Allelopathy is an important non-resource interaction in terrestrial plant communities that may affect invasions by non-indigenous plants. The ‘novel weapons hypothesis’ (NWH) predicts that non-indigenous plants will become invasive if they have allelopathic compounds that assemblages in the new range are not adapted to. Recently, the non-indigenous, chemically rich macroalga Bonnemaisonia hamifera (Hariot) has become one of the most abundant filamentous red algae in Scandinavian waters.
We used B. hamifera to specifically test the aspect of the NWH that concerns invasion success based on novel allelochemicals in the invaded range. Allelopathic interactions were tested through effects on the growth rate of adult native macroalgae in co-cultures with B. hamifera and through the settlement success of native macroalgal propagules and microalgae on surfaces coated with 1,1,3,3-tetrabromo-2-heptanone. We also investigated whether 1,1,3,3-tetrabromo-2-heptanone can be transferred from B. hamifera to its native host algae, as a means of pre-emptive competition.
The settlement of native macroalgal propagules and microalgae was strongly inhibited on surfaces coated with 1,1,3,3-tetrabromo-2-heptanone at ecologically relevant concentrations, but there were no effects of adult B. hamifera on growth rates of adults of the six native naturally co-occurring species. The compound was shown to be transferred from B. hamifera to the surface of its native host algae at inhibitory concentrations in both laboratory and field experiments.
By inhibiting the settlement of propagules on its thallus and on surrounding surfaces, B. hamifera achieves a competitive advantage over native macroalgae, a finding that parallels previous reports on soil- and litter-mediated allelopathic interactions among vascular plants. Because competition for available substrata in marine benthic systems is intense, the ability to reserve space may be vital for B. hamifera's successful invasion. This is the first example of an allelopathic compound that can be transferred by direct contact from an exotic to a native species, with an active and unaltered function.
Synthesis. Our results clearly show that the main secondary metabolite of the invasive red alga B. hamifera has strong allelopathic effects towards native competitors, suggesting that its novel chemical weapon is important for the highly successful invasion of new ranges.
Since the 19th century, ecologists have considered the possibility that plants can achieve competitive advantages through secretion of phytotoxic chemicals in terrestrial communities (Darwin 1859). The negative effect of one plant on another plant through the release of chemical compounds into the environment is termed allelopathy (sensu Muller 1966) and is an important non-resource interaction in terrestrial plant ecology (Callaway 2002). In marine systems, the importance of chemically mediated interactions among benthic competitors has taken longer to be recognized. The pioneering marine experiments on allelopathy concerned competition among sessile invertebrates (Goodbody 1961; Jackson & Buss 1975; Porter & Targett 1988), but a few recent studies have demonstrated that seaweeds also have the ability to suppress natural competitors through allelopathic agents (de Nys, Coll & Price 1991; Paul et al. 2011; Rasher et al. 2011).
Allelopathic activity has long been suggested to be an important characteristic of invasive plants (Steenhagen & Zimdahl 1979), but it is not until recently that it has been more generally acknowledged as an influential mechanism in invasion biology (Inderjit et al. 2011b). An important driving force behind this development was the formulation of the ‘novel weapons hypothesis’ (NWH: Callaway & Aschehoug 2000), which predicts that exotic invaders should become successful if they bring unique biochemicals, that is, secondary metabolites, to their new ranges. The native species are not adapted to these novel chemicals and will consequently be out-competed through allelopathic interactions, allowing the invader to establish and proliferate in the new environment. The NWH has recently been expanded to include chemical defences against herbivory (Cappuccino & Carpenter 2005; Verhoeven et al. 2009), but it was originally formulated for allelopathic interactions among terrestrial plants (Callaway & Ridenour 2004). The simplicity of the NWH holds an intuitive promise for the development of predictive tools in invasion biology. By identifying the secondary compounds of invasive species, it is possible to create an index of phytochemical novelty that can be matched to the chemical properties of species in native communities and, thus, could help identify potential invaders in new ranges a priori (Carpenter & Cappuccino 2005). Hence, such a straightforward tool may aid management decisions in preventing the global spread of invasive exotics before the species become established in the new range (Carpenter & Cappuccino 2005).
