Trehalases and trehalose hydrolysis in fungi

Abstract

The simultaneous presence of two different trehalose-hydrolysing activities has been recognised in several fungal species. While these enzymes, known as acid and neutral trehalases, share a strict specificity for trehalose, they are nevertheless rather different in subcellular localisation and in several biochemical and regulatory properties. The function of these apparently redundant activities in the same cell was not completely understood until recently. Biochemical and genetic studies now suggest that these enzymes may have specialised and exclusive roles in fungal cells. It is thought that neutral trehalases mobilise cytosolic trehalose, under the control of developmental programs, chemical and nutrient signals, or stress responses. On the other hand, acid trehalases appear not to mobilise cytosolic trehalose, but to act as ‘carbon scavenger’ hydrolases enabling cells to utilise exogenous trehalose as a carbon source, under the control of carbon catabolic regulatory circuits. Although much needs to be learned about the molecular identity of trehalases, it seems that in fungi at least one class of acid trehalases evolved independently from the other trehalases.

Trehalose, Trehalase, Stress response, Fungi, Metabolism

1 Introduction

The disaccharide trehalose (α-d-glucopyranosyl-α-d-glucopyranoside) accumulates in resting cells of fungi, such as spores, sclerotia and stationary-phase cells, and also in vegetative cells subjected to various forms of stress. A wealth of information about trehalose metabolism and its regulation in yeasts cells, mostly Saccharomyces cerevisiae, shows that trehalose metabolism is closely linked to other important biological phenomena, suggesting principles which may be of universal significance among fungi. For instance: (i) the dual role of cytosolic trehalose, which acts both as a carbon reserve and as a stress protectant (reviewed in [1–6]); (ii) the complex biological activities of trehalose-6-phosphate synthase, a component of the trehalose biosynthetic complex, which also contributes to modulate glucose influx and glycolysis, and might participate in, or interfere with, important signalling pathways (reviewed in [7]); (iii) the existence of two classes of trehalases (EC 3.2.1.28). While these enzymes are strictly specific for trehalose as substrate, they are nevertheless strikingly different in terms of subcellular localisation, and in other significant biochemical and regulatory properties (reviewed in [2]). The exact involvement of these two trehalases in trehalose hydrolysis has been a matter of speculation for many years. The purpose of this review is to analyse the biological role of acid and neutral fungal trehalases, in the light of evidence obtained with both yeasts and filamentous fungi, in an attempt to find unifying concepts about the role of these enzymes.

2 Fungal trehalases

In practically all fungal species studied trehalose hydrolysis is carried out by the enzyme trehalase. The only known exception thus far is Pichia fermentans, a yeast with trehalose phosphorylase activity [8]. All well characterised trehalases are strictly specific for trehalose. Multifunctional glycosidases with ability to hydrolyse trehalose have not yet been demonstrated convincingly. Fungal trehalases have been grouped into two main classes, according to the pH optimum and regulatory properties, as acid or non-regulatory trehalases (extracellular or vacuolar glycoproteins) and neutral or regulatory trehalases (unglycosylated cytosolic proteins) [2, 9] (Table 1). In several fungal species, such as Saccharomyces cerevisiae[10], Schizosaccharomyces pombe[11], Candida utilis[12], Torulaspora delbrueckii[13], Kluyveromyces lactis[14], Fusarium oxysporum[15], Mucor rouxii (unpublished observations), and very likely in many others, the two trehalases coexist. This fact, which would in principle suggest a separate function for each enzyme, remained unexplained for many years, probably because most studies focused only on the hydrolysis of cytosolic trehalose. Little attention has been given to the fact that many fungi also use extracellular trehalose as a source of carbon, a property which might provide a clue to the understanding of the specialised physiological role of the two distinct trehalases.

