[To contribute to this conference, send your message to biotech-room2@mailserv.fao.org. For further information on the Electronic Forum on Biotechnology in Food and Agriculture see Forum website.
Note, participants are assumed to be speaking on their own behalf, unless they state otherwise.]

-----Original Message-----
From: Biotech-Mod2
Sent: 25 November 2003 11:49
To: 'biotech-room2@mailserv.fao.org'
Subject: 37: Successful use of MAS - cereal crops

This message is from Dr Kevin Williams, Molecular Geneticist at the South Australian Research and Development Institute. I am involved in developing marker-trait associations in several crops, both major and minor, and also conduct MAS for the development of disease resistant germplasm in my laboratory.

I would like to give a view on MAS from a country where the technology is well advanced in the major cereal crops wheat and barley. Australia has industry-funded national molecular marker programs for these crops, which fund phenotyping and genotyping of segregating populations and also contributes to the validation of markers and implementation of MAS in many breeding programs. There is also the Cooperative Research Centre for Molecular Plant Breeding, which includes South Australian, Victorian and Western Australian agriculture departments and universities, as well as the International Maize and Wheat Improvement Center (CIMMYT). These groupings have been instrumental in ensuring that the large amount of research conducted (see special issues of Australian Journal of Agricultural Research vols 52:11 and 12 and 54:11 and 12 for examples) is actually translated into outcomes for plant breeders. Many tens of thousands of marker assays for a wide range of loci are conducted in wheat and barley breeding programs annually. [It is obviously important that MAS research eventually results in useful germplasm for the breeder. Are figures available (precise or estimated), for any country, on the amount of hecterage cultivated with varieties developed using MAS - referring to Message 8 (November 19) of C.R. Bhatia who suggested that the area covered by the cultivar(s) developed using MAS would be the ultimate indicator of the utility of MAS in plant breeding...Moderator].

Many of the concerns of previous correspondents about the limitations of MAS are valid and, although I am not a plant breeder, I would like to contribute several observations from six years of active research in this field with both major crops and also with breeders of minor crops wanting to access the technology cheaply.

1. Molecular markers have many uses in addition to MAS. They can be used to verify the identity of lines, investigate heterogeneity and assess diversity of germplasm. Anonymous markers, with low development costs, can be used for these purposes.

2. Molecular markers have the greatest benefit when selecting for loci controlling traits of high value which are difficult and expensive to phenotype. Root nematodes are a classic example in Australia. Phenotyping is time-consuming and variable, so molecular markers have been widely used in wheat and barley breeding, and barley cultivars bred using MAS for nematode resistance are being commercialised. Interestingly, as the highest cost of MAS in plants is DNA extraction and extra markers can be assessed at low marginal cost, once the DNA is available, breeders will also often ask for additional marker assays for a wide range of loci with lower value, but still important in selection decisions. Importantly, MAS can produce outcomes not easily achieved by traditional breeding, such as strategically selecting for resistance to exotic pests, pyramiding of race-specific resistance genes for potential durable resistance, and selection for multiple traits from single plants such as BCF1's rather than later segregating populations. [Gene pyramiding is the combination of target genes identified in different parents into a single genotype...Moderator].

3. Intuitively, recombination between markers and the gene of interest seems important. However, in practice we have found that lack of marker polymorphism is a far greater limitation than the distance between marker and gene. Many theoretical studies are done using wide crosses to improve the polymorphism level, but a breeding program will often have closely related material showing low polymorphism. The feedback that I receive from breeders is that any marker within 10 cM is useful at present, as their main aim is to increase the frequency of favourable alleles in a large population at the lowest cost.

4. MAS need not be expensive. The recent huge increase in the public availability of genomic resources such as expressed sequence tags (ESTs) for a wide range of species has greatly reduced the cost of the development of markers such as simple sequence repeats (SSRs) which can often be identified in model species and transferred to minor crop species. [SSRs are also known as microsatellites...Moderator]. MAS itself need not use sophisticated infrastructure. SSR analysis for MAS in my laboratory uses a thermal cycler, vertical gel tanks, UV transilluminator and camera, total cost US$10,000. By far the largest operating cost is labour, and although I have access to automated systems or could outsource genotyping, a small team of technical officers is still the most efficient set-up for the varied demands of MAS throughout the season.

