The Effects of Hybridization in Plants on Secondary Chemistry: Implications for the Ecology and Evolution of Plant±Herbivore Interactions1
Total Page：16
File Type：pdf, Size：1020Kb
American Journal of Botany 87(12): 1749±1756. 2000. INVITED SPECIAL PAPER THE EFFECTS OF HYBRIDIZATION IN PLANTS ON SECONDARY CHEMISTRY: IMPLICATIONS FOR THE ECOLOGY AND EVOLUTION OF PLANT±HERBIVORE INTERACTIONS1 COLIN M. ORIANS2 Department of Biology, Tufts University, Medford, Massachusetts 02155 USA Natural hybridization is a frequent phenomenon in plants. It can lead to the formation of new species, facilitate introgression of plant traits, and affect the interactions between plants and their biotic and abiotic environments. An important consequence of hybrid- ization is the generation of qualitative and quantitative variation in secondary chemistry. Using the literature and my own results, I review the effects of hybridization on plant secondary chemistry, the mechanisms that generate patterns of chemical variation, and the possible consequences of this variation for plants and herbivores. Hybrids are immensely variable. Qualitatively, hybrids may express all of the secondary chemicals of the parental taxa, may fail to express certain parental chemicals, or may express novel chemicals that are absent in each parent. Quantitatively, concentrations of parental chemicals may vary markedly among hybrids. There are ®ve primary factors that contribute to variation: parental taxa, hybrid class (F1,F2, etc.), ploidy level, chemical class, and the genetics of expression (dominance, recessive vs. additive inheritance). This variation is likely to affect the process of chemical diversi®cation, the potential for introgression, the likelihood that hybrids will facilitate host shifts by herbivores, and the conditions that might lead to enhanced hybrid susceptibility and lower ®tness. Key words: genetics; hybridization; inheritance; introgression; novelty; plant±herbivore interactions; secondary chemistry. The consequences of hybridization have intrigued scientists Relatively little attention has been paid to their resistance to for centuries. In the late 1700s, Linnaeus suggested that hy- herbivores and pathogens, yet recent reviews suggest that re- bridization could lead to new species, and Mendel made sim- sistance is an important component of hybrid survival (Whi- ilar claims in the late 1800s (Arnold, 1997). More recently tham, 1989; Strauss, 1994; Fritz, Nichols-Orians, and Bruns- Anderson (1949) proposed that hybridization facilitates the feld, 1994; Fritz, 1999; Fritz, Moulia, and Newcombe, 1999). transfer of traits between species, a process known as intro- The production of secondary chemicals mediates plant re- gressive hybridization, and Stebbins (1959) and Lewontin and sistance to herbivores and pathogens (e.g., Rosenthal and Jan- Birch (1966) suggested that hybridization and introgression zen, 1979). Not surprisingly several studies have shown that could promote adaptive evolutionary change. Hybridization plant secondary chemistry alters the resistance of hybrid plants and introgression are widespread features of many natural (Huesing et al., 1989; Ben-Hammouda et al., 1995; Orians et plant populations (e.g., Levin, 1966; Harborne and Turner, al., 1997; Fritz and Orians, unpublished data). Although plant 1984; McArthur, Welch, and Sanderson, 1988; Keim et al., breeders use hybridization to obtain or introgress desired 1989; Smith and Sytsma, 1990; Soltis and Soltis, 1991; Rie- chemical traits (e.g., Maxwell and Jennings, 1980; Huesing et seberg and Wendel, 1993; Arnold, 1994; Fritz, Nichols-Orians, al., 1989; Altman, Stipanovic, and Bell, 1990; GoÂmez and and Brunsfeld, 1994; Rieseberg, 1995; Ellstrand, Whitkus, and Ledbetter, 1993), surprisingly little attention has been paid to Rieseberg, 1996; Arnold, 1997; Rieseberg and Carney, 1998), the effects of hybridization on secondary chemistry and the and ;70% of all ¯owering plant species are believed to have survival of hybrids in natural populations. In one study, Rod- originated via hybridization (Whitham, Morrow, and Potts, man (1980) suggested that selection against hybrid glucosi- 1991; Arnold, 1994). So what determines the ecological per- nolate phenotypes (Cakile, Brassicaceae) maintains a narrow formance and evolutionary fates of hybrids? Historically, most hybrid zone and minimizes gene ¯ow. Clearly, secondary research has focused on gametic barriers and viability, specif- chemistry may be an important component of hybrid ®tness ically, the ability of hybrids to survive the abiotic conditions and in¯uence speciation, introgression, and plant±herbivore in- present in parental or nearby nonparental habitats (Arnold and teractions. Hodges, 1995; Rieseberg, 1995; Rieseberg and Carney, 1998). We know that hybridization results in progeny that differ qualitatively and