1

Biochemical properties of some representative fungal trehalases

Organism	pH/temperature (°C) optimum	Km (mM)	Vmax (μmol min−1 mg protein−1)	Mr (kDa) (native)	Regulation by phosphorylation demonstrated	References
Neutral trehalases (Ca2+/Mn2+-activated)
S. cerevisiae	6.7/30	5.7	variable	170	yes	[10, 21]
M. rouxii	7.0/35	?	variable	?	yes	[33], unpub. res.
P. blakesleeanus	7.5/30	55	variable	210	yes	[32]
F. oxysporum	6.8/30	8.45	?	100	no	[15]
Acid trehalases (Ca2+/Mn2+-ATP-insensitive)
S. cerevisiaea	4.5/45	4.7	99	215	no	[20, 27]
N. crassab	5.0/50	0.36	79	360	no	[40], unpub. res.
C. aureumb	4.0/50	0.29	58	450	?	[38]
T. reeseib	4.4/50	3.1	50	88	?
F. oxysporuma	4.6/45	0.43	?	160	no	[15]
A. nidulansb	4.5/?	?	?	370	?	[42]
Acid trehalases (Ca2+/Mn2+-activated; ATP-inhibited)
H. grisea form Ib	5.6/60	2.3	176	580	no	[34]
H. grisea form IIa	5.5/60	0.86	392	360	no	[35]
S. thermophilum form Ia	6.0/60	3.58	1275	370	no	[36]
S. thermophilum form IIa	6.0/65	2.24	1908	398	no	[36]

^aIntracellular (vacuolar?) glycoproteins. ^bExtracellular or cell wall-bound glycoproteins. ?, no data available.

3 Catabolism of trehalose in fungi

3.1 Two uses for extracellular and cytosolic trehalose

Many fungi can utilise trehalose as the sole extracellular source of carbon. Since trehalose accumulates in many microbial cells, insects and some plant tissues, and is chemically quite stable, soils rich in decaying biomass may contain significant quantities of trehalose which can serve as a potential source of carbon for organisms endowed with trehalase activity.

On the other hand, the physiological significance of cytosolic trehalose appears more complex. Early studies regarded trehalose as just another carbohydrate reserve, like glycogen, to support the energy requirements during specific transitions of the life cycle (i.e. spore germination), or for long-term survival during storage and dispersal of fungal spores. In fact, some fungal spores have the capacity to germinate in water, apparently using their trehalose stock as sole carbon reserve (reviewed in [3]). Other authors, however, have questioned the energetic value of cytosolic trehalose mobilisation (see for instance [16]).

In another line of research, it was shown that vegetative cells subjected to various forms of stress, such as toxic substances, high temperatures, freezing, desiccation, and others, also accumulate high levels of trehalose. These observations support the concept that cytosolic trehalose is also part of a stress protection system [4, 5]. The properties of trehalose, as a stress protectant or as a stabilising agent, have been convincingly demonstrated in vivo, in situ with permeabilised cells, and also in vitro with liposomes, membranes or purified proteins, attracting considerable biotechnological interest (reviewed in [6]). Nevertheless, the exact mode of action of cytosolic trehalose as a participant of stress responses is still somewhat controversial. For instance, in specific situations trehalose has to act in concert with other unidentified factors to provide stress protection (see for instance [17–19] and references therein).

4 A specific role for neutral and acid trehalases, respectively, in mobilisation of cytosolic trehalose and hydrolysis of extracellular trehalose?

4.1 The case of yeast

The best characterised fungal trehalases so far are the vacuolar acid trehalase and the cytosolic neutral trehalase of S. cerevisiae[10, 20–26]. Neutral trehalase genes have been cloned from S. cerevisiae (NTH1 and NTH2[22]) and K. lactis (NTH1[14]). Another gene (ATH1), required for the activity of the vacuolar acid trehalase of S. cerevisiae, was cloned in Holzer's laboratory [25]. Its product (ATH1) turned out to be the enzyme itself [27], but surprisingly, the ATH1 sequence shows no homology with that of NTH1, or with sequences of acid trehalases from Escherichia coli, insects or rabbit [25].