In summary, in contrast to those who question the effectiveness of MAS, I have no doubt that MAS has the potential to increase the rate of genetic gain when used in conjunction with traditional breeding, and the adoption of MAS by cereal breeders in Australia and the subsequent commercialisation of cultivars bred using MAS is testimony to this.

Dr Kevin Williams
Senior Research Scientist-Molecular Genetics
South Australian Research and Development Institute
PO Box 397
Adelaide SA 5001
Australia
Phone +61 8 8303 9369
Fax +61 8 8303 9393
williams.kevin (at) saugov.sa.gov.au
http://www.sardi.sa.gov.au/field_crops/pathology_quarantine/cereal_molecular/cereal_molecular.htm

-----Original Message-----
From: Biotech-Mod2
Sent: 25 November 2003 11:59
To: 'biotech-room2@mailserv.fao.org'
Subject: 38: Re: When the marker is the gene

This is Hayde Galvez again from the Philippines.

This is in reply to message 36 (November 24) from Adilson Mota about the necessity for 'complete linkage' of the marker to the gene for effective use in MAS breeding.

I strongly agree with this. In fact this should be a general rule for MAS application in animals, crops etc. However, complete linkage means the marker is within the gene sequence itself and getting (isolating) the whole gene sequenced/characterized is not a 1-2-3 process - map-based or direct PCR-based isolation etc. The developing countries may not be financially ready to do this themselves. So again [see message 5, November 18...Moderator], I maintain my opinion that international (or external fund source) collaboration is needed to develop markers for MAS for the developing countries, i.e. shuttle research. Targeting major genes (genes involved in major resistance/biochemical pathways - e.g. resistance gene analog-derived) may give more return as the markers can be applied across a wide range of populations (even species), environments and pathogens.

Another point is that the marker may not be "completely linked (0 cM)" to the gene/trait but still effective for MAS breeding. As actual gene identification and isolation/sequencing takes much longer time than in theory, a certain degree of recombination between the marker and the gene can be acceptable as long as the breeder takes this into consideration in the selection process. It is still much much better than using a pure morphological marker that may end up totally dis-associated from the desired trait or target gene. But how much 'dis-linkage' can be allowed that will still give REAL economic advantage of employing MAS in breeding programs remains to be studied, worryingly in a case-to-case basis. Anyone who can give relevant references on this aspect is much appreciated.

Hayde F. Galvez
University Researcher
Institute of Plant Breeding
University of the Philippines Los Baños
College, Laguna
Philippines
hgalvez_4031 (at) yahoo.com

-----Original Message-----
From: Biotech-Mod2
Sent: 25 November 2003 12:04
To: 'biotech-room2@mailserv.fao.org'
Subject: 39: Re: When the marker is the gene

Regarding message 36 (November 24) from Adilson Mota:

When the marker is the gene, then the most cost effective way of diffusing the gene is by artificial insemination using sires homozygous for the favourable allele that also have merit for other traits of economic importance.

Don Nicol,
Brisbane
Australia
dnicol (at) b022.aone.net.au

-----Original Message-----
From: Biotech-Mod2
Sent: 25 November 2003 12:17
To: 'biotech-room2@mailserv.fao.org'
Subject: 40: Re: MAS and animal breeding

I am Miguel Toro from Spain. I work on quantitative genetics and QTL detection.

With repect to message 33 (November 24) from Juan Chavez, I cannot see why MAS will have a "less harmful effect on reproductive traits because of more precise targetting to the desirable alleles we are looking for". Is there a general consensus on this opinion? I would think just the opposite. When people find a big QTL at intermediate frequency in a breed selected for growth, they are usually reluctant to use it because they are afraid that the gene will have adverse effects on fitness (if this was not the case, why is the gene still segregating despite selection pressure favouring it?). On the other hand, a very 'precise targetting' could be transgenesis with not very good results with respect to fitness...

I will rather support the more classical ideas on selection and fitness. Both standard selection and MAS could deteriorate fitness. There are two courses of action. The first is to evaluate also the fitness traits and then calculate the effect of markers on both production and fitness traits. The other action (also valid in classical selection and MAS) is to keep the inbreeding below certain levels. There is a very well developed theory on optimisation of selection response with restricted inbreeding.