quantitatively from the parents in the ex- 1 Manuscript received 4 February 2000; revision accepted 6 June 2000. pression of secondary chemicals (Harborne and Turner, 1984). The author thanks Francie Chew, Bob Fritz, Clive Jones, Frank Messina, Gordon Orians, and an anonymous reviewer for their comments on earlier It is often incorrectly assumed that traits of hybrids will be versions of the manuscript, Loren Rieseberg and Diana Pilson for providing intermediate to the two parental taxa (Rieseberg, 1995). In parental and hybrid sun¯ower seeds, Rafael Ricco for assisting with the sun- fact, patterns of inheritance are quite complex. In ®rst-gener- ¯ower analyses, and all the students (especially Minh Dao, Megan Grif®ths, ation (F1) hybrids, the suite of chemicals may be: (1) similar Cynthia Huang, Rachel Samberg, Adam Welland, Alexander Wild, and Pa- to that in one of the two parental taxa (Fahselt and Ownbey, mela Zee) for putting in hours of their time measuring plant chemicals and 1968), (2) intermediate between those of the two parental taxa herbivore responses. This review was completed during a Mellon Research- Semester Fellowship and supported by a travel grant from Tufts University, (McMillan, Chavez, and Mabry, 1975; Orians et al., 2000), (3) and by the Mary Flagler Cary Charitable Trust. overexpressed or present in higher concentrations than in ei- 2 Reprint requests: FAX: 617-627-3805; e-mail: [email protected]. ther parent (Spring and Schilling, 1990), (4) underexpressed 1749 1750 AMERICAN JOURNAL OF BOTANY [Vol. 87 or present in lower concentrations than either parent (Court et produces phenolic glycosides in its leaves and one that pro- al., 1992), (5) de®cient, in that hybrids lack chemicals that duces condensed tannins in its leaves (Fritz, Nichols-Orians, both parents produce (Fahselt and Ownbey, 1968), or (6) nov- and Brunsfeld, 1994; Orians and Fritz, 1995). Also, Helianthus el, in that hybrids may contain chemicals lacking in both pa- annuus (Asteraceae), a sun¯ower species that has foliar glan- rental taxa (Levy and Levin, 1974; Rieseberg and Ellstrand, dular trichomes that contain sesquiterpene lactones (STLs), 1993; Buschmann and Spring, 1995). Later-generation hybrids hybridizes naturally with H. petiolaris, a species that lacks generate even more variability (Connor and Purdie, 1976; Rie- foliar glandular trichomes (Spring and Schilling, 1989; Rie- seberg and Ellstrand, 1993). Another important feature of hy- seberg, 1991). Although both these species produce STLs in bridization is that individuals may differ qualitatively and their ¯owers, the major STLs belong to different groups (the quantitatively from each other even within a hybrid class (e.g., niveusin and argophyllin types in H. annuus and the budlein F1,F2, and backcross hybrids) (Connor and Purdie, 1976; type in H. petiolaris). These two species of sun¯owers have Crins, Bohm, and Carr, 1988; Orians et al., 2000). apparently hybridized naturally for thousands of years and re- It is my goal to review what is known about the secondary sulted in the evolution of three hybrid-derived species (Rie- chemistry of hybrid plants and to discuss the importance of seberg, 1991; Rieseberg et al., 1996). variation in hybrid chemistry to resistance. Although morpho- Although chemical similarity does not appear to dictate pat- logical traits are also highly variable and may contribute to terns of hybridization, hybridization is nonrandom. Hybrids patterns of resistance, this review focuses solely on secondary are more common in some plant groups than others (Ellstrand, chemistry. I address the following questions. First, is hybrid- Whitkus, and Rieseberg, 1996). For example, hybrids are fre- ization more common between closely related but chemically quently found in Asteraceae, Poaceae, Rosaceae, Salicaceae, similar parental taxa compared to closely related chemically and Scrophulariaceae, but rare in Brassicaceae and Solanaceae dissimilar taxa? Second, what are the patterns of qualitative (Ellstrand, Whitkus, and Rieseberg, 1996). In general, stabi- and quantitative variation in hybrid chemistry and what do we lized hybrids are most common in outcrossing perennials that know about the genetic mechanisms underlying the patterns? exhibit vegetative spread and are insect pollinated. Neverthe- Third, how might chemical variation among hybrids in¯uence less, many annuals hybridize. the ecology and evolution of plant-herbivore interactions? For the purposes of this paper, I adopt Arnold's (1997) def- PATTERNS OF CHEMICAL VARIATION inition of hybridization: ``the successful mating in nature be- tween individuals from two populations or groups of popula- For this review, I drew heavily upon the data generated by tions which are distinguished on the basis of one or more chemosystematists. These researchers use chemistry as a tool heritable