Support for the NWH has almost exclusively been reported from studies in terrestrial communities (Carpenter & Cappuccino 2005; Cappuccino & Arnason 2006; Inderjit, Callaway & Vivanco 2006). One well-known example, which gave rise to the NWH, is the diffuse knapweed Centaurea diffusa, which through its root-exuded phytotoxins can achieve competitive dominance over plants native to North America (Callaway & Aschehoug 2000; Vivanco et al. 2004). More recent examples of invasive plants that show allelopathic activity towards native plants in North America include the Brazilian pepper, Schinus terebinthifolius, (Donnelly, Green & Walters 2008), the Australian common reed, Phragmites australis, (Rudrappa, Bonsall & Bais 2007), the European narrow-leaved cattail, Typha angustifolia (Jarchow & Cook 2009) and the Mexican devil, Ageratina adenophora (Inderjit et al. 2011a). Empirical evidence of the role of allelopathy and novel chemistry in aquatic invasions is so far scarce and circumstantial. Conditioned water from the highly invasive floating primrose willow, Ludwigia peploides, has been shown to diminish native seedling survival (Dandelot et al. 2008), and the non-indigenous ponds weeds, Elodea nuttallii and Elodea canadensis, have allelopathic effects on epiphytes and phytoplankton in their new ranges (Erhard & Gross 2006). In the marine environment, Raniello et al. (2007) found that a secondary metabolite (caulerpenyne) from the invasive green algae Caulerpa racemosa var. cylindracea affected the optimal quantum yield of the native seagrass Cymodocea nodosa negatively, lowering its photosynthetic capacity.
To our knowledge, the only explicit test of the NWH in aquatic environments concerns interactions between native grazers and the highly invasive red alga Bonnemaisonia hamifera (Enge et al. 2012), which has recently become one of the most conspicuous red algae in Scandinavia (Thomsen et al. 2007). The alga produces a number of halogenated secondary compounds, and through a bioassay-guided fractionation approach, it was revealed that its major secondary compound (1,1,3,3-tetrabromo-2-heptanone) strongly deters native herbivores (Enge et al. 2012). This compound is considered to be novel in invaded ranges because B. hamifera, like all Bonnemaisoniaceans, have been shown to have species-specific secondary metabolites (Kladi, Vagias & Roussis 2004), and the compound has consequently never been reported for any species other than B. hamifera in either scientific journals or chemical compound databases (Enge et al. 2012). The compound has also previously been shown through bioassay-guided fractionation to have significant effects on bacterial abundance and community composition in the invaded range (Nylund et al. 2008; Persson et al. 2011).
In this study, we go further by investigating whether 1,1,3,3-tetrabromo-2-heptanone also have allelopathic effects on the native algal competitors of B. hamifera in the invaded range. Hence, we test the aspect of the NWH that concerns community effects of novel compounds in the invaded range and will not include biogeographical comparisons in this study. More specifically, we tested the potential allelopathic effects of adult B. hamifera on six native species of adult fine- or thick-branched green, brown and red macroalgae by comparing the effects on growth of the native algae in cultures with B. hamifera to those in cultures with conspecifics. Settlement success of propagules of native macroalgae and one species of benthic diatom was also tested at ecologically relevant concentrations of 1,1,3,3-tetrabromo-2-heptanone, that is, ranges that include the natural surface concentrations of B. hamifera (Nylund et al. 2008). Finally, because B. hamifera commonly grow epiphytically in the study area, we tested if the lipophilic 1,1,3,3-tetrabromo-2-heptanone can be transferred from the surface B. hamifera to the surface of native host algae in sufficient concentrations to inhibit the recruitment of other epiphytic competitors. Thus, by investigating the effects of a potential chemical weapon from a non-indigenous macroalga on different life stages of native competitors, as well as the potential to extend this weapon to surfaces beyond the alga itself, we here provide the first direct test of allelopathy as a potential mechanism for macroalgal invasions, within the context of the NWH.
