Host-exclusivity and host-recurrence by wood decay fungi ( Basidiomycota-Agaricomycetes ) in Brazilian mangroves

Th is study aimed to investigate for the fi rst time the ecological interactions between species of Agaricomycetes and their host plants in Brazilian mangroves. Th irty-two fi eld trips were undertaken to four mangroves in the state of Pernambuco, Brazil, from April 2009 to March 2010. One 250 x 40 m stand was delimited in each mangrove and six categories of substrates were artifi cially established: living Avicennia schaueriana (LA), dead A. schaueriana (DA), living Rhizophora mangle (LR), dead R. mangle (DR), living Laguncularia racemosa (LL) and dead L. racemosa (DL). Th irty-three species of Agaricomycetes were collected, 13 of which had more than fi ve reports and so were used in statistical analyses. Twelve species showed signifi cant values for fungal-plant interaction: one of them was hostexclusive in DR, while fi ve were host-recurrent on A. schauerianna; six occurred more in dead substrates, regardless the host species. Overall, the results were as expected for environments with low plant species richness, and where specifi city, exclusivity and/or recurrence are more easily seen. However, to properly evaluate these relationships, mangrove ecosystems cannot be considered homogeneous since they can possess diff erent plant communities, and thus diff erent types of fungal-plant interactions.


Introduction
Wood decay fungi are mostly macroscopic Basidiomycota usually known as mushrooms, bracket fungi and earthstars.They occur in different habitats, being found with higher frequency and abundance in forests.They are able to degrade lignin and/or cellulose, having an essential role to the nutrient cycle of the environment.Some of them establish close relationship with the substrate and can be considered host-specific to the plant that they decay (Kendrick 2000;Deacon 2006;Webster & Weber 2007).
About the terms used for the ecological relationships between saprobe fungi and living or dead hosts, Zhou & Hyde (2001) proposed a redefi nition of these terms.Th e term "specifi city" would imply in the relationship between a living host and a fungus, thus, would not be applied to the saprophytic species; the term "exclusivity" would be the exclusive occurrence of a saprobe in a particular host groups, while "recurrence" would be the frequent or predominant occurrence of a parasitic or saprobe fungus in a particular host or host groups in the same habitat.Th is last term was previously referred to "preference" (Lindblad 2000;Gilbert and Sousa 2002;Gilbert et al. 2008).However, Zhou & Hyde (2001) suggested that "preference" should not be used for fungi, because it would imply in "act of volition on the part of the fungus".
Th e fungal community from a particular environment is related, among other factors, to the substrate availability 1 Departamento de Micologia, Universidade Federal de Pernambuco, Av.Prof. Nelson Chaves s/n, 50760-420, Recife, PE, Brazil 2 Departamento de Zoologia, Universidade Federal de Pernambuco, Av.Prof. Nelson Chaves s/n, 50760-420, Recife, PE, Brazil Host-exclusivity and host-recurrence by wood decay fungi (Basidiomycota -Agaricomycetes) in Brazilian mangroves (plant species composition, for example).So, in low diversity environments, like mangroves, the relationships between fungi and plants would be more easily observed (Gilbert and Sousa 2002).
Thus, this study aimed to investigate and report for the first time the ecological interactions between lignolytic Agaricomycetes and host plants in Brazilian mangroves and to verify if these interactions are significantly different from other mangroves.
The climate in these areas is defined by Köepen-Geiger classification as a borderline between tropical monsoon (Am) and tropical wet and dry savanna (As instead of Aw, when the dry season occurs during the time of higher sun and longer days) (Peel et al. 2007;Chen & Chen 2013).For the estimate of the proportion of available living and dead substrate, an imaginary line of 250 m was delimited inside each stand.On the starting (0 m), intermediate (125 m) and final points (250 m), three other lines of 40 m long (20 m on each side) were delimitated perpendicularly to the main line.The living or dead substrate present in these lines, with or without basidiomata, were quantified.The values for each line were summed and the proportion of each species and condition of the substrate were calculated for each stand.Based on the plant species in the studied areas, six categories of substrate were proposed: living A. schaueriana (LA), dead A. schaueriana (DA), living R. mangle (LR), dead R. mangle (DR), living L. racemosa (LV) and dead L. racemosa (LM).Fungi found on dead parts of living plants were considered as degrading dead tissue.

Statistical Analyses
Species abundance values were represented by the number of occurrence of specimens/individuals on each substrate; one specimen/individual may be represented by several basidiomata (Nogueira-Melo et al. 2014).For the analysis, we considered the species with more than five specimens to reduce error type II probability.The binomial probability test was applied to the species occurrences using BioEstat 5.0 program (Ayres et al. 2007).The level of significance was set at p < 0.05 for all analyses.

