in: The Genetics and Exploitation of Heterosis in Crops (1997)
Trees and Heterosis
J. L. Brewbaker and W. G. Sun

Scope of Review

This review grandly assumes its embrace of all species of trees (defined as woody perennials, largely single-stemmed, exceeding 3 m in height). It thus ranges from forest trees to industrial crops, shade trees, fruit trees and trees like leucaena that are harvested bimonthly for fodder.

Manifestation of Hybrid Superiority

A useful term applied to the vigor of hybrid trees is "hybrid superiority" (Dieters et al. 1995), that recognizes the enhanced economic value of hybrids (vs. midpoint of parents) as having two major aspects: 1) complementarity, the complementary recombination of parental traits; and 2) true hybrid vigor or heterosis, sensu strictu. This review considers both aspects of hybrid superiority, since they are often inseperable. Hybrid superiority appears to characterize one or another trait in all tree species that have been studied, making this review a formidable task. Even more daunting is the fact that yield (e.g., wood, total biomass, fodder) of trees has many components that are virtually unknown (e.g., root biomass), and that can change with time during growth. Given these provisos, hybrid superiority is a commercial fact in most tree genera and notable in eucalypts, poplars, pines, casuarinas, oil palms and leucaenas. It is primarily evident from interspecific crosses, which occur with considerable ease in woody genera. Typically more than 50% of interspecific tree crosses produce hybrids (Sorensson and Brewbaker 14). Seed set and sterility of hybrids is rarely of concern in forest species. Hybrid superiority is common among inter-provenance crosses where genetic distances are relatively large, but generally is reduced as such distances diminish (Wang et al. 1996).

Complementarity in Hybrid Superiority

Complementarity is perhaps a major component of hybrid vigor in most trees, since they are customarily grown for long duration in polymorphic ecosystems with very limited environmental control (Wang et al. 1996). The homeostatic properties of hybrid systems thus become of great importance. An example from species of Larix, the larch, where 33-year old European hybrids have three times the biomass of the parent species:

Larix species L. decidua L. leptolepis Hybrids
Needle size Small Large Large
Needle Number High Low High

A second example from the tropical genus, Leucaena, where hybrids exceed the parents in range of adaptability:

Leucaena species L pallida L. leucocephala Hybrids
Cold tolerance High Low High
Heat tolerance Low High High
Psyllid tolerance High Low High

Heterosis in Hybrid Superiority

Heterosis in tree growth is evident in many hybrids and perhaps best illuminated in studies of Populus, the genus of poplars and aspens. New analytical methods to estimate the number of quantitative trait loci (QTLs) affecting traits in trees and their additive and dominance effects are being developed (Li and Wu 1996). Increasing importance is being attributed to the role of a few major QTLs in heterosis that have a higher probability of dominance and epistatic effects (Bradshaw and Stettler 1995). It is clear that many minor QTLs affect traits involved in the wood yield increases of hybrids. Multiple allelism is recognized increasingly as a fact for most QTLs in highly-outcrossed tree species, in alleles that differ greatly between species. There is continuing debate about overdominance among major QTLs, and suspicion that pseudo-overdominance (POD) is in fact the cause. POD involves interactions in hybrids between beneficial dominant alleles and deleterious recessives at closely-linked loci, an automatic feature for example of chromosomes bearing the S locus in self-incompatible species (common among woody perennials). Additive effects characterize most hybrid vigor, but the role and importance of dominance varies greatly. There is much evidence that multiplicative gene interactions affect composite traits such as wood yield (Li and Hu 1996). Standardization or transformation can minimize such scale effects (Dieters et al. 1995), but can also obscure the true nature of gene interactions both within and between loci (epistasis). Thus, the nature of gene action underlying heterosis in trees is not well or easily understood (Namkoong and Kang 1990).

Utilization of Hybrid Superiority

The ability to market cloned hybrids is essential to commercialization of many: frees, and breeding for clonability becomes a requisite exercise. The use of vegetatively propagated F1 hybrids is extensive in Eucalyptus, Populus and Pinus spp. In many instances seedlessness is associated with hybrid superiority of ornamental or fruit trees; e.g., sterility of 3N hybrids in bananas or in leguminous ornamentals (e.g., Cassia, Plumeria). Seedless triploid interspecific hybrids show superior wood production in Leucaena, where the legume pods represent a major carbon sink (Sorensson and Brewbaker 1994). The use of F2 populations can be economic; e.g., interim use of non-inbred hybrid of Pinus elliotii x P. caribaea (Nikles 1993). Advanced generation populations of hybrids, analogous to synthetics (or composites) in maize, are being marketed from inbred-derived F2s of the polyploid Leucaena leucocephala, where heterosis is only slightly less than F1 hybrids and a broad-germplasm base is required.

Breeding Strategies for Hybrid Superiority

Tree improvement goes through phases familiar to all breeders, but hybrid development normally lacks an extended phase of inbreeding due to high genetic loads (Wang et al. 1996). Maize breeders are familiar with similar loads in native racial materials, and must resynthesize from inbreds to create "inbreeding-tolerant" populations. However, selfed families can be a good index of hybrid performance. Sublining has been practiced within tree populations to minimize inbreeding (Namkoong and Kang 1990). The question of the value of reciprocal recurrent selection in tree improvement is open to debate, since contributions of dominance seem generally to be small. Added to the concepts of general combining ability (GCA), foresters speak of general hybridizing ability (GHA) in relation to interspecific hybrids. GCA within species is not always a good predictor of GHA for stern form and yield in trees. Breeding schemes that combine recurrent selection for GCA with cycles of GHA evaluation, if they can be performed economically, perhaps best optimize genetic gains (Dieters et al. 1995).

Literature Cited