Materials and methods
The marine subtidal macroalgae B. hamifera is native to the West Pacific, but has in recent years been found from Africa to North America (Harris & Tyrrell 2001; Cormaci et al. 2004) and is common in Europe (Thomsen et al. 2007). The life history of B. hamifera involves an alternation between the morphologically different gametophyte (sexual) and tetrasporophyte (asexual) life stages. Individuals of both life stages are brownish-red, but the tetrasporophyte forms dense cotton-like tufts, compared to the larger, thick-branched gametophyte (Rueness 1977). In Sweden, B. hamifera occurs only as the tetrasporophyte stage, which is also called the Trailiella phase as it was previously thought to be a separate species (e.g. Trailliella intricata; Batters 1896). It primarily grows epiphytically on other algae, and the most common host algae in Sweden are Furcellaria lumbricalis (Hudson) and Corallina officinalis (L.) (personal observation).
Bonnemaisonia hamifera is a chemically rich alga, and through two different bioassay-guided fractionations, its major secondary metabolite 1,1,3,3-tetrabromo-2-heptanone has been found to strongly affect native consumers (Enge et al. 2012) and bacteria (Nylund et al. 2008). The natural concentration on the surface of B. hamifera of 1,1,3,3-tetrabromo-2-heptanone has been measured in two different populations on the Swedish west coast to 4.4 (±0.49) and 2.8 (±0.65 SD) μg cm−2 by surface extractions (Nylund et al. 2008). The 1,1,3,3-tetrabromo-2-heptanone was discovered in the mid-1970s (Siuda et al. 1975) and is unique to B. hamifera (Kladi, Vagias & Roussis 2004; Enge et al. 2012). The compound is highly lipophilic, that is, not soluble in water, and it is stored in special gland cells between the vegetative cells, from which the compound is transported to the surface of the thallus of B. hamifera (Nylund et al. 2008; Abbas et al. 2012).
Co-cultures of Bonnemaisonia hamifera and potential competitors
To investigate the allelopathic capacity of B. hamifera, we cultured individuals of this alga together with potential seaweed competitors. Previous pilot studies have shown that there was no effect of exudates from B. hamifera on the growth of co-existing seaweeds (G.M. Nylund, unpublished). Therefore, we cultured B. hamifera in direct contact with six other seaweed species: the red algae Ceramium virgatum (Roth), Cystoclonium purpureum (Hudson) and Polysiphonia stricta (Dillwyn), the brown algae Pilayella littoralis (L.) and Sphacelaria cirrosa (Roth), and the green alga Cladophora rupestris (L.). These species are all potential competitors of B. hamifera because they are commonly found on the same host algae as B. hamifera. These species were also chosen because they are either fine-branched (apical cell diameter 20–30 μm: P. stricta, Pilayella littoralis and S. cirrosa) or thick-branched (apical cell diameter 100–200 μm: Cystoclonium purpureum, Cladophora rupestris and C. virgatum), and we wanted to investigate whether allelopathy through physical contact may affect more robust and thicker seaweeds, consisting of many cell layers, differently than delicate seaweeds. Individuals of the different species were collected by SCUBA diving from different localities in the archipelago west of Tjärnö Marine Biological Laboratory. Following arrival to the laboratory, the algal samples were weighed and placed separately in containers together with B. hamifera of similar weight. To ensure direct contact between B. hamifera and the opposing species, the seaweeds were tied together with sewing thread. Containers with two equally sized (50 g wet weight) individuals of the same species were also tied together and served as controls. Control and treatment containers were replicated six times (n = 6). The seaweeds had constant flux of seawater and were grown for 13 days at 16 °C with a 16/8 light : dark cycle, after which the seaweeds were weighed, and the growth rate was determined as relative weight increase (change in weight divided by initial biomass).
