Over the last decades, the organization of the secretory and endocytic pathways has been studied in a variety of different eukaryotic organisms. These studies have revealed that the principal organization of the endomembrane system and the molecular machineries involved in vesicular trafficking are conserved among all eukaryotes. In general, the transport between different intracellular organelles is mediated by cargo-laden vesicles that bud from a donor and then specifically fuse with an acceptor compartment. Key players during the final fusion step are the so-called SNARE proteins. These proteins are a large family of small cytoplasmically orientated membrane proteins that are typically tail-anchored. Their key characteristic is the so-called SNARE domain, an extended stretch of heptad repeats that is usually connected to a single transmembrane domain by a short linker. SNARE domains of heterologous sets of SNARE proteins have the ability to assemble into tight, parallel four-helix bundle complexes. It is thought that complex formation between SNARE proteins from opposing membranes provide the energy that drives membrane fusion (reviewed in [1
], Fig. &).
Figure 1 SNARE proteins are key factors in vesicle trafficking in yeast. A) Current model of the role of SNARE proteins in vesicle fusion. Heterologous sets of SNARE proteins, here exemplified by the secretory set of SNAREs, assemble in a zipper-like fashion into (more ...)
Although the concept of SNARE-mediated membrane fusion emerged in the 1990s [5
], several SNARE proteins had been discovered earlier. The field of membrane trafficking was greatly inspired by ingenious genetic screens in the single-cell fungi Saccharomyces cerevisiae
for mutants with defects in the secretory pathway in the late 1970s [8
]. In these mutants, secretion was blocked at higher temperatures. These so-called SEC
mutants accumulate secretory proteins at the point in the secretory pathway that is blocked and often, they could be distinguished by their phenotype. In the following years, it became clear that SEC
genes generally encode for key components of the machineries that mediate trafficking of a transport vesicle, including a few SNARE proteins. All at once, straightforward genetic tools had catapulted baker's yeast to the front row for studying vesicle trafficking pathways. In fact, other genetic screens for mutants showing defects in other transport processes (reviewed in [10
]) have turned up novel yeast SNAREs. The last yeast SNARE was tracked down in 2003, yielding an overall repertoire of 24 different SNARE genes in baker's yeast [14
] [see Additional file 1
]. For most trafficking steps in baker's yeast, distinct units of four interacting SNARE domains have now been assigned ([1
], Fig. ). However, not all details of the trafficking routes in yeast are resolved. Generally, different SNARE family members are localized on distinct organelles and vesicles that demarcate particular membranes. However, it is becoming clear that some trafficking steps cannot be identified by a unique set of SNAREs, since some SNAREs are involved in more than one trafficking step. In addition, they may participate in different SNARE complexes (for a discussion, see [16
]). To appraise S. cerevisiae
as model organism for studying vesicle trafficking pathways better, it is also necessary to understand the limits of the organism, since baker's yeast appears to be a secondarily reduced organism. For example, S. cerevisiae
has only a limited ability to produce hyphae (long branching multicellular filaments). The capability to produce polarized hyphae, however, is a hallmark of the fungal kingdom, allowing them to colonize and exploit new substrates efficiently. In addition, whereas most fungi are multicellular organisms, S. cerevisiae
most of the times grows as single cells that reproduce asexually. It thus seems possible that adaptation to a relatively simple lifestyle was accompanied by a degeneration of the intracellular trafficking itinerary together with the involved machineries.
A comparison of the SNARE gene repertoires of S. cerevisiae
with that of five other fungal species, including three species of filamentous fungi, revealed that, in general, the members of the SNARE family are largely conserved in fungi [18
]. This notion has been corroborated by subsequent inspections of the SNARE repertoire of few other fungal species [19
]. Together, these studies put forward that the yeast lifestyle did not entail a radical change in the intracellular trafficking pathways. The bioinformatic strategies used to identify the homologs of the yeast SNAREs, however, did not provide a universal classification scheme. Consequently, it was impossible to compare the SNARE sets of different fungi with each other in an unambiguous manner, let alone comparing them with SNARE sets from other eukaryotes like animals. Hence it remains unclear whether SNARE types, which are possibly linked to a novel transport step, evolved or degenerated in particular fungal lineages.
Fungi are thought to be more closely related to animals than to plants and, currently, are placed with animals and several protistan taxa into the monophyletic group of Opisthokonta. Notably, morphological observations show that the appearance of analogous intracellular organelles such as the vacuoles/lysosomes and the Golgi apparatus are markedly different between yeast and animal cells. It is unclear, however, whether these differences are reflected in their respective SNARE inventory. Recently, using Hidden Markov Models (HMM) we have established a precise classification for all eukaryotic SNARE proteins [21
]. Based on this classification, we were able to deduce that the emergence of multicellularity in animals went along with an expansion of the set of SNARE proteins, in particular the expansion of the SNAREs involved in endosomal trafficking [22
In order to analyze the differences between the SNARE repertoires of fungi and animals and the evolutionary changes in the SNARE inventory of different fungal lineages, we have collected and classified the SNAREs from a broad range of different fungi according to our HMMs, making use of the substantial number of available genomic sequences. To identify specification and duplication events, and to pinpoint the differences between the set of SNARE proteins in fungi and animals, we used phylogenetic information to reveal the history of SNARE genes. Afterwards, we mapped features of the fungal SNARE evolution onto the species tree.