Neutral trehalases are tightly regulated enzymes. The neutral trehalase of S. cerevisiae, as well as those from K. lactis, C. utilis, T. delbrueckii, S. pombe and Pachysolen tannophilus, are all controlled by signalling pathways which result in reversible activation by phosphorylation of the enzyme. These signalling pathways can be triggered in vivo by glucose, nitrogen sources, heat shock, and some chemicals like protonophores which produce intracellular acidification. cAMP-dependent and -independent phosphorylation has been demonstrated (reviewed in [9]). The activity of the neutral trehalases of S. cerevisiae[10, 21] and K. lactis[14] is stimulated in vitro by the ions Ca²⁺ and Mn²⁺, and possibly slightly inhibited by ATP [28]. Their protein sequences exhibit N-terminal domains with two consensus sites for cAMP-dependent protein phosphorylation [14, 24], and a putative Ca²⁺-binding sequence [14].

The acid trehalase of S. cerevisiae (ATH1) is a vacuolar glycoprotein which is also immunologically unrelated to the neutral trehalase [20]. The acid trehalase is not activated by phosphorylation, neither is it stimulated by Ca²⁺ or Mn²⁺ or inhibited by ATP. Instead, this enzyme seems to be controlled by carbon catabolite repression [20] and catabolite inactivation [29], and is inducible by trehalose [26]. In contrast to neutral trehalase activity, which is high in exponential-phase cells and drops in stationary phase, the activity of acid trehalase rises in stationary phase coincident with the accumulation of cytosolic trehalose [29]. This pattern of expression might be common for acid and neutral trehalases of other fungi.

Recent genetic studies using neutral trehalase-deficient mutants of S. cerevisiae[23] and K. lactis[14] contribute to clarify the physiological role of this enzyme. These mutants exhibit a high constitutive trehalose level and do not mobilise trehalose upon addition of glucose or nitrogen sources to correspondingly starved cells [14, 19], in spite of the normal activity of acid trehalase. Furthermore, cells of the neutral trehalase-deficient mutant (nth1Δ) of S. cerevisiae, after being subjected to a heat shock and returned to a physiological temperature, do not mobilise the cytosolic trehalose accumulated during the heat shock and show delayed growth recovery (‘poor-heat-shock-recovery phenotype’[23]). Altogether these results provide a clear-cut demonstration for the specific role of the neutral trehalase in mobilisation of cytosolic trehalose. On the other hand, it needs clarification whether the ‘poor-heat-shock-recovery phenotype’ results from the absence of trehalose hydrolysis releasing glucose for cell recovery, or whether high levels of cytosolic trehalose somehow interfere with the resumption of growth.

The biological role of the acid trehalase of S. cerevisiae has remained unclear until recently. It was speculated that this enzyme may hydrolyse cytosolic trehalose during autolysis [29]. Although this is very likely also true, recent studies by Holzer's group using ath1Δ and nth1Δ mutants of S. cerevisiae demonstrated that the ath1Δ mutant, which is not defective in the mobilisation of cytosolic trehalose [23], is nevertheless unable to use trehalose as a carbon source. In contrast, the nth1Δ mutant grows normally in trehalose [26]. These authors concluded that the acid trehalase in yeast cells was responsible for hydrolysis of extracellular trehalose, while the neutral trehalase was responsible for the hydrolysis of the cytosolic endogenous pool [26].