Settlement tests with 1,1,3,3-tetrabromo-2-heptanone
Allelopathic effects of B. hamifera on juvenile stages of potential competitors were assessed by investigating the antifouling effect of 1,1,3,3-tetrabromo-2-heptanone. The inhibition of algal settlement and germling development was investigated in a series of laboratory settlement assays using the benthic diatom Cylindrotheca fusiformis (Reiman & Lewin), spores from the red algae C. virgatum (tetrasporophyte) and P. stricta (carposporophyte) and gametes from the green alga Ulva lactuca. Artificial surfaces for settlement were prepared by applying 0.5 mL of synthesized 1,1,3,3-tetrabromo-2-heptanone (Nylund et al. 2008) dissolved in hexane to untreated 48-mm diameter polystyrene Petri dishes (Nunc A/S, Roskilde, Denmark). The Petri dishes were placed on a shaking table for 30 min to evaporate the hexane and used immediately afterwards in the bioassays. In the case of Cylindrotheca fusiformis, 25.8 μL of 1,1,3,3-tetrabromo-2-heptanone dissolved in hexane was applied to 96-well Nunc plates. A total of eight concentrations of 1,1,3,3-tetrabromo-2-heptanone were prepared: 0 (solvent control), 0.06, 0.125, 0.25, 0.5, 1.0, 2.0 and 4.0 μg cm−2. This range of concentrations was specifically chosen to include the natural surface concentration of the compound, that is, 2–5 μg cm−2, which has previously been measured in Swedish populations of B. hamifera by surface extractions (Nylund et al. 2008).
Fertile individuals of C. virgatum and P. stricta were collected by SCUBA diving in the archipelago west of Tjärnö Marine Biological Laboratory. In the laboratory, fertile filaments of C. virgatum and P. stricta were removed by forceps under a dissecting microscope and placed in a glass container filled with sterile-filtered seawater (FSW). The filaments were left until spores appeared on the bottom of the container (10–30 min). The spores were removed with a plastic pipette, and 20–40 spores were transferred to treatment and control dishes filled with 10 mL of FSW. Treatment and control dishes were replicated six times (n = 6). The dishes were incubated overnight at 20–22 °C at a light/dark cycle of 16/8 h, and after the incubation, per cent settlement was determined by counting the number of settled and unsettled spores under a dissecting microscope.
Fertile U. lactuca thalli were placed in individual glass beakers containing FSW, following arrival to the laboratory. The beakers were exposed to a bright light source, and within a few minutes, positively phototactic swarmers were concentrated at the light side of the beaker. The swarmers were identified as gametes based on the number of flagella (Fletcher 1989). Swarmers confirmed to be gametes were pipetted to a beaker containing 100 mL FSW and stirred with a magnetic stirrer to obtain a uniform suspension of gametes. Gametes were added to the beaker until the colour of the solution turned light green. A 2-mL aliquot of the resulting gamete suspension and 8 mL of nutrient-enriched FSW were added to treatment and control dishes (n = 8) (McLachlan 1973). Dishes were placed in the dark for 2 h to allow for even settlement of gametes. After 6 day of incubation at 15 °C and a light/dark cycle of 16/8 h, the number of growing germlings was counted. Ten fields of view (0.95 mm2) were counted for each dish using an inverted binocular microscope.
The diatom species Cylindrotheca fusiformis was taken from laboratory stock culture, grown under 15–18°C and a light/dark cycle of 16/8 h, in F2 + Si medium. The stock culture was diluted five times with F2 + Si medium, and 200 μL was added to each well. After 48 h, the growth medium and unattached Cylindrotheca fusiformis were removed, and the wells were rinsed one time with 200 μL filtered seawater. One hundred microlitres of ethanol was added to each well, and settlement of Cylindrotheca fusiformis was measured as amount chlorophyll using a plate reader (Wallac 1420, Victor3 V; Perkin Elmer, Turku, Finland).
