Polyphloroglucinol phenolics are the best known example of chemical deterrents against herbivores in temperate marine systems. However, most of the research on these compounds has been done in North America, where phenolic levels in algae are often low. I show here that algae in the Orders Fucales and Laminariales in temperate Australia and New Zealand typically contain very high levels of polyphenolics-much higher than species in these orders in North America. The median value for the distribution of mean phenolic levels for 25 North American species is 1.33% total phenolics (dry wt.); for 37 Australasian species, the median is 6.20%. Significant spatial, temporal, and intraplant variation in phenolic content occurs in a number of species in Australasia, but this does not significantly alter my major conclusion.
Phenolic levels in drift algae (an important food source for some herbivores) detached for up to two weeks are also not significantly different from living, attached plants. Many species in the Fucales in Australasia also contain non-polyphenolic secondary metabolites that are not found in North American species. Thus herbivores in Australasia face greater amounts, and a greater range,of putative chemical defenses in brown algae than do herbivores in similar systems in North America. Any general theory for the evolution of marine plant/herbivore interactions must take into account such broad-scale biogeographical (and taxonomic) patterns.
The diversity and intensity of selection pressure on organisms has been described as varying predictably across biogeographic zones, generally increasing with decreasing latitude. Apparent effects of this increased selective pressure in the tropics include patterns in species diversity (Pianka 1966), diversity and quantity of allelochemicals (Bakus and Green 1974; Levin 1971, 1976; Levin and York 1978; Hay and Fenical 1988; Coley and Aide 1991; Hay and Steinberg in press), predation intensity, inferred antipredatory morphological structures (Vermeij 1978; Jeanne 1979; Bertness et al. 1981; Heck and Wilson 1987) and herbivory (Levin 1976; Lubchenco and Gaines 1981; Gaines and Lubchenco 1982; Steneck 1988; Hay and Steinberg in press). In the marine environment, herbivory is most intense at low latitudes (Vermeij 1978; Lubchenco and Gaines 1981 ; Steneck 1988 ; Hay and Steinberg in press). Algal chemical defenses generally Offprint requests to: N.M.
Targett parallel this latitudinal variation in herbivory with increasing concentration and diversity of allelochemicals with decreasing latitude (Hay and Fenical 1988; Hay and Steinberg in press). Phloroglucinol-based polyphenolics (phlorotannins, see Ragan and Glombitza 1986), antifeedant allelochemicals which occur exclusively in brown algae (Phaeophyta), have been reported as an exception to this trend, increasing in concentration with increasing latitude (Steinberg and Paul 1990; Van Alstyne and Paul 1990; Steinberg et al. 1991 ; Steinberg and Van Altena in press).
This is in contrast to terrestrial data on condensed tannins for broad leaved forest species, which show that temperate species have lower phenolic concentrations than tropical ones (Coley and Aide, 1991 and references therein). In marine temperate regions, where considerable data exist on phlorotannin concentrations, brown algae are characterized as either high or low phenolic species. High phenolic species are defined as those with phenolic concentrations > 2% of their dry weight, the threshold level at which herbivory is typically deterred (Swain 1979; Geiselman and McConnell 1981 ; Steinberg 1988). However, a high degree of inter- and intraspecific variability in phenolic concentration has been observed (Ragan and Glombitza 1986; Steinberg 1986, 1988, 1989).
Phenolic concentrations in tropical phaeophytes have been reported only from Indo-Pacific species; therefore, a less complete picture is available for tropical than for temperate browns (Hay and Fenical 1988; Steinberg and Paul 1990; Van Alstyne and Paul 1990; Steinberg 1986, 1989; Steinberg et al. 1991 ; Steinberg and Van Altena in press). In the Indo-Pacific, phenolic concentrations have been found to be low, although Padina spp. from Magnetic Island, Australia (Steinberg et al. 1991) and individual plants in two other genera (Lobophora and Dictyopteris) were noted as exceptions (Steinberg and Paul 1990). Although brown algal phenolic concentrations have not been determined for any Caribbean species, patterns of herbivore preference have been used to suggest that polyphenolics might be high in some species (Norris and Fenical 1982). We examined phenolic levels in tropical phaeophytes from both the Caribbean and western Pacific and in selected species from the temperate western Atlantic and eastern Pacific (i.e. both coasts of North America). These levels were compared to literature values to determine: 1. Are there latitudinal trends in phenolic concentrations? 2. Are there interoceanic differences between tropical Old World (Indo-Pacific)and Neotropical (Caribbean) phaeophyte phenolic concentrations? 3. Does the magnitude of intraspecific phenolic variability in tropical and temperate phaeophytes overshadow observed latitudinal trends?