With respect to a second point by Juan Chavez: "Within our "Criollo" cattle there are some valuable genes related to traits that could be in demand from countries with well developed livestock industries (i.e. dairy cattle). These genes could be identified and concentrated by MAS into a breed developed from Criollo cattle and become a "live library" of desirable genes". I do not undertand the point. There COULD BE valuable genes.., but the problem is that we do not know them. In such circumstances, I cannot see how to accummulate them in a live library.

It seems that in plants, genetic resources from developing countries have been useful to improve commercial populations but in animals I think that the situation is not as clear. Is there examples? [This seems to me to be also the case. Use of genetic resources from developing countries to improve commercial populations in developed countries seems to have been more limited in animals. One example, would be the introduction of genetic material from the fat, but highly prolific, Meishan pig (from China) into lean commercial pig breeding lines...Moderator].

Miguel Angel Toro Ibañez
Departamento de Mejora Genética Animal
Instituto Nacional de Investigación y Tecnología Agraria y Agroalimentaria (INIA)
Carretera La Coruña km. 7
28040 Madrid
ESPAÑA
Telf: 34 913476807
Fax: 34 913572293
e-mail: toro (at) inia.es

-----Original Message-----
From: Biotech-Mod2
Sent: 25 November 2003 13:53
To: 'biotech-room2@mailserv.fao.org'
Subject: 41: Crops - India

This is from Dr Kuldeep Singh. I am primarily a plant breeder and now use the molecular biology techniques for rice and wheat improvement. We at Punjab Agricultural University, Ludhiana, are among the first who used MAS for pyramiding bacterial blight resistance genes in two of our popular rice varieties.

The conference has raised many issues and I would like to talk on a few issues and these are based on my experience and knowledge in the context of Indian agriculture.

During the last decade the emphasis was on training manpower and the Rockefeller Foundation, Asian Development Bank (ADB) and international agricultural research centres (especially the International Rice Research Institute (IRRI), in relation to rice) primarily contributed to this. Now I feel sufficient people are trained and who can train more people in molecular marker technology.

The feasibilty of MAS has been demonstrated, especially for pyramiding disease resistance genes, not only at one place but at several institutes in India like PAU, CRRI, UAS, IGAU, TNAU (as was mentioned by Dr R. Sridhar (message 35, November 24) as well). This was achieved more or less independent of plant breeders and mostly in well adapted varieties. Plant breeders simultaneously came up with new varieties that may be higher yielding, hence the pyramid lines may not find their way to the farmers` fields (although the newly developed pyramid lines yield at par with the recurrent parents). The MAS was adapted for disease resistance that plant breeders feel are relatively simple traits to work with and even single genes can do the job (an issue that is debatable).

The question now arises how to make MAS operational in the developing world. My opinion is:

1. Since MAS is expensive and breeding programmes are mostly funded by the local governments, the federal governments may start some pilot MAS projects with committed funding. In India, for example, the Indian Council of Agricultural Research (ICAR) can identify a few universities/institutes and identify the key traits, like quality, for which MAS can be started. This must be an integral part of the breeding programme and may not be the backcross programme. Once successful it can be extended to other institutes and finally the local governments can take over.

2. Breeders are not much thrilled about MAS for simply inherited traits, and not many QTL (especially the productivity related ones) with tightly linked markers are available. This I believe may take some more time, especially the productivity related QTL from the wild species germplasm, to become available to breeders. Any QTL mapped in one environment will need validation before it is involved in MAS.

3. QTL for abiotic stresses are available and MAS can be used for their incorporation but unfortunately the areas where abiotic stresses are more prevalent are poor as well and it may not be possible for institutes from this regions to begin MAS without funding from the federal governments or the international funding agencies.

Finally, as a student of plant breeding, I foresee a good scope for MAS for several traits.

Dr Kuldeep Singh
Molecular Geneticist
Dept. Genetics and Biotechnlogy
Punjab Agricultural University
Ludhiana 141 004,
India
Tel. +91-161-243 30 81 (R)
kuldeep35 (at) yahoo.com