Coating experiment in the laboratory
To test whether B. hamifera can affect the surrounding surface (i.e. its host algae) by spreading its secondary metabolites, two coating experiments were performed. In the first and second experiments, the native alga F. lumbricalis and the native C. officinalis, respectively, were attached adjacent to B. hamifera in a container with fresh seawater placed on a multipurpose horizontal shaking table in order to mimic wave action. Individuals of B. hamifera were placed at each side of individuals of the native macroalgal species on plastic strips and fastened using cable ties. Individuals were separated by a distance of approximately 1.5 cm, to make sure they were not in contact unless brought together by the motion of waves, in order to simulate natural conditions. The experiments were conducted in a constant temperature room (16 °C) with a light : dark cycle of 14 : 10 and with access to fresh seawater that was changed daily. After 1 week, the native algae were rinsed in seawater and extracted in hexane for 24 h, after which the concentration of 1,1,3,3-tetrabromo-2-heptanone were measured by gas chromatography-mass spectrometry according to Nylund et al. (2008) and divided by the previously measured area of the native algae.
Coating experiment in the field
Due to the great difficulty of finding appropriate specimens of F. lumbricalis and C. officinalis growing adjacently to, while not yet being overgrown by, B. hamifera, we manipulated coating by B. hamifera at an ecologically relevant depth and site (i.e. where these species of algae naturally occur) using methods equivalent to those in the laboratory experiment. In this experiment, we chose to focus on C. officinalis due to practical constraints in the field. Specimens of C. officinalis and B. hamifera were collected from Yttre Vattenholmen, a small island in the Tjärnö archipelago that has large populations of both algal species. Individuals of the native alga C. officinalis were attached adjacent to B. hamifera on a plastic strip at a distance of approximately 1.5 cm using cable ties to make sure they did not touch each other, but still close enough that they could come in contact by waves and currents under natural conditions. The cable-tied native and exotic algae were then attached to stainless steel screws drilled into a block of concrete. To avoid confounding factors, the experiment was placed in an area with natural populations of both algal species. After one week in the field, the algae were brought back to the laboratory, and the C. officinalis were placed in vials with hexane for extraction and subsequent quantification of 1,1,3,3-tetrabromo-2-heptanone, as mentioned previously.
The data obtained from the experiments were analysed with analysis of variance (anova) using Statistica 6.0 (Statsoft Inc., Tulsa, OK, USA). Hypotheses about effects of main factors and interactions on algal growth in the co-cultures using adult B. hamifera and native competitors were tested using the general linear model:
where μ is the overall mean, B. hamifera treatment (Ti) and morphology (Mj) are fixed factors with two levels, algal species (A[M]k(j)) is a nested random factor with three levels and εijk is a random deviation. The hypothesis that the exotic algae will affect growth rates of the native algae is supported if there are significant differences in growth rates between adults of the native algae P. stricta, Pilayella littoralis, S. cirrosa, Cystoclonium purpureum, Cladophora rupestris and C. virgatum grown together with conspecifics compared with those grown with adults of the B. hamifera. Hypotheses about the effect of different concentrations of 1,1,3,3-tetrabromo-2-heptanone on settlement success of propagules of native species were tested using separate one-way anovas for each species. In the settlement experiment, support for the hypothesis that increasing concentration of 1,1,3,3-tetrabromo-2-heptanone will inhibit settlement of native algal spores is provided if there is a significant effect of the treatment and if higher concentrations have fewer settled spores compared to the controls. Post hoc analyses on differences between levels of concentration and controls were performed using Dunnett's test (Underwood 1997).
Competitive relationship between exotic and native adults
There were statistically no significant differences in growth rate between adults of the native algae P. stricta, Pilayella littoralis, S. cirrosa, Cystoclonium purpureum, Cladophora rupestris and C. virgatum grown together with conspecifics (controls) compared with those grown with adults of the exotic B. hamifera (Table 1 and Fig. 1). The morphology of the native algae had no effect on the growth rate, and there were no statistically significant interactions between the main factors (Table 1). The only significant effect in the analysis of variance was that growth rate generally differed among native species, which is to be expected in this system. More specifically, Pilayella littoralis and C. virgatum had markedly higher growth rates compared with the other native species (Fig. 1). Hence, the results from this experiment do not support our hypothesis that adults of B. hamifera affect growth rates of adult natives in competitive environments.
