A review on the efficacy tests and risk analyses conducted on Chondrostereum purpureum , a potential biocontrol agent , in Finland

Hardwood sprouting is a problem in forest regeneration areas, under electric lines, on roadsides and railways. In Finland, isolates of Chondrostereum purpureum were screened by field experiments for their efficiency to control sprouting. The proportion of dead stumps with the best isolates exceeded 80% on birch (Betula pendula and B. pubescens), and C. purpureum was also found to affect the sprouting of aspen (Populus tremula) and rowan (Sorbus aucuparia). The risks of C. purpureum based biocontrol were evaluated by population genetic analysis. It showed that C. purpureum is a geographically undifferentiated species that does not reproduce clonally. The risk of infection of non-target trees was found to be highest in early spring. These findings suggest that the risks of using C. purpureum in biocontrol are small.


Introduction
Fast and abundant stump sprouting of broadleaved trees causes problems in conifer regeneration areas, where undesirable vegetation competes for light, water, nutrients, and space with more valuable conifers.Hardwood sprouting also causes problems under power transmission lines, as well as at roadsides and railways.Downy and silver birches (Betula pubescens Ehrh.and Betula pendula Roth.) are the most common hardwood species in Finland, and both of them regenerate vegetatively and sprout rapidly after cuttings.
Chondrostereum purpureum (Pers.ex Fr.) Pouzar is the causative agent of the silver-leaf disease on fruit and ornamental trees (Brooks and Moore, 1926;Peace, 1962;Setliff and Wade, 1973).In addition, as an eco-Mean diameter of the treated stumps was relatively large, about 30 mm (Vartiamäki et al. (2008a(Vartiamäki et al. ( , 2009a)).It is, however, known that stumps of older birch trees sprout less vigorously than stumps of younger trees.Johansson (1992) reported that 40-65% of stumps with diameter of 81-131 mm sprouted in mixed stands of downy and silver birches five years after cutting, and Andersson (1966) reported that 30-80% living stumps of 25-50-year-old downy birch sprouted two years after cutting.On the other hand, stump-sprouting ability of B. papyrifera has been shown to increase with the size of the stump up to an intermediate diameter (Perala and Alm, 1990).Therefore, it would be important to test further the efficacy of C. purpureum treatment on birch stumps with very small diameter.
In Finland birch is by far the most important tree species to be controlled.However, also other hardwood species cause problems locally.Therefore, it would be desirable if the control agent would be efficient also against them.It is known that aspen (Populus tremula L.), for example, is able to inhibit the growth of C. purpureum to a greater degree than some other broadleaved tree species (Wall. 1990;Becker et al., 1999;Harper et al., 1999;Pitt et al., 1999).Hamberg et al. (2011a,b) tested the efficacy of C. purpureum against rowan (Sorbus aucuparia L.) and aspen.Both of these species seemed to be sensitive to the treatment, although to a lesser extent than the birches.However, it should be noted, that the treated saplings were smaller in diameter (ca.17 and 18 mm respectively), and that the C. purpureum isolate used in these experiments was relatively weak, and is not among the potential strains to be selected for the commercial preparation.

Timing of control
The second question in using a biocontrol product is to find the best time for controlling.This was tested by cutting birch seedlings and treating them with a biocontrol product throughout a summer.The results showed that the best control level in Finnish conditions was obtained when the treatment was conducted in late May, June or July (Vartiamäki et al., 2009a).The treatment was less effective towards the end of the growing season.This is in accordance with the view that the resistance of birch against C. purpureum is greatest in the spring and decreases towards the midsummer, after which it increases again (Wall 1991).Also Dumas et nomically important pathogen, the fungus attacks many deciduous trees, shrubs and fruit trees throughout the temperate zone, and causes white rot on them (Rayner and Boddy, 1986).
C. purpureum is a rather common pathogen also on hardwoods in Fennoscandia, where it is found e.g. on Betula, Populus and Alnus species (Mazelaitis, 1976;Kauppila and Niemelä, 1986).The basidiospores of C. purpureum infect only fresh wounds (Spiers and Hopcroft, 1988) and no infection occurs on unwounded trees.
Studies on C. purpureum as a biocontrol agent for unwanted broadleaved trees and shrubs have been carried out in Canada and Netherlands during the recent decades, and commercial preparations are already available (Scheepens and Hoogerbrugge, 1989;Wall, 1990;Wall 1994;Gosselin, 1996;Dumas et al., 1997;Jobidon, 1998;Shamoun and Hintz, 1998;Harper et al., 1999;Pitt et al., 1999;De Jong, 2000).The use of these preparations in Finland makes the risk of introducing alien pathogens; hence investigations on local strains of C. purpureum were started.The purpose of this review is to describe the developments in analyzing the potential of C. purpureum as a mycoherbicide in Finland.

