- Top of page
- Materials and Methods
Microfungi, along with other microorganisms including bacteria, form a highly diverse and abundant community of organisms within the plants. Many microorganisms, for example, mycorrhizal fungi and nitrogen-fixing bacteria, form associations with the host plant below ground in the root system. The above-ground parts of the plants are inhabited by epiphytic microorganisms on plant surfaces, and endophytic microorganisms that live at least part of their life cycle inside plant tissues without causing visible symptoms to the host (Wilson, 1995). The taxonomy of the endophytic fungi suggests that they are often closely related to pathogenic fungi (Clay, 1988; Gennaro et al., 2003). In some cases, these symbionts may benefit the host plant, for example by producing herbivore-deterrent mycotoxins. Some endophytes may, however, turn out to be pathogens, if the environmental factors, or genotypic or phenotypic condition of the host plant or their interactions are favorable for disease expression. Thus, the symbiotic endophytic fungi represent a continuum from antagonistic to mutualistic interactions (Saikkonen et al., 1998, 2004, 2006).
Foliar tree endophytes are nonsystemic and transmitted horizontally by spores (Helander et al., 1993; Wilson & Carroll, 1994). Tree leaves are endophyte-free when they are unfurling from buds; however, they soon become infected with airborne, rainborne or insect-vectored fungal spores, and the frequency of infections increases towards the end of the growing season (Helander et al., 1993; Ahlholm et al., 2002a), and in perennial plant organs it increases continuously over the years (Helander et al., 1994; Carroll, 1995; Hata et al., 1998; Kumaresan & Suryanarayanan, 2002). Although many tree endophytes are known to be host-specific, such as Rhabdocline parkeri in Douglas fir needles (Stone, 1987), Lophodermium piceae in Norway spruce needles (Barklund, 1987), and Neohendersonia kickxii in beech twigs (Danti et al., 2002), mutualistic relationships, such as in systemic and vertically transmitted grass endophytes and their hosts (Clay & Schardl, 2002; Saikkonen et al., 2004, 2006), should not be expected between woody plants and their foliar endophytes, because the fungus is not tightly coupled with and highly dependent on the host.
Abiotic and biotic environmental factors modify the frequency and composition of horizontally transmitted endophytic fungi in host plants (Helander, 1995; Saikkonen et al., 1996; Ahlholm et al., 2002a,b). In addition to availability of the spores, biotic environmental factors include, for example, the host plant phenotype and genotype, interactions with other microorganisms in the plant, and the herbivores using the same host plant and/or acting as vectors of the endophytes (Ahlholm et al., 2002a,b). The most important abiotic factors influencing the frequency and composition of endophytic fungi in the plant foliage are temperature and humidity (Colhoun, 1973). Germination of fungal spores occurs only over a limited range of temperature and requires access to water or high air humidity. In addition, release and dispersal of fungal spores are modified by weather conditions, especially by rain and wind (Fitt et al., 1989). Coincidence of viable spores and infection sites is another requirement for successful establishment of endophytic fungal thalli.
Islands provide an excellent opportunity to study factors affecting the dispersal ability of endophyte species and their relative frequencies in fragmented environments. According to the theory of island biogeography (MacArthur & Wilson, 1967), the size of an island and its distance from the mainland, relating to the immigration and extinction rates of species, are the two main predictors determining the number of species inhabiting it. An ‘archipelago’ for endophytic fungi is, however, hierarchically structured, comprising individual leaves (‘islands’ for single spore origin infections) and trees (genets) distributed across true islands. The number of infection sources and microclimatic conditions of an ‘archipelago’ are largely determined by growth form of individual trees (monocormic, polycormic or clonally growing cluster of trees) and whether individual tree genets grow solitarily or in clusters of trees in open areas, homogenous forests, or mixed forests (K. Saikkonen, unpublished). In this study Betula pendula and Betula pubescens were chosen as host species and the island-rich Finnish Archipelago Sea as the fragmented environment. Specifically, we wanted to examine if the number of fungal endophyte species is dependent on size and remoteness of the island; and if the common foliar endophyte species differ in their ability to disperse and infect birch leaves, depending on the birch species and environmental isolation.
