OUP user menu

The appearances of autolytic and apoptotic markers are concomitant but differently regulated in carbon-starving Aspergillus nidulans cultures

Tamás Emri, Zsolt Molnár, István Pócsi
DOI: http://dx.doi.org/10.1016/j.femsle.2005.08.015 297-303 First published online: 1 October 2005


In ageing, carbon-depleted cultures of Aspergillus nidulans strain FGSC 26 progressing apoptotic-type cell death was detected, characterised by increasing numbers of Annexin V and TUNEL stained cells after protoplastation. DAPI staining of autolysing mycelia revealed numerous nuclei with elongated, stick-like morphology, which was not observed in surviving hyphal fragments representing a cell population adapted to carbon starvation. Apoptotic cell death was also progressing in aging cultures of the non-autolysing loss-of-function fluG and ΔbrlA mutants, indicating that apoptotic cell death and autolysis were regulated independently. In accordance with this, sphingosine derivatives added to A. nidulans cultures increased cell death rates without influencing autolytic biomass losses and hydrolase production.

  • Aspergillus nidulans
  • Apoptosis
  • Autolysis
  • fluG
  • brlA

1 Introduction

In spite of its primary importance in numerous biotechnological processes including secondary metabolite and heterologous protein production, autolysis is a relatively poorly studied and understood aspect of fungal growth in submerged cultures [1,2]. In surface cultures, breakdown of vegetative tissue of filamentous fungi is tightly coupled to conidiogenesis and, not surprisingly, the regulatory networks behind these phenomena share many common elements [35]. For example, FluG–BrlA signalling plays a pivotal role in the initialisation of sporulation in Aspergillus nidulans [4,6] and has also been shown to contribute to the initialisation of autolysis in submerged cultures [7]. Tightly coupled and highly concerted autolytic events such as cell-wall hydrolase production, fragmentation of hyphae, disorganisation of pellets and autolytic loss of biomass were disconnected by loss-of-function mutations in some genes of the FluG–BrlA regulatory network [7]. Moreover, the tight connection between pellet morphology and proteinase production was also erased by these mutations [7].

There are several lines of evidence clearly indicating that fungal autolysis is an energy and de novo protein synthesis-dependent process [810]. Entry into the stationary phase of growth was associated with the appearance of apoptosis markers, including phosphatidylserine exposure to the outer leaflet of the membrane and TUNEL staining of the nuclei, in Aspergillus fumigatus cultures [9]. This apoptotic-like phenotype should be beneficial for the surviving cell population by removing genetically damaged cells, supplying replicative younger cells with nutrients and promoting the re-growth of mutants better-adapted to the changing environment [11,12].

Although cell death and self-digestion by autolytic enzymes seem to be inherently coupled and concomitant events taking place in autolysing cultures of filamentous fungi, we demonstrate here that these processes are regulated independently in A. nidulans.

2 Materials and methods

2.1 Organism and growth conditions

A. nidulans strain FGSC 26 (biA1, veA1) was a kind gift of Dr. S. Rosén (University of Lund, Sweden). The loss-of-function fluG mutant FGSC 744 (fluG1, pabaA1, yA2) and the ΔbrlA mutant FGSC 1079 (biA1, pabaA1, pyroA4, ΔbrlA, veA+) strains were purchased from the Fungal Genetic Stock Center [13]. Strains were grown in shake flasks (500 ml) containing 100 ml minimal-nitrate medium, pH 6.5 [14] supplemented with 0.5% yeast extract. Culture media were inoculated with 5 × 107 spores (FGSC 26) or mycelia from a fresh agar slope (FGSC 744, FGSC 1079) and were incubated at 37 °C, 200 rpm for up to 96 h. To induce apoptosis, selected cultures were also supplemented with 0.05 g L−1d, l-dihydrosphingosine or 0.03 g L−1 phytosphingosine at 30 h of incubation. Both compounds have already been used to induce apoptosis in germinating spores and young hyphae in A. nidulans [15].

2.2 Detection of glucose consumption and the viability of the cultures

The consumption of glucose was monitored in culture filtrates by the rate assay of Leary et al. [16].

A modification of the method of Lee et al. [17] was used to measure the specific MTT (methylthiazoletetrazolium) reducing activity of the cells, a marker of cell viability. Mycelia from 1 ml aliquots of cultures were transferred into test tubes containing 2 ml fresh media supplemented with 5 mg ml−1 MTT. The mixtures were incubated for 4 h at 37 °C, then 0.6 ml of 200 g L−1 SDS in 20 mmol L−1 HCl solution were added and the incubation continued for another 24 h. After centrifugation (10,000g, 5 min) the MTT-formazan content of the supernatant was measured spectrophotometrically at 570 nm (A570).