Table 1. Multifactorial anova on differences in growth rate of adult native species that were grown in co-cultures with either conspecifics or the non-indigenous red alga Bonnemaisonia hamifera
T × M
T × A (M)
Settlement of native algal propagules
There was a significant inhibitory effect of 1,1,3,3-tetrabromo-2-heptanone on the amount of settled spores/germlings from the native macroalgal species C. virgatum, P. stricta and U. lactuca and settled cells from the native microalgal species Cylindrotheca fusiformis (Table 2 and Fig. 2). Post hoc tests showed that the lowest concentration for inhibitory effects were found at 0.06 μg cm−2 for C. virgatum and U. lactuca and 0.125 μg cm−2 for P. stricta and Cylindrotheca fusiformis (Dunnett's test at α = 0.05, Fig. 2). Further graphical examination revealed that the amount of successfully settled native spores were reduced to a minimum for all species tested already at concentrations well below B. hamifera's natural surface concentrations of 2–4 μg of 1,1,3,3-tetrabromo-2-heptanone per cm2 (Fig. 2). The compound also had a stronger effect on P. stricta at higher concentrations, compared to the other natives, reducing the amount of settled spores to a mere 1% at 0.5 μg cm−2, and no spores settled at the higher concentrations (Fig. 2a). Similarly, at 1 μg cm−2, the mean number of settled germlings of U. lactuca was below 1 and reduced to 0 at 2 and 4 μg cm−2, whereas C. virgatum and Cylindrotheca fusiformis continued to colonize the panels, although to low extents, also at the highest concentration (Fig. 2b–d).
Table 2. One-way anova on effects of 1,1,3,3-tetrabromo-2-heptanone on settlement of native algal propagules
Transfer of 1,1,3,3-tetrabromo-2-heptanone from the exotic species to its native host algae
In the laboratory experiment, the measured mean surface concentrations of 1,1,3,3-tetrabromo-2-heptanone on the native host algae F. lumbricalis and C. officinalis placed adjacent to B. hamifera were 0.103 ± 0.013 and 0.194 ± 0.033 μg cm−2 (mean ± SE, n = 16), respectively (Fig. 3). In comparison, the mean surface concentration on C. officinalis in the field experiment was high (mean ± SE: 0.534 ± 0.191 μg cm−2; n = 8, Fig. 3). 1,1,3,3-Tetrabromo-2-heptanone was not present on C. officinalis collected in the field (i.e. controls, 0 μg cm−2; n = 8). When the mean surface concentrations from these experiments are compared to the levels of this compound affecting the settlement of native species (Fig. 2) described previously, it becomes clear that there would be a very large reduction in settlement for all four native species under these surface concentrations. The high concentration measured on C. officinalis in the field experiment corresponds to the 0.5 μg level, although slightly underestimated, where the settlement of Cylindrotheca fusiformis would be decreased by 68%, C. virgatum by 70%, U. lactuca by 87% and P. stricta by 97% compared to controls. These results show that the transfer of B. hamifera's secondary compounds to other surfaces has a high potential to reduce the settlement of native algal competitors.
In this study, we found that the settlement of native macroalgal propagules and microalgae is strongly inhibited on surfaces coated with the major defence compound from B. hamifera, 1,1,3,3-tetrabromo-2-heptanone, at concentrations much lower than the natural surface concentration of B. hamifera. We also show that this compound can be transferred from B. hamifera to the surface of its primary host algae, F. lumbricalis and C. officinalis, by both laboratory and field experiments. Comparisons between the results from these two experiments revealed that the amounts of 1,1,3,3-tetrabromo-2-heptanone transferred to the host algae is more than sufficient to prevent settlement by the native species tested. Together, these results provide support for the main prediction of the NWH, that is, that non-indigenous species become successful in the invaded range if they possess novel chemical weapons that can suppress native competitors.
