by an endogenous contamination. In spring of 2002 the stock plant was heavily pruned
and re-potted, to encourage vigorous, healthy vegetative growth for stock re-initiation in 2003.

Three tetraploid clones of Sorbus aucuparia ‘Biessneri’ produced during the previous project were
multiplied together with diploid controls. Replicates of each were weaned and plants are being raised
for lining out for grafting to assess their dwarfing effect as rootstocks. Five more tetraploid clones
were produced using oryzalin and colchicine. These have been weaned into the greenhouse and will
be lined out.

CSG 15 (Rev. 6/02) 9

 Project Error: Reference source not found DEFRA Error:
title Error: Reference source not found project code Reference
source not
found
Tetraploid clones of the pear/apple selection PyMa2 (red leaves, columnar habit) were identified
cytometrically among clones produced during the previous project. Several attempts at weaning
approximately 15 clones failed (except for one plant), i.e. the shoots became dormant and eventually
died. Recently these were again rooted, treated with gibberellic acid (which it is hoped will help
prevent dormancy) and weaned into the glasshouse. Tetraploid clones of PyMa4 were worked on to
rootstocks to advance flowering, for backcrossing to pear. However these grafts failed and are to be
repeated.

Tetraploid forms of the sterile lilac hybrid Syringa vulgaris ‘Vestale’ × S. pinnatifolia were produced in
this and the previous project and two papers were published. In this project, two selections (104-1
and 104-32) were treated with colchicine and a number of the resulting clones were identified
cytometrically as tetraploids (14 clones of 104-1 and 4 clones of 104-32). Two selections, 104-18 (3
clones) and 104-25 (2 clones), produced in the previous project, were budded onto S. vulgaris
rootstocks to hasten flowering for use in crossing. Four of these clones flowered in spring 2002,
allowing preliminary observations of morphology and assessment of fertility. Flowers of 104-25 1Da1
and 104-25 03-3 were not obviously larger than the diploid control. However, flowers of clones 104-
18 2Bc1 and 104-18 1Cb2 were significantly wider than the diploid control (17.6 mm and 15.2 mm
respectively, compared to 13.2 mm) but not longer. Pollen viability of all four tetraploid clones was
tested on germination plates. Pollen of the diploid control and two of the four tetraploid clones was
not viable; however approximately 5% of the pollen of clones 104-18 2Bc1 and 104-25 03-3
germinated (compared to approximately 50% germination of pollen of the non-hybrid diploid S.
vulgaris 'Aurea'). Moreover, clone 104-18 2Bc1 set a few seed pods after pollination with S. vulgaris
'Aurea'. These results demonstrate that fertility has been restored to some of the tetraploid clones,
which will allow a second generation of crossing. More extensive observations and hybridisations, of
these and further tetraploid clones from selection 104-18 and selection 104-1, will be carried out in
project HH3709SHN.

In place of ploidy doubling of Sorbus aria 'Crysophylla', a seedling of Physocarpus 'Diabolo' x 'Dart's
Gold' with attractive tawny leaves was initiated in vitro and bulked up. Two ploidy experiments were
carried out in 2002 using various oryzalin levels and duration of exposure. Ploidy levels in shoots
derived from these experiments will be assessed in 2003 and it is hoped that any tetraploids will be
usefully more compact than the original diploid seedling.

The sterile intergeneric pear × quince hybrid (x Pyronia veitchii) was introduced in vitro. Good
proliferation rates were achieved following modification of the proliferation medium, and in vitro shoot
quality was improved following addition of supplementary calcium, allowing sufficient material to be
produced for ploidy doubling experiments using oryzalin. Initial experiments using a range of levels
of oryzalin in liquid medium (for 2 and 3 days) were unsuccessful, and there were losses of axillary
shoots caused by hyperhydry. In a subsequent experiment, liquid treatments for 2 days with
increased oryzalin levels were compared with incubation for 7 and 14 days on semi-solid medium
containing a range of oryzalin levels. Initially, six tetraploid clones were identified cytometrically, two
from a semi-solid oryzalin treatment, and four from the highest oryzalin liquid treatment. However,
subsequent re-analysis of replicate rooted plantlets prior to weaning confirmed tetraploidy in only the
four tetraploid clones from the liquid treatment. The shoots derived from the semi-solid treatment
may have been diploid/tetraploid sectorial chimaeras, the leaves originally assayed coming from a
tetraploid sector which was lost during subsequent sub-cultures. The four tetraploid clones and
diploid control have been weaned into the greenhouse.