Search for an efficient biocontrol strain
The first task in developing any biocontrol product is to find an efficient isolate to be used in commercial product.For satisfactory control of hardwood sprouting, reduction of at least 80% of sprouts are needed as their density may be extremely high.Vartiamäki et al. (2008a) screened 21 C. purpureum strains for several enzymatic activities and biomass production ability.Based on these analyses, two groups of isolates were selected for field trials.The first group included isolates with high enzymatic activity and biomass production whereas the isolates of the second group showed lower enzymatic activity and biomass production.In field trials considerable differences were observed between the isolates.The best isolates killed ca.90% of birch stumps during the first two years after the treatment, and reduced the mean number of sprouts by more than 80%.The above-described grouping of isolates did not predict their performance in the field, but some enzymatic activities, the laccase activity in particular, seemed to correlate with the biocontrol efficacy (Vartiamäki et al. 2008a).Hardwood sprout control al. (1997) demonstrated on aspen, that a higher number of sprouts were observed after cuttings in late September than in late August.

Biosafety aspects
As already stated above, the use of pathogenic organisms is potentially dangerous.Therefore a good body of information about the population biology of the control fungus should be collected before its spread to nature.This information was gathered by conducting a population study on the genetic diversity of C. purpureum.
The populations showed only a very low degree of geographical differentiation (Vartiamäki et al., 2008b).Therefore, either a gene flow occurs naturally throughout Finland and the Baltic countries, or the fungus has such a high population size that the frequencies of selectively neutral markers studied change extremely slowly.Therefore, distributing any local genotype of C. purpureum as a biocontrol agent should not lead to introduction of novel genes or genotypes in Finland, and the use of any native genotype as a biocontrol agent should therefore be safe.
The population analysis conducted by Vartiamäki et al. (2008b) also suggested that C. purpureum has no asexual reproduction, a result in accordance with previous studies showing no support for clonal reproduction in C. purpureum (Gosselin et al. 1996, 1999, Ramsfield et al. 1999).Thus, in the lack of clonal propagation the genes of the biocontrol strain will mix with those of other individuals in the same population as soon as basidiospores are produced, and the original genotype will not spread outside the treated areas.
However, if the high pathogenicity of the control isolate is due to a single allele, the frequency of this allele might increase.Thus, if the sprout control by C. purpureum becomes a common practice, it would be important to figure out the genetics of the pathogenicity of C. purpureum, and follow the development of genetic diversity of this fungus also in the future.
Another problem might be the infection of nontarget trees.Vartiamäki et al. (2009b) tested this risk by pruning and inoculating birches in one month intervals from spring to autumn, and showed that pruning causes highest risk for C. purpureum infection in early spring.However, the infection in neighboring trees in control plots was very low throughout the experiment.This indicates that the risk is not very high, but it cannot be ruled out that the basidiospores in favorable environmental conditions might infect freshly wounded non-target trees, although a long distance spread of spores has been shown to be unlikely (De Jong et al. 1990a).De Jong et al. (1990b) concluded that there is a little risk that added C. purpureum would significantly affect non-target, unwounded forest and horticultural trees further than 500 m from the target area.In Canada, where commercial products (Chontrol ® and Myco-Tech TM Paste) are already available, no buffer zone around treated trees is required (Anonymous, 2007).

Practical issues
When an efficient biocontrol strain is available, many practical questions arise.The most important of them relates to the production and form of the commercial product.This problem has not been solved in Finland.The growth of C. purpureum on artificial medium is, however, relatively good, and this fungus has already been commercialized in Canada.Therefore, we do not see big practical problems in this issue.
Also the question of dispersal machinery should be solved, as it would be cost-efficient only to spread the control agent at the same time as the trees/sprouts are cut.This issue might be solved by using the same device as already applied for the biocontrol of Heterobasidion root rot with Phlebiopsis gigantea (Korhonen et al. 1994, Pratt et al. 2000).Trace amounts of P. gigantea in C. purpureum suspension, and vice versa, do no reduce the biocontrol efficiency (Hamberg et al., unpublished).

Conclusion
The use of Chondrostereum purpureum in biocontrol of hardwood sprouting in forest regeneration areas, under electric lines and at the road sides looks promising in Finland.Highly efficient isolates are available and the possible risks have been evaluated.The next step will be to solve the practical issues: the development of a commercial preparation and finding out costefficient methods for its application to stumps.