How fungi utilise extracellular trehalose as a carbon source constitutes a separate problem. Apparently, in S. cerevisiae hydrolysis of trehalose takes place in vacuoles [26]. Such a way of utilising an exogenous carbon source seems strange because it would first require incorporation of trehalose, either through pinocytosis [26] or by transport activity [30], and also some way of communication between the vacuolar compartment and the cytosol. In fact, S. cerevisiae grows very slowly in trehalose [26], in contrast to N. crassa[31], M. rouxii and many other fungi (unpublished observations) endowed with periplasmic trehalases. Perhaps the growth of S. cerevisiae in trehalose is limited by trehalose uptake and intracellular transfer, rather than by acid trehalase activity. S. cerevisiae acid trehalase differs from other vacuolar hydrolases in that it is heavily glycosylated, a characteristic of secreted proteins [27]. Furthermore, overexpression of ATH1 increases acid trehalase activity 10-fold, and causes the vacuolar enzyme to be secreted in an active form in the periplasm [25]. Perhaps a transfer of the vacuolar enzyme to the periplasm might also occur in cells of the wild-type strain under specific conditions. In that case, it would be interesting to know if periplasmic trehalase activity has any influence under those conditions on the rate of growth with trehalose as a carbon source.

4.2 The case of filamentous fungi

The vast majority of trehalases reported for filamentous fungi belong to the class of acid trehalases (Table 1). Exceptions are the neutral trehalases in the Zygomycetes, specifically Phycomyces blakesleeanus[32] and M. rouxii[33]. Neutral trehalases of Zygomycetes show clear similarity with yeast neutral trehalases: (i) activation by cAMP-dependent phosphorylation in vitro, and by stimuli triggering spore germination, i.e. heat shock or glucose, in vivo; (ii) stimulation by Ca²⁺ and Mn²⁺ and slight inhibition by ATP in vitro ([2, 3, 9], unpublished observations) (Table 1). Examples of Zygomycetes with both acid and neutral trehalases, such as found in S. cerevisiae and in other yeasts, have not been reported up to now. However, we have recently obtained evidence for the existence of an acid (pH 4.5) trehalase in M. rouxii. This enzyme is in the periplasm, it exhibits a high temperature optimum (45°C), and is repressed by glucose and induced by exogenous trehalose (unpublished results).

Acid trehalases from filamentous fungi are glycoproteins generally present at the surface of spores or mycelium, or in vacuoles, and less frequently freely secreted into the medium. They exhibit a high temperature optimum and elevated thermostability. None of these enzymes seems to be controlled by reversible phosphorylation. Some biochemical properties of acid trehalases from filamentous fungi suggest that they might represent a heterogeneous group (Table 1). Distinctive properties are the activation by Ca²⁺ and Mn²⁺ and the clear inhibition by ATP, exhibited by the acid trehalases of the thermophilic fungi Humicola grisea[34, 35] and Scytalidium thermophilum[36]. Interestingly, at least the activation by Ca²⁺ and Mn²⁺ (and inhibition by EDTA), and perhaps inhibition by ATP, are shared with the neutral trehalases of yeasts and Zygomycetes. On the other hand, the acid trehalases of Dictyostelium discoideum[37], Chaetomium aureum[38], Pilobolus longipes[39], Neurospora crassa and M. rouxii (unpublished observations) are insensitive to these effectors, and could represent a different class among acid trehalases. Future work will have to show whether these enzymes are related to the acid trehalase of yeast, or to the neutral trehalase and the acid trehalases in other organisms.