In contrast to the results from our experiments with the recruiting stages, we found no effect of adult B. hamifera on growth rates of the adult native competitors C. virgatum, Cystoclonium purpureum, P. stricta, Pilayella littoralis, S. cirrosa and Cladophora rupestris. In general, there are remarkably few, if any, well-established examples of allelopathy among adult stages of marine macroalgae (Granéli & Pavia 2006), in strong contrast to the large number of reports on chemical defences against herbivory (Pavia et al. 2012) and biofouling (Steinberg, de Nys & Kjelleberg 2002;) in marine systems. There is, however, some circumstantial evidence for allelopathic effects among adult macroalgae. For instance, the green algae Chaetomorpha linum and Chaetomorpha aerea are never found at the same time in tide pools, most likely because they hinder the growth of one another, and both species have been reported to inhibit growth of the red algae C. virgatum (see Harlin 1987 and references therein). Furthermore, in a laboratory experiment, Cho et al. (2001) found that extracts from a few species, out of a wide range of green, brown and red algae tested, significantly inhibited the growth of the commonly epiphytic green algae Enteromorpha prolifera. It is also noteworthy that some of the most rigorous studies of allelopathy among benthic marine organisms have shown that macroalgae have the potential to suppress coral competitors through allelopathy (de Nys, Coll & Price 1991; Paul et al. 2011; Rasher et al. 2011), but similar well-documented examples for interactions among macroalgal adults are still lacking.
Allelopathic effects of adult macroalgae are commonly reported to have effects on early life stages of macroalgae, that is, propagules, spores and germlings (Råberg et al. 2005; Dworjanyn, de Nys & Steinberg 2006; Nylund et al. 2007), as well as on microalgae, bacteria and fungi (Harlin 1987; Ervin & Wetzel 2003; Gross 2003; Lane et al. 2009). Therefore, it is possible or even likely that the earliest life stages of native macroalgae are more heavily affected than adults by novel weapons in non-indigenous invasive macroalgae, in accordance with the results of the second experiment in this study. There was a very strong inhibitory effect of 1,1,3,3-tetrabromo-2-heptanone on the settlement of macroalgal spores of the native red algae C. virgatum and P. stricta and gametes from the green alga U. lactuca, as well as cells of the epiphytic microalga Cylindrotheca fusiformis. The surface concentration of 1,1,3,3-tetrabromo-2-heptanone in B. hamifera range from approximately 2 to 5 μg cm−2 in natural populations (Nylund et al. 2008). There were, however, significant negative effects on settlement of propagules from C. virgatum and U. lactuca already at 0.06 μg cm−2, which is about one-hundredth of the natural surface concentration. At concentrations of 0.5–1 μg cm−2, the settlement was reduced to a minimum, and at the natural surface concentration, no spores from C. virgatum and P. stricta settled on the coated surfaces.
This highly effective inhibition of settlement of natives parallels the reported negative effects on native plant propagules of compounds from the non-indigenous plants in North America, which gave rise to the NWH (Callaway & Ridenour 2004). Even though the specific mechanism causing the striking dominance of Centaurea stoebe (formerly maculosa) in the invaded range remains to be elucidated (Callaway et al. 2011), as the high levels of (−)-catechin initially reported for Centaurea stoebe are now revised (Bais et al. 2010), there are many other examples. For instance, Vivanco et al. (2004) showed that the active compound in root exudates of C. diffusa, 8-hydroxyquinoline, inhibits the root and shoot differentiation and germination of all the seven native species included in the study. Similarly, in the recent study by Inderjit et al. (2011a), volatile organic compounds (VOCs) from the leaf litter of the invader A. adenophora significantly reduced the seedling length and germination of the native plants Bidens biternata and Bambusa arundinacea. The invasive plant S. terebinthifolius has also been shown to decrease germination of seeds and seedling biomass for the native species Bidens alba and Rivina humilis through aqueous extracts from its leaves (Morgan & Overholt 2005), as well as shown to reduce the number of leaves on seedlings of the native mangroves Rhizophora mangle and Avicennia germinans grown in soil with natural levels of crushed S. terebinthifolius fruits (Donnelly, Green & Walters 2008). Thus, similar to previous studies on non-indigenous terrestrial plants, the chemical weapon of the macroalga in this study also shows very strong effects on early life stages of native species in the new range.