- Top of page
- Materials and Methods
In concordance with the theory of island biogeography (MacArthur & Wilson, 1967) and empirical evidence with pathogenic fungi (Carlsson-Granér & Thrall, 2002), overall frequencies of endophytic fungi in birch leaves depend on the size of the island and its distance from the mainland. The largest islands close to the mainland had the highest overall foliar endophyte infection frequencies, and the frequencies decreased with increasing distance to the mainland and decreasing size of the island. This might be because of the low availability of inocula, as the number of host plants is sparse and they are infrequently distributed in small and distant islands. Because some of the studied islands are fairly small and far from the nearest potential inoculation source, the seasonally infected birch leaves need to receive new inocula by air from other islands. In the autumn the deciduous birch trees shed their endophyte-infected leaves and the new spores are formed in the fallen leaves during the following spring. The new inocula are usually spread by rain splash, insect vectors or wind to the unfurling leaves during the following growing season. However, in remote islands with sparse vegetation and harsh climatic conditions, it is likely that the fallen leaves will be wiped out under the birch trees during the winter, and thus the endophyte infections in the following season are highly dependent on immigration rate and adaptive radiation of the fungi.
The proportional frequencies of common endophyte species changed along the distance gradient and varied between the two birch species. The distribution of endophytic fungal species, however, appears not to be restricted in fragmented environment in terms of the size of the island and the number of nearby islands, indicating that the windborne endophyte spores are able to colonize the distant islands.
The three most frequently isolated fungal species in birch foliage in archipelago area, similarly to other studies conducted elsewhere (Helander et al., 1993; Elamo et al., 1999; Lappalainen et al., 1999), were F. betulae, G. setace and M. betulinum. F. betulae is the most abundant endophyte in birch leaves throughout the studied environments from the arctic tree line in northern Finland to natural and managed boreal forests in southern Finland (Helander et al., 1993; Elamo et al., 1999; Lappalainen et al., 1999; Helander et al., 2006) and it is believed to be widespread throughout the distribution range of Betula spp. (Barr, 1968). The species is abundant in B. pendula and B. pubescens as well as in several birch hybrids (Saikkonen et al., 2003). It was also the most frequent endophyte in leaves of B. pubescens collected a few weeks after bud burst in Switzerland (Barengo et al., 2000). The Venturia spp. type ascospores (teleomorphic state of F. betulae) are commonly detected from the air samples after rainy periods and their occurrence is correlated with the increase of endophyte infections in birch trees (M. Helander, unpublished). The airborne ascospores are easily dispersed over large areas, and the success of the infection is assured by humid conditions after rain.
In accordance with previous studies (Lappalainen et al., 1999; Saikkonen et al., 2003), G. setacea was frequently isolated from B. pubescens leaves (25% of all the isolates), but only rarely from B. pendula foliage (3% of all the isolates). B. pendula may have some morphological or chemical barriers that prevent the success of G. setacea infections on its leaves. However, there may be some genetically different strains of G. setacea that are able to overcome these barriers. A recent study by Helander et al. (2006) showing that G. setacea was frequently isolated from B. pendula leaves in natural and managed forests, but only rarely from trees growing in sapling sites, suggests that the genetic background of the host plant or some environmental factors may also affect the success of infection. However, in the present study, only 2–6% of the variation in G. setacea frequencies was explained by distance to the mainland or size of the land area around the study site (Fig. 3). Thus, it seems that G. setacea is able to survive from year to year or effectively re-immigrate to a certain area in a fragmented environment. This may hold true, because G. setacea is also able to infect shoots (Kessler, 1978) where the fungus is able to survive the winter and other harsh environmental conditions.
Similarly to G. setacea, less than 6% of the variation in M. betulinum frequencies was explained by distance to the mainland or size of the land area around the study site (Fig. 3). Thus, in contrast to island biogeography theory (MacArthur & Wilson, 1967), the distribution of G. setacea and M. betulinum endophytes is not predominantly determined by distance to mainland and size of the island. Proportional infection frequencies of M. betulinum were highest in the two Isokari islands, which are isolated from the other islands approx. 18 km from the mainland. One island is fairly large, while the other is a small island in the vicinity of the larger one. Proportionally high infection frequencies of M. betulinum compared with other common endophytes may reflect adaptive differences among the fungal species to environmental conditions in the outer archipelago. In addition to the availability and dispersal ability of the fungal inocula, the early moments of the germination of the fungus are also critical for the realization of the infection. Successful infection requires, for example, the correct temperature, windiness and humidity.
In conclusion, the foliar endophytes of birch trees are able to disperse to fairly fragmented areas, but their frequencies seem to depend on the environmental isolation and size of the island rather than microclimatic conditions. However, fungal species show differences in their adaptive ability to persist in harsh conditions in the outer archipelago. Since many of these fungal endophytes or their close relatives are dormant saprophytes or latent pathogens (Saikkonen et al., 2004), recognizing the driving forces of patterns and frequencies of endophytic fungi enables us to understand changes in mycoflora caused by fragmenting previously continuous habitat through forest practices. Contrary to the conventional wisdom of negative effects of habitat fragmentation on biodiversity and ecosystem functions, forest fragmentation appears to decrease pathogen risk.