Cell viability was also tested by transferring samples taken at different cultivation times into fresh media and measuring gains in DCM [18].

2.3 Determination of autolytic markers

Autolysis induced by carbon starvation was characterised by decreases in DCM values and pellet diameters as well as by increases in the specific extracellular proteinase and chitinase activities [10,19]. Extracellular chitinase activities were determined in culture filtrates using CM-chitin-RBV (Loewe Biochimica GmbH, Sauerlach, Germany) as substrate according to the manufacturer's protocol. Extracellular proteinase activities were characterized by the velocity constant of the enzymic reaction (K) according to Tomarelli et al. [20]. DCM was determined as described [21].

2.4 Determination of apoptotic markers

Among the apoptotic markers, we studied nuclear morphology, externalisation of phosphatidylserine and DNA fragmentation.

To investigate changes in the morphology of nuclei, mycelia were collected, re-suspended in 70% (v/v) ethanol for brief fixation and permeabilisation, then stained with DAPI (4′,6′-diamidino-2-phenylindole) solution and observed under the epifluorescence microscope [22].

For detecting the externalisation of phosphatidylserine, mycelium was harvested on sintered glass and washed with 0.6 M MgSO4 before protoplasts were prepared according to Vágvölgyi and Ferenczy [23]. Protoplasts (4 × 106) were resuspended in isotonic solution (1 M sorbitol, 10 mM Tris–HCl, pH 7.5) and phosphatidylserine exposure on protoplast surface was determined by Annexin V assay using VybrantApoptosis Assay Kit #2 (Molecular Probes, Eugene, OR, USA) [24]. The protocol also included propidium iodide staining for filtering necrotic cells, and only Annexin V positive but propidium iodide negative protoplasts were regarded as “apoptotic”.

To identify cells undergoing DNA fragmentation, protoplasts were subjected to Terminal Deoxynucleotidyl Transferase-mediated dUTP Nick End Labelling (TUNEL) using the APO-BrdU™ TUNEL Assay Kit (Molecular Probes, Eugene, OR, USA) according to the manufacturer's instructions. Nucleus-containing protoplasts were counted using DAPI staining, and ratios of TUNEL positive and DAPI positive protoplasts were calculated [9,24].

In each phosphatidylserine exposure and TUNEL staining experiment, 5000 protoplasts were analysed using the epifluorescence microscope [24].

2.5 Statistics

Variation between experiments was estimated by standard deviations (SD) and the statistical significance of changes in physiological parameters was estimated using the Student's t-test. Only probability levels of p 5% were regarded as statistically significant. Means and SD values calculated from 3 to 6 independent experiments are presented.

2.6 Chemicals

Unless otherwise indicated, all the chemicals were purchased from Sigma–Aldrich Ltd., Budapest, Hungary.

3 Results

After glucose had been consumed in culture media, both the biomass value expressed in DCM and cell viability decreased markedly (Fig. 1). Decreases in the MTT reducing capability of the cultures (Fig. 1) were always accompanied by low growth rates of autolysing fungal tissues transferred into fresh media (data not shown). Concomitantly, significant cultivation time-dependent increases in the ratios of protoplasts with externalised phosphatidylserine (Annexin V positive protoplasts) and fragmented DNA (TUNEL positive protoplasts) were recorded (Figs. 2 and 3). Between 48 and 72 h of cultivation, the ratios of protoplasts positively stained in Annexin V and TUNEL labelings were 10–12% and 20–22%, respectively. The ratio of propidium iodide stained necrotic protoplasts never exceeded 1%, irrespective of the cultivation time tested.

Figure 1

Decrease in the DCM (□) and MTT reducing activity (Δ) after glucose (□) depletion in autolysing A. nidulans FGSC 26 cultures. Symbols represent means calculated from four independent experiments and the bars show the SD values.

Figure 2

Incubation time-dependent changes of apoptotic markers in ageing A. nidulans FGSC 26 cultures. Cells with externalised phosphatidylserine (Annexin V positive cells; black columns) and fragmented DNA (TUNEL positive cells; white columns) were counted after protoplastation of cultures. Columns represent means calculated from four parallel experiments and the bars show the SD values.

Figure 3

Aspergillus nidulans FGSC 26 protoplasts after Annexin V (Part A) and TUNEL (Parts B, C) stainings. Protoplasts were visualised by phase contrast and fluorescence (Parts A, B) or fluorescence (Part C) microscopies (Leiter et al. [4]). Bar = 5 μm (Part A) or 10 μm (Parts B, C).