Results and discussion
Three hundred seventy two plants were counted, 135 belonging to LR, 111 to DR, 86 to LA, 35 to DA, four to LL and one to DL.It was observed that the mangrove areas differ in the proportion of plant substrate, with predominance of LR in Rio Formoso (74), DR in Maracaípe (63) and LA in Maria Farinha and Itamaracá (48 and 38 respectively).
The occurrence values of Agaricomycetes in L. racemosa were not considered in the analysis, since the low number of individuals of this plant species may cause distortion or overestimation of p values.Thirteen species occurred more than five times and were sufficiently abundant for the analysis (Tab.2).Except for C. molle, whose occurrence values did not differ between the substrate categories, 12 species presented significant p-values.Coriolopsis hostmanii, H. amethystea, S. commune e T. biforme occurred mainly in DA, while G. striatum, in DR.Besides, F. nivosa, H. hydnoides, H. iguazuense, P. gilvus, P. guyanensis, S. paradoxa and T. detrita occurred more than expected in dead substrates, regardless of the plant species (Tab.2).
Except for C. molle, all the abundant species for ecological analysis occurred significantly more in dead substrates (Tab.2).This is expected for lignolytic Agaricomycetes which are, in general, saprophytic and degrade dead plant tissue being the main agents decomposing trees of the forest (Kendrick 2000;Webster and Weber 2007).
Additionally, the wood characteristics may also influence the occurrence of wood decay fungi found in this study.For example, even with the higher frequency of LR on the studied transects, which could have influenced the hostspecificity, the test showed that no species occurred in living substrates.This observation was more evident in R. mangle than A. schauerianna.Rhizophora species are known as great producers of tannin, a compound not produced by Avicennia species (Erickson et al. 2004).Rhizophora mangle has 20 to 30 % of the compound in the bark, which characterizes the reddish color of the trunk (Haslam 1966).The tannin is a phenolic compound produced by the plant, considered a potent inhibitor of enzymes, of processes of decay and of attack by herbivores and phytopathogenic microorganisms.When the plant dies, the tannin levels fall, enabling the growth of decomposer fungi and other organisms (Silva & Silva 1999).Based on the concepts proposed by Zhou & Hyde (2001), in our study, host-exclusivity and host-recurrence were found.No host-specificity was observed.Six of the 13 analyzed species showed predominance in one of the host categories.Gloeophyllum striatum was host-exclusive in R. mangle, while C. hostmanii, H. amethystea, P. gilvus, S. commune and T. biforme were host-recurrence in A. schauerianna.

Host-exclusivity and host-recurrence by wood decay fungi (Basidiomycota -Agaricomycetes) in Brazilian mangroves
Similar results were also obtained by Gilbert & Sousa (2002) and Gilbert et al. (2008).In mangroves of Panama, Gilbert & Sousa (2002) found nine sufficiently abundant Agaricomycetes for statistical analysis, of which five showed host-preference (host-recurrence sensu Zhou & Hyde 2001) in three plant species, while in mangroves of Micronesia, Gilbert et al. (2008) reported host-recurrence between five species of Agaricomycetes and three plant species.
The data here presented may support the hypothesis of Gilbert & Sousa (2002) that in environments of low diversity, such as mangroves, the differential occurrence of a fungus in a host may happen more easily, since in those environments plant diversity is low and, therefore, the number of suitable host plants is higher.
Overall, the results were those expected for environments with low richness of plants, since about half of the species sufficiently abundant for the statistical analyses occupied differentially one of the established categories of substrate.However, to evaluate these relationships, the mangrove cannot be considered as a homogeneous ecosystem because there are many factors that influence locally the distribution and composition of plant species, which, in turn, will influence the distribution of fungi.

For
basidiomata collection, one stand of 10 000 m² (40 × 250 m) was established in each mangrove using the Global Positioning System -GPS.Eight surveys in each stand were undertaken (April to September 2009, December 2009 and March 2010, totaling 32 surveys) and all basidiomata were collected.
(2010) in the Caatinga [a Brazilian ecoregion characterized mostly by the xerophytic vegetation and hot and dry (BSh) climate], where Phellinotus piptadeniae was differentially frequent on Piptadenia moniliformes and Phellinotus neoaridus on Caesalpinia microphylla.

Table 2 .
Lignolytic Agaricomycetes considered for the analysis of predominance by substrate category.DR = dead Rhizophora mangle; DA = dead Avicennia schaueriana; LA = living A. schaueriana; LR = living R. mangle; pHS = level of significance for host species; pCS = level of significance for condition of the substrate (dead or living).