985 LUCIANO AVIO 1 AND MANUELA GIOVANNETTI 2  Centro di Studio per la Microbiologia del Suolo, C.N.R., Universita  di Pisa, Via del Borghetto 80, 56124 Pisa, Italy  Dipartimento di Chimica e Biotecnologie Agrarie, Universita  di Pisa, Via del Borghetto 80, 56124 Pisa, Italy Spore proteins of different species and isolates of arbuscular mycorrhizal (AM) fungi were compared by PAGE. Reproducibility of protein patterns was assessed by using cultures of the same species either grown on different host plants, or produced during successive propagation cycles and stored up to 5 years. The results consistently showed that host species, different generations and storage, did not affect protein profiles, thus validating the accuracy of the method. Comparison among different geographical isolates of the same species revealed consistent protein patterns. The stability and diversity of spore protein profiles suggested that PAGE could be used to discriminate and identify AM fungal species and isolates. By contrast, the physiological state of spores affected the quality and quantity of bands, with germinating spores showing marked profile changes, as compared to quiescent spores, both in denaturating and native analytical conditions. The disappearance of some polypeptides in germinated spores might be related to the occurrence of storage proteins in AM fungi. Arbuscular mycorrhizal (AM) fungi, belonging to Glomales, include more than 150 species which live symbiotically in plant roots (Walker & Trappe, 1993). They exert a positive effect on plant growth and fitness (Harley & Smith, 1983 ; Newsham, Fitter & Watkinson, 1994) and play an important role in regulation of the structure and development of plant ecosystems (Grime et al., 1987 ; Read, 1991). Knowledge of the diversity of AM fungi is still poor, though it could prove crucial for germplasm conservation and economic exploitation of these fungi (Giovannetti & Gianinazzi-Pearson, 1994). Traditionally, the taxonomy of AM fungi is based on morphological characters of their asexual spores (Gerdemann & Trappe, 1974 ; Walker, 1983). Recent works have stressed the importance of ontogenetic characters for the effective evaluation of their biodiversity (Giovannetti, Avio & Salutini, 1991 ; Bentivenga & Morton, 1995 ; Morton, 1995). In- formation obtained in this manner may, however, mask biological, genetical and physiological variations occurring among AM fungi, and in several reports no correlation has been observed between morphological features of spores and functional properties of mycorrhizas (Morton, 1993 ; Allen et al., 1995). For a more comprehensive characterization of these fungi, Giovannetti & Gianinazzi-Pearson (1994) proposed the application of a multidisciplinary approach, and recent investi- gations have shown the usefulness of combining multiple methodologies for unraveling the biodiversity of different populations, species or isolates of AM fungi (Dodd et al., 1996). DNA sequences and polymorphisms, isozyme patterns and fatty acid profiles have been obtained by a variety of molecular and biochemical technologies, and are utilized for taxonomic and phylogenetic studies of AM fungi (Hepper et al., 1988 ; Jabaji-Hare, 1988 ; Simon et al., 1993; Wyss & Bonfante, 1993 ; Clapp et al., 1995 ; Sanders et al., 1995; LloydMacgilp et al., 1996; Ze  ze  et al., 1996). One-dimensional sodium-dodecyl-sulphate polyacrylamide gel electrophoresis (1-D SDS–PAGE) of soluble polypeptides is considered an effective systematic tool capable of giving good discrimination at the morphospecies level (Brasier, 1991) and of showing identity between isolates (Busse, Denner & Lubitz, 1996). It has been widely used to solve taxonomic problems in plants (Aiken & Gardiner, 1991), bacteria (Busse, El-Banna & Auling, 1989), and fungi (Sieber et al., 1991; Burgess, Malajczuk & Dell, 1995). The same technique has on the other hand been used to investigate meaningful physio- logical changes during the life cycle of many fungi (Huang & Staples, 1982 ; Linz & Orlowski, 1984 ; Novak & Kohn, 1988 ; Toth & Lacy, 1991 ; Dyer, Ingram & Johnstone, 1993) as well as molecular responses to induced stresses (Gianinazzi, 1984 ; Zoeger et al., 1992; Gro  pper & Rensing, 1993). However, only in a few studies has this technique been applied to AM fungal spores (Schellenbaum, Gianinazzi, & Gianinazzi- Pearson, 1992 ; Thingstrup et al., 1995), particularly in order to detect inter- and intra-specific variations among different isolates (Giovannetti & Gianinazzi-Pearson, 1994 ; Dodd et al., 1996). Nevertheless, these works have not studied the reliability of 1-D SDS–PAGE. A major problem in protein analysis is the production of reproducible and stable patterns, which can be obtained by the standardization of culture conditions and extraction procedures. This experimental approach may reveal technical difficulties since AM fungi are obligate symbionts. Mycol. Res. 102 (8) : 985–990 (1998) Printed in the United Kingdom The protein pattern of spores of arbuscular mycorrhizal fungi : comparison of species, isolates and physiological stages