In our coating experiments, we could show both in the laboratory and in the field that 1,1,3,3-tetrabromo-2-heptanone can be transferred to a remarkably high extent to the surfaces of the native host algae F. lumbricalis and C. officinalis, on which B. hamifera commonly grows as an epiphyte. The concentration of the metabolite on C. officinalis measured in the field experiment was high enough to reduce virtually all settlement of all the native species that were tested. By spreading its highly inhibitory compound to surrounding surfaces, B. hamifera can thus prevent native species to settle in adjacent areas, which could be regarded as a form of effective pre-emptive competition. Pre-emptive competition is a fundamental process in ecological communities (Schoener 1983) and has been shown to facilitate the invasion by amphipod species as they rapidly occupy available habitats and make them inaccessible to natives (van Riel et al. 2007). To our knowledge, the only related aquatic example of transferred allelochemicals concerns fischerellin A from the benthic cyanobacteria Fischerella musicola, which inhibits photosynthesis of other cyanobacteria (Gross, Wolk & Jüttner 1991). This compound was argued to be directly transferred by cell–cell contact, because it could be extracted from lipophilic beads but not the surrounding culture medium (Gross 1999). Hence, although not discussed by the authors, if the compound adheres to lipophilic surfaces and remains active long enough to affect settled cells, this could be a form of pre-emptive competition among microorganisms.
The ability of B. hamifera to spread its secondary metabolites to the surrounding environment also shows interesting possible similarities to mechanisms of invasion by exotic plants in terrestrial systems. Recent studies show that invading plants can alter the composition of microbes in the soil through exudation of defence compounds and thereby affect soil–plant interactions (Reinhart & Callaway 2006). More specifically, Alliaria petiolata has been shown to reduce both the survival and settlement of native plant species by inhibiting the growth of soil microbes through its root exudates (Stinson et al. 2006; Callaway et al. 2008). In marine systems, biofilms of substrata could be regarded as the equivalence to soil microbe communities, and the composition of biofilms has been shown to affect settlement of marine invertebrate and algal propagules (Steinberg, de Nys & Kjelleberg 2002). In natural populations of B. hamifera, settlement inhibition of epiphytes could also be related to the fact that the 1,1,3,3-tetrabromo-2-heptanone has strong negative effects against marine bacteria (Nylund et al. 2007;; Nylund et al. 2008) and alters the bacterial composition of biofilms in field experiments (Persson et al. 2011). However, regardless if the settlement inhibition of natives is related to altered biofilms or more direct allelopathic effects, the ability to ‘reserve’ space is likely to be of equal, or higher, importance in benthic sessile communities, where free substratum is a major limiting resource and competition for space is severe (Paine 1966; Lubchenco & Menge 1978; Petraitis, Latham & Niesenbaum 1989).
In conclusion, we show that the non-indigenous invasive red alga B. hamifera has strong allelopathic effects on its common native competitors. The secondary compound 1,1,3,3-tetrabromo-2-heptanone of B. hamifera strongly inhibited the settlement of native propagules, and it has previously been shown to strongly deter native grazers and to have significant effects on native bacteria and biofilms. We also show, for the first time, that an ecologically active secondary metabolite can be transferred from a macroalga to adjacent surfaces at inhibitory concentrations. Due to the strong effects on different groups of native species of the multipurpose defence compound 1,1,3,3-tetrabromo-2-heptanone, we conclude that the NWH is a viable explanation for the invasive success of B. hamifera in the Northern Atlantic.
This work was performed within the Linnaeus Centre for Marine Evolutionary Biology (http://www.cemeb.science.gu.se/), and the Centre for Marine Chemical Ecology (http://www.cemace.science.gu.se/), at the University of Gothenburg and supported by a Linnaeus grant from the Swedish Research Councils FORMAS and VR. Support was also provided by the Swedish Research Council through contracts no. 621-2007-5779 and 621-2011-5630