The Caribbean phaeophytes (orders Fucales and Dictyotales) that we examined had phenolic concentration levels ranging from 1.34% to 14.92% dry weight. Caribbean Fucales all had high phenolic concentrations (> 2 %) and tested positive for phloroglucinol derivatives. Two of the four Caribbean Dictyotales examined (Stypopodium zonale and Lobophora variegata in each of its three forms, Coen and Tanner 1989) also had high phenolic concentrations. The dry weight phenolic concentration values for the Lobophora forms and for Stypopodium (8.33% 13.39% and 14.92% respectively) far exceeded all previously reported values for species in this order. Thus, our data from Caribbean phaeophytes in which six of eight species produced phenolic concentrations in excess of 2% (3.37%-14.92%) show that high phlorotannin concentrations are not limited to temperate species. Tropical interocean differences are evident when comparing phenolic concentration values for phaeophyte genera common to the Indo-Pacific and Caribbean. Lobophora variegata, the only species examined in this study that is common to the tropical areas listed, showed low phenolic values for Pacific (0.81%, Hawaii, Table 1) and Indo-Pacific (<2%, p="0.006)," p="0.002)"> 2% dry weight), while species in the orders Laminariales and Dictyotales typically have low levels (Steinberg 1985, 1989; Ragan and Glombitza 1986; Steinberg and Paul 1990; Van Alstyne and Paul 1990; Steinberg et al. 1991 ; Steinberg and Van Altena in press). Earlier studies also show that polyphenolic levels in tropical Fucales and Dictyotales from the Indo-Pacific are low (Steinberg 1986, 1989; Hay and Fenical 1988; Steinberg and Paul 1990; Van Alstyne and Paul 1990; Steinberg et al. 1991). Our results clearly demonstrate that species in the tropics can be high in phenolics. Data from our study and from the literature, where extraction and quantification procedures allow comparisons to be made, also indicate that intraspecific variability may be as high as interspecific variability. Of the phenolic values obtained, most striking were the values obtained for the three forms of Lobophora (Coen and Tanner 1989) and for Stypopodium (8.33-13.39% and 14.92% dry weight respectively). Phloroglucinol and a variety of phenolic acids have been reported from Lobophora variegata (form not specified) collected in Brazil (Legaz et al. 1985). Acetate-derived phenolic lipids have been described from Lobophora papenfussii (Gerwick and Fenical 1982) and their presence in Lobophora variegata (form unspecified) from the Florida Keys has been hypothesized (Paul and Hay 1986). However, higher-molecular-weight phlorotannins are also evident (Folin-Denis test, Lindt test, 1H NMR) in molecular sizing
experiments done on aqueous extracts from each of the three forms of Lobophora collected in Belize (Boettcher and Targett unpublished data). Fresh Stypopodium is known to contain a series of primarily mevalonatederived low-molecular-weight phenolic metabolites (<> 100,000Da (Boettcher and Targett unpublished data).
The absence of a latitudinal phlorotannin trend between Caribbean tropical and temperate phaeophytes and the presence of site-specific phlorotannin variability within species suggests that other factors may play a more critical role within a biogeographic region in determining phenolic concentrations. Variability in phenolic concentration may be influenced by extrinsic factors such as salinity (Ragan and Glombitza 1986), nutrient availability (Ilvessalo and Tuomi 1989), herbivore intensity (Van Alstyne 1988) and season (Ragan and Jensen 1978 ; Ragan and Glombitza 1986). Intrinsic plant factors such as size (Denton et al. 1990), age (Pederson 1984), and tissue type (Steinberg 1984; Tugwell and Branch 1989; Tuomi et al. 1989) may also play a role. The magnitude of the variation in phenolic concentration is often high enough to alter the palatibility of plants to herbivores (Swain 1979; Geiselman and McConnell 1981; Steinberg 1988). Studies regarding the effect of plant phlorotannin concentration on marine tropical herbivores have yielded conflicting results. Van Alstyne and Paul (1990) showed that phlorotannins from temperate species are effective deterrents against some tropical herbivores when they are incorporated into tropical analogues at concentrations > 2% dry wt, while Steinberg et al. (1991) showed that quantitative variability in temperate and tropical algal phenolic levels do not correlate with the susceptibility of these algae to herbivory. These and other studies make supportive arguments based upon the assumption that all tropical phaeophytes are typically low in phenolics and invoke physiological and cost arguments as explanations for their low concentrations (Steinberg 1986; Hay and Fenical 1988; Van Alstyne 1988; Steinberg and Paul 1990; Van Alstyne and Paul 1990; Steinberg et al. 1991; Hay and Steinberg in press; Steinberg and Van Altena in press). Our study shows that phenolics can be present in high and variable concentrations in tropical Caribbean phaeophytes, and thus the assumption that phlorotannins are of little consequence in tropical marine plant-herbivore interactions because of their low concentrations (Steinberg 1986; Hay and Fenical 1988; Van Alstyne 1988; Steinberg and Paul 1990; Van Alstyne and Paul 1990; Steinberg et al. 1991) is premature.
Our results in combination with previously published data suggest that there is an interoceanic difference between phenolic concentrations in tropical Indo-Pacific phaeophytes and Caribbean phaeophytes. Grazing intensity is higher in the Pacific and Indian oceans than in ecologically similar communities in the Atlantic (Vermeij 1978; Estes and Steinberg 1988; Steinberg and Van A1-tena in press). Grazing intensity thus correlates negatively with our observations of phaeophyte phenolic concentrations and calls into question the role of phlorotannins in defense. However, other extrinsic and intrinsic factors may confound the interpretation and should be examined (Bernays et al. 1989). To explain these differences and those in temperate and tropical marine plant-herbivore interactions, it is now evident that more subtle within-plant (Zucker 1983; Gaines 1985; Lewis et al. 1987; Coen and Tanner 1989) and among-herbivore (Gaines 1985; Hay et al. 1987; Horn 1989) characteristics must be examined. These include phenotypic variation in phenolic concentrations, phlorotannin size distributions, induction of phenolics by herbivores, as well as herbivore features such as gut pH and the presence of gut surfactants. Studies such as these will allow a further comparison between the importance of biological/physical factors and geographic distribution in determining phlorotannin concentrations.
Nb* Dari banyak sumber bacaan.