As far as the cell nucleus morphology is concerned, large numbers of nuclei with elongated, stick-like morphology were observed in autolysing cultures (Fig. 4A and B). After glucose depletion, the ratio of stick-like nuclei increased steadily and reached 10–20% by 48 h of cultivation, but remained constant thereafter. The nuclei in autolysing cultures always lined up along hyphae in series, with the first ones starting well behind the hyphal tips (Fig. 4A). This type of morphology was represented in less than 1% of nuclei in exponentially growing cultures normally characterised with spherical cell nuclei evenly distributed along the hyphae (Fig. 4C). Similarly, autolytic hyphal fragments normally consisting of 1–2 cells and produced by breakage of hyphae peeling off autolysing pellets [10] typically contained nuclei with the normal spherical morphology (Fig. 4B).

Figure 4

Changes in nuclear morphology in autolytic phase Aspergillus nidulans FGSC 26 cultures. (Parts A, B) Elongated, stick-like nuclei were visualised with DAPI staining in hyphae taken from 72 h cultures. (Parts C, D) DAPI-stained normal morphology nuclei in exponential growth phase (20 h) hypha and in an autolytic phase (72 h) hyphal fragment, respectively. Bar = 10 μm (Part A), 5 μm (Part B) or 3 μm (Parts C, D).

Cultivation time-dependent increases in the ratio of Annexin V positive protoplasts (Fig. 5A) was also recorded in submerged cultures of the loss-of-function fluG and the ΔbrlA strains [7]. Similar results were found with the TUNEL test: At 72 h, the ratios of the apoptotic protoplasts in the FGSC 26, fluG1 and ΔbrlA strains were 22 ± 5%, 19 ± 6% and 25 ± 6%, respectively, while at 16 h they were 1 ± 0.5%, 0.5 ± 0.3% and 0.8 ± 0.3%, respectively. In accordance with these data, the decreases in cell viability were also very similar for the three strains (Fig. 5B). The ratio of nuclei with elongated, stick-like morphology increased in the mutants similarly to the kinetics described for the control strain with a maximum value of 10–20% recorded starting from 48 h of cultivation.

Figure 5

Progressing cell death and decreasing cell vitality in carbon-depleted cultures of the FGSC 744 loss-of-function fluG and the FGSC 1079 ΔbrlA mutant strains. (Part A) Cultivation time-dependent increases in the ratios of Annexin V positive protoplasts prepared from A. nidulans FGSC 26 (control; black columns), loss-of-function fluG mutant (white columns) and ΔbrlA (grey columns) cultures. Columns represent means calculated from four parallel experiments and the bars show the SD values. (Part B) Decreases in the MTT reducing activity of the A. nidulans FGSC 26 (control; □), loss-of-function fluG mutant (□) and ΔbrlA (Δ) cultures after glucose depletion. Symbols represent means calculated from three independent experiments. The SD values were less than 20%. No significant differences were calculated among the three strains using the Student's t-test.

Sphingosine derivatives induce apoptosis in A. nidulans [15]. Addition of either 0.05 g L−1d, l-dihydrosphingosine or 0.03 g L−1 phytosphingosine to A. nidulans FGSC 26 cultures at 30 h cultivation increased the number of apoptotic cells more than twofold. The ratios of Annexin V positive protoplasts at 48 h of cultivation were: FGSC 26 control, 11 ± 3%; d, l-dihydrosphingosine treated cultures, 25 ± 4%; phytosphingosine treated cultures, 28 ± 5%; all calculated from four independent experiments. Interestingly, sphingosine treatments did not accelerate either the autolytic loss of biomass or the disintegration of pellets, and did not result in any elevated release of extracellular chitinase or proteinase activities (Fig. 6A–C).

Figure 6

Decrease in the DCM (A) as well as induction of extracellular chitinase (B) and proteinase (C) activities in control, untreated Aspergillus nidulans FGSC 26 cultures (□) and in the presence of 0.05 g L−1d, l-dihydrosphingosine (□) or 0.03 g L−1 phytosphingosine (Δ). Sphingosine derivatives were added to the cultures at 30 h of cultivation. Symbols represent means calculated from four independent experiments and the bars show the SD values. No significant differences were calculated among the three types of cultures using the Student's t-test.

4 Discussion

As described previously by Emri et al. [10], three growth phases were distinguished in A. nidulans cultures after glucose depletion: stationary (24–34 h), early autolytic (34–100 h) and late autolytic (after 100 h) phases. This paper focuses mainly on the physiological changes taking place in the early autolytic phase of growth up to 72 h of cultivation.