CSG 15 (Rev. 6/02) 10

 Project Error: Reference source not found DEFRA Error:
title Error: Reference source not found project code Reference
source not
found
Following several attempts, clean shoot tips of three hybrids of Sambucus nigra 'Guincho Purple' × S.
racemosa 'Tenuifolia' were successfully initiated in vitro. One of these multiplied more quickly than
the other two allowing sufficient material to be obtained for a ploidy doubling experiment using
oryzalin. Ploidy levels have not yet been assessed in shoots derived from these experiments.

Transformation

Preliminary experiments were carried out with B. davidii ‘Nanhoensis Alba’ to optimise regeneration of
leaf pieces in vitro and a transformation experiment was carried out which failed to yield any
transformants. Following this, further experiments were performed to improve regeneration before
continuing with transformation experiments. Various levels and combinations of cytokinin (BAP, TDZ)
and auxin (NAA, IBA) were tested. Other variables optimised were carbohydrate source (sucrose,
fructose, glucose, maltose), basal media (MS and DKW) and gelling agent (Sigma phytagel, agargel
and agar). Routinely, between 70% and 80% of leaf pieces regenerate shoots on the following
regeneration medium: DKW basal medium, 3% sucrose, BAP 1mg/l, IBA 0.1mg/l, solidified with
Sigma agar (0.8%) or Sigma agargel (0.6%).

B. davidii cultivars 'Black Knight' and 'Pink Delight' were initiated in vitro and bulked up. In a
regeneration experiment using the protocol established for Nanhoensis Alba, approximately 75% of
'Black Knight' leaf explants regenerated shoots.

For both 'Nanhoensis Alba' and 'Black Knight', a range of kanamycin concentrations were assessed
for inhibition of callus formation, prior to use of kanamycin for selection in transformation
experiments. For both cultivars, callus formation was inhibited on agar or agargel with 100 mg/l
kanamycin (K100).

Eight transformation experiments were carried out, using super-virulent disarmed Agrobacterium
strain EHA101 (or variant EHA105, depending on the antibiotic selection required). T-DNA constructs
included the following constitutively expressed genes: RolC (dwarfing) or uid A (GUS marker gene),
and nptII for selection of transformants on kanamycin or paromomycin. Six experiments were
performed with kanamycin selection and two with paromomycin selection.

In the first experiment, before the result of the kanamycin range experiment was fully known, 460
'Nanhoensis Alba' (NA) explants were co-cultivated using the RolC construct (PY09), with selection
initially on K75 (agargel) before transferring to K100 (agar). Nearly 70 putative transformed shoots
(transformants) were regenerated, and cultured on proliferation medium with K100. Control shoots
bleached and died on this medium, but approximately 20 of the putative transformants remained
green. However, PCR analysis failed to confirm presence of transgenes in these shoots. It was
hoped that maintenance on high levels of kanamycin would favour growth of transformed tissue and
prevent growth of un-transformed tissue; however following successive sub-cultures on this medium
all the shoots eventually died. It is suspected that the shoots were chimaeric, i.e. a small proportion
of the tissue may have been transformed, insufficient to detect by PCR but enough to allow the non-
transformed tissue to survive on selection.

In a second experiment NA explants were co-cultivated with EHA101 and EHA105 containing both
PY09 and pSCV1.6 (with uid A marker gene construct). Selection was increased to K100 (agar was
used in this and all subsequent experiments) throughout and the number of putative transformants
regenerated was much reduced. From a total of 650 explants, six putative 'PY09' and two putative
'pSCV1.6' transformants were regenerated. However they failed to survive. For experiment 3, NA
was co-ultivated with pSCV1.6 and none of the explants callused on K100.