5 Hydrolysis of cytosolic trehalose in filamentous fungi: do the acid trehalases participate?

Spores of many filamentous fungi are rich in cytosolic trehalose, which is mobilised during germination. In the Zygomycetes mobilisation of trehalose coincides with the activation of the cytosolic neutral trehalase. On the other hand, for most filamentous fungi which also accumulate and mobilise cytosolic trehalose such as N. crassa[40], P. longipes[39] and others, only acid trehalase activity has been reported. These acid trehalases, which do not display activation by phosphorylation, are separated from cytosolic trehalose by a membranous barrier (plasma membrane or tonoplast). Thus, in order to explain the mobilisation of cytosolic trehalose during germination in these fungi, it was suggested that germinative stimuli may somehow alter plasma membrane permeability allowing cytosolic trehalose to reach the compartment (periplasm) with the acid trehalase [40]. This compartmentation/decompartmentation model [2] never received direct experimental support. More recent studies on the mobilisation of cytosolic trehalose in N. crassa argue against the compartmentation/decompartmentation hypothesis. They demonstrate that cytosolic trehalose is hydrolysed to glucose in the cytosol, and not in the periplasm. Germlings of N. crassa wild-type strains accumulate trehalose at 45°C and hydrolyse it again upon return to physiological temperatures [41]. The strongest evidence against the compartmentation/decompartmentation hypothesis was obtained using an acid trehalase-deficient (tre) mutant of N. crassa. This strain shows less than 5% of the trehalase activity of the wild-type strain and is unable to utilise trehalose as a carbon source. Nevertheless, it hydrolyses cytosolic trehalose to glucose as effectively as the wild-type [31]. Similar findings have been reported by d'Enfert and Fontaine [42] in Aspergillus nidulans. These authors succeeded in cloning the acid trehalase (treA) gene of this fungus and demonstrated that its product is required for growth on trehalose, but is not involved in the mobilisation of the intracellular pool.

Altogether, these results show that the acid trehalases of N. crassa, A. nidulans, and probably many other fungi, only function as ‘carbon scavenger’ hydrolases which do not contribute to the metabolism of cytosolic trehalose. These results would also imply the existence in N. crassa and A. nidulans of cytosolic (neutral?) trehalases, involved in the mobilisation of cytosolic trehalose. Thus far their activity has escaped biochemical detection, in spite of intense efforts [31, 42]. However, d'Enfert and Fontaine [42] also mention the cloning of a gene in A. nidulans that encodes a protein with strong homology to the S. cerevisiae neutral trehalase.

6 Concluding remarks

Evidence obtained from various genera of yeasts and filamentous fungi strongly suggests that acid and neutral trehalases have specific and independent roles, specialised in the hydrolysis of extracellular and cytosolic trehalose, respectively. Therefore, the coexistence of the two classes of trehalases may be rather common, if not the rule, in fungi which accumulate and mobilise cytosolic trehalose, and also utilise extracellular trehalose as a carbon source.

Other aspects concerning fungal trehalases may also be worth comparative study. (i) To assume that acid and neutral trehalases have exclusive specialised functions implies that all fungal species which accumulate and mobilise cytosolic trehalose should possess cytosolic (neutral?) trehalases regulated by developmental programs, chemical and nutrient signals, or stress responses. Few of these enzymes have been recognised up to date. The best characterised neutral trehalases are regulated by reversible phosphorylation mechanisms, but additional regulatory mechanisms might still be discovered. (ii) The evolutionary origin of trehalases also poses an interesting problem. Considering the narrow catalytic specificity of these enzymes, one would expect to find homology at the molecular level for all trehalases, suggesting a common evolutionary origin. Indeed, the neutral trehalases from S. cerevisiae and K. lactis, and the acid trehalases from E. coli, the mealworm Tenebrio molitor, the silkworm Bombix mori and rabbit small intestine show sufficient homology to suggest evolutionary relationships [14]. Surprisingly, the vacuolar acid trehalase of S. cerevisiae[25, 27] and the acid trehalase of A. nidulans[42], which share 53.8% similarity [42], show no homology with other acid and neutral trehalases, suggesting an independent evolutionary origin.

Another interesting problem concerns the trehalose synthase system of filamentous fungi and its possible involvement in the control of glucose metabolism and cellular signalling pathways. Very little is known about this issue in organisms other than yeasts, but preliminary studies suggest that N. crassa trehalose-6-phosphate synthase [43] shares some properties with the S. cerevisiae enzyme [44] such as a rather high (55°C) temperature optimum, and temperature-conditional inhibition by phosphate and activation by fructose-6-phosphate. A deeper insight into the involvement of trehalose metabolism in diverse basic phenomena of microbial physiology may benefit from additional comparative studies on a wider range of representatives of the fungal kingdom.