Similarly to A. fumigatus [9], stationary and early autolytic phase A. nidulans cells underwent apoptosis-like cell death (Figs. 2 and 3) with decreasing biomass and cell viability (Fig. 1). Carbon starvation represents a frequently occurring threat for fungi in their natural habitats [11] and, therefore, programmed cell death processes are likely to be triggered under these harsh environmental conditions. Apoptosis triggered by environmental changes will allow fungal cell populations to survive by selection of mutants that are better-adapted to the changing environment [11]. Moreover, cell autolysates are likely to provide surviving cells with a minimum level of nutrient supply [5,10,12,21].

Nuclei in starving hyphae show distorted morphology (Fig. 4), which is likely to be connected to the progressive breakage of DNA, as indicated by the similar ratios (about 20%) of TUNEL positive cells and cells with stick-like nuclei. It is noteworthy that hyphal fragments, which retain viability under carbon starvation in submerged cultures [10,21,25], typically included nuclei with normal spherical morphology (Fig. 4). It is reasonable to hypothesize that the adaptive re-growth selection mechanism described for the yeast Saccharomyces cerevisiae [11] may also take place in the hyphal fragments of autolysing filamentous fungi. Unfortunately, the appearance of uni- or bicellular hyphal fragments and their re-growth are relatively rare events in autolysing A. nidulans cultures [10]. By contrast, a high β-lactam producer strain of Penicillium chrysogenum was shown to fragment mainly into round-ended two-celled fragments, many of which were outgrowing at any incubation time tested (“cryptic growth”), which clearly demonstrated the adaptation of surviving fragments to carbon starvation [21].

When growing on solid surfaces, a fine-tuning between growth, autolysis and conidiation is crucially important to support the development of conidiophores [4], and the participation of FluG–BrlA signalling in the initialisation of both conidiation and autolysis has been demonstrated [4,7]. The gene fluG encodes a cytoplasmically localised protein with homology to the prokaryotic glutamine synthetase [6,26,27]. Its hypothesised, small diffusible product is thought to activate a cascade of Flb proteins leading to the blockage of vegetative growth and the induction of conidiogenesis [4]. The transcriptional factor encoded by brlA is also activated via Flb proteins and is the primary activator of conidiation-specific genes [4,27]. As shown by Emri et al. [7], both the loss-of-function fluG and the ΔbrlA strains possessed a non-autolysing phenotype and, as a consequence, the DCM decreased only very slowly, and no hyphal fragmentation and disintegration of pellets were observed in submerged cultures of these strains. Moreover, the loss-of-function fluG and ΔbrlA cultures did not produce significant extracellular chitinases activities either [7]. Importantly, the kinetics of apoptotic cell death was not affected by these mutations (Fig. 5) indicating that cell death and autolysis, although occurring concomitantly in aging cultures, are regulated independently.

Sphingosine derivatives efficiently induce apoptosis in germinating spores and young hyphae of A. nidulans [15]. Although aged cultures are generally less sensitive to environmental stress than exponentially growing cells [28,29], the number of apoptotic cells was doubled by the addition of sphingosine derivatives to stationary phase A. nidulans cultures. As shown in Fig. 6, the elevated level of apoptotic cells did not influence the appearance of autolysis markers including DCW declination and hydrolase production. These findings further strengthen the view that apoptosis and autolysis are regulated independently and the development of autolytic markers is not a self-evident consequence of cell death in carbon-depleted cultures of filamentous fungi.

In summary, the development of genetically engineered morphological mutants and the addition of chemicals selectively accelerating or hindering cell death and autolysis may provide us with a tool to control fungal cell morphology and vitality in stationary and autolytic phase submerged cultures, which are frequently used in the bioprocess industry. To reach this goal, a deeper understanding of the molecular background of fungal cell death is crucially important.


The Hungarian Ministry of Education awarded a Széchenyi Scholarship for Professors to I.P.; and T.E. was a grantee of the Bolyai János Scholarship. This project was supported financially by the Office for Higher Education Programmes (Grant Reference No. 0092/2001) and by the OTKA (Grant Reference Nos. T034315, T037473 and D034568).


  1. [1].
  2. [2].
  3. [3].
  4. [4].
  5. [5].
  6. [6].
  7. [7].
  8. [8].
  9. [9].
  10. [10].
  11. [11].
  12. [12].
  13. [13].
  14. [14].
  15. [15].
  16. [16].
  17. [17].
  18. [18].
  19. [19].
  20. [20].
  21. [21].
  22. [22].
  23. [23].
  24. [24].
  25. [25].
  26. [26].
  27. [27].
  28. [28].
  29. [29].
View Abstract