CSG 15 (Rev. 6/02) 11

 Project Error: Reference source not found DEFRA Error:
title Error: Reference source not found project code Reference
source not
found
For the fourth experiment, transformation of a different cultivar ('Black Knight') was attempted, and
the binary vector pSCV1.6 was compared with a second binary vector (pFAV RB7) which also has a
constitutively expressed GUS gene construct in the T-DNA. After selection on K100, four putative
transformants were regenerated from 316 'pSCV1.6 explants' and three from 409 'pSCV FAV RB7
explants'. Again, these failed to survive on kanamycin selection..

A fifth experiment, with cultivar NA, was started before the results of experiment 4 were known, and
again binary vectors pSCV1.6 (446 explants) and pSCV FAV RB7 (376) were used. This time, before
shoots were regenerated, a histochemical assay was performed on calli growing in contact with the
selection medium. In this assay, cleavage of the substrate 'X-Gluc' by the GUS enzyme yields a blue
product. Eight calli were tested from each set of explants, only a bit of each callus being sampled
where possible. The edges and basal regions of two 'pSCV1.6' and one 'pSCV FAV RB7' calli
appeared blue, strengthening the view that chimaeric callus and shoots were being regenerated on
kanamycin. In an attempt to surround each callus with selection medium and prevent regeneration of
chimaeric tissue, remaining calli were transferred to liquid medium with K75 (a level in liquid shown to
inhibit callus formation). However, although control calli regenerated shoots on liquid medium without
kanamycin, transformed shoots were not regenerated from selection medium.

Transformation experiment 6 (cultivar NA, binary vector pSCV1.6) compared selection on kanamycin
(75 mg/l) in semi-solid medium initially (195 explants) before transfer to liquid selection medium, with
kanamycin (75 mg/l) in liquid medium (540 explants). On semi-solid medium small calli appeared
and were transferred to liquid medium, K75. Eighty shoots were regenerated and transferred to semi
solid proliferation medium K100. None of the calli or shoots survived. Of the explants in liquid
selection medium only one piece regenerated green shoots. These were transferred to semi solid
selection medium but died.

To try and increase the chances of obtaining non-chimaeric transformed calli and shoots, two
transformation experiments were performed using paromomycin as the selective agent instead of
kanamycin. Paromomycin causes a higher level of misreading during RNA transcription in plant cells,
and can be a better selective agent than kanamycin. First, a range of paromomycin levels were
tested to determine the optimum concentration for inhibition of regeneration of cultivars 'Nanhoensis
Alba' and 'Black Knight', both on semi-solid medium and liquid medium. On semi-solid medium, 5
mg/l allowed minimal callus formation on approximately 75% of explants and shoots regenerated
from the explant surfaces (but not from the callus or from tissue in contact with the medium) of
approx. 70% explants. On 10 mg/l, callus formation was completely inhibited and 25% of explants
regenerated shoots distant from the selection medium. Inhibition in liquid medium was at lower
concentrations: at 5 mg/l no calli formed, with approximately 5% explants forming a few shoots but no
callus, and at 10 mg/ all explants died. In transformation experiments (cultivar NA, binary vector
pSCV1.6) using paromomycin selection no calli formed in contact with the selection medium.
Selection strategies using semi-solid medium were 5 mg/l paromomycin for one week, then transfer
to 10 mg/l paromomycin, compared with 10 paromomycin throughout. From approximately 1600
explants no calli were obtained.

Given that regeneration is highly efficient in the cultivars chosen for transformation experiments, we
expected to obtain transformed shoots following co-cultivation with Agrobacterium. The lack of
success following extensive experiments with different cultivars and selection strategies based on
resistance conferred by the nptII gene, suggests that the transformation strategy for Buddleia needs
to be revised. Either, selection based on resistance to aminoglycoside antiobiotics is ineffective in
Buddleia, or the cells that are being transformed are not those that subsequently regenerate to form
shoots (although when kanamycin was used for selection it is possible that chimaeric shoots may

CSG 15 (Rev. 6/02) 12

 Project Error: Reference source not found DEFRA Error:
title Error: Reference source not found project code Reference
source not
found
have regenerated). Use of non-antibiotic and marker-free selection systems, in either Buddleia or
another model subject, is to be investigated in project HH3709SHN.

Technology transfer

Wilt-resistant clones of Acer platanoides were passed to the industry for establishment of a seed
orchard. The Sambucus nigra variety, Black Lace, a result of collaboration between the Defra
genetics programme and the HDC breeding programme, was prepared for release in 2003.
Selections of Rhododendron have been identified for trial for garden merit and lime tolerance.
Breeding lines of Sorbus have been identified for use as parents. Tetraploid lines of Buddleia and
Syringa may be worth trialling directly and will certainly be valuable breeding parents.

Papers were published describing work on tetraploidy and microsatellites, manuscripts prepared on
Jasminum and Sorbus and several presentations were given.

Discussion and Implications

The Acer work has led to 10 resistant clones passed to the industry, another 20 propagated for
transfer and 5 more to be offered. The intended seed orchard will be a good way to exploit this
material but there could be a case for certain resistant clones being tried as clonal rootstocks. We
imagine the industry would undertake this work either directly or via an HDC project. Sometimes it is
possible to re-isolate V. dahliae from apparently healthy plants indicating tolerance rather than
resistance. It would be interesting to dissect the biochemical and molecular basis of resistance e.g.
by screening seedlings from controlled crosses.

The genetics work was an area in which particular progress was made. The success of Sambucus
nigra 'Black Beauty' and the imminent release of 'Black Lace' demonstrates how the research to
clarify nature of inheritance of particular characters leads, in conjunction with a breeding programme,
to new varieties benefiting the UK trade and gardener, and also earning royalties from overseas. The
knowledge gained and breeding lines identified in Sorbus, and perhaps Ribes, should lead to similar
outcomes. The Jasminum studies should allow a small crossing programme to be planned to
produce a showier winter jasmine and a brighter summer jasmine. Also, the initiation of a small
collection of Rosa could lead in due course to collaborative work on this important horticultural genus.

The Rhododendron crossing and screening work has identified several selections worthy of trial for
lime tolerance and horticultural merit. The mapping work suffered from the ever decreasing progeny
size, which meant it was unrealistic to detect molecular markers for components of lime tolerance.

The tetraploidy research was another area in which great progress was made. The intercrosses of
the fertile hybrids derived from crossing Buddleia davidii with the induced tetraploid form of Buddleia
globosa promise to recombine the large flower panicles of the former with the yellow colour of the
latter to produce a plant with as much appeal as Sambucus 'Black Lace'. The measurements on the
induced tetraploid forms of the hybrids of Syringa vulgaris x S. pinnatifolia show that, in at least one
clone, the flowers are indeed larger and the pollen fertile - paving the way for a range of large-
flowered hybrids with pinnate leaves. The Sorbus tetraploids need to be assessed as dwarfing
rootstocks and the tetraploid pear x apple hybrids, if fertile, need to be crossed to tetraploid pears to
introgress red leaves and columnar growth habit into that species. The production of the tetraploid
form of the pear x quince hybrid opens up exciting opportunities in breeding pear rootstocks that
combine dwarfing and easy rooting (quince characters) with compatibility and resistance to lime-

CSG 15 (Rev. 6/02) 13

 Project Error: Reference source not found DEFRA Error:
title Error: Reference source not found project code Reference
source not
found
induced chlorosis (pear characters). If the tetraploids of the intraspecific Sambucus hybrid is
confirmed it will be selfed in the hope that some seedlings will be homozyous for purle leaf (from S.
nigra 'Guincho Purple' and for dissected foliage and dwarf habit (from S. racemosa 'Tenuifolia').

The transformation work on Buddleia davidii, which regenerates readily, has proved to be more
challenging than expected. Despite an exhaustive and well-planned approach, no transformed
shoots were obtained using selection based on resistance to aminoglycoside antibiotics. However,
valuable information was obtained and in project HH3709SHN work will concentrate on alternative
selection methods including 'clean gene' transformation approaches, possibly with another model
subject, to be decided.

Options for further work

A CSG7 has recently been agreed with Defra for a five-year continuation project. The new project
will build on the success of HH1026 and will concentrate on developing breeding lines and tools. It
has the following objectives:

 To clarify genetics of horticulturally important characters in Sambucus, Buddleia, Sorbus and

Jasminum to inform breeding programmes.
 To identify sources of resistance to blackspot, and perhaps aphids, in Rosa, and transfer breeding
lines to a commercial breeding programme, with a view to subsequently developing a LINK
research programme.
 To demonstrate that tetraploidy has restored fertility to interspecific hybrids of Syringa and Pyrus x
Malus and check for self compatibility and mode of inheritance, and complete backcrossing
programme of tetraploid Buddleia globosa to B. davidii to introgress yellow flower colour.
 To determine if tetraploidy in the climbers Clematis and honeysuckle (Lonicera), and shrubs such
as Physocarpus induces more compact habit, useful for small gardens.
 To develop transformation methods for hardy ornamentals, assessing the xylose isomerase and
MAT vector systems in Buddleia davidii, and establishing regeneration methods for two additional
subjects, for use in subsequent transformation experiments and technology transfer to the
industry.

CSG 15 (Rev. 6/02) 14

 Project Error: Reference source not found DEFRA Error:
title Error: Reference source not found project code Reference
source not
found
Actions arising - IP and Technology transfer

European Plant Breeders Rights have been obtained for Sambucus nigra 'Black Beauty', under the
name 'Gerda', and are being sought for 'Black Lace', under the name 'Eva'. They arise from joint
Defra/HDC funding and it had been agreed between HRI and HDC that the royalties should be
shared between HRI and HDC and also that the royalties would be used primarily to fund a
continuation of the HNS breeding programme. However, although we submitted proposals to HDC in
2001 for further HNS breeding, these have not been approved. So the royalty-funded breeding
programme to which we are passing selections from the Defra-funded work is HRI owned. As
explained, selections of Sorbus have been identified for crossing and selections of Rhododendron
have been identified for trial.

Several papers were published describing aspects of the work and several presentations made.
These are detailed below.

Publications and presentations

Publications

ROSE J. (2001). Rose 'fingerprinting'. In: The Royal National Rose Society Historic Roses Group
Journal, 21: 16-?d.
ROSE J.B., KUBBA J. & TOBUTT K.R. (2001). Induction of tetraploidy in Buddleia globosa. Plant
Cell, Tissue and Organ Culture 63:121-125.
ROSE J.B., KUBBA J. & TOBUTT K.R. (2001). Chromosome doubling in sterile Syringa vulgaris × S.
pinnatifolia hybrids by in vitro culture of nodal explants. Plant Cell, Tissue and Organ Culture
63:127-132.
ROSE J.B., KUBBA J. & TOBUTT K.R. (2001). Induction of tetraploids for breeding hardy
ornamentals. Acta Horticulturae, 560:109-112.
DE LA ROSA R., JAMES C.M. AND TOBUTT K.R. (2002). Isolation and characterization of
polymorphic microsatellites in olive (Olea europaea L.) and their transferability to other genera
in the Oleaceae. Molecular Ecology Notes 2:265-267.

Presentations

Rose J.B., Kubba J.& Tobutt K.R. 'Induction of tetraploids for breeding hardy ornamentals'. Poster at
the 4th International Symposium on In vitro Culture and horticultural Breeding, Finland, 2-7 July
2000.
Tobutt K.R. 'Woody plant collections at HRI-East Malling - management and use'. Invited lecture to
PlantNet 2000, Royal Botanical Garden, Edinburgh, May 2000.
Tobutt K.R. 'Developing markers for use in breeding perennials'. Talk to BBSRC visitors, HRI-
Wellesbourne, May 2000.
Tobutt, K.R. 'Hardy Nursery Stock Improvement'. Talk and Exhibit at VIP Day, HRI East-Malling,
September 2002.
Wilson, F.M. Marks T.R., Bates C.H.A., James C.M., Meyers P.E. Rose J.B. & Tobutt K.R. (2002).
'Breeding and selection for lime-tolerant rhododendrons'. Poster presented at "'Rhodo'02'
(Rhododendrons in Horticulture and Science)", Royal Botanic Garden Edinburgh, 17-19 May
2002.
Wilson, F.M. 'Hardy Nursery Stock Improvement'. Exhibit at Open Day, HRI-East-Malling, September 2002.

CSG 15 (Rev. 6/02) 15